repo_id
stringclasses 55
values | file_path
stringlengths 42
186
| content
stringlengths 1
333k
| __index_level_0__
int64 0
0
|
|---|---|---|---|
mavonic_private_repos/transformers/src/transformers/kernels
|
mavonic_private_repos/transformers/src/transformers/kernels/mra/torch_extension.cpp
|
#include <torch/extension.h>
#include <ATen/ATen.h>
#include "cuda_launch.h"
#include <vector>
std::vector<at::Tensor> index_max(
at::Tensor index_vals,
at::Tensor indices,
int A_num_block,
int B_num_block
) {
return index_max_kernel(
index_vals,
indices,
A_num_block,
B_num_block
);
}
at::Tensor mm_to_sparse(
at::Tensor dense_A,
at::Tensor dense_B,
at::Tensor indices
) {
return mm_to_sparse_kernel(
dense_A,
dense_B,
indices
);
}
at::Tensor sparse_dense_mm(
at::Tensor sparse_A,
at::Tensor indices,
at::Tensor dense_B,
int A_num_block
) {
return sparse_dense_mm_kernel(
sparse_A,
indices,
dense_B,
A_num_block
);
}
at::Tensor reduce_sum(
at::Tensor sparse_A,
at::Tensor indices,
int A_num_block,
int B_num_block
) {
return reduce_sum_kernel(
sparse_A,
indices,
A_num_block,
B_num_block
);
}
at::Tensor scatter(
at::Tensor dense_A,
at::Tensor indices,
int B_num_block
) {
return scatter_kernel(
dense_A,
indices,
B_num_block
);
}
PYBIND11_MODULE(TORCH_EXTENSION_NAME, m) {
m.def("index_max", &index_max, "index_max (CUDA)");
m.def("mm_to_sparse", &mm_to_sparse, "mm_to_sparse (CUDA)");
m.def("sparse_dense_mm", &sparse_dense_mm, "sparse_dense_mm (CUDA)");
m.def("reduce_sum", &reduce_sum, "reduce_sum (CUDA)");
m.def("scatter", &scatter, "scatter (CUDA)");
}
| 0
|
mavonic_private_repos/transformers/src/transformers/kernels
|
mavonic_private_repos/transformers/src/transformers/kernels/mra/cuda_kernel.cu
|
#include "cuda_kernel.h"
//////////////////////////////////////////////////////////////////////////////////////////////////
//////////////////////////////////////////////////////////////////////////////////////////////////
__global__ void index_max_cuda_kernel(
float *index_vals, // [batch_size, 32, num_block]
int *indices, // [batch_size, num_block]
float *max_vals, // [batch_size, A_num_block * 32]
float *max_vals_scatter, // [batch_size, 32, num_block]
long batch_size,
long A_num_block,
long B_num_block,
long num_block
) {
long batch_idx = blockIdx.x;
long thread_idx = threadIdx.x;
long num_thread = blockDim.x;
extern __shared__ float buffer[];
int *max_buffer = (int*)buffer;
for (int i = 0; i < A_num_block * 32; i = i + num_thread) {
int idx = i + thread_idx;
if (idx < A_num_block * 32) {
max_buffer[idx] = -1e8;
}
}
__syncthreads();
int *indices_pt = &indices[batch_idx * num_block];
float *index_vals_pt = &index_vals[batch_idx * num_block * 32];
for (int idx_start = 0; idx_start < 32 * num_block; idx_start = idx_start + num_thread) {
int idx = idx_start + thread_idx;
int A_block_idx = indices_pt[idx % num_block] / B_num_block;
atomicMax(&max_buffer[A_block_idx * 32 + idx / num_block], (int)(index_vals_pt[idx] * 1000));
}
__syncthreads();
float *max_vals_pt = &max_vals[batch_idx * A_num_block * 32];
for (int i = 0; i < A_num_block * 32; i = i + num_thread) {
int idx = i + thread_idx;
if (idx < A_num_block * 32) {
max_vals_pt[idx] = (float)max_buffer[idx] / 1000.;
}
}
float *max_vals_scatter_pt = &max_vals_scatter[batch_idx * num_block * 32];
for (int idx_start = 0; idx_start < 32 * num_block; idx_start = idx_start + num_thread) {
int idx = idx_start + thread_idx;
int A_block_idx = indices_pt[idx % num_block] / B_num_block;
max_vals_scatter_pt[idx] = (float)max_buffer[A_block_idx * 32 + idx / num_block] / 1000.;
}
}
__global__ void mm_to_sparse_cuda_kernel(
float *dense_A, // [batch_size, A_num_block, dim, 32]
float *dense_B, // [batch_size, B_num_block, dim, 32]
int *indices, // [batch_size, num_block]
float *sparse_C, // [batch_size, num_block, 32, 32]
long batch_size,
long A_num_block,
long B_num_block,
long dim,
long num_block
) {
long batch_idx = blockIdx.y;
long block_idx = blockIdx.x * blockDim.y + threadIdx.y;
long thread_idx = threadIdx.x;
__shared__ float buffer[4096];
float *A_buffer = &buffer[threadIdx.y * 1024]; // [2, 8, 32]
float *B_buffer = &buffer[threadIdx.y * 1024 + 512]; // [2, 8, 32]
long batch_idx__block_idx = batch_idx * num_block + block_idx;
long AB_block_idx = indices[batch_idx__block_idx];
float *dense_A_pt = &dense_A[(batch_idx * A_num_block + AB_block_idx / B_num_block) * dim * 32];
float *dense_B_pt = &dense_B[(batch_idx * B_num_block + AB_block_idx % B_num_block) * dim * 32];
int reg_1_idx = thread_idx / 8; // [0000000011111111222222223333333344444444555555556666666677777777]
int reg_2_idx = thread_idx % 8; // [0123456701234567012345670123456701234567012345670123456701234567]
float reg_1[8];
float reg_2[8];
float reg_array[16] = {0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0};
#pragma unroll
for (int i = 0; i < 4; i++) {
A_buffer[i * 64 + thread_idx] = dense_A_pt[i * 64 + thread_idx];
B_buffer[i * 64 + thread_idx] = dense_B_pt[i * 64 + thread_idx];
}
__syncthreads();
#pragma unroll
for (int i = 0; i < 4; i++) {
reg_1[i] = A_buffer[reg_1_idx * 4 + i];
reg_2[i] = B_buffer[reg_2_idx * 4 + i];
}
for (int dim_stride = 1; dim_stride < (dim / 8); dim_stride++) {
#pragma unroll
for (int i = 0; i < 4; i++) {
A_buffer[(dim_stride % 2) * 256 + i * 64 + thread_idx] = dense_A_pt[dim_stride * 256 + i * 64 + thread_idx];
B_buffer[(dim_stride % 2) * 256 + i * 64 + thread_idx] = dense_B_pt[dim_stride * 256 + i * 64 + thread_idx];
}
#pragma unroll
for (int mini_dim_idx = 1; mini_dim_idx < 8; mini_dim_idx++) {
#pragma unroll
for (int i = 0; i < 4; i++) {
reg_1[(mini_dim_idx % 2) * 4 + i] = A_buffer[((dim_stride - 1) % 2) * 256 + mini_dim_idx * 32 + reg_1_idx * 4 + i];
reg_2[(mini_dim_idx % 2) * 4 + i] = B_buffer[((dim_stride - 1) % 2) * 256 + mini_dim_idx * 32 + reg_2_idx * 4 + i];
}
#pragma unroll
for (int i = 0; i < 4; i++) {
#pragma unroll
for (int j = 0; j < 4; j++) {
reg_array[i * 4 + j] += reg_1[((mini_dim_idx - 1) % 2) * 4 + i] * reg_2[((mini_dim_idx - 1) % 2) * 4 + j];
}
}
}
__syncthreads();
#pragma unroll
for (int i = 0; i < 4; i++) {
reg_1[i] = A_buffer[(dim_stride % 2) * 256 + reg_1_idx * 4 + i];
reg_2[i] = B_buffer[(dim_stride % 2) * 256 + reg_2_idx * 4 + i];
}
#pragma unroll
for (int i = 0; i < 4; i++) {
#pragma unroll
for (int j = 0; j < 4; j++) {
reg_array[i * 4 + j] += reg_1[4 + i] * reg_2[4 + j];
}
}
}
#pragma unroll
for (int mini_dim_idx = 1; mini_dim_idx < 8; mini_dim_idx++) {
#pragma unroll
for (int i = 0; i < 4; i++) {
reg_1[(mini_dim_idx % 2) * 4 + i] = A_buffer[256 + mini_dim_idx * 32 + reg_1_idx * 4 + i];
reg_2[(mini_dim_idx % 2) * 4 + i] = B_buffer[256 + mini_dim_idx * 32 + reg_2_idx * 4 + i];
}
#pragma unroll
for (int i = 0; i < 4; i++) {
#pragma unroll
for (int j = 0; j < 4; j++) {
reg_array[i * 4 + j] += reg_1[((mini_dim_idx - 1) % 2) * 4 + i] * reg_2[((mini_dim_idx - 1) % 2) * 4 + j];
}
}
}
#pragma unroll
for (int i = 0; i < 4; i++) {
#pragma unroll
for (int j = 0; j < 4; j++) {
reg_array[i * 4 + j] += reg_1[4 + i] * reg_2[4 + j];
}
}
__syncthreads();
float *C_buffer = &buffer[threadIdx.y * 1024]; // [32, 32]
#pragma unroll
for (int i = 0; i < 4; i++) {
#pragma unroll
for (int j = 0; j < 4; j++) {
C_buffer[(reg_2_idx * 4 + j) * 32 + reg_1_idx * 4 + i] = reg_array[i * 4 + j];
}
}
__syncthreads();
float *sparse_C_pt = &sparse_C[batch_idx__block_idx * 1024];
#pragma unroll
for (int i = 0; i < 16; i++) {
sparse_C_pt[i * 64 + thread_idx] = C_buffer[i * 64 + thread_idx];
}
}
__global__ void sparse_dense_mm_cuda_kernel(
float *sparse_A, // [batch_size, num_block, 32, 32]
int *indices, // [batch_size, num_block]
float *dense_B, // [batch_size, B_num_block, dim, 32]
float *dense_C, // [batch_size, A_num_block, dim, 32]
long batch_size,
long A_num_block,
long B_num_block,
long dim,
long num_block
) {
long batch_idx = blockIdx.y;
long block_idx = blockIdx.x * blockDim.y + threadIdx.y;
long thread_idx = threadIdx.x;
__shared__ float buffer[6144];
float *A_buffer = &buffer[threadIdx.y * 3072]; // [32, 32]
float *B_buffer = &buffer[threadIdx.y * 3072 + 1024]; // [32, 64]
long batch_idx__block_idx = batch_idx * num_block + block_idx;
float *sparse_A_pt = &sparse_A[batch_idx__block_idx * 1024];
#pragma unroll
for (int i = 0; i < 8; i++) {
A_buffer[i * 128 + thread_idx] = sparse_A_pt[i * 128 + thread_idx];
}
long AB_block_idx = indices[batch_idx__block_idx];
float *dense_B_pt = &dense_B[(batch_idx * B_num_block + AB_block_idx % B_num_block) * 32 * dim];
float *dense_C_pt = &dense_C[(batch_idx * A_num_block + AB_block_idx / B_num_block) * 32 * dim];
// [0000000011111111222222223333333344444444555555556666666677777777]
// [0123456701234567012345670123456701234567012345670123456701234567]
int reg_1_idx = thread_idx / 8;
int reg_2_idx = thread_idx % 8;
float reg_1[8];
float reg_2[8];
float reg_array[16];
for (int dim_stride = 0; dim_stride < dim; dim_stride = dim_stride + 64) {
#pragma unroll
for (int i = 0; i < 16; i++) {
B_buffer[i * 128 + thread_idx] = dense_B_pt[dim_stride * 32 + i * 128 + thread_idx];
}
#pragma unroll
for (int i = 0; i < 16; i++) {
reg_array[i] = 0;
}
__syncthreads();
#pragma unroll
for (int i = 0; i < 4; i++) {
reg_1[i] = B_buffer[(reg_1_idx * 4 + i) * 32];
reg_2[i] = A_buffer[reg_2_idx * 4 + i];
}
#pragma unroll
for (int mini_dim_idx = 1; mini_dim_idx < 32; mini_dim_idx++) {
#pragma unroll
for (int i = 0; i < 4; i++) {
reg_1[(mini_dim_idx % 2) * 4 + i] = B_buffer[(reg_1_idx * 4 + i) * 32 + mini_dim_idx];
reg_2[(mini_dim_idx % 2) * 4 + i] = A_buffer[mini_dim_idx * 32 + reg_2_idx * 4 + i];
}
#pragma unroll
for (int i = 0; i < 4; i++) {
#pragma unroll
for (int j = 0; j < 4; j++) {
reg_array[i * 4 + j] += reg_1[((mini_dim_idx - 1) % 2) * 4 + i] * reg_2[((mini_dim_idx - 1) % 2) * 4 + j];
}
}
}
#pragma unroll
for (int i = 0; i < 4; i++) {
#pragma unroll
for (int j = 0; j < 4; j++) {
reg_array[i * 4 + j] += reg_1[4 + i] * reg_2[4 + j];
}
}
__syncthreads();
float *C_buffer = &buffer[threadIdx.y * 3072 + 1024]; // [64, 32]
#pragma unroll
for (int i = 0; i < 4; i++) {
#pragma unroll
for (int j = 0; j < 4; j++) {
C_buffer[(reg_1_idx * 4 + i) * 32 + reg_2_idx * 4 + j] = reg_array[i * 4 + j];
}
}
__syncthreads();
#pragma unroll
for (int i = 0; i < 16; i++) {
atomicAdd(&dense_C_pt[dim_stride * 32 + i * 128 + thread_idx], C_buffer[i * 128 + thread_idx]);
}
__syncthreads();
}
}
__global__ void reduce_sum_cuda_kernel(
float *sparse_A, // [batch_size, num_block, 32, 32]
int *indices, // [batch_size, num_block]
float *dense_C, // [batch_size, A_num_block, 32]
long batch_size,
long A_num_block,
long B_num_block,
long num_block
) {
long batch_idx = blockIdx.y;
long block_idx = blockIdx.x * blockDim.y + threadIdx.y;
long thread_idx = threadIdx.x;
long batch_idx__block_idx = batch_idx * num_block + block_idx;
long AB_block_idx = indices[batch_idx__block_idx];
float *sparse_A_pt = &sparse_A[batch_idx__block_idx * 1024];
float reg_array[16];
float value = 0;
#pragma unroll
for (int i = 0; i < 8; i++) {
reg_array[i] = sparse_A_pt[i * 32 + thread_idx];
}
#pragma unroll
for (int stride = 8; stride < 32; stride = stride + 8) {
#pragma unroll
for (int i = 0; i < 8; i++) {
reg_array[(stride + i) % 16] = sparse_A_pt[(stride + i) * 32 + thread_idx];
}
#pragma unroll
for (int i = 0; i < 8; i++) {
value = value + reg_array[(stride - 8 + i) % 16];
}
}
#pragma unroll
for (int i = 0; i < 8; i++) {
value = value + reg_array[8 + i];
}
float *dense_C_pt = &dense_C[(batch_idx * A_num_block + AB_block_idx / B_num_block) * 32];
atomicAdd(&dense_C_pt[thread_idx], value);
}
__global__ void scatter_cuda_kernel(
float *dense_A, // [batch_size, A_num_block, 32]
int *indices, // [batch_size, num_block]
float *sparse_C, // [batch_size, num_block, 32, 32]
long batch_size,
long A_num_block,
long B_num_block,
long num_block
) {
long batch_idx = blockIdx.y;
long block_idx = blockIdx.x * blockDim.y + threadIdx.y;
long thread_idx = threadIdx.x;
long batch_idx__block_idx = batch_idx * num_block + block_idx;
long AB_block_idx = indices[batch_idx__block_idx];
float *dense_A_pt = &dense_A[(batch_idx * A_num_block + AB_block_idx / B_num_block) * 32];
float *sparse_C_pt = &sparse_C[(batch_idx * num_block + block_idx) * 1024];
float value = dense_A_pt[thread_idx];
#pragma unroll
for (int i = 0; i < 32; i++) {
sparse_C_pt[i * 32 + thread_idx] = value;
}
}
| 0
|
mavonic_private_repos/transformers/src/transformers/kernels
|
mavonic_private_repos/transformers/src/transformers/kernels/yoso/fast_lsh_cumulation_cuda.h
|
__global__ void fast_hash_ver1_cuda_kernel(
int *mask, // [batch_size, num_vector]
float *vector, // [batch_size, num_vector, vector_dim]
int *Dmat, // [3, num_part, vector_dim]
int *hash_code, // [batch_size, num_vector, num_hash_f]
int batch_size,
int num_vector,
int vector_dim,
int num_part,
int num_hash_f,
int hash_code_len
);
__global__ void lsh_cumulation_ver1_step1_cuda_kernel(
int *key_mask, // [batch_size, num_key]
int *key_hash_code, // [batch_size, num_key, num_hash_f]
float *value, // [batch_size, num_key, value_dim]
float *hashtable_value, // [batch_size, num_hash_f, hashtable_capacity, value_dim]
int batch_size,
int num_hash_f,
int hashtable_capacity,
int num_key,
int value_dim,
int offset_warp
);
__global__ void lsh_cumulation_ver1_step2_cuda_kernel(
int *query_mask, // [batch_size, num_query]
int *query_hash_code, // [batch_size, num_query, num_hash_f]
float *hashtable_value, // [batch_size, num_hash_f, hashtable_capacity, value_dim]
float *cumulation_value, // [batch_size, num_query, value_dim]
int batch_size,
int num_hash_f,
int hashtable_capacity,
int num_query,
int value_dim,
int offset_warp
);
__global__ void lsh_weighted_cumulation_ver1_step1_cuda_kernel(
int *key_mask, // [batch_size, num_key]
int *key_hash_code, // [batch_size, num_key, num_hash_f]
float *key_weight, // [batch_size, num_key, weight_dim]
float *value, // [batch_size, num_key, value_dim]
float *hashtable_value, // [batch_size, num_hash_f, hashtable_capacity, WARP_SIZE]
int batch_size,
int num_hash_f,
int hashtable_capacity,
int num_key,
int value_dim,
int weight_dim,
int offset_warp,
int weight_idx
);
__global__ void lsh_weighted_cumulation_ver1_step2_cuda_kernel(
int *query_mask, // [batch_size, num_query]
int *query_hash_code, // [batch_size, num_query, num_hash_f]
float *query_weight, // [batch_size, num_query, weight_dim]
float *hashtable_value, // [batch_size, num_hash_f, hashtable_capacity, WARP_SIZE]
float *cumulation_value, // [batch_size, num_query, value_dim]
int batch_size,
int num_hash_f,
int hashtable_capacity,
int num_query,
int value_dim,
int weight_dim,
int offset_warp,
int weight_idx
);
__global__ void count_sort_step1_cuda_kernel(
int *key_mask, // [batch_size, num_key]
int *key_hash_code, // [batch_size, num_key, num_hash_f]
int *count_sort_table, // [batch_size, num_hash_f, hashtable_capacity]
int batch_size,
int num_hash_f,
int hashtable_capacity,
int num_key
);
__global__ void count_sort_step2_cuda_kernel(
int *count_sort_table, // [batch_size, num_hash_f, hashtable_capacity]
int batch_size,
int num_hash_f,
int hashtable_capacity
);
__global__ void count_sort_step3_cuda_kernel(
int *key_mask, // [batch_size, num_key]
int *key_hash_code, // [batch_size, num_key, num_hash_f]
int *count_sort_table, // [batch_size, num_hash_f, hashtable_capacity]
int *key_sorted_idxes, // [batch_size, num_hash_f, num_key]
int batch_size,
int num_hash_f,
int hashtable_capacity,
int num_key
);
__global__ void extract_query_info_cuda_kernel(
int *query_mask, // [batch_size, num_query]
int *query_hash_code, // [batch_size, num_query, num_hash_f]
int *count_sort_table, // [batch_size, num_hash_f, hashtable_capacity]
int *query_info, // [batch_size, num_query, 2, num_hash_f]
int batch_size,
int num_hash_f,
int hashtable_capacity,
int num_query
);
__global__ void lsh_weighted_cumulation_ver2_step2_cuda_kernel(
int *query_mask, // [batch_size, num_query]
int *query_info, // [batch_size, num_query, 2, num_hash_f]
int *key_sorted_idxes, // [batch_size, num_hash_f, num_key]
float *query_weight, // [batch_size, num_query, weight_dim]
float *key_weight, // [batch_size, num_key, weight_dim]
float *value, // [batch_size, num_key, value_dim]
float *cumulation_value, // [batch_size, num_query, value_dim]
int batch_size,
int num_hash_f,
int num_query,
int num_key,
int value_dim,
int weight_dim
);
__global__ void lsh_weighted_cumulation_ver3_step2_cuda_kernel(
int *query_sorted_idxes, // [batch_size, num_hash_f, num_query]
int *key_mask, // [batch_size, num_key]
int *key_info, // [batch_size, num_key, 2, num_hash_f]
float *query_weight, // [batch_size, num_query, weight_dim]
float *key_weight, // [batch_size, num_key, weight_dim]
float *value, // [batch_size, num_key, value_dim]
float *cumulation_value, // [batch_size, num_query, value_dim]
int batch_size,
int num_hash_f,
int num_query,
int num_key,
int value_dim,
int weight_dim
);
__global__ void lsh_weighted_cumulation_ver4_step2_cuda_kernel(
int *query_sorted_idxes, // [batch_size, num_hash_f, num_query]
int *key_mask, // [batch_size, num_key]
int *key_info, // [batch_size, num_key, 2, num_hash_f]
float *query_weight, // [batch_size, num_query, weight_dim]
float *key_weight, // [batch_size, num_key, weight_dim]
float *value, // [batch_size, num_key, value_dim]
float *cumulation_value, // [batch_size, num_query, value_dim]
int batch_size,
int num_hash_f,
int num_query,
int num_key,
int value_dim,
int weight_dim
);
| 0
|
mavonic_private_repos/transformers/src/transformers/kernels
|
mavonic_private_repos/transformers/src/transformers/kernels/yoso/fast_lsh_cumulation_cuda.cu
|
// File from https://github.com/mlpen/YOSO/blob/main/encoders/backbones/efficient_attentions/yoso/yoso_v1/cuda/fast_lsh_cumulation_cuda.cu
#include "fast_lsh_cumulation_cuda.h"
#include "common_cuda_device.h"
#include "common_cuda.h"
#include "common.h"
#include <stdio.h>
//////////////////////////////////////////////////////////////////////////////////////////////////
//////////////////////////////////////////////////////////////////////////////////////////////////
inline __device__ void fast_hadamard_transform(float *vector_buffer, int vector_dim, int dim_idx) {
int stride = vector_dim / 2;
while (stride > (WARP_SIZE / 2)) {
__syncthreads();
int sign = 1 - ((dim_idx / stride) % 2) * 2;
float val1 = vector_buffer[dim_idx];
float val2 = vector_buffer[dim_idx + sign * stride];
__syncthreads();
vector_buffer[dim_idx] = float(sign) * val1 + val2;
stride = stride / 2;
}
float val = vector_buffer[dim_idx];
#pragma unroll
for (stride = (WARP_SIZE / 2); stride > 0; stride = stride / 2) {
int sign = 1 - ((dim_idx / stride) % 2) * 2;
val = float(sign) * val + __shfl_xor_sync(FULL_MASK, val, stride);
}
vector_buffer[dim_idx] = val;
}
__global__ void fast_hash_ver1_cuda_kernel(
int *mask, // [batch_size, num_vector]
float *vector, // [batch_size, num_vector, vector_dim]
int *Dmat, // [batch_size, 3, num_part, vector_dim]
int *hash_code, // [batch_size, num_vector, num_hash_f]
int batch_size,
int num_vector,
int vector_dim,
int num_part,
int num_hash_f,
int hash_code_len
) {
int batch_idx = blockIdx.z;
int vector_idx = blockIdx.y;
int part_idx = blockIdx.x;
int dim_idx = threadIdx.x;
int batch_idx__vector_idx = batch_idx * num_vector + vector_idx;
if (mask[batch_idx__vector_idx] == 0) {
return;
}
extern __shared__ float buffer[];
float *vector_buffer = buffer;
vector_buffer[dim_idx] = vector[batch_idx__vector_idx * vector_dim + dim_idx];
vector_buffer[dim_idx] = vector_buffer[dim_idx] * (float)Dmat[((batch_idx * 3 + 0) * num_part + part_idx) * vector_dim + dim_idx];
fast_hadamard_transform(vector_buffer, vector_dim, dim_idx);
vector_buffer[dim_idx] = vector_buffer[dim_idx] * (float)Dmat[((batch_idx * 3 + 1) * num_part + part_idx) * vector_dim + dim_idx];
fast_hadamard_transform(vector_buffer, vector_dim, dim_idx);
vector_buffer[dim_idx] = vector_buffer[dim_idx] * (float)Dmat[((batch_idx * 3 + 2) * num_part + part_idx) * vector_dim + dim_idx];
fast_hadamard_transform(vector_buffer, vector_dim, dim_idx);
int num_hash_per_part = vector_dim / hash_code_len;
if (hash_code_len == 8 || hash_code_len == 16) {
int code = select(vector_buffer[dim_idx] > 0, 1 << (dim_idx % hash_code_len), 0);
for (int offset = 1; offset < hash_code_len; offset = offset * 2) {
code += __shfl_xor_sync(FULL_MASK, code, offset);
}
if (dim_idx % hash_code_len == 0) {
int hash_f_idx = part_idx * num_hash_per_part + dim_idx / hash_code_len;
if (hash_f_idx < num_hash_f) {
hash_code[batch_idx__vector_idx * num_hash_f + hash_f_idx] = code;
}
}
} else {
vector_buffer[dim_idx] = select(vector_buffer[dim_idx] > 0, 1 << (dim_idx % hash_code_len), 0);
__syncthreads();
if (dim_idx < num_hash_per_part) {
int code = 0;
for (int i = 0; i < hash_code_len; i++) {
code += vector_buffer[dim_idx * hash_code_len + i];
}
int hash_f_idx = part_idx * num_hash_per_part + dim_idx;
if (hash_f_idx < num_hash_f) {
hash_code[batch_idx__vector_idx * num_hash_f + hash_f_idx] = code;
}
}
}
}
__global__ void lsh_cumulation_ver1_step1_cuda_kernel(
int *key_mask, // [batch_size, num_key]
int *key_hash_code, // [batch_size, num_key, num_hash_f]
float *value, // [batch_size, num_key, value_dim]
float *hashtable_value, // [batch_size, num_hash_f, hashtable_capacity, WARP_SIZE]
int batch_size,
int num_hash_f,
int hashtable_capacity,
int num_key,
int value_dim,
int offset_warp
) {
int warp_thread_idx = threadIdx.x;
int batch_idx = blockIdx.y;
int key_idx = blockIdx.x * blockDim.y + threadIdx.y;
int batch_idx__key_idx = batch_idx * num_key + key_idx;
if (key_mask[batch_idx__key_idx] == 0) {
return;
}
if (num_hash_f > WARP_SIZE) {
float warp_value = value[batch_idx__key_idx * value_dim + offset_warp + warp_thread_idx];
for (int hash_f_start = 0; hash_f_start < num_hash_f; hash_f_start = hash_f_start + WARP_SIZE) {
int warp_hashcode = key_hash_code[batch_idx__key_idx * num_hash_f + hash_f_start + warp_thread_idx];
#pragma unroll
for (int hash_f_offset = 0; hash_f_offset < WARP_SIZE; hash_f_offset++) {
int current_hashcode = warp_hashcode;
current_hashcode = __shfl_sync(FULL_MASK, current_hashcode, hash_f_offset);
int hashtable_idx = (batch_idx * num_hash_f + (hash_f_start + hash_f_offset)) * hashtable_capacity + current_hashcode;
atomicAdd(&hashtable_value[hashtable_idx * WARP_SIZE + warp_thread_idx], warp_value);
}
}
} else {
float warp_value = value[batch_idx__key_idx * value_dim + offset_warp + warp_thread_idx];
int warp_hashcode = 0;
if (warp_thread_idx < num_hash_f) {
warp_hashcode = key_hash_code[batch_idx__key_idx * num_hash_f + warp_thread_idx];
}
for (int hash_f_idx = 0; hash_f_idx < num_hash_f; hash_f_idx++) {
int current_hashcode = warp_hashcode;
current_hashcode = __shfl_sync(FULL_MASK, current_hashcode, hash_f_idx);
int hashtable_idx = (batch_idx * num_hash_f + hash_f_idx) * hashtable_capacity + current_hashcode;
atomicAdd(&hashtable_value[hashtable_idx * WARP_SIZE + warp_thread_idx], warp_value);
}
}
}
__global__ void lsh_cumulation_ver1_step2_cuda_kernel(
int *query_mask, // [batch_size, num_query]
int *query_hash_code, // [batch_size, num_query, num_hash_f]
float *hashtable_value, // [batch_size, num_hash_f, hashtable_capacity, WARP_SIZE]
float *cumulation_value, // [batch_size, num_query, value_dim]
int batch_size,
int num_hash_f,
int hashtable_capacity,
int num_query,
int value_dim,
int offset_warp
) {
int warp_thread_idx = threadIdx.x;
int batch_idx = blockIdx.y;
int query_idx = blockIdx.x * blockDim.y + threadIdx.y;
int batch_idx__query_idx = batch_idx * num_query + query_idx;
if (query_mask[batch_idx__query_idx] == 0) {
return;
}
if (num_hash_f > WARP_SIZE) {
float warp_value = 0;
for (int hash_f_start = 0; hash_f_start < num_hash_f; hash_f_start = hash_f_start + WARP_SIZE) {
int warp_hashcode = query_hash_code[batch_idx__query_idx * num_hash_f + hash_f_start + warp_thread_idx];
#pragma unroll
for (int hash_f_offset = 0; hash_f_offset < WARP_SIZE; hash_f_offset++) {
int current_hashcode = warp_hashcode;
current_hashcode = __shfl_sync(FULL_MASK, current_hashcode, hash_f_offset);
int hashtable_idx = (batch_idx * num_hash_f + (hash_f_start + hash_f_offset)) * hashtable_capacity + current_hashcode;
warp_value = warp_value + hashtable_value[hashtable_idx * WARP_SIZE + warp_thread_idx];
}
}
cumulation_value[batch_idx__query_idx * value_dim + offset_warp + warp_thread_idx] = warp_value / float(num_hash_f);
} else {
float warp_value = 0;
int warp_hashcode = 0;
if (warp_thread_idx < num_hash_f) {
warp_hashcode = query_hash_code[batch_idx__query_idx * num_hash_f + warp_thread_idx];
}
for (int hash_f_idx = 0; hash_f_idx < num_hash_f; hash_f_idx++) {
int current_hashcode = warp_hashcode;
current_hashcode = __shfl_sync(FULL_MASK, current_hashcode, hash_f_idx);
int hashtable_idx = (batch_idx * num_hash_f + hash_f_idx) * hashtable_capacity + current_hashcode;
warp_value = warp_value + hashtable_value[hashtable_idx * WARP_SIZE + warp_thread_idx];
}
cumulation_value[batch_idx__query_idx * value_dim + offset_warp + warp_thread_idx] = warp_value / float(num_hash_f);
}
}
__global__ void lsh_weighted_cumulation_ver1_step1_cuda_kernel(
int *key_mask, // [batch_size, num_key]
int *key_hash_code, // [batch_size, num_key, num_hash_f]
float *key_weight, // [batch_size, num_key, weight_dim]
float *value, // [batch_size, num_key, value_dim]
float *hashtable_value, // [batch_size, num_hash_f, hashtable_capacity, WARP_SIZE]
int batch_size,
int num_hash_f,
int hashtable_capacity,
int num_key,
int value_dim,
int weight_dim,
int offset_warp,
int weight_idx
) {
int warp_thread_idx = threadIdx.x;
int batch_idx = blockIdx.y;
int key_idx = blockIdx.x * blockDim.y + threadIdx.y;
int batch_idx__key_idx = batch_idx * num_key + key_idx;
if (key_mask[batch_idx__key_idx] == 0) {
return;
}
if (num_hash_f > WARP_SIZE) {
float warp_value = key_weight[batch_idx__key_idx * weight_dim + weight_idx] * value[batch_idx__key_idx * value_dim + offset_warp + warp_thread_idx];
for (int hash_f_start = 0; hash_f_start < num_hash_f; hash_f_start = hash_f_start + WARP_SIZE) {
int warp_hashcode = key_hash_code[batch_idx__key_idx * num_hash_f + hash_f_start + warp_thread_idx];
#pragma unroll
for (int hash_f_offset = 0; hash_f_offset < WARP_SIZE; hash_f_offset++) {
int current_hashcode = warp_hashcode;
current_hashcode = __shfl_sync(FULL_MASK, current_hashcode, hash_f_offset);
int hashtable_idx = (batch_idx * num_hash_f + (hash_f_start + hash_f_offset)) * hashtable_capacity + current_hashcode;
atomicAdd(&hashtable_value[hashtable_idx * WARP_SIZE + warp_thread_idx], warp_value);
}
}
} else {
float warp_value = key_weight[batch_idx__key_idx * weight_dim + weight_idx] * value[batch_idx__key_idx * value_dim + offset_warp + warp_thread_idx];
int warp_hashcode = 0;
if (warp_thread_idx < num_hash_f) {
warp_hashcode = key_hash_code[batch_idx__key_idx * num_hash_f + warp_thread_idx];
}
for (int hash_f_idx = 0; hash_f_idx < num_hash_f; hash_f_idx++) {
int current_hashcode = warp_hashcode;
current_hashcode = __shfl_sync(FULL_MASK, current_hashcode, hash_f_idx);
int hashtable_idx = (batch_idx * num_hash_f + hash_f_idx) * hashtable_capacity + current_hashcode;
atomicAdd(&hashtable_value[hashtable_idx * WARP_SIZE + warp_thread_idx], warp_value);
}
}
}
__global__ void lsh_weighted_cumulation_ver1_step2_cuda_kernel(
int *query_mask, // [batch_size, num_query]
int *query_hash_code, // [batch_size, num_query, num_hash_f]
float *query_weight, // [batch_size, num_query, weight_dim]
float *hashtable_value, // [batch_size, num_hash_f, hashtable_capacity, WARP_SIZE]
float *cumulation_value, // [batch_size, num_query, value_dim]
int batch_size,
int num_hash_f,
int hashtable_capacity,
int num_query,
int value_dim,
int weight_dim,
int offset_warp,
int weight_idx
) {
int warp_thread_idx = threadIdx.x;
int batch_idx = blockIdx.y;
int query_idx = blockIdx.x * blockDim.y + threadIdx.y;
int batch_idx__query_idx = batch_idx * num_query + query_idx;
if (query_mask[batch_idx__query_idx] == 0) {
return;
}
if (num_hash_f > WARP_SIZE) {
float warp_value = 0;
for (int hash_f_start = 0; hash_f_start < num_hash_f; hash_f_start = hash_f_start + WARP_SIZE) {
int warp_hashcode = query_hash_code[batch_idx__query_idx * num_hash_f + hash_f_start + warp_thread_idx];
#pragma unroll
for (int hash_f_offset = 0; hash_f_offset < WARP_SIZE; hash_f_offset++) {
int current_hashcode = warp_hashcode;
current_hashcode = __shfl_sync(FULL_MASK, current_hashcode, hash_f_offset);
int hashtable_idx = (batch_idx * num_hash_f + (hash_f_start + hash_f_offset)) * hashtable_capacity + current_hashcode;
warp_value = warp_value + hashtable_value[hashtable_idx * WARP_SIZE + warp_thread_idx];
}
}
float warp_weight = query_weight[batch_idx__query_idx * weight_dim + weight_idx];
cumulation_value[batch_idx__query_idx * value_dim + offset_warp + warp_thread_idx] += warp_weight * warp_value / float(num_hash_f);
} else {
float warp_value = 0;
int warp_hashcode = 0;
if (warp_thread_idx < num_hash_f) {
warp_hashcode = query_hash_code[batch_idx__query_idx * num_hash_f + warp_thread_idx];
}
for (int hash_f_idx = 0; hash_f_idx < num_hash_f; hash_f_idx++) {
int current_hashcode = warp_hashcode;
current_hashcode = __shfl_sync(FULL_MASK, current_hashcode, hash_f_idx);
int hashtable_idx = (batch_idx * num_hash_f + hash_f_idx) * hashtable_capacity + current_hashcode;
warp_value = warp_value + hashtable_value[hashtable_idx * WARP_SIZE + warp_thread_idx];
}
float warp_weight = query_weight[batch_idx__query_idx * weight_dim + weight_idx];
cumulation_value[batch_idx__query_idx * value_dim + offset_warp + warp_thread_idx] += warp_weight * warp_value / float(num_hash_f);
}
}
__global__ void count_sort_step1_cuda_kernel(
int *key_mask, // [batch_size, num_key]
int *key_hash_code, // [batch_size, num_key, num_hash_f]
int *count_sort_table, // [batch_size, num_hash_f, hashtable_capacity]
int batch_size,
int num_hash_f,
int hashtable_capacity,
int num_key
) {
int batch_idx = blockIdx.y;
int key_idx = blockIdx.x * blockDim.y + threadIdx.y;
int hash_f_idx = threadIdx.x;
int batch_idx__key_idx = batch_idx * num_key + key_idx;
if (key_mask[batch_idx__key_idx] == 0) {
return;
}
int hash_code = key_hash_code[batch_idx__key_idx * num_hash_f + hash_f_idx];
atomicAdd(&count_sort_table[(batch_idx * num_hash_f + hash_f_idx) * hashtable_capacity + hash_code], 1);
}
__global__ void count_sort_step2_cuda_kernel(
int *count_sort_table, // [batch_size, num_hash_f, hashtable_capacity]
int batch_size,
int num_hash_f,
int hashtable_capacity
) {
int batch_idx = blockIdx.y;
int hash_f_idx = blockIdx.x;
int num_threads = blockDim.x;
int thread_id = threadIdx.x;
int batch_idx__hash_f_idx = batch_idx * num_hash_f + hash_f_idx;
extern __shared__ float buffer[];
int *table_buffer = (int*)buffer;
if (thread_id == 0) {
table_buffer[0] = 0;
}
copy_data<int>(&count_sort_table[batch_idx__hash_f_idx * hashtable_capacity], &table_buffer[1], hashtable_capacity - 1, num_threads, thread_id);
for (int table_idx_start = 0; table_idx_start < hashtable_capacity; table_idx_start = table_idx_start + num_threads) {
int thread_value = table_buffer[table_idx_start + thread_id];
int next_thread_value = 0;
for (int offset = 1; offset < WARP_SIZE; offset = offset << 1) {
next_thread_value = __shfl_up_sync(FULL_MASK, thread_value, offset);
if (thread_id % WARP_SIZE >= offset) {
thread_value = thread_value + next_thread_value;
}
}
table_buffer[table_idx_start + thread_id] = thread_value;
}
__syncthreads();
if (hashtable_capacity > WARP_SIZE) {
if (thread_id < WARP_SIZE) {
for (int table_idx_start = WARP_SIZE; table_idx_start < hashtable_capacity; table_idx_start = table_idx_start + WARP_SIZE) {
table_buffer[table_idx_start + thread_id] += table_buffer[table_idx_start - 1];
}
}
}
copy_data<int>(table_buffer, &count_sort_table[batch_idx__hash_f_idx * hashtable_capacity], hashtable_capacity, num_threads, thread_id);
}
__global__ void count_sort_step3_cuda_kernel(
int *key_mask, // [batch_size, num_key]
int *key_hash_code, // [batch_size, num_key, num_hash_f]
int *count_sort_table, // [batch_size, num_hash_f, hashtable_capacity]
int *key_sorted_idxes, // [batch_size, num_hash_f, num_key]
int batch_size,
int num_hash_f,
int hashtable_capacity,
int num_key
) {
int batch_idx = blockIdx.y;
int key_idx = blockIdx.x * blockDim.y + threadIdx.y;
int hash_f_idx = threadIdx.x;
int batch_idx__key_idx = batch_idx * num_key + key_idx;
if (key_mask[batch_idx__key_idx] == 0) {
return;
}
int batch_idx__hash_f_idx = batch_idx * num_hash_f + hash_f_idx;
int hash_code = key_hash_code[batch_idx__key_idx * num_hash_f + hash_f_idx];
int sort_idx = atomicAdd(&count_sort_table[batch_idx__hash_f_idx * hashtable_capacity + hash_code], 1);
key_sorted_idxes[batch_idx__hash_f_idx * num_key + sort_idx] = key_idx;
}
__global__ void extract_query_info_cuda_kernel(
int *query_mask, // [batch_size, num_query]
int *query_hash_code, // [batch_size, num_query, num_hash_f]
int *count_sort_table, // [batch_size, num_hash_f, hashtable_capacity]
int *query_info, // [batch_size, num_query, 2, num_hash_f]
int batch_size,
int num_hash_f,
int hashtable_capacity,
int num_query
) {
int batch_idx = blockIdx.y;
int query_idx = blockIdx.x * blockDim.y + threadIdx.y;
int hash_f_idx = threadIdx.x;
int batch_idx__query_idx = batch_idx * num_query + query_idx;
if (query_mask[batch_idx__query_idx] == 0) {
return;
}
int hash_code = query_hash_code[batch_idx__query_idx * num_hash_f + hash_f_idx];
int batch_idx__hash_f_idx__hash_code = (batch_idx * num_hash_f + hash_f_idx) * hashtable_capacity + hash_code;
int key_offset = select(hash_code == 0, 0, count_sort_table[batch_idx__hash_f_idx__hash_code - 1]);
int key_count = count_sort_table[batch_idx__hash_f_idx__hash_code] - key_offset;
query_info[batch_idx__query_idx * 2 * num_hash_f + hash_f_idx] = key_offset;
query_info[(batch_idx__query_idx * 2 + 1) * num_hash_f + hash_f_idx] = key_count;
}
__global__ void lsh_weighted_cumulation_ver2_step2_cuda_kernel(
int *query_mask, // [batch_size, num_query]
int *query_info, // [batch_size, num_query, 2, num_hash_f]
int *key_sorted_idxes, // [batch_size, num_hash_f, num_key]
float *query_weight, // [batch_size, num_query, weight_dim]
float *key_weight, // [batch_size, num_key, weight_dim]
float *value, // [batch_size, num_key, value_dim]
float *cumulation_value, // [batch_size, num_query, value_dim]
int batch_size,
int num_hash_f,
int num_query,
int num_key,
int value_dim,
int weight_dim
) {
int batch_idx = blockIdx.z;
int hash_f_idx = blockIdx.y;
int query_idx = blockIdx.x;
int num_threads = blockDim.y * blockDim.x;
int thread_id = threadIdx.y * blockDim.x + threadIdx.x;
int num_warps = blockDim.y;
int warp_idx = threadIdx.y;
int warp_thread_idx = threadIdx.x;
int batch_idx__query_idx = batch_idx * num_query + query_idx;
if (query_mask[batch_idx__query_idx] == 0) {
return;
}
int key_offset = query_info[batch_idx__query_idx * 2 * num_hash_f + hash_f_idx];
int key_count = query_info[(batch_idx__query_idx * 2 + 1) * num_hash_f + hash_f_idx];
if (key_count == 0) {
return;
}
extern __shared__ float buffer[];
if (key_count == 1) {
if (warp_idx == 0) {
int key_idx = key_sorted_idxes[(batch_idx * num_hash_f + hash_f_idx) * num_key + key_offset];
int batch_idx__key_idx = batch_idx * num_key + key_idx;
float weight = 0;
for (int weight_offset = 0; weight_offset < weight_dim; weight_offset = weight_offset + WARP_SIZE) {
int weight_dim_idx = weight_offset + warp_thread_idx;
float val = query_weight[batch_idx__query_idx * weight_dim + weight_dim_idx] * key_weight[batch_idx__key_idx * weight_dim + weight_dim_idx];
#pragma unroll
for (int offset = 1; offset < WARP_SIZE; offset = offset << 1) {
val += __shfl_xor_sync(FULL_MASK, val, offset);
}
weight = weight + val;
}
weight = weight / float(num_hash_f);
for (int value_offset = 0; value_offset < value_dim; value_offset = value_offset + WARP_SIZE) {
int value_dim_idx = value_offset + warp_thread_idx;
float val = value[batch_idx__key_idx * value_dim + value_dim_idx];
atomicAdd(&cumulation_value[batch_idx__query_idx * value_dim + value_dim_idx], weight * val);
}
}
} else {
float *weight_buffer = buffer;
int *key_idxes_buffer = (int*)&buffer[weight_dim];
copy_data_nonblocking<float>(&query_weight[batch_idx__query_idx * weight_dim], weight_buffer, weight_dim, num_threads, thread_id);
while (key_count > 0) {
int work_size = min(WARP_SIZE, key_count);
copy_data_nonblocking<int>(&key_sorted_idxes[(batch_idx * num_hash_f + hash_f_idx) * num_key + key_offset], key_idxes_buffer, work_size, num_threads, thread_id);
__syncthreads();
for (int work_offset = 0; work_offset < WARP_SIZE; work_offset = work_offset + num_warps) {
int work_idx = work_offset + warp_idx;
if (work_idx < key_count) {
int key_idx = key_idxes_buffer[work_idx];
int batch_idx__key_idx = batch_idx * num_key + key_idx;
float weight = 0;
for (int weight_offset = 0; weight_offset < weight_dim; weight_offset = weight_offset + WARP_SIZE) {
int weight_dim_idx = weight_offset + warp_thread_idx;
float val = weight_buffer[weight_dim_idx] * key_weight[batch_idx__key_idx * weight_dim + weight_dim_idx];
#pragma unroll
for (int offset = 1; offset < WARP_SIZE; offset = offset << 1) {
val += __shfl_xor_sync(FULL_MASK, val, offset);
}
weight = weight + val;
}
weight = weight / float(num_hash_f);
for (int value_offset = 0; value_offset < value_dim; value_offset = value_offset + WARP_SIZE) {
int value_dim_idx = value_offset + warp_thread_idx;
float val = value[batch_idx__key_idx * value_dim + value_dim_idx];
atomicAdd(&cumulation_value[batch_idx__query_idx * value_dim + value_dim_idx], weight * val);
}
}
}
key_count = key_count - work_size;
key_offset = key_offset + work_size;
}
}
}
__global__ void lsh_weighted_cumulation_ver3_step2_cuda_kernel(
int *query_sorted_idxes, // [batch_size, num_hash_f, num_query]
int *key_mask, // [batch_size, num_key]
int *key_info, // [batch_size, num_key, 2, num_hash_f]
float *query_weight, // [batch_size, num_query, weight_dim]
float *key_weight, // [batch_size, num_key, weight_dim]
float *value, // [batch_size, num_key, value_dim]
float *cumulation_value, // [batch_size, num_query, value_dim]
int batch_size,
int num_hash_f,
int num_query,
int num_key,
int value_dim,
int weight_dim
) {
int batch_idx = blockIdx.z;
int hash_f_idx = blockIdx.y;
int key_idx = blockIdx.x;
int num_threads = blockDim.y * blockDim.x;
int thread_id = threadIdx.y * blockDim.x + threadIdx.x;
int num_warps = blockDim.y;
int warp_idx = threadIdx.y;
int warp_thread_idx = threadIdx.x;
int batch_idx__key_idx = batch_idx * num_key + key_idx;
if (key_mask[batch_idx__key_idx] == 0) {
return;
}
int query_offset = key_info[batch_idx__key_idx * 2 * num_hash_f + hash_f_idx];
int query_count = key_info[(batch_idx__key_idx * 2 + 1) * num_hash_f + hash_f_idx];
if (query_count == 0) {
return;
}
extern __shared__ float buffer[];
if (query_count == 1) {
if (warp_idx == 0) {
int query_idx = query_sorted_idxes[(batch_idx * num_hash_f + hash_f_idx) * num_query + query_offset];
int batch_idx__query_idx = batch_idx * num_query + query_idx;
float weight = 0;
for (int weight_offset = 0; weight_offset < weight_dim; weight_offset = weight_offset + WARP_SIZE) {
int weight_dim_idx = weight_offset + warp_thread_idx;
float val = key_weight[batch_idx__key_idx * weight_dim + weight_dim_idx] * query_weight[batch_idx__query_idx * weight_dim + weight_dim_idx];
#pragma unroll
for (int offset = 1; offset < WARP_SIZE; offset = offset << 1) {
val += __shfl_xor_sync(FULL_MASK, val, offset);
}
weight = weight + val;
}
weight = weight / float(num_hash_f);
for (int value_offset = 0; value_offset < value_dim; value_offset = value_offset + WARP_SIZE) {
int value_dim_idx = value_offset + warp_thread_idx;
float val = value[batch_idx__key_idx * value_dim + value_dim_idx];
atomicAdd(&cumulation_value[batch_idx__query_idx * value_dim + value_dim_idx], weight * val);
}
}
} else {
float *weight_buffer = buffer;
float *value_buffer = &buffer[weight_dim];
int *query_idxes_buffer = (int*)&buffer[weight_dim + value_dim];
copy_data_nonblocking<float>(&key_weight[batch_idx__key_idx * weight_dim], weight_buffer, weight_dim, num_threads, thread_id);
copy_data_nonblocking<float>(&value[batch_idx__key_idx * value_dim], value_buffer, value_dim, num_threads, thread_id);
while (query_count > 0) {
int work_size = min(WARP_SIZE, query_count);
copy_data_nonblocking<int>(&query_sorted_idxes[(batch_idx * num_hash_f + hash_f_idx) * num_query + query_offset], query_idxes_buffer, work_size, num_threads, thread_id);
__syncthreads();
for (int work_offset = 0; work_offset < WARP_SIZE; work_offset = work_offset + num_warps) {
int work_idx = work_offset + warp_idx;
if (work_idx < query_count) {
int query_idx = query_idxes_buffer[work_idx];
int batch_idx__query_idx = batch_idx * num_query + query_idx;
float weight = 0;
for (int weight_offset = 0; weight_offset < weight_dim; weight_offset = weight_offset + WARP_SIZE) {
int weight_dim_idx = weight_offset + warp_thread_idx;
float val = weight_buffer[weight_dim_idx] * query_weight[batch_idx__query_idx * weight_dim + weight_dim_idx];
#pragma unroll
for (int offset = 1; offset < WARP_SIZE; offset = offset << 1) {
val += __shfl_xor_sync(FULL_MASK, val, offset);
}
weight = weight + val;
}
weight = weight / float(num_hash_f);
for (int value_offset = 0; value_offset < value_dim; value_offset = value_offset + WARP_SIZE) {
int value_dim_idx = value_offset + warp_thread_idx;
float val = value_buffer[value_dim_idx];
atomicAdd(&cumulation_value[batch_idx__query_idx * value_dim + value_dim_idx], weight * val);
}
}
}
query_count = query_count - work_size;
query_offset = query_offset + work_size;
}
}
}
__global__ void lsh_weighted_cumulation_ver4_step2_cuda_kernel(
int *query_sorted_idxes, // [batch_size, num_hash_f, num_query]
int *key_mask, // [batch_size, num_key]
int *key_info, // [batch_size, num_key, 2, num_hash_f]
float *query_weight, // [batch_size, num_query, weight_dim]
float *key_weight, // [batch_size, num_key, weight_dim]
float *value, // [batch_size, num_key, value_dim]
float *cumulation_value, // [batch_size, num_query, value_dim]
int batch_size,
int num_hash_f,
int num_query,
int num_key,
int value_dim,
int weight_dim
) {
int batch_idx = blockIdx.y;
int key_idx = blockIdx.x;
int num_threads = blockDim.y * blockDim.x;
int thread_id = threadIdx.y * blockDim.x + threadIdx.x;
int num_warps = blockDim.y;
int warp_idx = threadIdx.y;
int warp_thread_idx = threadIdx.x;
int batch_idx__key_idx = batch_idx * num_key + key_idx;
if (key_mask[batch_idx__key_idx] == 0) {
return;
}
extern __shared__ float buffer[];
float *weight_buffer = buffer;
float *value_buffer = &buffer[weight_dim];
int *key_info_buffer = (int*)&buffer[weight_dim + value_dim];
copy_data_nonblocking<float>(&key_weight[batch_idx__key_idx * weight_dim], weight_buffer, weight_dim, num_threads, thread_id);
copy_data_nonblocking<float>(&value[batch_idx__key_idx * value_dim], value_buffer, value_dim, num_threads, thread_id);
copy_data_nonblocking<int>(&key_info[batch_idx__key_idx * 2 * num_hash_f], key_info_buffer, 2 * num_hash_f, num_threads, thread_id);
int *query_offset_buffer = key_info_buffer;
int *query_count_buffer = &key_info_buffer[num_hash_f];
const int hashtable_size = 1024 + OPTIMAL_THREADS_PER_BLOCK;
__shared__ int hashtable_query[hashtable_size];
__shared__ int hashtable_count[hashtable_size];
__shared__ int inserted_query[hashtable_size];
__shared__ int query_counter[1];
int hash_f_idx_base = 0;
while (true) {
init_buffer_nonblocking<int>(EMPTY_VALUE, hashtable_query, hashtable_size, num_threads, thread_id);
init_buffer_nonblocking<int>(0, hashtable_count, hashtable_size, num_threads, thread_id);
init_buffer_nonblocking<int>(EMPTY_VALUE, inserted_query, hashtable_size, num_threads, thread_id);
init_buffer_nonblocking<int>(0, query_counter, 1, num_threads, thread_id);
__syncthreads();
while (hash_f_idx_base < num_hash_f) {
int hash_f_idx = hash_f_idx_base + warp_idx;
int batch_idx__hash_f_idx = batch_idx * num_hash_f + hash_f_idx;
int stop_flag = 0;
int query_offset = query_offset_buffer[hash_f_idx];
int query_count = query_count_buffer[hash_f_idx];
while (query_count > 0) {
int work_size = min(query_count, WARP_SIZE);
// try inserting query to set and check whether the query is new
int found_new_query = 0;
int query_idx = -1;
if (warp_thread_idx < work_size) {
query_idx = query_sorted_idxes[batch_idx__hash_f_idx * num_query + query_offset + warp_thread_idx];
int slot = set_insert<int>(hashtable_query, hashtable_size, query_idx);
if (slot >= 0) {
found_new_query = atomicAdd(&hashtable_count[slot], 1) == 0;
}
}
// compute cumulative offset
int position_offset = found_new_query;
int next_position_offset = 0;
#pragma unroll
for (int offset = 1; offset < WARP_SIZE; offset = offset << 1) {
next_position_offset = __shfl_up_sync(FULL_MASK, position_offset, offset);
if (thread_id % WARP_SIZE >= offset) {
position_offset = position_offset + next_position_offset;
}
}
// get the inserted query list end index
int inserted_query_base = 0;
if (thread_id % WARP_SIZE == WARP_SIZE - 1) {
inserted_query_base = atomicAdd(query_counter, position_offset);
}
inserted_query_base = __shfl_sync(FULL_MASK, inserted_query_base, WARP_SIZE - 1);
// insert new queries to list
int insert_idx = inserted_query_base + position_offset - 1;
if (found_new_query) {
inserted_query[insert_idx] = query_idx;
}
// remove inserted queries from list
query_offset_buffer[hash_f_idx] += work_size;
query_count_buffer[hash_f_idx] -= work_size;
query_offset += work_size;
query_count -= work_size;
// if list is almost full, stop inserting
if (inserted_query_base + OPTIMAL_THREADS_PER_BLOCK > hashtable_size) {
stop_flag = 1;
break;
}
}
if (stop_flag) {
break;
}
hash_f_idx_base = hash_f_idx_base + num_warps;
}
__syncthreads();
int num_distint_query = query_counter[0];
if (num_distint_query > 0) {
for (int idx_base = 0; idx_base < num_distint_query; idx_base = idx_base + num_warps) {
int idx = idx_base + warp_idx;
if (idx < num_distint_query) {
int query_idx = inserted_query[idx];
int batch_idx__query_idx = batch_idx * num_query + query_idx;
int slot = set_lookup<int>(hashtable_query, hashtable_size, query_idx);
int duplicate_count = hashtable_count[slot];
float weight = 0;
for (int weight_idx_base = 0; weight_idx_base < weight_dim; weight_idx_base = weight_idx_base + WARP_SIZE) {
int weight_dim_idx = weight_idx_base + warp_thread_idx;
float val = weight_buffer[weight_dim_idx] * query_weight[batch_idx__query_idx * weight_dim + weight_dim_idx];
#pragma unroll
for (int offset = 1; offset < WARP_SIZE; offset = offset << 1) {
val += __shfl_xor_sync(FULL_MASK, val, offset);
}
weight = weight + val;
}
weight = (float)duplicate_count * weight / float(num_hash_f);
for (int value_idx_base = 0; value_idx_base < value_dim; value_idx_base = value_idx_base + WARP_SIZE) {
int value_dim_idx = value_idx_base + warp_thread_idx;
float val = value_buffer[value_dim_idx];
atomicAdd(&cumulation_value[batch_idx__query_idx * value_dim + value_dim_idx], weight * val);
}
}
}
} else {
// all computation is completed if num_distint_query == 0
break;
}
__syncthreads();
}
}
| 0
|
mavonic_private_repos/transformers/src/transformers/kernels
|
mavonic_private_repos/transformers/src/transformers/kernels/yoso/fast_lsh_cumulation_torch.cpp
|
#include <torch/extension.h>
#include <ATen/ATen.h>
#include "fast_lsh_cumulation.h"
#include "common_cuda.h"
#include <vector>
std::vector<at::Tensor> fast_hash(
at::Tensor query_mask,
at::Tensor query_vector,
at::Tensor key_mask,
at::Tensor key_vector,
int num_hash_f,
int hash_code_len,
bool use_cuda,
int version
) {
return fast_hash_ver1_kernel(
query_mask,
query_vector,
key_mask,
key_vector,
num_hash_f,
hash_code_len,
use_cuda
);
}
at::Tensor lsh_cumulation(
at::Tensor query_mask, // [batch_size, num_query]
at::Tensor query_hash_code, // [batch_size, num_query, num_hash_f]
at::Tensor key_mask, // [batch_size, num_key]
at::Tensor key_hash_code, // [batch_size, num_key, num_hash_f]
at::Tensor value, // [batch_size, num_key, value_dim]
int hashtable_capacity,
bool use_cuda,
int version
) {
return lsh_cumulation_ver1_kernel(
query_mask,
query_hash_code,
key_mask,
key_hash_code,
value,
hashtable_capacity,
use_cuda
);
}
at::Tensor lsh_weighted_cumulation(
at::Tensor query_mask, // [batch_size, num_query]
at::Tensor query_hash_code, // [batch_size, num_query, num_hash_f]
at::Tensor query_weight, // [batch_size, num_query, weight_dim]
at::Tensor key_mask, // [batch_size, num_key]
at::Tensor key_hash_code, // [batch_size, num_key, num_hash_f]
at::Tensor key_weight, // [batch_size, num_key, weight_dim]
at::Tensor value, // [batch_size, num_key, value_dim]
int hashtable_capacity,
bool use_cuda,
int version
) {
if (version == 1) {
return lsh_weighted_cumulation_ver1_kernel(
query_mask,
query_hash_code,
query_weight,
key_mask,
key_hash_code,
key_weight,
value,
hashtable_capacity,
use_cuda
);
} else if (version == 2) {
return lsh_weighted_cumulation_ver2_kernel(
query_mask,
query_hash_code,
query_weight,
key_mask,
key_hash_code,
key_weight,
value,
hashtable_capacity,
use_cuda
);
} else if (version == 3) {
return lsh_weighted_cumulation_ver3_kernel(
query_mask,
query_hash_code,
query_weight,
key_mask,
key_hash_code,
key_weight,
value,
hashtable_capacity,
use_cuda
);
} else if (version == 4) {
return lsh_weighted_cumulation_ver4_kernel(
query_mask,
query_hash_code,
query_weight,
key_mask,
key_hash_code,
key_weight,
value,
hashtable_capacity,
use_cuda
);
} else {
return lsh_weighted_cumulation_ver3_kernel(
query_mask,
query_hash_code,
query_weight,
key_mask,
key_hash_code,
key_weight,
value,
hashtable_capacity,
use_cuda
);
}
}
PYBIND11_MODULE(TORCH_EXTENSION_NAME, m) {
m.def("fast_hash", &fast_hash, "Fast Hash (CUDA)");
m.def("lsh_cumulation", &lsh_cumulation, "LSH Cumulation (CUDA)");
m.def("lsh_weighted_cumulation", &lsh_weighted_cumulation, "LSH Weighted Cumulation (CUDA)");
}
| 0
|
mavonic_private_repos/transformers/src/transformers/kernels
|
mavonic_private_repos/transformers/src/transformers/kernels/yoso/common_cuda_device.h
|
#include "common.h"
template<typename T>
__device__ int set_insert(T *set, int set_size, T value) {
int slot = value % set_size;
int start_slot = slot;
while (true) {
T prev = atomicCAS(&set[slot], EMPTY_VALUE, value);
if (prev == EMPTY_VALUE || prev == value) {
return slot;
}
slot = (slot + 1) % set_size;
if (slot == start_slot) {
return -1;
}
}
return -1;
}
template<typename T>
__device__ int set_lookup(T *set, int set_size, T value) {
int slot = value % set_size;
int start_slot = slot;
while (true) {
if (set[slot] == value) {
return slot;
}
slot = (slot + 1) % set_size;
if (slot == start_slot) {
return -1;
}
}
return -1;
}
template<typename T>
__device__ void init_buffer(T init_value, T *buffer, int buffer_size, int num_threads, int thread_id) {
__syncthreads();
for (int i = 0; i < buffer_size; i = i + num_threads) {
int offset_idx = i + thread_id;
if (offset_idx < buffer_size) {
buffer[offset_idx] = init_value;
}
}
__syncthreads();
}
template<typename T>
__device__ void copy_data(T *src_pt, T *dist_pt, int data_length, int num_threads, int thread_id) {
__syncthreads();
for (int i = 0; i < data_length; i = i + num_threads) {
int offset_idx = i + thread_id;
if (offset_idx < data_length) {
dist_pt[offset_idx] = src_pt[offset_idx];
}
}
__syncthreads();
}
template<typename T>
__device__ void init_buffer_nonblocking(T init_value, T *buffer, int buffer_size, int num_threads, int thread_id) {
for (int i = 0; i < buffer_size; i = i + num_threads) {
int offset_idx = i + thread_id;
if (offset_idx < buffer_size) {
buffer[offset_idx] = init_value;
}
}
}
template<typename T>
__device__ void copy_data_nonblocking(T *src_pt, T *dist_pt, int data_length, int num_threads, int thread_id) {
for (int i = 0; i < data_length; i = i + num_threads) {
int offset_idx = i + thread_id;
if (offset_idx < data_length) {
dist_pt[offset_idx] = src_pt[offset_idx];
}
}
}
| 0
|
mavonic_private_repos/transformers/src/transformers/kernels
|
mavonic_private_repos/transformers/src/transformers/kernels/yoso/common_cuda.h
|
#define MAX_THREADS_PER_BLOCK 1024
#define OPTIMAL_THREADS_PER_BLOCK 256
#define WARP_SIZE 32
#define MAX_NUM_BLOCK_X 2147483647
#define MAX_NUM_BLOCK_Y 65535
#define MAX_NUM_BLOCK_Z 65535
#define MAX_SHARED_MEM_PER_BLOCK 48000
#define FULL_MASK 0xffffffff
| 0
|
mavonic_private_repos/transformers/src/transformers/kernels
|
mavonic_private_repos/transformers/src/transformers/kernels/yoso/fast_lsh_cumulation.cu
|
// File from https://github.com/mlpen/YOSO/blob/main/encoders/backbones/efficient_attentions/yoso/yoso_v1/cuda/fast_lsh_cumulation.cu
#include <torch/extension.h>
#include <ATen/ATen.h>
#include "fast_lsh_cumulation.h"
#include "fast_lsh_cumulation_cuda.h"
#include "common_cuda.h"
#include "common.h"
#include <vector>
//////////////////////////////////////////////////////////////////////////////////////////////////
//////////////////////////////////////////////////////////////////////////////////////////////////
std::vector<at::Tensor> fast_hash_ver1_kernel(
at::Tensor query_mask,
at::Tensor query_vector,
at::Tensor key_mask,
at::Tensor key_vector,
int num_hash_f,
int hash_code_len,
bool use_cuda
) {
int batch_size = query_vector.size(0);
int num_query = query_vector.size(1);
int num_key = key_vector.size(1);
int vector_dim = query_vector.size(2);
int num_hash_per_part = vector_dim / hash_code_len;
int num_part = max(1, ceil_divide(num_hash_f, num_hash_per_part));
at::Tensor Dmat = 2 * at::randint(0, 2, {batch_size, 3, num_part, vector_dim}, query_mask.options()) - 1;
at::Tensor query_hash_code = at::zeros({batch_size, num_query, num_hash_f}, query_mask.options());
at::Tensor key_hash_code = at::zeros({batch_size, num_key, num_hash_f}, key_mask.options());
int *query_mask_ptr = query_mask.data_ptr<int>();
float *query_vector_ptr = query_vector.data_ptr<float>();
int *key_mask_ptr = key_mask.data_ptr<int>();
float *key_vector_ptr = key_vector.data_ptr<float>();
int *Dmat_ptr = Dmat.data_ptr<int>();
int *query_hash_code_ptr = query_hash_code.data_ptr<int>();
int *key_hash_code_ptr = key_hash_code.data_ptr<int>();
if (use_cuda) {
{
dim3 threads(vector_dim);
dim3 blocks(num_part, num_query, batch_size);
int shared_mem = vector_dim * sizeof(float);
fast_hash_ver1_cuda_kernel<<<blocks, threads, shared_mem>>>(
query_mask_ptr,
query_vector_ptr,
Dmat_ptr,
query_hash_code_ptr,
batch_size,
num_query,
vector_dim,
num_part,
num_hash_f,
hash_code_len
);
}
{
dim3 threads(vector_dim);
dim3 blocks(num_part, num_key, batch_size);
int shared_mem = vector_dim * sizeof(float);
fast_hash_ver1_cuda_kernel<<<blocks, threads, shared_mem>>>(
key_mask_ptr,
key_vector_ptr,
Dmat_ptr,
key_hash_code_ptr,
batch_size,
num_key,
vector_dim,
num_part,
num_hash_f,
hash_code_len
);
}
}
return {query_hash_code, key_hash_code};
}
at::Tensor lsh_cumulation_ver1_kernel(
at::Tensor query_mask,
at::Tensor query_hash_code,
at::Tensor key_mask,
at::Tensor key_hash_code,
at::Tensor value,
int hashtable_capacity,
bool use_cuda
) {
int batch_size = query_hash_code.size(0);
int num_hash_f = query_hash_code.size(2);
int num_query = query_hash_code.size(1);
int num_key = key_hash_code.size(1);
int value_dim = value.size(2);
at::Tensor hashtable_value = at::empty({batch_size, num_hash_f, hashtable_capacity, WARP_SIZE}, value.options());
at::Tensor cumulation_value = at::zeros({batch_size, num_query, value_dim}, value.options());
if (use_cuda) {
int threads_x = WARP_SIZE;
int threads_y = OPTIMAL_THREADS_PER_BLOCK / WARP_SIZE;
int block_x_step1 = num_key / threads_y;
int block_x_step2 = num_query / threads_y;
int block_y = batch_size;
dim3 threads(threads_x, threads_y);
dim3 blocks_step1(block_x_step1, block_y);
dim3 blocks_step2(block_x_step2, block_y);
int *query_mask_ptr = query_mask.data_ptr<int>();
int *query_hash_code_ptr = query_hash_code.data_ptr<int>();
int *key_mask_ptr = key_mask.data_ptr<int>();
int *key_hash_code_ptr = key_hash_code.data_ptr<int>();
float *value_ptr = value.data_ptr<float>();
float *hashtable_value_ptr = hashtable_value.data_ptr<float>();
float *cumulation_value_ptr = cumulation_value.data_ptr<float>();
for (int value_offset = 0; value_offset < value_dim; value_offset = value_offset + WARP_SIZE) {
cudaMemset(hashtable_value_ptr, 0, (batch_size * num_hash_f * hashtable_capacity * WARP_SIZE) * sizeof(float));
lsh_cumulation_ver1_step1_cuda_kernel<<<blocks_step1, threads>>>(
key_mask_ptr,
key_hash_code_ptr,
value_ptr,
hashtable_value_ptr,
batch_size,
num_hash_f,
hashtable_capacity,
num_key,
value_dim,
value_offset
);
lsh_cumulation_ver1_step2_cuda_kernel<<<blocks_step2, threads>>>(
query_mask_ptr,
query_hash_code_ptr,
hashtable_value_ptr,
cumulation_value_ptr,
batch_size,
num_hash_f,
hashtable_capacity,
num_query,
value_dim,
value_offset
);
}
}
return cumulation_value;
}
at::Tensor lsh_weighted_cumulation_ver1_kernel(
at::Tensor query_mask,
at::Tensor query_hash_code,
at::Tensor query_weight,
at::Tensor key_mask,
at::Tensor key_hash_code,
at::Tensor key_weight,
at::Tensor value,
int hashtable_capacity,
bool use_cuda
) {
int batch_size = query_hash_code.size(0);
int num_hash_f = query_hash_code.size(2);
int num_query = query_hash_code.size(1);
int num_key = key_hash_code.size(1);
int value_dim = value.size(2);
int weight_dim = query_weight.size(2);
at::Tensor hashtable_value = at::zeros({batch_size, num_hash_f, hashtable_capacity, WARP_SIZE}, value.options());
at::Tensor cumulation_value = at::zeros({batch_size, num_query, value_dim}, value.options());
if (use_cuda) {
int threads_x = WARP_SIZE;
int threads_y = OPTIMAL_THREADS_PER_BLOCK / WARP_SIZE;
int block_x_step1 = num_key / threads_y;
int block_x_step2 = num_query / threads_y;
int block_y = batch_size;
dim3 threads(threads_x, threads_y);
dim3 blocks_step1(block_x_step1, block_y);
dim3 blocks_step2(block_x_step2, block_y);
int *query_mask_ptr = query_mask.data_ptr<int>();
int *query_hash_code_ptr = query_hash_code.data_ptr<int>();
float *query_weight_ptr = query_weight.data_ptr<float>();
int *key_mask_ptr = key_mask.data_ptr<int>();
int *key_hash_code_ptr = key_hash_code.data_ptr<int>();
float *key_weight_ptr = key_weight.data_ptr<float>();
float *value_ptr = value.data_ptr<float>();
float *hashtable_value_ptr = hashtable_value.data_ptr<float>();
float *cumulation_value_ptr = cumulation_value.data_ptr<float>();
for (int value_offset = 0; value_offset < value_dim; value_offset = value_offset + WARP_SIZE) {
for (int weight_idx = 0; weight_idx < weight_dim; weight_idx++) {
cudaMemset(hashtable_value_ptr, 0, (batch_size * num_hash_f * hashtable_capacity * WARP_SIZE) * sizeof(float));
lsh_weighted_cumulation_ver1_step1_cuda_kernel<<<blocks_step1, threads>>>(
key_mask_ptr,
key_hash_code_ptr,
key_weight_ptr,
value_ptr,
hashtable_value_ptr,
batch_size,
num_hash_f,
hashtable_capacity,
num_key,
value_dim,
weight_dim,
value_offset,
weight_idx
);
lsh_weighted_cumulation_ver1_step2_cuda_kernel<<<blocks_step2, threads>>>(
query_mask_ptr,
query_hash_code_ptr,
query_weight_ptr,
hashtable_value_ptr,
cumulation_value_ptr,
batch_size,
num_hash_f,
hashtable_capacity,
num_query,
value_dim,
weight_dim,
value_offset,
weight_idx
);
}
}
}
return cumulation_value;
}
at::Tensor lsh_weighted_cumulation_ver2_kernel(
at::Tensor query_mask,
at::Tensor query_hash_code,
at::Tensor query_weight,
at::Tensor key_mask,
at::Tensor key_hash_code,
at::Tensor key_weight,
at::Tensor value,
int hashtable_capacity,
bool use_cuda
) {
int batch_size = query_hash_code.size(0);
int num_hash_f = query_hash_code.size(2);
int num_query = query_hash_code.size(1);
int num_key = key_hash_code.size(1);
int value_dim = value.size(2);
int weight_dim = query_weight.size(2);
at::Tensor count_sort_table = at::zeros({batch_size, num_hash_f, hashtable_capacity}, query_hash_code.options());
at::Tensor key_sorted_idxes = at::zeros({batch_size, num_hash_f, num_key}, query_hash_code.options());
at::Tensor query_info = at::zeros({batch_size, num_query, 2, num_hash_f}, query_hash_code.options());
at::Tensor cumulation_value = at::zeros({batch_size, num_query, value_dim}, value.options());
if (use_cuda) {
int *query_mask_ptr = query_mask.data_ptr<int>();
int *query_hash_code_ptr = query_hash_code.data_ptr<int>();
float *query_weight_ptr = query_weight.data_ptr<float>();
int *key_mask_ptr = key_mask.data_ptr<int>();
int *key_hash_code_ptr = key_hash_code.data_ptr<int>();
float *key_weight_ptr = key_weight.data_ptr<float>();
float *value_ptr = value.data_ptr<float>();
int *count_sort_table_ptr = count_sort_table.data_ptr<int>();
int *key_sorted_idxes_ptr = key_sorted_idxes.data_ptr<int>();
int *query_info_ptr = query_info.data_ptr<int>();
float *cumulation_value_ptr = cumulation_value.data_ptr<float>();
{
dim3 threads_step13(num_hash_f, max(1, OPTIMAL_THREADS_PER_BLOCK / num_hash_f));
dim3 blocks_step13(num_key / max(1, OPTIMAL_THREADS_PER_BLOCK / num_hash_f), batch_size);
dim3 threads_step2(min(hashtable_capacity, OPTIMAL_THREADS_PER_BLOCK));
dim3 blocks_step2(num_hash_f, batch_size);
int shared_mem = hashtable_capacity * sizeof(float);
count_sort_step1_cuda_kernel<<<blocks_step13, threads_step13>>>(
key_mask_ptr,
key_hash_code_ptr,
count_sort_table_ptr,
batch_size,
num_hash_f,
hashtable_capacity,
num_key
);
count_sort_step2_cuda_kernel<<<blocks_step2, threads_step2, shared_mem>>>(
count_sort_table_ptr,
batch_size,
num_hash_f,
hashtable_capacity
);
count_sort_step3_cuda_kernel<<<blocks_step13, threads_step13>>>(
key_mask_ptr,
key_hash_code_ptr,
count_sort_table_ptr,
key_sorted_idxes_ptr,
batch_size,
num_hash_f,
hashtable_capacity,
num_key
);
}
{
dim3 threads(num_hash_f, max(1, OPTIMAL_THREADS_PER_BLOCK / num_hash_f));
dim3 blocks(num_query / max(1, OPTIMAL_THREADS_PER_BLOCK / num_hash_f), batch_size);
extract_query_info_cuda_kernel<<<blocks, threads>>>(
query_mask_ptr,
query_hash_code_ptr,
count_sort_table_ptr,
query_info_ptr,
batch_size,
num_hash_f,
hashtable_capacity,
num_query
);
}
{
dim3 threads(WARP_SIZE, OPTIMAL_THREADS_PER_BLOCK / WARP_SIZE);
dim3 blocks(num_query, num_hash_f, batch_size);
int shared_mem = (weight_dim + WARP_SIZE) * sizeof(float);
lsh_weighted_cumulation_ver2_step2_cuda_kernel<<<blocks, threads, shared_mem>>>(
query_mask_ptr,
query_info_ptr,
key_sorted_idxes_ptr,
query_weight_ptr,
key_weight_ptr,
value_ptr,
cumulation_value_ptr,
batch_size,
num_hash_f,
num_query,
num_key,
value_dim,
weight_dim
);
}
}
return cumulation_value;
}
at::Tensor lsh_weighted_cumulation_ver3_kernel(
at::Tensor query_mask,
at::Tensor query_hash_code,
at::Tensor query_weight,
at::Tensor key_mask,
at::Tensor key_hash_code,
at::Tensor key_weight,
at::Tensor value,
int hashtable_capacity,
bool use_cuda
) {
int batch_size = query_hash_code.size(0);
int num_hash_f = query_hash_code.size(2);
int num_query = query_hash_code.size(1);
int num_key = key_hash_code.size(1);
int value_dim = value.size(2);
int weight_dim = query_weight.size(2);
at::Tensor count_sort_table = at::zeros({batch_size, num_hash_f, hashtable_capacity}, query_hash_code.options());
at::Tensor query_sorted_idxes = at::zeros({batch_size, num_hash_f, num_query}, query_hash_code.options());
at::Tensor key_info = at::zeros({batch_size, num_key, 2, num_hash_f}, query_hash_code.options());
at::Tensor cumulation_value = at::zeros({batch_size, num_query, value_dim}, value.options());
if (use_cuda) {
int *query_mask_ptr = query_mask.data_ptr<int>();
int *query_hash_code_ptr = query_hash_code.data_ptr<int>();
float *query_weight_ptr = query_weight.data_ptr<float>();
int *key_mask_ptr = key_mask.data_ptr<int>();
int *key_hash_code_ptr = key_hash_code.data_ptr<int>();
float *key_weight_ptr = key_weight.data_ptr<float>();
float *value_ptr = value.data_ptr<float>();
int *count_sort_table_ptr = count_sort_table.data_ptr<int>();
int *query_sorted_idxes_ptr = query_sorted_idxes.data_ptr<int>();
int *key_info_ptr = key_info.data_ptr<int>();
float *cumulation_value_ptr = cumulation_value.data_ptr<float>();
{
dim3 threads_step13(num_hash_f, max(1, OPTIMAL_THREADS_PER_BLOCK / num_hash_f));
dim3 blocks_step13(num_query / max(1, OPTIMAL_THREADS_PER_BLOCK / num_hash_f), batch_size);
dim3 threads_step2(min(hashtable_capacity, OPTIMAL_THREADS_PER_BLOCK));
dim3 blocks_step2(num_hash_f, batch_size);
int shared_mem = hashtable_capacity * sizeof(float);
count_sort_step1_cuda_kernel<<<blocks_step13, threads_step13>>>(
query_mask_ptr,
query_hash_code_ptr,
count_sort_table_ptr,
batch_size,
num_hash_f,
hashtable_capacity,
num_query
);
count_sort_step2_cuda_kernel<<<blocks_step2, threads_step2, shared_mem>>>(
count_sort_table_ptr,
batch_size,
num_hash_f,
hashtable_capacity
);
count_sort_step3_cuda_kernel<<<blocks_step13, threads_step13>>>(
query_mask_ptr,
query_hash_code_ptr,
count_sort_table_ptr,
query_sorted_idxes_ptr,
batch_size,
num_hash_f,
hashtable_capacity,
num_query
);
}
{
dim3 threads(num_hash_f, max(1, OPTIMAL_THREADS_PER_BLOCK / num_hash_f));
dim3 blocks(num_key / max(1, OPTIMAL_THREADS_PER_BLOCK / num_hash_f), batch_size);
extract_query_info_cuda_kernel<<<blocks, threads>>>(
key_mask_ptr,
key_hash_code_ptr,
count_sort_table_ptr,
key_info_ptr,
batch_size,
num_hash_f,
hashtable_capacity,
num_key
);
}
{
dim3 threads(WARP_SIZE, OPTIMAL_THREADS_PER_BLOCK / WARP_SIZE);
dim3 blocks(num_key, num_hash_f, batch_size);
int shared_mem = (weight_dim + value_dim + WARP_SIZE) * sizeof(float);
lsh_weighted_cumulation_ver3_step2_cuda_kernel<<<blocks, threads, shared_mem>>>(
query_sorted_idxes_ptr,
key_mask_ptr,
key_info_ptr,
query_weight_ptr,
key_weight_ptr,
value_ptr,
cumulation_value_ptr,
batch_size,
num_hash_f,
num_query,
num_key,
value_dim,
weight_dim
);
}
}
return cumulation_value;
}
at::Tensor lsh_weighted_cumulation_ver4_kernel(
at::Tensor query_mask,
at::Tensor query_hash_code,
at::Tensor query_weight,
at::Tensor key_mask,
at::Tensor key_hash_code,
at::Tensor key_weight,
at::Tensor value,
int hashtable_capacity,
bool use_cuda
) {
int batch_size = query_hash_code.size(0);
int num_hash_f = query_hash_code.size(2);
int num_query = query_hash_code.size(1);
int num_key = key_hash_code.size(1);
int value_dim = value.size(2);
int weight_dim = query_weight.size(2);
at::Tensor count_sort_table = at::zeros({batch_size, num_hash_f, hashtable_capacity}, query_hash_code.options());
at::Tensor query_sorted_idxes = at::zeros({batch_size, num_hash_f, num_query}, query_hash_code.options());
at::Tensor key_info = at::zeros({batch_size, num_key, 2, num_hash_f}, query_hash_code.options());
at::Tensor cumulation_value = at::zeros({batch_size, num_query, value_dim}, value.options());
if (use_cuda) {
int *query_mask_ptr = query_mask.data_ptr<int>();
int *query_hash_code_ptr = query_hash_code.data_ptr<int>();
float *query_weight_ptr = query_weight.data_ptr<float>();
int *key_mask_ptr = key_mask.data_ptr<int>();
int *key_hash_code_ptr = key_hash_code.data_ptr<int>();
float *key_weight_ptr = key_weight.data_ptr<float>();
float *value_ptr = value.data_ptr<float>();
int *count_sort_table_ptr = count_sort_table.data_ptr<int>();
int *query_sorted_idxes_ptr = query_sorted_idxes.data_ptr<int>();
int *key_info_ptr = key_info.data_ptr<int>();
float *cumulation_value_ptr = cumulation_value.data_ptr<float>();
{
dim3 threads_step13(num_hash_f, max(1, OPTIMAL_THREADS_PER_BLOCK / num_hash_f));
dim3 blocks_step13(num_query / max(1, OPTIMAL_THREADS_PER_BLOCK / num_hash_f), batch_size);
dim3 threads_step2(min(hashtable_capacity, OPTIMAL_THREADS_PER_BLOCK));
dim3 blocks_step2(num_hash_f, batch_size);
int shared_mem = hashtable_capacity * sizeof(float);
count_sort_step1_cuda_kernel<<<blocks_step13, threads_step13>>>(
query_mask_ptr,
query_hash_code_ptr,
count_sort_table_ptr,
batch_size,
num_hash_f,
hashtable_capacity,
num_query
);
count_sort_step2_cuda_kernel<<<blocks_step2, threads_step2, shared_mem>>>(
count_sort_table_ptr,
batch_size,
num_hash_f,
hashtable_capacity
);
count_sort_step3_cuda_kernel<<<blocks_step13, threads_step13>>>(
query_mask_ptr,
query_hash_code_ptr,
count_sort_table_ptr,
query_sorted_idxes_ptr,
batch_size,
num_hash_f,
hashtable_capacity,
num_query
);
}
{
dim3 threads(num_hash_f, max(1, OPTIMAL_THREADS_PER_BLOCK / num_hash_f));
dim3 blocks(num_key / max(1, OPTIMAL_THREADS_PER_BLOCK / num_hash_f), batch_size);
extract_query_info_cuda_kernel<<<blocks, threads>>>(
key_mask_ptr,
key_hash_code_ptr,
count_sort_table_ptr,
key_info_ptr,
batch_size,
num_hash_f,
hashtable_capacity,
num_key
);
}
{
dim3 threads(WARP_SIZE, OPTIMAL_THREADS_PER_BLOCK / WARP_SIZE);
dim3 blocks(num_key, batch_size);
int shared_mem = (weight_dim + value_dim + 2 * num_hash_f) * sizeof(float);
lsh_weighted_cumulation_ver4_step2_cuda_kernel<<<blocks, threads, shared_mem>>>(
query_sorted_idxes_ptr,
key_mask_ptr,
key_info_ptr,
query_weight_ptr,
key_weight_ptr,
value_ptr,
cumulation_value_ptr,
batch_size,
num_hash_f,
num_query,
num_key,
value_dim,
weight_dim
);
}
}
return cumulation_value;
}
| 0
|
mavonic_private_repos/transformers/src/transformers/kernels
|
mavonic_private_repos/transformers/src/transformers/kernels/yoso/fast_lsh_cumulation.h
|
#include <torch/extension.h>
#include <ATen/ATen.h>
#include <vector>
std::vector<at::Tensor> fast_hash_ver1_kernel(
at::Tensor query_mask,
at::Tensor query_vector,
at::Tensor key_mask,
at::Tensor key_vector,
int num_hash_f,
int hash_code_len,
bool use_cuda
);
at::Tensor lsh_cumulation_ver1_kernel(
at::Tensor query_mask,
at::Tensor query_hash_code,
at::Tensor key_mask,
at::Tensor key_hash_code,
at::Tensor value,
int hashtable_capacity,
bool use_cuda
);
at::Tensor lsh_weighted_cumulation_ver1_kernel(
at::Tensor query_mask,
at::Tensor query_hash_code,
at::Tensor query_weight,
at::Tensor key_mask,
at::Tensor key_hash_code,
at::Tensor key_weight,
at::Tensor value,
int hashtable_capacity,
bool use_cuda
);
at::Tensor lsh_weighted_cumulation_ver2_kernel(
at::Tensor query_mask,
at::Tensor query_hash_code,
at::Tensor query_weight,
at::Tensor key_mask,
at::Tensor key_hash_code,
at::Tensor key_weight,
at::Tensor value,
int hashtable_capacity,
bool use_cuda
);
at::Tensor lsh_weighted_cumulation_ver3_kernel(
at::Tensor query_mask,
at::Tensor query_hash_code,
at::Tensor query_weight,
at::Tensor key_mask,
at::Tensor key_hash_code,
at::Tensor key_weight,
at::Tensor value,
int hashtable_capacity,
bool use_cuda
);
at::Tensor lsh_weighted_cumulation_ver4_kernel(
at::Tensor query_mask,
at::Tensor query_hash_code,
at::Tensor query_weight,
at::Tensor key_mask,
at::Tensor key_hash_code,
at::Tensor key_weight,
at::Tensor value,
int hashtable_capacity,
bool use_cuda
);
| 0
|
mavonic_private_repos/transformers/src/transformers/kernels
|
mavonic_private_repos/transformers/src/transformers/kernels/yoso/common.h
|
#define min(a, b) ((a)<(b)?(a):(b))
#define max(a, b) ((a)>(b)?(a):(b))
#define ceil_divide(a, b) ((a)/(b)+((a)%(b)!=0))
#define select(cond, a, b) ((cond)?(a):(b))
#define PI 3.141592
#define EPSILON 1e-8
#define MAX_VAL 1e12
#define MIN_VAL -1e12
#define EMPTY_VALUE -1
| 0
|
mavonic_private_repos/transformers/src/transformers/kernels
|
mavonic_private_repos/transformers/src/transformers/kernels/rwkv/wkv_op.cpp
|
#include <torch/extension.h>
#include "ATen/ATen.h"
typedef at::BFloat16 bf16;
void cuda_forward(int B, int T, int C, float *w, float *u, float *k, float *v, float *y);
void cuda_forward_bf16(int B, int T, int C, float *w, bf16 *u, bf16 *k, bf16 *v, bf16 *y);
void cuda_forward_with_state(int B, int T, int C, float *w, float *u, float *k, float *v, float *y, float *s);
void cuda_forward_with_state_bf16(int B, int T, int C, float *w, bf16 *u, bf16 *k, bf16 *v, bf16 *y, float *s);
void cuda_backward(int B, int T, int C, float *w, float *u, float *k, float *v, float *y, float *gy, float *gw, float *gu, float *gk, float *gv);
void cuda_backward_bf16(int B, int T, int C, float *w, bf16 *u, bf16 *k, bf16 *v, bf16 *y, bf16 *gy, bf16 *gw, bf16 *gu, bf16 *gk, bf16 *gv);
void forward(torch::Tensor &w, torch::Tensor &u, torch::Tensor &k, torch::Tensor &v, torch::Tensor &y) {
const int B = k.size(0);
const int T = k.size(1);
const int C = k.size(2);
cuda_forward(B, T, C, w.data_ptr<float>(), u.data_ptr<float>(), k.data_ptr<float>(), v.data_ptr<float>(), y.data_ptr<float>());
}
void forward_bf16(torch::Tensor &w, torch::Tensor &u, torch::Tensor &k, torch::Tensor &v, torch::Tensor &y) {
const int B = k.size(0);
const int T = k.size(1);
const int C = k.size(2);
cuda_forward_bf16(B, T, C, w.data_ptr<float>(), u.data_ptr<bf16>(), k.data_ptr<bf16>(), v.data_ptr<bf16>(), y.data_ptr<bf16>());
}
void forward_with_state(torch::Tensor &w, torch::Tensor &u, torch::Tensor &k, torch::Tensor &v, torch::Tensor &y, torch::Tensor &s) {
const int B = k.size(0);
const int T = k.size(1);
const int C = k.size(2);
cuda_forward_with_state(B, T, C, w.data_ptr<float>(), u.data_ptr<float>(), k.data_ptr<float>(), v.data_ptr<float>(), y.data_ptr<float>(), s.data_ptr<float>());
}
void forward_with_state_bf16(torch::Tensor &w, torch::Tensor &u, torch::Tensor &k, torch::Tensor &v, torch::Tensor &y, torch::Tensor &s) {
const int B = k.size(0);
const int T = k.size(1);
const int C = k.size(2);
cuda_forward_with_state_bf16(B, T, C, w.data_ptr<float>(), u.data_ptr<bf16>(), k.data_ptr<bf16>(), v.data_ptr<bf16>(), y.data_ptr<bf16>(), s.data_ptr<float>());
}
void backward(torch::Tensor &w, torch::Tensor &u, torch::Tensor &k, torch::Tensor &v, torch::Tensor &y, torch::Tensor &gy, torch::Tensor &gw, torch::Tensor &gu, torch::Tensor &gk, torch::Tensor &gv) {
const int B = k.size(0);
const int T = k.size(1);
const int C = k.size(2);
cuda_backward(B, T, C, w.data_ptr<float>(), u.data_ptr<float>(), k.data_ptr<float>(), v.data_ptr<float>(), y.data_ptr<float>(), gy.data_ptr<float>(), gw.data_ptr<float>(), gu.data_ptr<float>(), gk.data_ptr<float>(), gv.data_ptr<float>());
}
void backward_bf16(torch::Tensor &w, torch::Tensor &u, torch::Tensor &k, torch::Tensor &v, torch::Tensor &y, torch::Tensor &gy, torch::Tensor &gw, torch::Tensor &gu, torch::Tensor &gk, torch::Tensor &gv) {
const int B = k.size(0);
const int T = k.size(1);
const int C = k.size(2);
cuda_backward_bf16(B, T, C, w.data_ptr<float>(), u.data_ptr<bf16>(), k.data_ptr<bf16>(), v.data_ptr<bf16>(), y.data_ptr<bf16>(),
gy.data_ptr<bf16>(), gw.data_ptr<bf16>(), gu.data_ptr<bf16>(), gk.data_ptr<bf16>(), gv.data_ptr<bf16>());
}
PYBIND11_MODULE(TORCH_EXTENSION_NAME, m) {
m.def("forward", &forward, "wkv forward");
m.def("forward_bf16", &forward_bf16, "wkv forward bf16");
m.def("forward_with_state", &forward_with_state, "wkv forward with state");
m.def("forward_with_state_bf16", &forward_with_state_bf16, "wkv forward with state bf16");
m.def("backward", &backward, "wkv backward");
m.def("backward_bf16", &backward_bf16, "wkv backward bf16");
}
TORCH_LIBRARY(wkv, m) {
m.def("forward", forward);
m.def("forward_bf16", forward_bf16);
m.def("forward_with_state", forward_with_state);
m.def("forward_with_state_bf16", forward_with_state_bf16);
m.def("backward", backward);
m.def("backward_bf16", backward_bf16);
}
| 0
|
mavonic_private_repos/transformers/src/transformers/kernels
|
mavonic_private_repos/transformers/src/transformers/kernels/rwkv/wkv_cuda.cu
|
#include <stdio.h>
#include <assert.h>
#define MIN_VALUE (-1e38)
template <typename F>
__global__ void kernel_forward(
const int B, const int T, const int C, const F *__restrict__ const _w, const F *__restrict__ const _u,
const F *__restrict__ const _k, const F *__restrict__ const _v, F *__restrict__ const _y
) {
const int idx = blockIdx.x * blockDim.x + threadIdx.x;
const int _b = idx / C;
const int _c = idx % C;
const int _offset = _b * T * C + _c;
F u = _u[_c];
F w = _w[_c];
const F *__restrict__ const k = _k + _offset;
const F *__restrict__ const v = _v + _offset;
F *__restrict__ const y = _y + _offset;
// aa and bb are running sums divided by exp(pp) (to avoid overflow)
F aa = 0, bb = 0, pp = MIN_VALUE;
for (int i = 0; i < T; i++) {
const int ii = i * C;
const F kk = k[ii];
const F vv = v[ii];
F ww = u + kk;
F p = max(pp, ww);
F e1 = exp(pp - p);
F e2 = exp(ww - p);
y[ii] = (e1 * aa + e2 * vv) / (e1 * bb + e2);
ww = w + pp;
p = max(ww, kk);
e1 = exp(ww - p);
e2 = exp(kk - p);
aa = e1 * aa + e2 * vv;
bb = e1 * bb + e2;
pp = p;
}
}
template <typename F>
__global__ void kernel_forward_with_state(
const int B, const int T, const int C, const F *__restrict__ const _w, const F *__restrict__ const _u,
const F *__restrict__ const _k, const F *__restrict__ const _v, F *__restrict__ const _y, F *__restrict__ const _s
) {
const int idx = blockIdx.x * blockDim.x + threadIdx.x;
const int _b = idx / C;
const int _c = idx % C;
const int _offset_s = _b * C * 3 + _c * 3;
const int _offset = _b * T * C + _c;
F u = _u[_c];
F w = _w[_c];
const F *__restrict__ const k = _k + _offset;
const F *__restrict__ const v = _v + _offset;
F *__restrict__ const y = _y + _offset;
F *__restrict__ const s = _s + _offset_s;
// aa and bb are running sums divided by exp(pp) (to avoid overflow)
F aa = s[0], bb = s[1], pp = s[2];
for (int i = 0; i < T; i++) {
const int ii = i * C;
const F kk = k[ii];
const F vv = v[ii];
F ww = u + kk;
F p = max(pp, ww);
F e1 = exp(pp - p);
F e2 = exp(ww - p);
y[ii] = (e1 * aa + e2 * vv) / (e1 * bb + e2);
ww = w + pp;
p = max(ww, kk);
e1 = exp(ww - p);
e2 = exp(kk - p);
aa = e1 * aa + e2 * vv;
bb = e1 * bb + e2;
pp = p;
}
s[0] = aa;
s[1] = bb;
s[2] = pp;
}
template <typename F>
__global__ void kernel_backward(
const int B, const int T, const int C, const F *__restrict__ const _w, const F *__restrict__ const _u,
const F *__restrict__ const _k, const F *__restrict__ const _v, const F *__restrict__ const _y,
const F *__restrict__ const _gy, F *__restrict__ const _gw, F *__restrict__ const _gu, F *__restrict__ const _gk,
F *__restrict__ const _gv
) {
const int idx = blockIdx.x * blockDim.x + threadIdx.x;
const int _b = idx / C;
const int _c = idx % C;
const int _offset = _b * T * C + _c;
F u = _u[_c];
F w = _w[_c];
const F *__restrict__ const k = _k + _offset;
const F *__restrict__ const v = _v + _offset;
const F *__restrict__ const y = _y + _offset;
const F *__restrict__ const gy = _gy + _offset;
F *__restrict__ const gk = _gk + _offset;
F *__restrict__ const gv = _gv + _offset;
F q[Tmax], r[Tmax];
F gw = 0, gu = 0, aa = 0, bb = 0, ga = 0, gb = 0, pp = MIN_VALUE;
for (int i = 0; i < T; i++) {
const int ii = i * C;
const F kk = k[ii];
const F vv = v[ii];
const F yy = y[ii];
F ww = u + kk;
F p = max(pp, ww);
F e1 = exp(pp - p);
F e2 = exp(ww - p);
const F qq = gy[ii] / (e1 * bb + e2);
gw += (ga - gb * yy) * e1 * qq;
gu += (vv - yy) * e2 * qq;
q[i] = qq;
r[i] = ww - p;
ww = w + pp;
p = max(ww, kk);
e1 = exp(ww - p);
e2 = exp(kk - p);
ga = e1 * (aa + ga);
gb = e1 * (bb + gb);
aa = e1 * aa + e2 * vv;
bb = e1 * bb + e2;
pp = p;
}
const int _offsetBC = _b * C + _c;
_gw[_offsetBC] = gw * _w[_c]; // multiply by w because of w -> -exp(w) in python forward()
_gu[_offsetBC] = gu;
aa = 0, bb = 0, pp = MIN_VALUE;
for (int i = T - 1; i >= 0; i--) {
const int ii = i * C;
const F kk = k[ii];
const F vv = v[ii];
const F yy = y[ii];
const F qq = q[i];
const F rr = r[i];
F e1 = qq * exp(rr);
F e2 = exp(kk + pp);
gk[ii] = e1 * (vv - yy) + e2 * (aa * vv + bb);
gv[ii] = e1 + e2 * aa;
const F ww = w + pp;
const F www = rr - u - kk;
const F p = max(ww, www);
e1 = exp(ww - p);
e2 = qq * exp(www - p);
aa = e1 * aa + e2;
bb = e1 * bb - e2 * yy;
pp = p;
}
}
void cuda_forward(int B, int T, int C, float *w, float *u, float *k, float *v, float *y) {
dim3 threadsPerBlock( min(C, 32) ); // requires --maxrregcount 60 for optimal performance
assert(B * C % threadsPerBlock.x == 0);
dim3 numBlocks(B * C / threadsPerBlock.x);
kernel_forward<<<numBlocks, threadsPerBlock>>>(B, T, C, w, u, k, v, y);
}
void cuda_forward_with_state(int B, int T, int C, float *w, float *u, float *k, float *v, float *y, float *s) {
dim3 threadsPerBlock( min(C, 32) ); // requires --maxrregcount 60 for optimal performance
assert(B * C % threadsPerBlock.x == 0);
dim3 numBlocks(B * C / threadsPerBlock.x);
kernel_forward_with_state<<<numBlocks, threadsPerBlock>>>(B, T, C, w, u, k, v, y, s);
}
void cuda_backward(int B, int T, int C, float *w, float *u, float *k, float *v, float *y, float *gy, float *gw, float *gu, float *gk, float *gv) {
dim3 threadsPerBlock( min(C, 32) ); // requires --maxrregcount 60 for optimal performance
assert(B * C % threadsPerBlock.x == 0);
dim3 numBlocks(B * C / threadsPerBlock.x);
kernel_backward<<<numBlocks, threadsPerBlock>>>(B, T, C, w, u, k, v, y, gy, gw, gu, gk, gv);
}
| 0
|
mavonic_private_repos/transformers/src/transformers/kernels
|
mavonic_private_repos/transformers/src/transformers/kernels/rwkv/wkv_cuda_bf16.cu
|
#include <stdio.h>
#include <assert.h>
#include "ATen/ATen.h"
#define MIN_VALUE (-1e38)
typedef at::BFloat16 bf16;
__global__ void kernel_forward_bf16(
const int B, const int T, const int C, const float *__restrict__ const _w, const bf16 *__restrict__ const _u,
const bf16 *__restrict__ const _k, const bf16 *__restrict__ const _v, bf16 *__restrict__ const _y
) {
const int idx = blockIdx.x * blockDim.x + threadIdx.x;
const int _b = idx / C;
const int _c = idx % C;
const int _offset = _b * T * C + _c;
float u = float(_u[_c]);
float w = _w[_c];
const bf16 *__restrict__ const k = _k + _offset;
const bf16 *__restrict__ const v = _v + _offset;
bf16 *__restrict__ const y = _y + _offset;
// aa and bb are running sums divided by exp(pp) (to avoid overflow)
float aa = 0, bb = 0, pp = MIN_VALUE;
for (int i = 0; i < T; i++) {
const int ii = i * C;
const float kk = float(k[ii]);
const float vv = float(v[ii]);
float ww = u + kk;
float p = max(pp, ww);
float e1 = exp(pp - p);
float e2 = exp(ww - p);
y[ii] = bf16((e1 * aa + e2 * vv) / (e1 * bb + e2));
ww = w + pp;
p = max(ww, kk);
e1 = exp(ww - p);
e2 = exp(kk - p);
aa = e1 * aa + e2 * vv;
bb = e1 * bb + e2;
pp = p;
}
}
__global__ void kernel_forward_with_state_bf16(
const int B, const int T, const int C, const float *__restrict__ const _w, const bf16 *__restrict__ const _u,
const bf16 *__restrict__ const _k, const bf16 *__restrict__ const _v, bf16 *__restrict__ const _y,
float *__restrict__ const _s
) {
const int idx = blockIdx.x * blockDim.x + threadIdx.x;
const int _b = idx / C;
const int _c = idx % C;
const int _offset_s = _b * C * 3 + _c * 3;
const int _offset = _b * T * C + _c;
float u = float(_u[_c]);
float w = _w[_c];
const bf16 *__restrict__ const k = _k + _offset;
const bf16 *__restrict__ const v = _v + _offset;
bf16 *__restrict__ const y = _y + _offset;
float *__restrict__ const s = _s + _offset_s;
// aa and bb are running sums divided by exp(pp) (to avoid overflow)
float aa = s[0], bb = s[1], pp = s[2];
for (int i = 0; i < T; i++) {
const int ii = i * C;
const float kk = float(k[ii]);
const float vv = float(v[ii]);
float ww = u + kk;
float p = max(pp, ww);
float e1 = exp(pp - p);
float e2 = exp(ww - p);
y[ii] = bf16(e1 * aa + e2 * vv) / (e1 * bb + e2);
ww = w + pp;
p = max(ww, kk);
e1 = exp(ww - p);
e2 = exp(kk - p);
aa = e1 * aa + e2 * vv;
bb = e1 * bb + e2;
pp = p;
}
s[0] = aa;
s[1] = bb;
s[2] = pp;
}
__global__ void kernel_backward_bf16(
const int B, const int T, const int C, const float *__restrict__ const _w, const bf16 *__restrict__ const _u,
const bf16 *__restrict__ const _k, const bf16 *__restrict__ const _v, const bf16 *__restrict__ const _y,
const bf16 *__restrict__ const _gy, bf16 *__restrict__ const _gw, bf16 *__restrict__ const _gu,
bf16 *__restrict__ const _gk, bf16 *__restrict__ const _gv
) {
const int idx = blockIdx.x * blockDim.x + threadIdx.x;
const int _b = idx / C;
const int _c = idx % C;
const int _offset = _b * T * C + _c;
float u = float(_u[_c]);
float w = _w[_c];
const bf16 *__restrict__ const k = _k + _offset;
const bf16 *__restrict__ const v = _v + _offset;
const bf16 *__restrict__ const y = _y + _offset;
const bf16 *__restrict__ const gy = _gy + _offset;
bf16 *__restrict__ const gk = _gk + _offset;
bf16 *__restrict__ const gv = _gv + _offset;
float q[Tmax], r[Tmax];
float gw = 0, gu = 0, aa = 0, bb = 0, ga = 0, gb = 0, pp = MIN_VALUE;
for (int i = 0; i < T; i++) {
const int ii = i * C;
const float kk = float(k[ii]);
const float vv = float(v[ii]);
const float yy = float(y[ii]);
float ww = u + kk;
float p = max(pp, ww);
float e1 = exp(pp - p);
float e2 = exp(ww - p);
const float qq = float(gy[ii]) / (e1 * bb + e2);
gw += (ga - gb * yy) * e1 * qq;
gu += (vv - yy) * e2 * qq;
q[i] = qq;
r[i] = ww - p;
ww = w + pp;
p = max(ww, kk);
e1 = exp(ww - p);
e2 = exp(kk - p);
ga = e1 * (aa + ga);
gb = e1 * (bb + gb);
aa = e1 * aa + e2 * vv;
bb = e1 * bb + e2;
pp = p;
}
const int _offsetBC = _b * C + _c;
_gw[_offsetBC] = bf16(gw * _w[_c]); // multiply by w because of w -> -exp(w) in python forward()
_gu[_offsetBC] = bf16(gu);
aa = 0, bb = 0, pp = MIN_VALUE;
for (int i = T - 1; i >= 0; i--) {
const int ii = i * C;
const float kk = float(k[ii]);
const float vv = float(v[ii]);
const float yy = float(y[ii]);
const float qq = q[i];
const float rr = r[i];
float e1 = qq * exp(rr);
float e2 = exp(kk + pp);
gk[ii] = bf16(e1 * (vv - yy) + e2 * (aa * vv + bb));
gv[ii] = bf16(e1 + e2 * aa);
const float ww = w + pp;
const float www = rr - u - kk;
const float p = max(ww, www);
e1 = exp(ww - p);
e2 = qq * exp(www - p);
aa = e1 * aa + e2;
bb = e1 * bb - e2 * yy;
pp = p;
}
}
void cuda_forward_bf16(int B, int T, int C, float *w, bf16 *u, bf16 *k, bf16 *v, bf16 *y) {
dim3 threadsPerBlock( min(C, 32) ); // requires --maxrregcount 60 for optimal performance
assert(B * C % threadsPerBlock.x == 0);
dim3 numBlocks(B * C / threadsPerBlock.x);
kernel_forward_bf16<<<numBlocks, threadsPerBlock>>>(B, T, C, w, u, k, v, y);
}
void cuda_forward_with_state_bf16(int B, int T, int C, float *w, bf16 *u, bf16 *k, bf16 *v, bf16 *y, float *s) {
dim3 threadsPerBlock( min(C, 32) ); // requires --maxrregcount 60 for optimal performance
assert(B * C % threadsPerBlock.x == 0);
dim3 numBlocks(B * C / threadsPerBlock.x);
kernel_forward_with_state_bf16<<<numBlocks, threadsPerBlock>>>(B, T, C, w, u, k, v, y, s);
}
void cuda_backward_bf16(int B, int T, int C, float *w, bf16 *u, bf16 *k, bf16 *v, bf16 *y, bf16 *gy, bf16 *gw, bf16 *gu, bf16 *gk, bf16 *gv) {
dim3 threadsPerBlock( min(C, 32) ); // requires --maxrregcount 60 for optimal performance
assert(B * C % threadsPerBlock.x == 0);
dim3 numBlocks(B * C / threadsPerBlock.x);
kernel_backward_bf16<<<numBlocks, threadsPerBlock>>>(B, T, C, w, u, k, v, y, gy, gw, gu, gk, gv);
}
| 0
|
mavonic_private_repos/transformers/src/transformers/kernels
|
mavonic_private_repos/transformers/src/transformers/kernels/deta/ms_deform_attn.h
|
/*!
**************************************************************************************************
* Deformable DETR
* Copyright (c) 2020 SenseTime. All Rights Reserved.
* Licensed under the Apache License, Version 2.0 [see LICENSE for details]
**************************************************************************************************
* Modified from https://github.com/chengdazhi/Deformable-Convolution-V2-PyTorch/tree/pytorch_1.0.0
**************************************************************************************************
*/
#pragma once
#include "cpu/ms_deform_attn_cpu.h"
#ifdef WITH_CUDA
#include "cuda/ms_deform_attn_cuda.h"
#endif
at::Tensor
ms_deform_attn_forward(
const at::Tensor &value,
const at::Tensor &spatial_shapes,
const at::Tensor &level_start_index,
const at::Tensor &sampling_loc,
const at::Tensor &attn_weight,
const int im2col_step)
{
if (value.type().is_cuda())
{
#ifdef WITH_CUDA
return ms_deform_attn_cuda_forward(
value, spatial_shapes, level_start_index, sampling_loc, attn_weight, im2col_step);
#else
AT_ERROR("Not compiled with GPU support");
#endif
}
AT_ERROR("Not implemented on the CPU");
}
std::vector<at::Tensor>
ms_deform_attn_backward(
const at::Tensor &value,
const at::Tensor &spatial_shapes,
const at::Tensor &level_start_index,
const at::Tensor &sampling_loc,
const at::Tensor &attn_weight,
const at::Tensor &grad_output,
const int im2col_step)
{
if (value.type().is_cuda())
{
#ifdef WITH_CUDA
return ms_deform_attn_cuda_backward(
value, spatial_shapes, level_start_index, sampling_loc, attn_weight, grad_output, im2col_step);
#else
AT_ERROR("Not compiled with GPU support");
#endif
}
AT_ERROR("Not implemented on the CPU");
}
| 0
|
mavonic_private_repos/transformers/src/transformers/kernels
|
mavonic_private_repos/transformers/src/transformers/kernels/deta/vision.cpp
|
/*!
**************************************************************************************************
* Deformable DETR
* Copyright (c) 2020 SenseTime. All Rights Reserved.
* Licensed under the Apache License, Version 2.0 [see LICENSE for details]
**************************************************************************************************
* Modified from https://github.com/chengdazhi/Deformable-Convolution-V2-PyTorch/tree/pytorch_1.0.0
**************************************************************************************************
*/
#include "ms_deform_attn.h"
PYBIND11_MODULE(TORCH_EXTENSION_NAME, m) {
m.def("ms_deform_attn_forward", &ms_deform_attn_forward, "ms_deform_attn_forward");
m.def("ms_deform_attn_backward", &ms_deform_attn_backward, "ms_deform_attn_backward");
}
| 0
|
mavonic_private_repos/transformers/src/transformers/kernels/deta
|
mavonic_private_repos/transformers/src/transformers/kernels/deta/cpu/ms_deform_attn_cpu.h
|
/*!
**************************************************************************************************
* Deformable DETR
* Copyright (c) 2020 SenseTime. All Rights Reserved.
* Licensed under the Apache License, Version 2.0 [see LICENSE for details]
**************************************************************************************************
* Modified from https://github.com/chengdazhi/Deformable-Convolution-V2-PyTorch/tree/pytorch_1.0.0
**************************************************************************************************
*/
#pragma once
#include <torch/extension.h>
at::Tensor
ms_deform_attn_cpu_forward(
const at::Tensor &value,
const at::Tensor &spatial_shapes,
const at::Tensor &level_start_index,
const at::Tensor &sampling_loc,
const at::Tensor &attn_weight,
const int im2col_step);
std::vector<at::Tensor>
ms_deform_attn_cpu_backward(
const at::Tensor &value,
const at::Tensor &spatial_shapes,
const at::Tensor &level_start_index,
const at::Tensor &sampling_loc,
const at::Tensor &attn_weight,
const at::Tensor &grad_output,
const int im2col_step);
| 0
|
mavonic_private_repos/transformers/src/transformers/kernels/deta
|
mavonic_private_repos/transformers/src/transformers/kernels/deta/cpu/ms_deform_attn_cpu.cpp
|
/*!
**************************************************************************************************
* Deformable DETR
* Copyright (c) 2020 SenseTime. All Rights Reserved.
* Licensed under the Apache License, Version 2.0 [see LICENSE for details]
**************************************************************************************************
* Modified from https://github.com/chengdazhi/Deformable-Convolution-V2-PyTorch/tree/pytorch_1.0.0
**************************************************************************************************
*/
#include <vector>
#include <ATen/ATen.h>
#include <ATen/cuda/CUDAContext.h>
at::Tensor
ms_deform_attn_cpu_forward(
const at::Tensor &value,
const at::Tensor &spatial_shapes,
const at::Tensor &level_start_index,
const at::Tensor &sampling_loc,
const at::Tensor &attn_weight,
const int im2col_step)
{
AT_ERROR("Not implement on cpu");
}
std::vector<at::Tensor>
ms_deform_attn_cpu_backward(
const at::Tensor &value,
const at::Tensor &spatial_shapes,
const at::Tensor &level_start_index,
const at::Tensor &sampling_loc,
const at::Tensor &attn_weight,
const at::Tensor &grad_output,
const int im2col_step)
{
AT_ERROR("Not implement on cpu");
}
| 0
|
mavonic_private_repos/transformers/src/transformers/kernels/deta
|
mavonic_private_repos/transformers/src/transformers/kernels/deta/cuda/ms_deform_im2col_cuda.cuh
|
/*!
**************************************************************************
* Deformable DETR
* Copyright (c) 2020 SenseTime. All Rights Reserved.
* Licensed under the Apache License, Version 2.0 [see LICENSE for details]
**************************************************************************
* Modified from DCN (https://github.com/msracver/Deformable-ConvNets)
* Copyright (c) 2018 Microsoft
**************************************************************************
*/
#include <cstdio>
#include <algorithm>
#include <cstring>
#include <ATen/ATen.h>
#include <ATen/cuda/CUDAContext.h>
#include <THC/THCAtomics.cuh>
#define CUDA_KERNEL_LOOP(i, n) \
for (int i = blockIdx.x * blockDim.x + threadIdx.x; \
i < (n); \
i += blockDim.x * gridDim.x)
const int CUDA_NUM_THREADS = 1024;
inline int GET_BLOCKS(const int N, const int num_threads)
{
return (N + num_threads - 1) / num_threads;
}
template <typename scalar_t>
__device__ scalar_t ms_deform_attn_im2col_bilinear(const scalar_t* &bottom_data,
const int &height, const int &width, const int &nheads, const int &channels,
const scalar_t &h, const scalar_t &w, const int &m, const int &c)
{
const int h_low = floor(h);
const int w_low = floor(w);
const int h_high = h_low + 1;
const int w_high = w_low + 1;
const scalar_t lh = h - h_low;
const scalar_t lw = w - w_low;
const scalar_t hh = 1 - lh, hw = 1 - lw;
const int w_stride = nheads * channels;
const int h_stride = width * w_stride;
const int h_low_ptr_offset = h_low * h_stride;
const int h_high_ptr_offset = h_low_ptr_offset + h_stride;
const int w_low_ptr_offset = w_low * w_stride;
const int w_high_ptr_offset = w_low_ptr_offset + w_stride;
const int base_ptr = m * channels + c;
scalar_t v1 = 0;
if (h_low >= 0 && w_low >= 0)
{
const int ptr1 = h_low_ptr_offset + w_low_ptr_offset + base_ptr;
v1 = bottom_data[ptr1];
}
scalar_t v2 = 0;
if (h_low >= 0 && w_high <= width - 1)
{
const int ptr2 = h_low_ptr_offset + w_high_ptr_offset + base_ptr;
v2 = bottom_data[ptr2];
}
scalar_t v3 = 0;
if (h_high <= height - 1 && w_low >= 0)
{
const int ptr3 = h_high_ptr_offset + w_low_ptr_offset + base_ptr;
v3 = bottom_data[ptr3];
}
scalar_t v4 = 0;
if (h_high <= height - 1 && w_high <= width - 1)
{
const int ptr4 = h_high_ptr_offset + w_high_ptr_offset + base_ptr;
v4 = bottom_data[ptr4];
}
const scalar_t w1 = hh * hw, w2 = hh * lw, w3 = lh * hw, w4 = lh * lw;
const scalar_t val = (w1 * v1 + w2 * v2 + w3 * v3 + w4 * v4);
return val;
}
template <typename scalar_t>
__device__ void ms_deform_attn_col2im_bilinear(const scalar_t* &bottom_data,
const int &height, const int &width, const int &nheads, const int &channels,
const scalar_t &h, const scalar_t &w, const int &m, const int &c,
const scalar_t &top_grad,
const scalar_t &attn_weight,
scalar_t* &grad_value,
scalar_t* grad_sampling_loc,
scalar_t* grad_attn_weight)
{
const int h_low = floor(h);
const int w_low = floor(w);
const int h_high = h_low + 1;
const int w_high = w_low + 1;
const scalar_t lh = h - h_low;
const scalar_t lw = w - w_low;
const scalar_t hh = 1 - lh, hw = 1 - lw;
const int w_stride = nheads * channels;
const int h_stride = width * w_stride;
const int h_low_ptr_offset = h_low * h_stride;
const int h_high_ptr_offset = h_low_ptr_offset + h_stride;
const int w_low_ptr_offset = w_low * w_stride;
const int w_high_ptr_offset = w_low_ptr_offset + w_stride;
const int base_ptr = m * channels + c;
const scalar_t w1 = hh * hw, w2 = hh * lw, w3 = lh * hw, w4 = lh * lw;
const scalar_t top_grad_value = top_grad * attn_weight;
scalar_t grad_h_weight = 0, grad_w_weight = 0;
scalar_t v1 = 0;
if (h_low >= 0 && w_low >= 0)
{
const int ptr1 = h_low_ptr_offset + w_low_ptr_offset + base_ptr;
v1 = bottom_data[ptr1];
grad_h_weight -= hw * v1;
grad_w_weight -= hh * v1;
atomicAdd(grad_value+ptr1, w1*top_grad_value);
}
scalar_t v2 = 0;
if (h_low >= 0 && w_high <= width - 1)
{
const int ptr2 = h_low_ptr_offset + w_high_ptr_offset + base_ptr;
v2 = bottom_data[ptr2];
grad_h_weight -= lw * v2;
grad_w_weight += hh * v2;
atomicAdd(grad_value+ptr2, w2*top_grad_value);
}
scalar_t v3 = 0;
if (h_high <= height - 1 && w_low >= 0)
{
const int ptr3 = h_high_ptr_offset + w_low_ptr_offset + base_ptr;
v3 = bottom_data[ptr3];
grad_h_weight += hw * v3;
grad_w_weight -= lh * v3;
atomicAdd(grad_value+ptr3, w3*top_grad_value);
}
scalar_t v4 = 0;
if (h_high <= height - 1 && w_high <= width - 1)
{
const int ptr4 = h_high_ptr_offset + w_high_ptr_offset + base_ptr;
v4 = bottom_data[ptr4];
grad_h_weight += lw * v4;
grad_w_weight += lh * v4;
atomicAdd(grad_value+ptr4, w4*top_grad_value);
}
const scalar_t val = (w1 * v1 + w2 * v2 + w3 * v3 + w4 * v4);
*grad_attn_weight = top_grad * val;
*grad_sampling_loc = width * grad_w_weight * top_grad_value;
*(grad_sampling_loc + 1) = height * grad_h_weight * top_grad_value;
}
template <typename scalar_t>
__device__ void ms_deform_attn_col2im_bilinear_gm(const scalar_t* &bottom_data,
const int &height, const int &width, const int &nheads, const int &channels,
const scalar_t &h, const scalar_t &w, const int &m, const int &c,
const scalar_t &top_grad,
const scalar_t &attn_weight,
scalar_t* &grad_value,
scalar_t* grad_sampling_loc,
scalar_t* grad_attn_weight)
{
const int h_low = floor(h);
const int w_low = floor(w);
const int h_high = h_low + 1;
const int w_high = w_low + 1;
const scalar_t lh = h - h_low;
const scalar_t lw = w - w_low;
const scalar_t hh = 1 - lh, hw = 1 - lw;
const int w_stride = nheads * channels;
const int h_stride = width * w_stride;
const int h_low_ptr_offset = h_low * h_stride;
const int h_high_ptr_offset = h_low_ptr_offset + h_stride;
const int w_low_ptr_offset = w_low * w_stride;
const int w_high_ptr_offset = w_low_ptr_offset + w_stride;
const int base_ptr = m * channels + c;
const scalar_t w1 = hh * hw, w2 = hh * lw, w3 = lh * hw, w4 = lh * lw;
const scalar_t top_grad_value = top_grad * attn_weight;
scalar_t grad_h_weight = 0, grad_w_weight = 0;
scalar_t v1 = 0;
if (h_low >= 0 && w_low >= 0)
{
const int ptr1 = h_low_ptr_offset + w_low_ptr_offset + base_ptr;
v1 = bottom_data[ptr1];
grad_h_weight -= hw * v1;
grad_w_weight -= hh * v1;
atomicAdd(grad_value+ptr1, w1*top_grad_value);
}
scalar_t v2 = 0;
if (h_low >= 0 && w_high <= width - 1)
{
const int ptr2 = h_low_ptr_offset + w_high_ptr_offset + base_ptr;
v2 = bottom_data[ptr2];
grad_h_weight -= lw * v2;
grad_w_weight += hh * v2;
atomicAdd(grad_value+ptr2, w2*top_grad_value);
}
scalar_t v3 = 0;
if (h_high <= height - 1 && w_low >= 0)
{
const int ptr3 = h_high_ptr_offset + w_low_ptr_offset + base_ptr;
v3 = bottom_data[ptr3];
grad_h_weight += hw * v3;
grad_w_weight -= lh * v3;
atomicAdd(grad_value+ptr3, w3*top_grad_value);
}
scalar_t v4 = 0;
if (h_high <= height - 1 && w_high <= width - 1)
{
const int ptr4 = h_high_ptr_offset + w_high_ptr_offset + base_ptr;
v4 = bottom_data[ptr4];
grad_h_weight += lw * v4;
grad_w_weight += lh * v4;
atomicAdd(grad_value+ptr4, w4*top_grad_value);
}
const scalar_t val = (w1 * v1 + w2 * v2 + w3 * v3 + w4 * v4);
atomicAdd(grad_attn_weight, top_grad * val);
atomicAdd(grad_sampling_loc, width * grad_w_weight * top_grad_value);
atomicAdd(grad_sampling_loc + 1, height * grad_h_weight * top_grad_value);
}
template <typename scalar_t>
__global__ void ms_deformable_im2col_gpu_kernel(const int n,
const scalar_t *data_value,
const int64_t *data_spatial_shapes,
const int64_t *data_level_start_index,
const scalar_t *data_sampling_loc,
const scalar_t *data_attn_weight,
const int batch_size,
const int spatial_size,
const int num_heads,
const int channels,
const int num_levels,
const int num_query,
const int num_point,
scalar_t *data_col)
{
CUDA_KERNEL_LOOP(index, n)
{
int _temp = index;
const int c_col = _temp % channels;
_temp /= channels;
const int sampling_index = _temp;
const int m_col = _temp % num_heads;
_temp /= num_heads;
const int q_col = _temp % num_query;
_temp /= num_query;
const int b_col = _temp;
scalar_t *data_col_ptr = data_col + index;
int data_weight_ptr = sampling_index * num_levels * num_point;
int data_loc_w_ptr = data_weight_ptr << 1;
const int qid_stride = num_heads * channels;
const int data_value_ptr_init_offset = b_col * spatial_size * qid_stride;
scalar_t col = 0;
for (int l_col=0; l_col < num_levels; ++l_col)
{
const int level_start_id = data_level_start_index[l_col];
const int spatial_h_ptr = l_col << 1;
const int spatial_h = data_spatial_shapes[spatial_h_ptr];
const int spatial_w = data_spatial_shapes[spatial_h_ptr + 1];
const scalar_t *data_value_ptr = data_value + (data_value_ptr_init_offset + level_start_id * qid_stride);
for (int p_col=0; p_col < num_point; ++p_col)
{
const scalar_t loc_w = data_sampling_loc[data_loc_w_ptr];
const scalar_t loc_h = data_sampling_loc[data_loc_w_ptr + 1];
const scalar_t weight = data_attn_weight[data_weight_ptr];
const scalar_t h_im = loc_h * spatial_h - 0.5;
const scalar_t w_im = loc_w * spatial_w - 0.5;
if (h_im > -1 && w_im > -1 && h_im < spatial_h && w_im < spatial_w)
{
col += ms_deform_attn_im2col_bilinear(data_value_ptr, spatial_h, spatial_w, num_heads, channels, h_im, w_im, m_col, c_col) * weight;
}
data_weight_ptr += 1;
data_loc_w_ptr += 2;
}
}
*data_col_ptr = col;
}
}
template <typename scalar_t, unsigned int blockSize>
__global__ void ms_deformable_col2im_gpu_kernel_shm_blocksize_aware_reduce_v1(const int n,
const scalar_t *grad_col,
const scalar_t *data_value,
const int64_t *data_spatial_shapes,
const int64_t *data_level_start_index,
const scalar_t *data_sampling_loc,
const scalar_t *data_attn_weight,
const int batch_size,
const int spatial_size,
const int num_heads,
const int channels,
const int num_levels,
const int num_query,
const int num_point,
scalar_t *grad_value,
scalar_t *grad_sampling_loc,
scalar_t *grad_attn_weight)
{
CUDA_KERNEL_LOOP(index, n)
{
__shared__ scalar_t cache_grad_sampling_loc[blockSize * 2];
__shared__ scalar_t cache_grad_attn_weight[blockSize];
unsigned int tid = threadIdx.x;
int _temp = index;
const int c_col = _temp % channels;
_temp /= channels;
const int sampling_index = _temp;
const int m_col = _temp % num_heads;
_temp /= num_heads;
const int q_col = _temp % num_query;
_temp /= num_query;
const int b_col = _temp;
const scalar_t top_grad = grad_col[index];
int data_weight_ptr = sampling_index * num_levels * num_point;
int data_loc_w_ptr = data_weight_ptr << 1;
const int grad_sampling_ptr = data_weight_ptr;
grad_sampling_loc += grad_sampling_ptr << 1;
grad_attn_weight += grad_sampling_ptr;
const int grad_weight_stride = 1;
const int grad_loc_stride = 2;
const int qid_stride = num_heads * channels;
const int data_value_ptr_init_offset = b_col * spatial_size * qid_stride;
for (int l_col=0; l_col < num_levels; ++l_col)
{
const int level_start_id = data_level_start_index[l_col];
const int spatial_h_ptr = l_col << 1;
const int spatial_h = data_spatial_shapes[spatial_h_ptr];
const int spatial_w = data_spatial_shapes[spatial_h_ptr + 1];
const int value_ptr_offset = data_value_ptr_init_offset + level_start_id * qid_stride;
const scalar_t *data_value_ptr = data_value + value_ptr_offset;
scalar_t *grad_value_ptr = grad_value + value_ptr_offset;
for (int p_col=0; p_col < num_point; ++p_col)
{
const scalar_t loc_w = data_sampling_loc[data_loc_w_ptr];
const scalar_t loc_h = data_sampling_loc[data_loc_w_ptr + 1];
const scalar_t weight = data_attn_weight[data_weight_ptr];
const scalar_t h_im = loc_h * spatial_h - 0.5;
const scalar_t w_im = loc_w * spatial_w - 0.5;
*(cache_grad_sampling_loc+(threadIdx.x << 1)) = 0;
*(cache_grad_sampling_loc+((threadIdx.x << 1) + 1)) = 0;
*(cache_grad_attn_weight+threadIdx.x)=0;
if (h_im > -1 && w_im > -1 && h_im < spatial_h && w_im < spatial_w)
{
ms_deform_attn_col2im_bilinear(
data_value_ptr, spatial_h, spatial_w, num_heads, channels, h_im, w_im, m_col, c_col,
top_grad, weight, grad_value_ptr,
cache_grad_sampling_loc+(threadIdx.x << 1), cache_grad_attn_weight+threadIdx.x);
}
__syncthreads();
if (tid == 0)
{
scalar_t _grad_w=cache_grad_sampling_loc[0], _grad_h=cache_grad_sampling_loc[1], _grad_a=cache_grad_attn_weight[0];
int sid=2;
for (unsigned int tid = 1; tid < blockSize; ++tid)
{
_grad_w += cache_grad_sampling_loc[sid];
_grad_h += cache_grad_sampling_loc[sid + 1];
_grad_a += cache_grad_attn_weight[tid];
sid += 2;
}
*grad_sampling_loc = _grad_w;
*(grad_sampling_loc + 1) = _grad_h;
*grad_attn_weight = _grad_a;
}
__syncthreads();
data_weight_ptr += 1;
data_loc_w_ptr += 2;
grad_attn_weight += grad_weight_stride;
grad_sampling_loc += grad_loc_stride;
}
}
}
}
template <typename scalar_t, unsigned int blockSize>
__global__ void ms_deformable_col2im_gpu_kernel_shm_blocksize_aware_reduce_v2(const int n,
const scalar_t *grad_col,
const scalar_t *data_value,
const int64_t *data_spatial_shapes,
const int64_t *data_level_start_index,
const scalar_t *data_sampling_loc,
const scalar_t *data_attn_weight,
const int batch_size,
const int spatial_size,
const int num_heads,
const int channels,
const int num_levels,
const int num_query,
const int num_point,
scalar_t *grad_value,
scalar_t *grad_sampling_loc,
scalar_t *grad_attn_weight)
{
CUDA_KERNEL_LOOP(index, n)
{
__shared__ scalar_t cache_grad_sampling_loc[blockSize * 2];
__shared__ scalar_t cache_grad_attn_weight[blockSize];
unsigned int tid = threadIdx.x;
int _temp = index;
const int c_col = _temp % channels;
_temp /= channels;
const int sampling_index = _temp;
const int m_col = _temp % num_heads;
_temp /= num_heads;
const int q_col = _temp % num_query;
_temp /= num_query;
const int b_col = _temp;
const scalar_t top_grad = grad_col[index];
int data_weight_ptr = sampling_index * num_levels * num_point;
int data_loc_w_ptr = data_weight_ptr << 1;
const int grad_sampling_ptr = data_weight_ptr;
grad_sampling_loc += grad_sampling_ptr << 1;
grad_attn_weight += grad_sampling_ptr;
const int grad_weight_stride = 1;
const int grad_loc_stride = 2;
const int qid_stride = num_heads * channels;
const int data_value_ptr_init_offset = b_col * spatial_size * qid_stride;
for (int l_col=0; l_col < num_levels; ++l_col)
{
const int level_start_id = data_level_start_index[l_col];
const int spatial_h_ptr = l_col << 1;
const int spatial_h = data_spatial_shapes[spatial_h_ptr];
const int spatial_w = data_spatial_shapes[spatial_h_ptr + 1];
const int value_ptr_offset = data_value_ptr_init_offset + level_start_id * qid_stride;
const scalar_t *data_value_ptr = data_value + value_ptr_offset;
scalar_t *grad_value_ptr = grad_value + value_ptr_offset;
for (int p_col=0; p_col < num_point; ++p_col)
{
const scalar_t loc_w = data_sampling_loc[data_loc_w_ptr];
const scalar_t loc_h = data_sampling_loc[data_loc_w_ptr + 1];
const scalar_t weight = data_attn_weight[data_weight_ptr];
const scalar_t h_im = loc_h * spatial_h - 0.5;
const scalar_t w_im = loc_w * spatial_w - 0.5;
*(cache_grad_sampling_loc+(threadIdx.x << 1)) = 0;
*(cache_grad_sampling_loc+((threadIdx.x << 1) + 1)) = 0;
*(cache_grad_attn_weight+threadIdx.x)=0;
if (h_im > -1 && w_im > -1 && h_im < spatial_h && w_im < spatial_w)
{
ms_deform_attn_col2im_bilinear(
data_value_ptr, spatial_h, spatial_w, num_heads, channels, h_im, w_im, m_col, c_col,
top_grad, weight, grad_value_ptr,
cache_grad_sampling_loc+(threadIdx.x << 1), cache_grad_attn_weight+threadIdx.x);
}
__syncthreads();
for (unsigned int s=blockSize/2; s>0; s>>=1)
{
if (tid < s) {
const unsigned int xid1 = tid << 1;
const unsigned int xid2 = (tid + s) << 1;
cache_grad_attn_weight[tid] += cache_grad_attn_weight[tid + s];
cache_grad_sampling_loc[xid1] += cache_grad_sampling_loc[xid2];
cache_grad_sampling_loc[xid1 + 1] += cache_grad_sampling_loc[xid2 + 1];
}
__syncthreads();
}
if (tid == 0)
{
*grad_sampling_loc = cache_grad_sampling_loc[0];
*(grad_sampling_loc + 1) = cache_grad_sampling_loc[1];
*grad_attn_weight = cache_grad_attn_weight[0];
}
__syncthreads();
data_weight_ptr += 1;
data_loc_w_ptr += 2;
grad_attn_weight += grad_weight_stride;
grad_sampling_loc += grad_loc_stride;
}
}
}
}
template <typename scalar_t>
__global__ void ms_deformable_col2im_gpu_kernel_shm_reduce_v1(const int n,
const scalar_t *grad_col,
const scalar_t *data_value,
const int64_t *data_spatial_shapes,
const int64_t *data_level_start_index,
const scalar_t *data_sampling_loc,
const scalar_t *data_attn_weight,
const int batch_size,
const int spatial_size,
const int num_heads,
const int channels,
const int num_levels,
const int num_query,
const int num_point,
scalar_t *grad_value,
scalar_t *grad_sampling_loc,
scalar_t *grad_attn_weight)
{
CUDA_KERNEL_LOOP(index, n)
{
extern __shared__ int _s[];
scalar_t* cache_grad_sampling_loc = (scalar_t*)_s;
scalar_t* cache_grad_attn_weight = cache_grad_sampling_loc + 2 * blockDim.x;
unsigned int tid = threadIdx.x;
int _temp = index;
const int c_col = _temp % channels;
_temp /= channels;
const int sampling_index = _temp;
const int m_col = _temp % num_heads;
_temp /= num_heads;
const int q_col = _temp % num_query;
_temp /= num_query;
const int b_col = _temp;
const scalar_t top_grad = grad_col[index];
int data_weight_ptr = sampling_index * num_levels * num_point;
int data_loc_w_ptr = data_weight_ptr << 1;
const int grad_sampling_ptr = data_weight_ptr;
grad_sampling_loc += grad_sampling_ptr << 1;
grad_attn_weight += grad_sampling_ptr;
const int grad_weight_stride = 1;
const int grad_loc_stride = 2;
const int qid_stride = num_heads * channels;
const int data_value_ptr_init_offset = b_col * spatial_size * qid_stride;
for (int l_col=0; l_col < num_levels; ++l_col)
{
const int level_start_id = data_level_start_index[l_col];
const int spatial_h_ptr = l_col << 1;
const int spatial_h = data_spatial_shapes[spatial_h_ptr];
const int spatial_w = data_spatial_shapes[spatial_h_ptr + 1];
const int value_ptr_offset = data_value_ptr_init_offset + level_start_id * qid_stride;
const scalar_t *data_value_ptr = data_value + value_ptr_offset;
scalar_t *grad_value_ptr = grad_value + value_ptr_offset;
for (int p_col=0; p_col < num_point; ++p_col)
{
const scalar_t loc_w = data_sampling_loc[data_loc_w_ptr];
const scalar_t loc_h = data_sampling_loc[data_loc_w_ptr + 1];
const scalar_t weight = data_attn_weight[data_weight_ptr];
const scalar_t h_im = loc_h * spatial_h - 0.5;
const scalar_t w_im = loc_w * spatial_w - 0.5;
*(cache_grad_sampling_loc+(threadIdx.x << 1)) = 0;
*(cache_grad_sampling_loc+((threadIdx.x << 1) + 1)) = 0;
*(cache_grad_attn_weight+threadIdx.x)=0;
if (h_im > -1 && w_im > -1 && h_im < spatial_h && w_im < spatial_w)
{
ms_deform_attn_col2im_bilinear(
data_value_ptr, spatial_h, spatial_w, num_heads, channels, h_im, w_im, m_col, c_col,
top_grad, weight, grad_value_ptr,
cache_grad_sampling_loc+(threadIdx.x << 1), cache_grad_attn_weight+threadIdx.x);
}
__syncthreads();
if (tid == 0)
{
scalar_t _grad_w=cache_grad_sampling_loc[0], _grad_h=cache_grad_sampling_loc[1], _grad_a=cache_grad_attn_weight[0];
int sid=2;
for (unsigned int tid = 1; tid < blockDim.x; ++tid)
{
_grad_w += cache_grad_sampling_loc[sid];
_grad_h += cache_grad_sampling_loc[sid + 1];
_grad_a += cache_grad_attn_weight[tid];
sid += 2;
}
*grad_sampling_loc = _grad_w;
*(grad_sampling_loc + 1) = _grad_h;
*grad_attn_weight = _grad_a;
}
__syncthreads();
data_weight_ptr += 1;
data_loc_w_ptr += 2;
grad_attn_weight += grad_weight_stride;
grad_sampling_loc += grad_loc_stride;
}
}
}
}
template <typename scalar_t>
__global__ void ms_deformable_col2im_gpu_kernel_shm_reduce_v2(const int n,
const scalar_t *grad_col,
const scalar_t *data_value,
const int64_t *data_spatial_shapes,
const int64_t *data_level_start_index,
const scalar_t *data_sampling_loc,
const scalar_t *data_attn_weight,
const int batch_size,
const int spatial_size,
const int num_heads,
const int channels,
const int num_levels,
const int num_query,
const int num_point,
scalar_t *grad_value,
scalar_t *grad_sampling_loc,
scalar_t *grad_attn_weight)
{
CUDA_KERNEL_LOOP(index, n)
{
extern __shared__ int _s[];
scalar_t* cache_grad_sampling_loc = (scalar_t*)_s;
scalar_t* cache_grad_attn_weight = cache_grad_sampling_loc + 2 * blockDim.x;
unsigned int tid = threadIdx.x;
int _temp = index;
const int c_col = _temp % channels;
_temp /= channels;
const int sampling_index = _temp;
const int m_col = _temp % num_heads;
_temp /= num_heads;
const int q_col = _temp % num_query;
_temp /= num_query;
const int b_col = _temp;
const scalar_t top_grad = grad_col[index];
int data_weight_ptr = sampling_index * num_levels * num_point;
int data_loc_w_ptr = data_weight_ptr << 1;
const int grad_sampling_ptr = data_weight_ptr;
grad_sampling_loc += grad_sampling_ptr << 1;
grad_attn_weight += grad_sampling_ptr;
const int grad_weight_stride = 1;
const int grad_loc_stride = 2;
const int qid_stride = num_heads * channels;
const int data_value_ptr_init_offset = b_col * spatial_size * qid_stride;
for (int l_col=0; l_col < num_levels; ++l_col)
{
const int level_start_id = data_level_start_index[l_col];
const int spatial_h_ptr = l_col << 1;
const int spatial_h = data_spatial_shapes[spatial_h_ptr];
const int spatial_w = data_spatial_shapes[spatial_h_ptr + 1];
const int value_ptr_offset = data_value_ptr_init_offset + level_start_id * qid_stride;
const scalar_t *data_value_ptr = data_value + value_ptr_offset;
scalar_t *grad_value_ptr = grad_value + value_ptr_offset;
for (int p_col=0; p_col < num_point; ++p_col)
{
const scalar_t loc_w = data_sampling_loc[data_loc_w_ptr];
const scalar_t loc_h = data_sampling_loc[data_loc_w_ptr + 1];
const scalar_t weight = data_attn_weight[data_weight_ptr];
const scalar_t h_im = loc_h * spatial_h - 0.5;
const scalar_t w_im = loc_w * spatial_w - 0.5;
*(cache_grad_sampling_loc+(threadIdx.x << 1)) = 0;
*(cache_grad_sampling_loc+((threadIdx.x << 1) + 1)) = 0;
*(cache_grad_attn_weight+threadIdx.x)=0;
if (h_im > -1 && w_im > -1 && h_im < spatial_h && w_im < spatial_w)
{
ms_deform_attn_col2im_bilinear(
data_value_ptr, spatial_h, spatial_w, num_heads, channels, h_im, w_im, m_col, c_col,
top_grad, weight, grad_value_ptr,
cache_grad_sampling_loc+(threadIdx.x << 1), cache_grad_attn_weight+threadIdx.x);
}
__syncthreads();
for (unsigned int s=blockDim.x/2, spre=blockDim.x; s>0; s>>=1, spre>>=1)
{
if (tid < s) {
const unsigned int xid1 = tid << 1;
const unsigned int xid2 = (tid + s) << 1;
cache_grad_attn_weight[tid] += cache_grad_attn_weight[tid + s];
cache_grad_sampling_loc[xid1] += cache_grad_sampling_loc[xid2];
cache_grad_sampling_loc[xid1 + 1] += cache_grad_sampling_loc[xid2 + 1];
if (tid + (s << 1) < spre)
{
cache_grad_attn_weight[tid] += cache_grad_attn_weight[tid + (s << 1)];
cache_grad_sampling_loc[xid1] += cache_grad_sampling_loc[xid2 + (s << 1)];
cache_grad_sampling_loc[xid1 + 1] += cache_grad_sampling_loc[xid2 + 1 + (s << 1)];
}
}
__syncthreads();
}
if (tid == 0)
{
*grad_sampling_loc = cache_grad_sampling_loc[0];
*(grad_sampling_loc + 1) = cache_grad_sampling_loc[1];
*grad_attn_weight = cache_grad_attn_weight[0];
}
__syncthreads();
data_weight_ptr += 1;
data_loc_w_ptr += 2;
grad_attn_weight += grad_weight_stride;
grad_sampling_loc += grad_loc_stride;
}
}
}
}
template <typename scalar_t>
__global__ void ms_deformable_col2im_gpu_kernel_shm_reduce_v2_multi_blocks(const int n,
const scalar_t *grad_col,
const scalar_t *data_value,
const int64_t *data_spatial_shapes,
const int64_t *data_level_start_index,
const scalar_t *data_sampling_loc,
const scalar_t *data_attn_weight,
const int batch_size,
const int spatial_size,
const int num_heads,
const int channels,
const int num_levels,
const int num_query,
const int num_point,
scalar_t *grad_value,
scalar_t *grad_sampling_loc,
scalar_t *grad_attn_weight)
{
CUDA_KERNEL_LOOP(index, n)
{
extern __shared__ int _s[];
scalar_t* cache_grad_sampling_loc = (scalar_t*)_s;
scalar_t* cache_grad_attn_weight = cache_grad_sampling_loc + 2 * blockDim.x;
unsigned int tid = threadIdx.x;
int _temp = index;
const int c_col = _temp % channels;
_temp /= channels;
const int sampling_index = _temp;
const int m_col = _temp % num_heads;
_temp /= num_heads;
const int q_col = _temp % num_query;
_temp /= num_query;
const int b_col = _temp;
const scalar_t top_grad = grad_col[index];
int data_weight_ptr = sampling_index * num_levels * num_point;
int data_loc_w_ptr = data_weight_ptr << 1;
const int grad_sampling_ptr = data_weight_ptr;
grad_sampling_loc += grad_sampling_ptr << 1;
grad_attn_weight += grad_sampling_ptr;
const int grad_weight_stride = 1;
const int grad_loc_stride = 2;
const int qid_stride = num_heads * channels;
const int data_value_ptr_init_offset = b_col * spatial_size * qid_stride;
for (int l_col=0; l_col < num_levels; ++l_col)
{
const int level_start_id = data_level_start_index[l_col];
const int spatial_h_ptr = l_col << 1;
const int spatial_h = data_spatial_shapes[spatial_h_ptr];
const int spatial_w = data_spatial_shapes[spatial_h_ptr + 1];
const int value_ptr_offset = data_value_ptr_init_offset + level_start_id * qid_stride;
const scalar_t *data_value_ptr = data_value + value_ptr_offset;
scalar_t *grad_value_ptr = grad_value + value_ptr_offset;
for (int p_col=0; p_col < num_point; ++p_col)
{
const scalar_t loc_w = data_sampling_loc[data_loc_w_ptr];
const scalar_t loc_h = data_sampling_loc[data_loc_w_ptr + 1];
const scalar_t weight = data_attn_weight[data_weight_ptr];
const scalar_t h_im = loc_h * spatial_h - 0.5;
const scalar_t w_im = loc_w * spatial_w - 0.5;
*(cache_grad_sampling_loc+(threadIdx.x << 1)) = 0;
*(cache_grad_sampling_loc+((threadIdx.x << 1) + 1)) = 0;
*(cache_grad_attn_weight+threadIdx.x)=0;
if (h_im > -1 && w_im > -1 && h_im < spatial_h && w_im < spatial_w)
{
ms_deform_attn_col2im_bilinear(
data_value_ptr, spatial_h, spatial_w, num_heads, channels, h_im, w_im, m_col, c_col,
top_grad, weight, grad_value_ptr,
cache_grad_sampling_loc+(threadIdx.x << 1), cache_grad_attn_weight+threadIdx.x);
}
__syncthreads();
for (unsigned int s=blockDim.x/2, spre=blockDim.x; s>0; s>>=1, spre>>=1)
{
if (tid < s) {
const unsigned int xid1 = tid << 1;
const unsigned int xid2 = (tid + s) << 1;
cache_grad_attn_weight[tid] += cache_grad_attn_weight[tid + s];
cache_grad_sampling_loc[xid1] += cache_grad_sampling_loc[xid2];
cache_grad_sampling_loc[xid1 + 1] += cache_grad_sampling_loc[xid2 + 1];
if (tid + (s << 1) < spre)
{
cache_grad_attn_weight[tid] += cache_grad_attn_weight[tid + (s << 1)];
cache_grad_sampling_loc[xid1] += cache_grad_sampling_loc[xid2 + (s << 1)];
cache_grad_sampling_loc[xid1 + 1] += cache_grad_sampling_loc[xid2 + 1 + (s << 1)];
}
}
__syncthreads();
}
if (tid == 0)
{
atomicAdd(grad_sampling_loc, cache_grad_sampling_loc[0]);
atomicAdd(grad_sampling_loc + 1, cache_grad_sampling_loc[1]);
atomicAdd(grad_attn_weight, cache_grad_attn_weight[0]);
}
__syncthreads();
data_weight_ptr += 1;
data_loc_w_ptr += 2;
grad_attn_weight += grad_weight_stride;
grad_sampling_loc += grad_loc_stride;
}
}
}
}
template <typename scalar_t>
__global__ void ms_deformable_col2im_gpu_kernel_gm(const int n,
const scalar_t *grad_col,
const scalar_t *data_value,
const int64_t *data_spatial_shapes,
const int64_t *data_level_start_index,
const scalar_t *data_sampling_loc,
const scalar_t *data_attn_weight,
const int batch_size,
const int spatial_size,
const int num_heads,
const int channels,
const int num_levels,
const int num_query,
const int num_point,
scalar_t *grad_value,
scalar_t *grad_sampling_loc,
scalar_t *grad_attn_weight)
{
CUDA_KERNEL_LOOP(index, n)
{
int _temp = index;
const int c_col = _temp % channels;
_temp /= channels;
const int sampling_index = _temp;
const int m_col = _temp % num_heads;
_temp /= num_heads;
const int q_col = _temp % num_query;
_temp /= num_query;
const int b_col = _temp;
const scalar_t top_grad = grad_col[index];
int data_weight_ptr = sampling_index * num_levels * num_point;
int data_loc_w_ptr = data_weight_ptr << 1;
const int grad_sampling_ptr = data_weight_ptr;
grad_sampling_loc += grad_sampling_ptr << 1;
grad_attn_weight += grad_sampling_ptr;
const int grad_weight_stride = 1;
const int grad_loc_stride = 2;
const int qid_stride = num_heads * channels;
const int data_value_ptr_init_offset = b_col * spatial_size * qid_stride;
for (int l_col=0; l_col < num_levels; ++l_col)
{
const int level_start_id = data_level_start_index[l_col];
const int spatial_h_ptr = l_col << 1;
const int spatial_h = data_spatial_shapes[spatial_h_ptr];
const int spatial_w = data_spatial_shapes[spatial_h_ptr + 1];
const int value_ptr_offset = data_value_ptr_init_offset + level_start_id * qid_stride;
const scalar_t *data_value_ptr = data_value + value_ptr_offset;
scalar_t *grad_value_ptr = grad_value + value_ptr_offset;
for (int p_col=0; p_col < num_point; ++p_col)
{
const scalar_t loc_w = data_sampling_loc[data_loc_w_ptr];
const scalar_t loc_h = data_sampling_loc[data_loc_w_ptr + 1];
const scalar_t weight = data_attn_weight[data_weight_ptr];
const scalar_t h_im = loc_h * spatial_h - 0.5;
const scalar_t w_im = loc_w * spatial_w - 0.5;
if (h_im > -1 && w_im > -1 && h_im < spatial_h && w_im < spatial_w)
{
ms_deform_attn_col2im_bilinear_gm(
data_value_ptr, spatial_h, spatial_w, num_heads, channels, h_im, w_im, m_col, c_col,
top_grad, weight, grad_value_ptr,
grad_sampling_loc, grad_attn_weight);
}
data_weight_ptr += 1;
data_loc_w_ptr += 2;
grad_attn_weight += grad_weight_stride;
grad_sampling_loc += grad_loc_stride;
}
}
}
}
template <typename scalar_t>
void ms_deformable_im2col_cuda(cudaStream_t stream,
const scalar_t* data_value,
const int64_t* data_spatial_shapes,
const int64_t* data_level_start_index,
const scalar_t* data_sampling_loc,
const scalar_t* data_attn_weight,
const int batch_size,
const int spatial_size,
const int num_heads,
const int channels,
const int num_levels,
const int num_query,
const int num_point,
scalar_t* data_col)
{
const int num_kernels = batch_size * num_query * num_heads * channels;
const int num_actual_kernels = batch_size * num_query * num_heads * channels;
const int num_threads = CUDA_NUM_THREADS;
ms_deformable_im2col_gpu_kernel<scalar_t>
<<<GET_BLOCKS(num_actual_kernels, num_threads), num_threads,
0, stream>>>(
num_kernels, data_value, data_spatial_shapes, data_level_start_index, data_sampling_loc, data_attn_weight,
batch_size, spatial_size, num_heads, channels, num_levels, num_query, num_point, data_col);
cudaError_t err = cudaGetLastError();
if (err != cudaSuccess)
{
printf("error in ms_deformable_im2col_cuda: %s\n", cudaGetErrorString(err));
}
}
template <typename scalar_t>
void ms_deformable_col2im_cuda(cudaStream_t stream,
const scalar_t* grad_col,
const scalar_t* data_value,
const int64_t * data_spatial_shapes,
const int64_t * data_level_start_index,
const scalar_t * data_sampling_loc,
const scalar_t * data_attn_weight,
const int batch_size,
const int spatial_size,
const int num_heads,
const int channels,
const int num_levels,
const int num_query,
const int num_point,
scalar_t* grad_value,
scalar_t* grad_sampling_loc,
scalar_t* grad_attn_weight)
{
const int num_threads = (channels > CUDA_NUM_THREADS)?CUDA_NUM_THREADS:channels;
const int num_kernels = batch_size * num_query * num_heads * channels;
const int num_actual_kernels = batch_size * num_query * num_heads * channels;
if (channels > 1024)
{
if ((channels & 1023) == 0)
{
ms_deformable_col2im_gpu_kernel_shm_reduce_v2_multi_blocks<scalar_t>
<<<GET_BLOCKS(num_actual_kernels, num_threads), num_threads,
num_threads*3*sizeof(scalar_t), stream>>>(
num_kernels,
grad_col,
data_value,
data_spatial_shapes,
data_level_start_index,
data_sampling_loc,
data_attn_weight,
batch_size,
spatial_size,
num_heads,
channels,
num_levels,
num_query,
num_point,
grad_value,
grad_sampling_loc,
grad_attn_weight);
}
else
{
ms_deformable_col2im_gpu_kernel_gm<scalar_t>
<<<GET_BLOCKS(num_actual_kernels, num_threads), num_threads,
0, stream>>>(
num_kernels,
grad_col,
data_value,
data_spatial_shapes,
data_level_start_index,
data_sampling_loc,
data_attn_weight,
batch_size,
spatial_size,
num_heads,
channels,
num_levels,
num_query,
num_point,
grad_value,
grad_sampling_loc,
grad_attn_weight);
}
}
else{
switch(channels)
{
case 1:
ms_deformable_col2im_gpu_kernel_shm_blocksize_aware_reduce_v1<scalar_t, 1>
<<<GET_BLOCKS(num_actual_kernels, num_threads), num_threads,
0, stream>>>(
num_kernels,
grad_col,
data_value,
data_spatial_shapes,
data_level_start_index,
data_sampling_loc,
data_attn_weight,
batch_size,
spatial_size,
num_heads,
channels,
num_levels,
num_query,
num_point,
grad_value,
grad_sampling_loc,
grad_attn_weight);
break;
case 2:
ms_deformable_col2im_gpu_kernel_shm_blocksize_aware_reduce_v1<scalar_t, 2>
<<<GET_BLOCKS(num_actual_kernels, num_threads), num_threads,
0, stream>>>(
num_kernels,
grad_col,
data_value,
data_spatial_shapes,
data_level_start_index,
data_sampling_loc,
data_attn_weight,
batch_size,
spatial_size,
num_heads,
channels,
num_levels,
num_query,
num_point,
grad_value,
grad_sampling_loc,
grad_attn_weight);
break;
case 4:
ms_deformable_col2im_gpu_kernel_shm_blocksize_aware_reduce_v1<scalar_t, 4>
<<<GET_BLOCKS(num_actual_kernels, num_threads), num_threads,
0, stream>>>(
num_kernels,
grad_col,
data_value,
data_spatial_shapes,
data_level_start_index,
data_sampling_loc,
data_attn_weight,
batch_size,
spatial_size,
num_heads,
channels,
num_levels,
num_query,
num_point,
grad_value,
grad_sampling_loc,
grad_attn_weight);
break;
case 8:
ms_deformable_col2im_gpu_kernel_shm_blocksize_aware_reduce_v1<scalar_t, 8>
<<<GET_BLOCKS(num_actual_kernels, num_threads), num_threads,
0, stream>>>(
num_kernels,
grad_col,
data_value,
data_spatial_shapes,
data_level_start_index,
data_sampling_loc,
data_attn_weight,
batch_size,
spatial_size,
num_heads,
channels,
num_levels,
num_query,
num_point,
grad_value,
grad_sampling_loc,
grad_attn_weight);
break;
case 16:
ms_deformable_col2im_gpu_kernel_shm_blocksize_aware_reduce_v1<scalar_t, 16>
<<<GET_BLOCKS(num_actual_kernels, num_threads), num_threads,
0, stream>>>(
num_kernels,
grad_col,
data_value,
data_spatial_shapes,
data_level_start_index,
data_sampling_loc,
data_attn_weight,
batch_size,
spatial_size,
num_heads,
channels,
num_levels,
num_query,
num_point,
grad_value,
grad_sampling_loc,
grad_attn_weight);
break;
case 32:
ms_deformable_col2im_gpu_kernel_shm_blocksize_aware_reduce_v1<scalar_t, 32>
<<<GET_BLOCKS(num_actual_kernels, num_threads), num_threads,
0, stream>>>(
num_kernels,
grad_col,
data_value,
data_spatial_shapes,
data_level_start_index,
data_sampling_loc,
data_attn_weight,
batch_size,
spatial_size,
num_heads,
channels,
num_levels,
num_query,
num_point,
grad_value,
grad_sampling_loc,
grad_attn_weight);
break;
case 64:
ms_deformable_col2im_gpu_kernel_shm_blocksize_aware_reduce_v2<scalar_t, 64>
<<<GET_BLOCKS(num_actual_kernels, num_threads), num_threads,
0, stream>>>(
num_kernels,
grad_col,
data_value,
data_spatial_shapes,
data_level_start_index,
data_sampling_loc,
data_attn_weight,
batch_size,
spatial_size,
num_heads,
channels,
num_levels,
num_query,
num_point,
grad_value,
grad_sampling_loc,
grad_attn_weight);
break;
case 128:
ms_deformable_col2im_gpu_kernel_shm_blocksize_aware_reduce_v2<scalar_t, 128>
<<<GET_BLOCKS(num_actual_kernels, num_threads), num_threads,
0, stream>>>(
num_kernels,
grad_col,
data_value,
data_spatial_shapes,
data_level_start_index,
data_sampling_loc,
data_attn_weight,
batch_size,
spatial_size,
num_heads,
channels,
num_levels,
num_query,
num_point,
grad_value,
grad_sampling_loc,
grad_attn_weight);
break;
case 256:
ms_deformable_col2im_gpu_kernel_shm_blocksize_aware_reduce_v2<scalar_t, 256>
<<<GET_BLOCKS(num_actual_kernels, num_threads), num_threads,
0, stream>>>(
num_kernels,
grad_col,
data_value,
data_spatial_shapes,
data_level_start_index,
data_sampling_loc,
data_attn_weight,
batch_size,
spatial_size,
num_heads,
channels,
num_levels,
num_query,
num_point,
grad_value,
grad_sampling_loc,
grad_attn_weight);
break;
case 512:
ms_deformable_col2im_gpu_kernel_shm_blocksize_aware_reduce_v2<scalar_t, 512>
<<<GET_BLOCKS(num_actual_kernels, num_threads), num_threads,
0, stream>>>(
num_kernels,
grad_col,
data_value,
data_spatial_shapes,
data_level_start_index,
data_sampling_loc,
data_attn_weight,
batch_size,
spatial_size,
num_heads,
channels,
num_levels,
num_query,
num_point,
grad_value,
grad_sampling_loc,
grad_attn_weight);
break;
case 1024:
ms_deformable_col2im_gpu_kernel_shm_blocksize_aware_reduce_v2<scalar_t, 1024>
<<<GET_BLOCKS(num_actual_kernels, num_threads), num_threads,
0, stream>>>(
num_kernels,
grad_col,
data_value,
data_spatial_shapes,
data_level_start_index,
data_sampling_loc,
data_attn_weight,
batch_size,
spatial_size,
num_heads,
channels,
num_levels,
num_query,
num_point,
grad_value,
grad_sampling_loc,
grad_attn_weight);
break;
default:
if (channels < 64)
{
ms_deformable_col2im_gpu_kernel_shm_reduce_v1<scalar_t>
<<<GET_BLOCKS(num_actual_kernels, num_threads), num_threads,
num_threads*3*sizeof(scalar_t), stream>>>(
num_kernels,
grad_col,
data_value,
data_spatial_shapes,
data_level_start_index,
data_sampling_loc,
data_attn_weight,
batch_size,
spatial_size,
num_heads,
channels,
num_levels,
num_query,
num_point,
grad_value,
grad_sampling_loc,
grad_attn_weight);
}
else
{
ms_deformable_col2im_gpu_kernel_shm_reduce_v2<scalar_t>
<<<GET_BLOCKS(num_actual_kernels, num_threads), num_threads,
num_threads*3*sizeof(scalar_t), stream>>>(
num_kernels,
grad_col,
data_value,
data_spatial_shapes,
data_level_start_index,
data_sampling_loc,
data_attn_weight,
batch_size,
spatial_size,
num_heads,
channels,
num_levels,
num_query,
num_point,
grad_value,
grad_sampling_loc,
grad_attn_weight);
}
}
}
cudaError_t err = cudaGetLastError();
if (err != cudaSuccess)
{
printf("error in ms_deformable_col2im_cuda: %s\n", cudaGetErrorString(err));
}
}
| 0
|
mavonic_private_repos/transformers/src/transformers/kernels/deta
|
mavonic_private_repos/transformers/src/transformers/kernels/deta/cuda/ms_deform_attn_cuda.h
|
/*!
**************************************************************************************************
* Deformable DETR
* Copyright (c) 2020 SenseTime. All Rights Reserved.
* Licensed under the Apache License, Version 2.0 [see LICENSE for details]
**************************************************************************************************
* Modified from https://github.com/chengdazhi/Deformable-Convolution-V2-PyTorch/tree/pytorch_1.0.0
**************************************************************************************************
*/
#pragma once
#include <torch/extension.h>
at::Tensor ms_deform_attn_cuda_forward(
const at::Tensor &value,
const at::Tensor &spatial_shapes,
const at::Tensor &level_start_index,
const at::Tensor &sampling_loc,
const at::Tensor &attn_weight,
const int im2col_step);
std::vector<at::Tensor> ms_deform_attn_cuda_backward(
const at::Tensor &value,
const at::Tensor &spatial_shapes,
const at::Tensor &level_start_index,
const at::Tensor &sampling_loc,
const at::Tensor &attn_weight,
const at::Tensor &grad_output,
const int im2col_step);
| 0
|
mavonic_private_repos/transformers/src/transformers/kernels/deta
|
mavonic_private_repos/transformers/src/transformers/kernels/deta/cuda/ms_deform_attn_cuda.cu
|
/*!
**************************************************************************************************
* Deformable DETR
* Copyright (c) 2020 SenseTime. All Rights Reserved.
* Licensed under the Apache License, Version 2.0 [see LICENSE for details]
**************************************************************************************************
* Modified from https://github.com/chengdazhi/Deformable-Convolution-V2-PyTorch/tree/pytorch_1.0.0
**************************************************************************************************
*/
#include <vector>
#include "cuda/ms_deform_im2col_cuda.cuh"
#include <ATen/ATen.h>
#include <ATen/cuda/CUDAContext.h>
#include <cuda.h>
#include <cuda_runtime.h>
#pragma once
#include <torch/extension.h>
at::Tensor ms_deform_attn_cuda_forward(
const at::Tensor &value,
const at::Tensor &spatial_shapes,
const at::Tensor &level_start_index,
const at::Tensor &sampling_loc,
const at::Tensor &attn_weight,
const int im2col_step)
{
AT_ASSERTM(value.is_contiguous(), "value tensor has to be contiguous");
AT_ASSERTM(spatial_shapes.is_contiguous(), "spatial_shapes tensor has to be contiguous");
AT_ASSERTM(level_start_index.is_contiguous(), "level_start_index tensor has to be contiguous");
AT_ASSERTM(sampling_loc.is_contiguous(), "sampling_loc tensor has to be contiguous");
AT_ASSERTM(attn_weight.is_contiguous(), "attn_weight tensor has to be contiguous");
AT_ASSERTM(value.type().is_cuda(), "value must be a CUDA tensor");
AT_ASSERTM(spatial_shapes.type().is_cuda(), "spatial_shapes must be a CUDA tensor");
AT_ASSERTM(level_start_index.type().is_cuda(), "level_start_index must be a CUDA tensor");
AT_ASSERTM(sampling_loc.type().is_cuda(), "sampling_loc must be a CUDA tensor");
AT_ASSERTM(attn_weight.type().is_cuda(), "attn_weight must be a CUDA tensor");
const int batch = value.size(0);
const int spatial_size = value.size(1);
const int num_heads = value.size(2);
const int channels = value.size(3);
const int num_levels = spatial_shapes.size(0);
const int num_query = sampling_loc.size(1);
const int num_point = sampling_loc.size(4);
const int im2col_step_ = std::min(batch, im2col_step);
AT_ASSERTM(batch % im2col_step_ == 0, "batch(%d) must divide im2col_step(%d)", batch, im2col_step_);
auto output = at::zeros({batch, num_query, num_heads, channels}, value.options());
const int batch_n = im2col_step_;
auto output_n = output.view({batch/im2col_step_, batch_n, num_query, num_heads, channels});
auto per_value_size = spatial_size * num_heads * channels;
auto per_sample_loc_size = num_query * num_heads * num_levels * num_point * 2;
auto per_attn_weight_size = num_query * num_heads * num_levels * num_point;
for (int n = 0; n < batch/im2col_step_; ++n)
{
auto columns = output_n.select(0, n);
AT_DISPATCH_FLOATING_TYPES(value.type(), "ms_deform_attn_forward_cuda", ([&] {
ms_deformable_im2col_cuda(at::cuda::getCurrentCUDAStream(),
value.data<scalar_t>() + n * im2col_step_ * per_value_size,
spatial_shapes.data<int64_t>(),
level_start_index.data<int64_t>(),
sampling_loc.data<scalar_t>() + n * im2col_step_ * per_sample_loc_size,
attn_weight.data<scalar_t>() + n * im2col_step_ * per_attn_weight_size,
batch_n, spatial_size, num_heads, channels, num_levels, num_query, num_point,
columns.data<scalar_t>());
}));
}
output = output.view({batch, num_query, num_heads*channels});
return output;
}
std::vector<at::Tensor> ms_deform_attn_cuda_backward(
const at::Tensor &value,
const at::Tensor &spatial_shapes,
const at::Tensor &level_start_index,
const at::Tensor &sampling_loc,
const at::Tensor &attn_weight,
const at::Tensor &grad_output,
const int im2col_step)
{
AT_ASSERTM(value.is_contiguous(), "value tensor has to be contiguous");
AT_ASSERTM(spatial_shapes.is_contiguous(), "spatial_shapes tensor has to be contiguous");
AT_ASSERTM(level_start_index.is_contiguous(), "level_start_index tensor has to be contiguous");
AT_ASSERTM(sampling_loc.is_contiguous(), "sampling_loc tensor has to be contiguous");
AT_ASSERTM(attn_weight.is_contiguous(), "attn_weight tensor has to be contiguous");
AT_ASSERTM(grad_output.is_contiguous(), "grad_output tensor has to be contiguous");
AT_ASSERTM(value.type().is_cuda(), "value must be a CUDA tensor");
AT_ASSERTM(spatial_shapes.type().is_cuda(), "spatial_shapes must be a CUDA tensor");
AT_ASSERTM(level_start_index.type().is_cuda(), "level_start_index must be a CUDA tensor");
AT_ASSERTM(sampling_loc.type().is_cuda(), "sampling_loc must be a CUDA tensor");
AT_ASSERTM(attn_weight.type().is_cuda(), "attn_weight must be a CUDA tensor");
AT_ASSERTM(grad_output.type().is_cuda(), "grad_output must be a CUDA tensor");
const int batch = value.size(0);
const int spatial_size = value.size(1);
const int num_heads = value.size(2);
const int channels = value.size(3);
const int num_levels = spatial_shapes.size(0);
const int num_query = sampling_loc.size(1);
const int num_point = sampling_loc.size(4);
const int im2col_step_ = std::min(batch, im2col_step);
AT_ASSERTM(batch % im2col_step_ == 0, "batch(%d) must divide im2col_step(%d)", batch, im2col_step_);
auto grad_value = at::zeros_like(value);
auto grad_sampling_loc = at::zeros_like(sampling_loc);
auto grad_attn_weight = at::zeros_like(attn_weight);
const int batch_n = im2col_step_;
auto per_value_size = spatial_size * num_heads * channels;
auto per_sample_loc_size = num_query * num_heads * num_levels * num_point * 2;
auto per_attn_weight_size = num_query * num_heads * num_levels * num_point;
auto grad_output_n = grad_output.view({batch/im2col_step_, batch_n, num_query, num_heads, channels});
for (int n = 0; n < batch/im2col_step_; ++n)
{
auto grad_output_g = grad_output_n.select(0, n);
AT_DISPATCH_FLOATING_TYPES(value.type(), "ms_deform_attn_backward_cuda", ([&] {
ms_deformable_col2im_cuda(at::cuda::getCurrentCUDAStream(),
grad_output_g.data<scalar_t>(),
value.data<scalar_t>() + n * im2col_step_ * per_value_size,
spatial_shapes.data<int64_t>(),
level_start_index.data<int64_t>(),
sampling_loc.data<scalar_t>() + n * im2col_step_ * per_sample_loc_size,
attn_weight.data<scalar_t>() + n * im2col_step_ * per_attn_weight_size,
batch_n, spatial_size, num_heads, channels, num_levels, num_query, num_point,
grad_value.data<scalar_t>() + n * im2col_step_ * per_value_size,
grad_sampling_loc.data<scalar_t>() + n * im2col_step_ * per_sample_loc_size,
grad_attn_weight.data<scalar_t>() + n * im2col_step_ * per_attn_weight_size);
}));
}
return {
grad_value, grad_sampling_loc, grad_attn_weight
};
}
| 0
|
mavonic_private_repos/transformers/src/transformers/kernels/deta
|
mavonic_private_repos/transformers/src/transformers/kernels/deta/cuda/ms_deform_attn_cuda.cuh
|
/*!
**************************************************************************************************
* Deformable DETR
* Copyright (c) 2020 SenseTime. All Rights Reserved.
* Licensed under the Apache License, Version 2.0 [see LICENSE for details]
**************************************************************************************************
* Modified from https://github.com/chengdazhi/Deformable-Convolution-V2-PyTorch/tree/pytorch_1.0.0
**************************************************************************************************
*/
#include <vector>
#include <cuda.h>
#include <cuda_runtime.h>
#include <cstdio>
#include <algorithm>
#include <cstring>
#include <ATen/ATen.h>
#include <ATen/cuda/CUDAContext.h>
#include <THC/THCAtomics.cuh>
#define CUDA_KERNEL_LOOP(i, n) \
for (int i = blockIdx.x * blockDim.x + threadIdx.x; \
i < (n); \
i += blockDim.x * gridDim.x)
at::Tensor ms_deform_attn_cuda_forward(
const at::Tensor &value,
const at::Tensor &spatial_shapes,
const at::Tensor &level_start_index,
const at::Tensor &sampling_loc,
const at::Tensor &attn_weight,
const int im2col_step)
{
AT_ASSERTM(value.is_contiguous(), "value tensor has to be contiguous");
AT_ASSERTM(spatial_shapes.is_contiguous(), "spatial_shapes tensor has to be contiguous");
AT_ASSERTM(level_start_index.is_contiguous(), "level_start_index tensor has to be contiguous");
AT_ASSERTM(sampling_loc.is_contiguous(), "sampling_loc tensor has to be contiguous");
AT_ASSERTM(attn_weight.is_contiguous(), "attn_weight tensor has to be contiguous");
AT_ASSERTM(value.type().is_cuda(), "value must be a CUDA tensor");
AT_ASSERTM(spatial_shapes.type().is_cuda(), "spatial_shapes must be a CUDA tensor");
AT_ASSERTM(level_start_index.type().is_cuda(), "level_start_index must be a CUDA tensor");
AT_ASSERTM(sampling_loc.type().is_cuda(), "sampling_loc must be a CUDA tensor");
AT_ASSERTM(attn_weight.type().is_cuda(), "attn_weight must be a CUDA tensor");
const int batch = value.size(0);
const int spatial_size = value.size(1);
const int num_heads = value.size(2);
const int channels = value.size(3);
const int num_levels = spatial_shapes.size(0);
const int num_query = sampling_loc.size(1);
const int num_point = sampling_loc.size(4);
const int im2col_step_ = std::min(batch, im2col_step);
AT_ASSERTM(batch % im2col_step_ == 0, "batch(%d) must divide im2col_step(%d)", batch, im2col_step_);
auto output = at::zeros({batch, num_query, num_heads, channels}, value.options());
const int batch_n = im2col_step_;
auto output_n = output.view({batch/im2col_step_, batch_n, num_query, num_heads, channels});
auto per_value_size = spatial_size * num_heads * channels;
auto per_sample_loc_size = num_query * num_heads * num_levels * num_point * 2;
auto per_attn_weight_size = num_query * num_heads * num_levels * num_point;
for (int n = 0; n < batch/im2col_step_; ++n)
{
auto columns = output_n.select(0, n);
AT_DISPATCH_FLOATING_TYPES(value.type(), "ms_deform_attn_forward_cuda", ([&] {
ms_deformable_im2col_cuda(at::cuda::getCurrentCUDAStream(),
value.data<scalar_t>() + n * im2col_step_ * per_value_size,
spatial_shapes.data<int64_t>(),
level_start_index.data<int64_t>(),
sampling_loc.data<scalar_t>() + n * im2col_step_ * per_sample_loc_size,
attn_weight.data<scalar_t>() + n * im2col_step_ * per_attn_weight_size,
batch_n, spatial_size, num_heads, channels, num_levels, num_query, num_point,
columns.data<scalar_t>());
}));
}
output = output.view({batch, num_query, num_heads*channels});
return output;
}
std::vector<at::Tensor> ms_deform_attn_cuda_backward(
const at::Tensor &value,
const at::Tensor &spatial_shapes,
const at::Tensor &level_start_index,
const at::Tensor &sampling_loc,
const at::Tensor &attn_weight,
const at::Tensor &grad_output,
const int im2col_step)
{
AT_ASSERTM(value.is_contiguous(), "value tensor has to be contiguous");
AT_ASSERTM(spatial_shapes.is_contiguous(), "spatial_shapes tensor has to be contiguous");
AT_ASSERTM(level_start_index.is_contiguous(), "level_start_index tensor has to be contiguous");
AT_ASSERTM(sampling_loc.is_contiguous(), "sampling_loc tensor has to be contiguous");
AT_ASSERTM(attn_weight.is_contiguous(), "attn_weight tensor has to be contiguous");
AT_ASSERTM(grad_output.is_contiguous(), "grad_output tensor has to be contiguous");
AT_ASSERTM(value.type().is_cuda(), "value must be a CUDA tensor");
AT_ASSERTM(spatial_shapes.type().is_cuda(), "spatial_shapes must be a CUDA tensor");
AT_ASSERTM(level_start_index.type().is_cuda(), "level_start_index must be a CUDA tensor");
AT_ASSERTM(sampling_loc.type().is_cuda(), "sampling_loc must be a CUDA tensor");
AT_ASSERTM(attn_weight.type().is_cuda(), "attn_weight must be a CUDA tensor");
AT_ASSERTM(grad_output.type().is_cuda(), "grad_output must be a CUDA tensor");
const int batch = value.size(0);
const int spatial_size = value.size(1);
const int num_heads = value.size(2);
const int channels = value.size(3);
const int num_levels = spatial_shapes.size(0);
const int num_query = sampling_loc.size(1);
const int num_point = sampling_loc.size(4);
const int im2col_step_ = std::min(batch, im2col_step);
AT_ASSERTM(batch % im2col_step_ == 0, "batch(%d) must divide im2col_step(%d)", batch, im2col_step_);
auto grad_value = at::zeros_like(value);
auto grad_sampling_loc = at::zeros_like(sampling_loc);
auto grad_attn_weight = at::zeros_like(attn_weight);
const int batch_n = im2col_step_;
auto per_value_size = spatial_size * num_heads * channels;
auto per_sample_loc_size = num_query * num_heads * num_levels * num_point * 2;
auto per_attn_weight_size = num_query * num_heads * num_levels * num_point;
auto grad_output_n = grad_output.view({batch/im2col_step_, batch_n, num_query, num_heads, channels});
for (int n = 0; n < batch/im2col_step_; ++n)
{
auto grad_output_g = grad_output_n.select(0, n);
AT_DISPATCH_FLOATING_TYPES(value.type(), "ms_deform_attn_backward_cuda", ([&] {
ms_deformable_col2im_cuda(at::cuda::getCurrentCUDAStream(),
grad_output_g.data<scalar_t>(),
value.data<scalar_t>() + n * im2col_step_ * per_value_size,
spatial_shapes.data<int64_t>(),
level_start_index.data<int64_t>(),
sampling_loc.data<scalar_t>() + n * im2col_step_ * per_sample_loc_size,
attn_weight.data<scalar_t>() + n * im2col_step_ * per_attn_weight_size,
batch_n, spatial_size, num_heads, channels, num_levels, num_query, num_point,
grad_value.data<scalar_t>() + n * im2col_step_ * per_value_size,
grad_sampling_loc.data<scalar_t>() + n * im2col_step_ * per_sample_loc_size,
grad_attn_weight.data<scalar_t>() + n * im2col_step_ * per_attn_weight_size);
}));
}
return {
grad_value, grad_sampling_loc, grad_attn_weight
};
}
const int CUDA_NUM_THREADS = 1024;
inline int GET_BLOCKS(const int N, const int num_threads)
{
return (N + num_threads - 1) / num_threads;
}
template <typename scalar_t>
__device__ scalar_t ms_deform_attn_im2col_bilinear(const scalar_t* &bottom_data,
const int &height, const int &width, const int &nheads, const int &channels,
const scalar_t &h, const scalar_t &w, const int &m, const int &c)
{
const int h_low = floor(h);
const int w_low = floor(w);
const int h_high = h_low + 1;
const int w_high = w_low + 1;
const scalar_t lh = h - h_low;
const scalar_t lw = w - w_low;
const scalar_t hh = 1 - lh, hw = 1 - lw;
const int w_stride = nheads * channels;
const int h_stride = width * w_stride;
const int h_low_ptr_offset = h_low * h_stride;
const int h_high_ptr_offset = h_low_ptr_offset + h_stride;
const int w_low_ptr_offset = w_low * w_stride;
const int w_high_ptr_offset = w_low_ptr_offset + w_stride;
const int base_ptr = m * channels + c;
scalar_t v1 = 0;
if (h_low >= 0 && w_low >= 0)
{
const int ptr1 = h_low_ptr_offset + w_low_ptr_offset + base_ptr;
v1 = bottom_data[ptr1];
}
scalar_t v2 = 0;
if (h_low >= 0 && w_high <= width - 1)
{
const int ptr2 = h_low_ptr_offset + w_high_ptr_offset + base_ptr;
v2 = bottom_data[ptr2];
}
scalar_t v3 = 0;
if (h_high <= height - 1 && w_low >= 0)
{
const int ptr3 = h_high_ptr_offset + w_low_ptr_offset + base_ptr;
v3 = bottom_data[ptr3];
}
scalar_t v4 = 0;
if (h_high <= height - 1 && w_high <= width - 1)
{
const int ptr4 = h_high_ptr_offset + w_high_ptr_offset + base_ptr;
v4 = bottom_data[ptr4];
}
const scalar_t w1 = hh * hw, w2 = hh * lw, w3 = lh * hw, w4 = lh * lw;
const scalar_t val = (w1 * v1 + w2 * v2 + w3 * v3 + w4 * v4);
return val;
}
template <typename scalar_t>
__device__ void ms_deform_attn_col2im_bilinear(const scalar_t* &bottom_data,
const int &height, const int &width, const int &nheads, const int &channels,
const scalar_t &h, const scalar_t &w, const int &m, const int &c,
const scalar_t &top_grad,
const scalar_t &attn_weight,
scalar_t* &grad_value,
scalar_t* grad_sampling_loc,
scalar_t* grad_attn_weight)
{
const int h_low = floor(h);
const int w_low = floor(w);
const int h_high = h_low + 1;
const int w_high = w_low + 1;
const scalar_t lh = h - h_low;
const scalar_t lw = w - w_low;
const scalar_t hh = 1 - lh, hw = 1 - lw;
const int w_stride = nheads * channels;
const int h_stride = width * w_stride;
const int h_low_ptr_offset = h_low * h_stride;
const int h_high_ptr_offset = h_low_ptr_offset + h_stride;
const int w_low_ptr_offset = w_low * w_stride;
const int w_high_ptr_offset = w_low_ptr_offset + w_stride;
const int base_ptr = m * channels + c;
const scalar_t w1 = hh * hw, w2 = hh * lw, w3 = lh * hw, w4 = lh * lw;
const scalar_t top_grad_value = top_grad * attn_weight;
scalar_t grad_h_weight = 0, grad_w_weight = 0;
scalar_t v1 = 0;
if (h_low >= 0 && w_low >= 0)
{
const int ptr1 = h_low_ptr_offset + w_low_ptr_offset + base_ptr;
v1 = bottom_data[ptr1];
grad_h_weight -= hw * v1;
grad_w_weight -= hh * v1;
atomicAdd(grad_value+ptr1, w1*top_grad_value);
}
scalar_t v2 = 0;
if (h_low >= 0 && w_high <= width - 1)
{
const int ptr2 = h_low_ptr_offset + w_high_ptr_offset + base_ptr;
v2 = bottom_data[ptr2];
grad_h_weight -= lw * v2;
grad_w_weight += hh * v2;
atomicAdd(grad_value+ptr2, w2*top_grad_value);
}
scalar_t v3 = 0;
if (h_high <= height - 1 && w_low >= 0)
{
const int ptr3 = h_high_ptr_offset + w_low_ptr_offset + base_ptr;
v3 = bottom_data[ptr3];
grad_h_weight += hw * v3;
grad_w_weight -= lh * v3;
atomicAdd(grad_value+ptr3, w3*top_grad_value);
}
scalar_t v4 = 0;
if (h_high <= height - 1 && w_high <= width - 1)
{
const int ptr4 = h_high_ptr_offset + w_high_ptr_offset + base_ptr;
v4 = bottom_data[ptr4];
grad_h_weight += lw * v4;
grad_w_weight += lh * v4;
atomicAdd(grad_value+ptr4, w4*top_grad_value);
}
const scalar_t val = (w1 * v1 + w2 * v2 + w3 * v3 + w4 * v4);
*grad_attn_weight = top_grad * val;
*grad_sampling_loc = width * grad_w_weight * top_grad_value;
*(grad_sampling_loc + 1) = height * grad_h_weight * top_grad_value;
}
template <typename scalar_t>
__device__ void ms_deform_attn_col2im_bilinear_gm(const scalar_t* &bottom_data,
const int &height, const int &width, const int &nheads, const int &channels,
const scalar_t &h, const scalar_t &w, const int &m, const int &c,
const scalar_t &top_grad,
const scalar_t &attn_weight,
scalar_t* &grad_value,
scalar_t* grad_sampling_loc,
scalar_t* grad_attn_weight)
{
const int h_low = floor(h);
const int w_low = floor(w);
const int h_high = h_low + 1;
const int w_high = w_low + 1;
const scalar_t lh = h - h_low;
const scalar_t lw = w - w_low;
const scalar_t hh = 1 - lh, hw = 1 - lw;
const int w_stride = nheads * channels;
const int h_stride = width * w_stride;
const int h_low_ptr_offset = h_low * h_stride;
const int h_high_ptr_offset = h_low_ptr_offset + h_stride;
const int w_low_ptr_offset = w_low * w_stride;
const int w_high_ptr_offset = w_low_ptr_offset + w_stride;
const int base_ptr = m * channels + c;
const scalar_t w1 = hh * hw, w2 = hh * lw, w3 = lh * hw, w4 = lh * lw;
const scalar_t top_grad_value = top_grad * attn_weight;
scalar_t grad_h_weight = 0, grad_w_weight = 0;
scalar_t v1 = 0;
if (h_low >= 0 && w_low >= 0)
{
const int ptr1 = h_low_ptr_offset + w_low_ptr_offset + base_ptr;
v1 = bottom_data[ptr1];
grad_h_weight -= hw * v1;
grad_w_weight -= hh * v1;
atomicAdd(grad_value+ptr1, w1*top_grad_value);
}
scalar_t v2 = 0;
if (h_low >= 0 && w_high <= width - 1)
{
const int ptr2 = h_low_ptr_offset + w_high_ptr_offset + base_ptr;
v2 = bottom_data[ptr2];
grad_h_weight -= lw * v2;
grad_w_weight += hh * v2;
atomicAdd(grad_value+ptr2, w2*top_grad_value);
}
scalar_t v3 = 0;
if (h_high <= height - 1 && w_low >= 0)
{
const int ptr3 = h_high_ptr_offset + w_low_ptr_offset + base_ptr;
v3 = bottom_data[ptr3];
grad_h_weight += hw * v3;
grad_w_weight -= lh * v3;
atomicAdd(grad_value+ptr3, w3*top_grad_value);
}
scalar_t v4 = 0;
if (h_high <= height - 1 && w_high <= width - 1)
{
const int ptr4 = h_high_ptr_offset + w_high_ptr_offset + base_ptr;
v4 = bottom_data[ptr4];
grad_h_weight += lw * v4;
grad_w_weight += lh * v4;
atomicAdd(grad_value+ptr4, w4*top_grad_value);
}
const scalar_t val = (w1 * v1 + w2 * v2 + w3 * v3 + w4 * v4);
atomicAdd(grad_attn_weight, top_grad * val);
atomicAdd(grad_sampling_loc, width * grad_w_weight * top_grad_value);
atomicAdd(grad_sampling_loc + 1, height * grad_h_weight * top_grad_value);
}
template <typename scalar_t>
__global__ void ms_deformable_im2col_gpu_kernel(const int n,
const scalar_t *data_value,
const int64_t *data_spatial_shapes,
const int64_t *data_level_start_index,
const scalar_t *data_sampling_loc,
const scalar_t *data_attn_weight,
const int batch_size,
const int spatial_size,
const int num_heads,
const int channels,
const int num_levels,
const int num_query,
const int num_point,
scalar_t *data_col)
{
CUDA_KERNEL_LOOP(index, n)
{
int _temp = index;
const int c_col = _temp % channels;
_temp /= channels;
const int sampling_index = _temp;
const int m_col = _temp % num_heads;
_temp /= num_heads;
const int q_col = _temp % num_query;
_temp /= num_query;
const int b_col = _temp;
scalar_t *data_col_ptr = data_col + index;
int data_weight_ptr = sampling_index * num_levels * num_point;
int data_loc_w_ptr = data_weight_ptr << 1;
const int qid_stride = num_heads * channels;
const int data_value_ptr_init_offset = b_col * spatial_size * qid_stride;
scalar_t col = 0;
for (int l_col=0; l_col < num_levels; ++l_col)
{
const int level_start_id = data_level_start_index[l_col];
const int spatial_h_ptr = l_col << 1;
const int spatial_h = data_spatial_shapes[spatial_h_ptr];
const int spatial_w = data_spatial_shapes[spatial_h_ptr + 1];
const scalar_t *data_value_ptr = data_value + (data_value_ptr_init_offset + level_start_id * qid_stride);
for (int p_col=0; p_col < num_point; ++p_col)
{
const scalar_t loc_w = data_sampling_loc[data_loc_w_ptr];
const scalar_t loc_h = data_sampling_loc[data_loc_w_ptr + 1];
const scalar_t weight = data_attn_weight[data_weight_ptr];
const scalar_t h_im = loc_h * spatial_h - 0.5;
const scalar_t w_im = loc_w * spatial_w - 0.5;
if (h_im > -1 && w_im > -1 && h_im < spatial_h && w_im < spatial_w)
{
col += ms_deform_attn_im2col_bilinear(data_value_ptr, spatial_h, spatial_w, num_heads, channels, h_im, w_im, m_col, c_col) * weight;
}
data_weight_ptr += 1;
data_loc_w_ptr += 2;
}
}
*data_col_ptr = col;
}
}
template <typename scalar_t, unsigned int blockSize>
__global__ void ms_deformable_col2im_gpu_kernel_shm_blocksize_aware_reduce_v1(const int n,
const scalar_t *grad_col,
const scalar_t *data_value,
const int64_t *data_spatial_shapes,
const int64_t *data_level_start_index,
const scalar_t *data_sampling_loc,
const scalar_t *data_attn_weight,
const int batch_size,
const int spatial_size,
const int num_heads,
const int channels,
const int num_levels,
const int num_query,
const int num_point,
scalar_t *grad_value,
scalar_t *grad_sampling_loc,
scalar_t *grad_attn_weight)
{
CUDA_KERNEL_LOOP(index, n)
{
__shared__ scalar_t cache_grad_sampling_loc[blockSize * 2];
__shared__ scalar_t cache_grad_attn_weight[blockSize];
unsigned int tid = threadIdx.x;
int _temp = index;
const int c_col = _temp % channels;
_temp /= channels;
const int sampling_index = _temp;
const int m_col = _temp % num_heads;
_temp /= num_heads;
const int q_col = _temp % num_query;
_temp /= num_query;
const int b_col = _temp;
const scalar_t top_grad = grad_col[index];
int data_weight_ptr = sampling_index * num_levels * num_point;
int data_loc_w_ptr = data_weight_ptr << 1;
const int grad_sampling_ptr = data_weight_ptr;
grad_sampling_loc += grad_sampling_ptr << 1;
grad_attn_weight += grad_sampling_ptr;
const int grad_weight_stride = 1;
const int grad_loc_stride = 2;
const int qid_stride = num_heads * channels;
const int data_value_ptr_init_offset = b_col * spatial_size * qid_stride;
for (int l_col=0; l_col < num_levels; ++l_col)
{
const int level_start_id = data_level_start_index[l_col];
const int spatial_h_ptr = l_col << 1;
const int spatial_h = data_spatial_shapes[spatial_h_ptr];
const int spatial_w = data_spatial_shapes[spatial_h_ptr + 1];
const int value_ptr_offset = data_value_ptr_init_offset + level_start_id * qid_stride;
const scalar_t *data_value_ptr = data_value + value_ptr_offset;
scalar_t *grad_value_ptr = grad_value + value_ptr_offset;
for (int p_col=0; p_col < num_point; ++p_col)
{
const scalar_t loc_w = data_sampling_loc[data_loc_w_ptr];
const scalar_t loc_h = data_sampling_loc[data_loc_w_ptr + 1];
const scalar_t weight = data_attn_weight[data_weight_ptr];
const scalar_t h_im = loc_h * spatial_h - 0.5;
const scalar_t w_im = loc_w * spatial_w - 0.5;
*(cache_grad_sampling_loc+(threadIdx.x << 1)) = 0;
*(cache_grad_sampling_loc+((threadIdx.x << 1) + 1)) = 0;
*(cache_grad_attn_weight+threadIdx.x)=0;
if (h_im > -1 && w_im > -1 && h_im < spatial_h && w_im < spatial_w)
{
ms_deform_attn_col2im_bilinear(
data_value_ptr, spatial_h, spatial_w, num_heads, channels, h_im, w_im, m_col, c_col,
top_grad, weight, grad_value_ptr,
cache_grad_sampling_loc+(threadIdx.x << 1), cache_grad_attn_weight+threadIdx.x);
}
__syncthreads();
if (tid == 0)
{
scalar_t _grad_w=cache_grad_sampling_loc[0], _grad_h=cache_grad_sampling_loc[1], _grad_a=cache_grad_attn_weight[0];
int sid=2;
for (unsigned int tid = 1; tid < blockSize; ++tid)
{
_grad_w += cache_grad_sampling_loc[sid];
_grad_h += cache_grad_sampling_loc[sid + 1];
_grad_a += cache_grad_attn_weight[tid];
sid += 2;
}
*grad_sampling_loc = _grad_w;
*(grad_sampling_loc + 1) = _grad_h;
*grad_attn_weight = _grad_a;
}
__syncthreads();
data_weight_ptr += 1;
data_loc_w_ptr += 2;
grad_attn_weight += grad_weight_stride;
grad_sampling_loc += grad_loc_stride;
}
}
}
}
template <typename scalar_t, unsigned int blockSize>
__global__ void ms_deformable_col2im_gpu_kernel_shm_blocksize_aware_reduce_v2(const int n,
const scalar_t *grad_col,
const scalar_t *data_value,
const int64_t *data_spatial_shapes,
const int64_t *data_level_start_index,
const scalar_t *data_sampling_loc,
const scalar_t *data_attn_weight,
const int batch_size,
const int spatial_size,
const int num_heads,
const int channels,
const int num_levels,
const int num_query,
const int num_point,
scalar_t *grad_value,
scalar_t *grad_sampling_loc,
scalar_t *grad_attn_weight)
{
CUDA_KERNEL_LOOP(index, n)
{
__shared__ scalar_t cache_grad_sampling_loc[blockSize * 2];
__shared__ scalar_t cache_grad_attn_weight[blockSize];
unsigned int tid = threadIdx.x;
int _temp = index;
const int c_col = _temp % channels;
_temp /= channels;
const int sampling_index = _temp;
const int m_col = _temp % num_heads;
_temp /= num_heads;
const int q_col = _temp % num_query;
_temp /= num_query;
const int b_col = _temp;
const scalar_t top_grad = grad_col[index];
int data_weight_ptr = sampling_index * num_levels * num_point;
int data_loc_w_ptr = data_weight_ptr << 1;
const int grad_sampling_ptr = data_weight_ptr;
grad_sampling_loc += grad_sampling_ptr << 1;
grad_attn_weight += grad_sampling_ptr;
const int grad_weight_stride = 1;
const int grad_loc_stride = 2;
const int qid_stride = num_heads * channels;
const int data_value_ptr_init_offset = b_col * spatial_size * qid_stride;
for (int l_col=0; l_col < num_levels; ++l_col)
{
const int level_start_id = data_level_start_index[l_col];
const int spatial_h_ptr = l_col << 1;
const int spatial_h = data_spatial_shapes[spatial_h_ptr];
const int spatial_w = data_spatial_shapes[spatial_h_ptr + 1];
const int value_ptr_offset = data_value_ptr_init_offset + level_start_id * qid_stride;
const scalar_t *data_value_ptr = data_value + value_ptr_offset;
scalar_t *grad_value_ptr = grad_value + value_ptr_offset;
for (int p_col=0; p_col < num_point; ++p_col)
{
const scalar_t loc_w = data_sampling_loc[data_loc_w_ptr];
const scalar_t loc_h = data_sampling_loc[data_loc_w_ptr + 1];
const scalar_t weight = data_attn_weight[data_weight_ptr];
const scalar_t h_im = loc_h * spatial_h - 0.5;
const scalar_t w_im = loc_w * spatial_w - 0.5;
*(cache_grad_sampling_loc+(threadIdx.x << 1)) = 0;
*(cache_grad_sampling_loc+((threadIdx.x << 1) + 1)) = 0;
*(cache_grad_attn_weight+threadIdx.x)=0;
if (h_im > -1 && w_im > -1 && h_im < spatial_h && w_im < spatial_w)
{
ms_deform_attn_col2im_bilinear(
data_value_ptr, spatial_h, spatial_w, num_heads, channels, h_im, w_im, m_col, c_col,
top_grad, weight, grad_value_ptr,
cache_grad_sampling_loc+(threadIdx.x << 1), cache_grad_attn_weight+threadIdx.x);
}
__syncthreads();
for (unsigned int s=blockSize/2; s>0; s>>=1)
{
if (tid < s) {
const unsigned int xid1 = tid << 1;
const unsigned int xid2 = (tid + s) << 1;
cache_grad_attn_weight[tid] += cache_grad_attn_weight[tid + s];
cache_grad_sampling_loc[xid1] += cache_grad_sampling_loc[xid2];
cache_grad_sampling_loc[xid1 + 1] += cache_grad_sampling_loc[xid2 + 1];
}
__syncthreads();
}
if (tid == 0)
{
*grad_sampling_loc = cache_grad_sampling_loc[0];
*(grad_sampling_loc + 1) = cache_grad_sampling_loc[1];
*grad_attn_weight = cache_grad_attn_weight[0];
}
__syncthreads();
data_weight_ptr += 1;
data_loc_w_ptr += 2;
grad_attn_weight += grad_weight_stride;
grad_sampling_loc += grad_loc_stride;
}
}
}
}
template <typename scalar_t>
__global__ void ms_deformable_col2im_gpu_kernel_shm_reduce_v1(const int n,
const scalar_t *grad_col,
const scalar_t *data_value,
const int64_t *data_spatial_shapes,
const int64_t *data_level_start_index,
const scalar_t *data_sampling_loc,
const scalar_t *data_attn_weight,
const int batch_size,
const int spatial_size,
const int num_heads,
const int channels,
const int num_levels,
const int num_query,
const int num_point,
scalar_t *grad_value,
scalar_t *grad_sampling_loc,
scalar_t *grad_attn_weight)
{
CUDA_KERNEL_LOOP(index, n)
{
extern __shared__ int _s[];
scalar_t* cache_grad_sampling_loc = (scalar_t*)_s;
scalar_t* cache_grad_attn_weight = cache_grad_sampling_loc + 2 * blockDim.x;
unsigned int tid = threadIdx.x;
int _temp = index;
const int c_col = _temp % channels;
_temp /= channels;
const int sampling_index = _temp;
const int m_col = _temp % num_heads;
_temp /= num_heads;
const int q_col = _temp % num_query;
_temp /= num_query;
const int b_col = _temp;
const scalar_t top_grad = grad_col[index];
int data_weight_ptr = sampling_index * num_levels * num_point;
int data_loc_w_ptr = data_weight_ptr << 1;
const int grad_sampling_ptr = data_weight_ptr;
grad_sampling_loc += grad_sampling_ptr << 1;
grad_attn_weight += grad_sampling_ptr;
const int grad_weight_stride = 1;
const int grad_loc_stride = 2;
const int qid_stride = num_heads * channels;
const int data_value_ptr_init_offset = b_col * spatial_size * qid_stride;
for (int l_col=0; l_col < num_levels; ++l_col)
{
const int level_start_id = data_level_start_index[l_col];
const int spatial_h_ptr = l_col << 1;
const int spatial_h = data_spatial_shapes[spatial_h_ptr];
const int spatial_w = data_spatial_shapes[spatial_h_ptr + 1];
const int value_ptr_offset = data_value_ptr_init_offset + level_start_id * qid_stride;
const scalar_t *data_value_ptr = data_value + value_ptr_offset;
scalar_t *grad_value_ptr = grad_value + value_ptr_offset;
for (int p_col=0; p_col < num_point; ++p_col)
{
const scalar_t loc_w = data_sampling_loc[data_loc_w_ptr];
const scalar_t loc_h = data_sampling_loc[data_loc_w_ptr + 1];
const scalar_t weight = data_attn_weight[data_weight_ptr];
const scalar_t h_im = loc_h * spatial_h - 0.5;
const scalar_t w_im = loc_w * spatial_w - 0.5;
*(cache_grad_sampling_loc+(threadIdx.x << 1)) = 0;
*(cache_grad_sampling_loc+((threadIdx.x << 1) + 1)) = 0;
*(cache_grad_attn_weight+threadIdx.x)=0;
if (h_im > -1 && w_im > -1 && h_im < spatial_h && w_im < spatial_w)
{
ms_deform_attn_col2im_bilinear(
data_value_ptr, spatial_h, spatial_w, num_heads, channels, h_im, w_im, m_col, c_col,
top_grad, weight, grad_value_ptr,
cache_grad_sampling_loc+(threadIdx.x << 1), cache_grad_attn_weight+threadIdx.x);
}
__syncthreads();
if (tid == 0)
{
scalar_t _grad_w=cache_grad_sampling_loc[0], _grad_h=cache_grad_sampling_loc[1], _grad_a=cache_grad_attn_weight[0];
int sid=2;
for (unsigned int tid = 1; tid < blockDim.x; ++tid)
{
_grad_w += cache_grad_sampling_loc[sid];
_grad_h += cache_grad_sampling_loc[sid + 1];
_grad_a += cache_grad_attn_weight[tid];
sid += 2;
}
*grad_sampling_loc = _grad_w;
*(grad_sampling_loc + 1) = _grad_h;
*grad_attn_weight = _grad_a;
}
__syncthreads();
data_weight_ptr += 1;
data_loc_w_ptr += 2;
grad_attn_weight += grad_weight_stride;
grad_sampling_loc += grad_loc_stride;
}
}
}
}
template <typename scalar_t>
__global__ void ms_deformable_col2im_gpu_kernel_shm_reduce_v2(const int n,
const scalar_t *grad_col,
const scalar_t *data_value,
const int64_t *data_spatial_shapes,
const int64_t *data_level_start_index,
const scalar_t *data_sampling_loc,
const scalar_t *data_attn_weight,
const int batch_size,
const int spatial_size,
const int num_heads,
const int channels,
const int num_levels,
const int num_query,
const int num_point,
scalar_t *grad_value,
scalar_t *grad_sampling_loc,
scalar_t *grad_attn_weight)
{
CUDA_KERNEL_LOOP(index, n)
{
extern __shared__ int _s[];
scalar_t* cache_grad_sampling_loc = (scalar_t*)_s;
scalar_t* cache_grad_attn_weight = cache_grad_sampling_loc + 2 * blockDim.x;
unsigned int tid = threadIdx.x;
int _temp = index;
const int c_col = _temp % channels;
_temp /= channels;
const int sampling_index = _temp;
const int m_col = _temp % num_heads;
_temp /= num_heads;
const int q_col = _temp % num_query;
_temp /= num_query;
const int b_col = _temp;
const scalar_t top_grad = grad_col[index];
int data_weight_ptr = sampling_index * num_levels * num_point;
int data_loc_w_ptr = data_weight_ptr << 1;
const int grad_sampling_ptr = data_weight_ptr;
grad_sampling_loc += grad_sampling_ptr << 1;
grad_attn_weight += grad_sampling_ptr;
const int grad_weight_stride = 1;
const int grad_loc_stride = 2;
const int qid_stride = num_heads * channels;
const int data_value_ptr_init_offset = b_col * spatial_size * qid_stride;
for (int l_col=0; l_col < num_levels; ++l_col)
{
const int level_start_id = data_level_start_index[l_col];
const int spatial_h_ptr = l_col << 1;
const int spatial_h = data_spatial_shapes[spatial_h_ptr];
const int spatial_w = data_spatial_shapes[spatial_h_ptr + 1];
const int value_ptr_offset = data_value_ptr_init_offset + level_start_id * qid_stride;
const scalar_t *data_value_ptr = data_value + value_ptr_offset;
scalar_t *grad_value_ptr = grad_value + value_ptr_offset;
for (int p_col=0; p_col < num_point; ++p_col)
{
const scalar_t loc_w = data_sampling_loc[data_loc_w_ptr];
const scalar_t loc_h = data_sampling_loc[data_loc_w_ptr + 1];
const scalar_t weight = data_attn_weight[data_weight_ptr];
const scalar_t h_im = loc_h * spatial_h - 0.5;
const scalar_t w_im = loc_w * spatial_w - 0.5;
*(cache_grad_sampling_loc+(threadIdx.x << 1)) = 0;
*(cache_grad_sampling_loc+((threadIdx.x << 1) + 1)) = 0;
*(cache_grad_attn_weight+threadIdx.x)=0;
if (h_im > -1 && w_im > -1 && h_im < spatial_h && w_im < spatial_w)
{
ms_deform_attn_col2im_bilinear(
data_value_ptr, spatial_h, spatial_w, num_heads, channels, h_im, w_im, m_col, c_col,
top_grad, weight, grad_value_ptr,
cache_grad_sampling_loc+(threadIdx.x << 1), cache_grad_attn_weight+threadIdx.x);
}
__syncthreads();
for (unsigned int s=blockDim.x/2, spre=blockDim.x; s>0; s>>=1, spre>>=1)
{
if (tid < s) {
const unsigned int xid1 = tid << 1;
const unsigned int xid2 = (tid + s) << 1;
cache_grad_attn_weight[tid] += cache_grad_attn_weight[tid + s];
cache_grad_sampling_loc[xid1] += cache_grad_sampling_loc[xid2];
cache_grad_sampling_loc[xid1 + 1] += cache_grad_sampling_loc[xid2 + 1];
if (tid + (s << 1) < spre)
{
cache_grad_attn_weight[tid] += cache_grad_attn_weight[tid + (s << 1)];
cache_grad_sampling_loc[xid1] += cache_grad_sampling_loc[xid2 + (s << 1)];
cache_grad_sampling_loc[xid1 + 1] += cache_grad_sampling_loc[xid2 + 1 + (s << 1)];
}
}
__syncthreads();
}
if (tid == 0)
{
*grad_sampling_loc = cache_grad_sampling_loc[0];
*(grad_sampling_loc + 1) = cache_grad_sampling_loc[1];
*grad_attn_weight = cache_grad_attn_weight[0];
}
__syncthreads();
data_weight_ptr += 1;
data_loc_w_ptr += 2;
grad_attn_weight += grad_weight_stride;
grad_sampling_loc += grad_loc_stride;
}
}
}
}
template <typename scalar_t>
__global__ void ms_deformable_col2im_gpu_kernel_shm_reduce_v2_multi_blocks(const int n,
const scalar_t *grad_col,
const scalar_t *data_value,
const int64_t *data_spatial_shapes,
const int64_t *data_level_start_index,
const scalar_t *data_sampling_loc,
const scalar_t *data_attn_weight,
const int batch_size,
const int spatial_size,
const int num_heads,
const int channels,
const int num_levels,
const int num_query,
const int num_point,
scalar_t *grad_value,
scalar_t *grad_sampling_loc,
scalar_t *grad_attn_weight)
{
CUDA_KERNEL_LOOP(index, n)
{
extern __shared__ int _s[];
scalar_t* cache_grad_sampling_loc = (scalar_t*)_s;
scalar_t* cache_grad_attn_weight = cache_grad_sampling_loc + 2 * blockDim.x;
unsigned int tid = threadIdx.x;
int _temp = index;
const int c_col = _temp % channels;
_temp /= channels;
const int sampling_index = _temp;
const int m_col = _temp % num_heads;
_temp /= num_heads;
const int q_col = _temp % num_query;
_temp /= num_query;
const int b_col = _temp;
const scalar_t top_grad = grad_col[index];
int data_weight_ptr = sampling_index * num_levels * num_point;
int data_loc_w_ptr = data_weight_ptr << 1;
const int grad_sampling_ptr = data_weight_ptr;
grad_sampling_loc += grad_sampling_ptr << 1;
grad_attn_weight += grad_sampling_ptr;
const int grad_weight_stride = 1;
const int grad_loc_stride = 2;
const int qid_stride = num_heads * channels;
const int data_value_ptr_init_offset = b_col * spatial_size * qid_stride;
for (int l_col=0; l_col < num_levels; ++l_col)
{
const int level_start_id = data_level_start_index[l_col];
const int spatial_h_ptr = l_col << 1;
const int spatial_h = data_spatial_shapes[spatial_h_ptr];
const int spatial_w = data_spatial_shapes[spatial_h_ptr + 1];
const int value_ptr_offset = data_value_ptr_init_offset + level_start_id * qid_stride;
const scalar_t *data_value_ptr = data_value + value_ptr_offset;
scalar_t *grad_value_ptr = grad_value + value_ptr_offset;
for (int p_col=0; p_col < num_point; ++p_col)
{
const scalar_t loc_w = data_sampling_loc[data_loc_w_ptr];
const scalar_t loc_h = data_sampling_loc[data_loc_w_ptr + 1];
const scalar_t weight = data_attn_weight[data_weight_ptr];
const scalar_t h_im = loc_h * spatial_h - 0.5;
const scalar_t w_im = loc_w * spatial_w - 0.5;
*(cache_grad_sampling_loc+(threadIdx.x << 1)) = 0;
*(cache_grad_sampling_loc+((threadIdx.x << 1) + 1)) = 0;
*(cache_grad_attn_weight+threadIdx.x)=0;
if (h_im > -1 && w_im > -1 && h_im < spatial_h && w_im < spatial_w)
{
ms_deform_attn_col2im_bilinear(
data_value_ptr, spatial_h, spatial_w, num_heads, channels, h_im, w_im, m_col, c_col,
top_grad, weight, grad_value_ptr,
cache_grad_sampling_loc+(threadIdx.x << 1), cache_grad_attn_weight+threadIdx.x);
}
__syncthreads();
for (unsigned int s=blockDim.x/2, spre=blockDim.x; s>0; s>>=1, spre>>=1)
{
if (tid < s) {
const unsigned int xid1 = tid << 1;
const unsigned int xid2 = (tid + s) << 1;
cache_grad_attn_weight[tid] += cache_grad_attn_weight[tid + s];
cache_grad_sampling_loc[xid1] += cache_grad_sampling_loc[xid2];
cache_grad_sampling_loc[xid1 + 1] += cache_grad_sampling_loc[xid2 + 1];
if (tid + (s << 1) < spre)
{
cache_grad_attn_weight[tid] += cache_grad_attn_weight[tid + (s << 1)];
cache_grad_sampling_loc[xid1] += cache_grad_sampling_loc[xid2 + (s << 1)];
cache_grad_sampling_loc[xid1 + 1] += cache_grad_sampling_loc[xid2 + 1 + (s << 1)];
}
}
__syncthreads();
}
if (tid == 0)
{
atomicAdd(grad_sampling_loc, cache_grad_sampling_loc[0]);
atomicAdd(grad_sampling_loc + 1, cache_grad_sampling_loc[1]);
atomicAdd(grad_attn_weight, cache_grad_attn_weight[0]);
}
__syncthreads();
data_weight_ptr += 1;
data_loc_w_ptr += 2;
grad_attn_weight += grad_weight_stride;
grad_sampling_loc += grad_loc_stride;
}
}
}
}
template <typename scalar_t>
__global__ void ms_deformable_col2im_gpu_kernel_gm(const int n,
const scalar_t *grad_col,
const scalar_t *data_value,
const int64_t *data_spatial_shapes,
const int64_t *data_level_start_index,
const scalar_t *data_sampling_loc,
const scalar_t *data_attn_weight,
const int batch_size,
const int spatial_size,
const int num_heads,
const int channels,
const int num_levels,
const int num_query,
const int num_point,
scalar_t *grad_value,
scalar_t *grad_sampling_loc,
scalar_t *grad_attn_weight)
{
CUDA_KERNEL_LOOP(index, n)
{
int _temp = index;
const int c_col = _temp % channels;
_temp /= channels;
const int sampling_index = _temp;
const int m_col = _temp % num_heads;
_temp /= num_heads;
const int q_col = _temp % num_query;
_temp /= num_query;
const int b_col = _temp;
const scalar_t top_grad = grad_col[index];
int data_weight_ptr = sampling_index * num_levels * num_point;
int data_loc_w_ptr = data_weight_ptr << 1;
const int grad_sampling_ptr = data_weight_ptr;
grad_sampling_loc += grad_sampling_ptr << 1;
grad_attn_weight += grad_sampling_ptr;
const int grad_weight_stride = 1;
const int grad_loc_stride = 2;
const int qid_stride = num_heads * channels;
const int data_value_ptr_init_offset = b_col * spatial_size * qid_stride;
for (int l_col=0; l_col < num_levels; ++l_col)
{
const int level_start_id = data_level_start_index[l_col];
const int spatial_h_ptr = l_col << 1;
const int spatial_h = data_spatial_shapes[spatial_h_ptr];
const int spatial_w = data_spatial_shapes[spatial_h_ptr + 1];
const int value_ptr_offset = data_value_ptr_init_offset + level_start_id * qid_stride;
const scalar_t *data_value_ptr = data_value + value_ptr_offset;
scalar_t *grad_value_ptr = grad_value + value_ptr_offset;
for (int p_col=0; p_col < num_point; ++p_col)
{
const scalar_t loc_w = data_sampling_loc[data_loc_w_ptr];
const scalar_t loc_h = data_sampling_loc[data_loc_w_ptr + 1];
const scalar_t weight = data_attn_weight[data_weight_ptr];
const scalar_t h_im = loc_h * spatial_h - 0.5;
const scalar_t w_im = loc_w * spatial_w - 0.5;
if (h_im > -1 && w_im > -1 && h_im < spatial_h && w_im < spatial_w)
{
ms_deform_attn_col2im_bilinear_gm(
data_value_ptr, spatial_h, spatial_w, num_heads, channels, h_im, w_im, m_col, c_col,
top_grad, weight, grad_value_ptr,
grad_sampling_loc, grad_attn_weight);
}
data_weight_ptr += 1;
data_loc_w_ptr += 2;
grad_attn_weight += grad_weight_stride;
grad_sampling_loc += grad_loc_stride;
}
}
}
}
template <typename scalar_t>
void ms_deformable_im2col_cuda(cudaStream_t stream,
const scalar_t* data_value,
const int64_t* data_spatial_shapes,
const int64_t* data_level_start_index,
const scalar_t* data_sampling_loc,
const scalar_t* data_attn_weight,
const int batch_size,
const int spatial_size,
const int num_heads,
const int channels,
const int num_levels,
const int num_query,
const int num_point,
scalar_t* data_col)
{
const int num_kernels = batch_size * num_query * num_heads * channels;
const int num_actual_kernels = batch_size * num_query * num_heads * channels;
const int num_threads = CUDA_NUM_THREADS;
ms_deformable_im2col_gpu_kernel<scalar_t>
<<<GET_BLOCKS(num_actual_kernels, num_threads), num_threads,
0, stream>>>(
num_kernels, data_value, data_spatial_shapes, data_level_start_index, data_sampling_loc, data_attn_weight,
batch_size, spatial_size, num_heads, channels, num_levels, num_query, num_point, data_col);
cudaError_t err = cudaGetLastError();
if (err != cudaSuccess)
{
printf("error in ms_deformable_im2col_cuda: %s\n", cudaGetErrorString(err));
}
}
template <typename scalar_t>
void ms_deformable_col2im_cuda(cudaStream_t stream,
const scalar_t* grad_col,
const scalar_t* data_value,
const int64_t * data_spatial_shapes,
const int64_t * data_level_start_index,
const scalar_t * data_sampling_loc,
const scalar_t * data_attn_weight,
const int batch_size,
const int spatial_size,
const int num_heads,
const int channels,
const int num_levels,
const int num_query,
const int num_point,
scalar_t* grad_value,
scalar_t* grad_sampling_loc,
scalar_t* grad_attn_weight)
{
const int num_threads = (channels > CUDA_NUM_THREADS)?CUDA_NUM_THREADS:channels;
const int num_kernels = batch_size * num_query * num_heads * channels;
const int num_actual_kernels = batch_size * num_query * num_heads * channels;
if (channels > 1024)
{
if ((channels & 1023) == 0)
{
ms_deformable_col2im_gpu_kernel_shm_reduce_v2_multi_blocks<scalar_t>
<<<GET_BLOCKS(num_actual_kernels, num_threads), num_threads,
num_threads*3*sizeof(scalar_t), stream>>>(
num_kernels,
grad_col,
data_value,
data_spatial_shapes,
data_level_start_index,
data_sampling_loc,
data_attn_weight,
batch_size,
spatial_size,
num_heads,
channels,
num_levels,
num_query,
num_point,
grad_value,
grad_sampling_loc,
grad_attn_weight);
}
else
{
ms_deformable_col2im_gpu_kernel_gm<scalar_t>
<<<GET_BLOCKS(num_actual_kernels, num_threads), num_threads,
0, stream>>>(
num_kernels,
grad_col,
data_value,
data_spatial_shapes,
data_level_start_index,
data_sampling_loc,
data_attn_weight,
batch_size,
spatial_size,
num_heads,
channels,
num_levels,
num_query,
num_point,
grad_value,
grad_sampling_loc,
grad_attn_weight);
}
}
else{
switch(channels)
{
case 1:
ms_deformable_col2im_gpu_kernel_shm_blocksize_aware_reduce_v1<scalar_t, 1>
<<<GET_BLOCKS(num_actual_kernels, num_threads), num_threads,
0, stream>>>(
num_kernels,
grad_col,
data_value,
data_spatial_shapes,
data_level_start_index,
data_sampling_loc,
data_attn_weight,
batch_size,
spatial_size,
num_heads,
channels,
num_levels,
num_query,
num_point,
grad_value,
grad_sampling_loc,
grad_attn_weight);
break;
case 2:
ms_deformable_col2im_gpu_kernel_shm_blocksize_aware_reduce_v1<scalar_t, 2>
<<<GET_BLOCKS(num_actual_kernels, num_threads), num_threads,
0, stream>>>(
num_kernels,
grad_col,
data_value,
data_spatial_shapes,
data_level_start_index,
data_sampling_loc,
data_attn_weight,
batch_size,
spatial_size,
num_heads,
channels,
num_levels,
num_query,
num_point,
grad_value,
grad_sampling_loc,
grad_attn_weight);
break;
case 4:
ms_deformable_col2im_gpu_kernel_shm_blocksize_aware_reduce_v1<scalar_t, 4>
<<<GET_BLOCKS(num_actual_kernels, num_threads), num_threads,
0, stream>>>(
num_kernels,
grad_col,
data_value,
data_spatial_shapes,
data_level_start_index,
data_sampling_loc,
data_attn_weight,
batch_size,
spatial_size,
num_heads,
channels,
num_levels,
num_query,
num_point,
grad_value,
grad_sampling_loc,
grad_attn_weight);
break;
case 8:
ms_deformable_col2im_gpu_kernel_shm_blocksize_aware_reduce_v1<scalar_t, 8>
<<<GET_BLOCKS(num_actual_kernels, num_threads), num_threads,
0, stream>>>(
num_kernels,
grad_col,
data_value,
data_spatial_shapes,
data_level_start_index,
data_sampling_loc,
data_attn_weight,
batch_size,
spatial_size,
num_heads,
channels,
num_levels,
num_query,
num_point,
grad_value,
grad_sampling_loc,
grad_attn_weight);
break;
case 16:
ms_deformable_col2im_gpu_kernel_shm_blocksize_aware_reduce_v1<scalar_t, 16>
<<<GET_BLOCKS(num_actual_kernels, num_threads), num_threads,
0, stream>>>(
num_kernels,
grad_col,
data_value,
data_spatial_shapes,
data_level_start_index,
data_sampling_loc,
data_attn_weight,
batch_size,
spatial_size,
num_heads,
channels,
num_levels,
num_query,
num_point,
grad_value,
grad_sampling_loc,
grad_attn_weight);
break;
case 32:
ms_deformable_col2im_gpu_kernel_shm_blocksize_aware_reduce_v1<scalar_t, 32>
<<<GET_BLOCKS(num_actual_kernels, num_threads), num_threads,
0, stream>>>(
num_kernels,
grad_col,
data_value,
data_spatial_shapes,
data_level_start_index,
data_sampling_loc,
data_attn_weight,
batch_size,
spatial_size,
num_heads,
channels,
num_levels,
num_query,
num_point,
grad_value,
grad_sampling_loc,
grad_attn_weight);
break;
case 64:
ms_deformable_col2im_gpu_kernel_shm_blocksize_aware_reduce_v2<scalar_t, 64>
<<<GET_BLOCKS(num_actual_kernels, num_threads), num_threads,
0, stream>>>(
num_kernels,
grad_col,
data_value,
data_spatial_shapes,
data_level_start_index,
data_sampling_loc,
data_attn_weight,
batch_size,
spatial_size,
num_heads,
channels,
num_levels,
num_query,
num_point,
grad_value,
grad_sampling_loc,
grad_attn_weight);
break;
case 128:
ms_deformable_col2im_gpu_kernel_shm_blocksize_aware_reduce_v2<scalar_t, 128>
<<<GET_BLOCKS(num_actual_kernels, num_threads), num_threads,
0, stream>>>(
num_kernels,
grad_col,
data_value,
data_spatial_shapes,
data_level_start_index,
data_sampling_loc,
data_attn_weight,
batch_size,
spatial_size,
num_heads,
channels,
num_levels,
num_query,
num_point,
grad_value,
grad_sampling_loc,
grad_attn_weight);
break;
case 256:
ms_deformable_col2im_gpu_kernel_shm_blocksize_aware_reduce_v2<scalar_t, 256>
<<<GET_BLOCKS(num_actual_kernels, num_threads), num_threads,
0, stream>>>(
num_kernels,
grad_col,
data_value,
data_spatial_shapes,
data_level_start_index,
data_sampling_loc,
data_attn_weight,
batch_size,
spatial_size,
num_heads,
channels,
num_levels,
num_query,
num_point,
grad_value,
grad_sampling_loc,
grad_attn_weight);
break;
case 512:
ms_deformable_col2im_gpu_kernel_shm_blocksize_aware_reduce_v2<scalar_t, 512>
<<<GET_BLOCKS(num_actual_kernels, num_threads), num_threads,
0, stream>>>(
num_kernels,
grad_col,
data_value,
data_spatial_shapes,
data_level_start_index,
data_sampling_loc,
data_attn_weight,
batch_size,
spatial_size,
num_heads,
channels,
num_levels,
num_query,
num_point,
grad_value,
grad_sampling_loc,
grad_attn_weight);
break;
case 1024:
ms_deformable_col2im_gpu_kernel_shm_blocksize_aware_reduce_v2<scalar_t, 1024>
<<<GET_BLOCKS(num_actual_kernels, num_threads), num_threads,
0, stream>>>(
num_kernels,
grad_col,
data_value,
data_spatial_shapes,
data_level_start_index,
data_sampling_loc,
data_attn_weight,
batch_size,
spatial_size,
num_heads,
channels,
num_levels,
num_query,
num_point,
grad_value,
grad_sampling_loc,
grad_attn_weight);
break;
default:
if (channels < 64)
{
ms_deformable_col2im_gpu_kernel_shm_reduce_v1<scalar_t>
<<<GET_BLOCKS(num_actual_kernels, num_threads), num_threads,
num_threads*3*sizeof(scalar_t), stream>>>(
num_kernels,
grad_col,
data_value,
data_spatial_shapes,
data_level_start_index,
data_sampling_loc,
data_attn_weight,
batch_size,
spatial_size,
num_heads,
channels,
num_levels,
num_query,
num_point,
grad_value,
grad_sampling_loc,
grad_attn_weight);
}
else
{
ms_deformable_col2im_gpu_kernel_shm_reduce_v2<scalar_t>
<<<GET_BLOCKS(num_actual_kernels, num_threads), num_threads,
num_threads*3*sizeof(scalar_t), stream>>>(
num_kernels,
grad_col,
data_value,
data_spatial_shapes,
data_level_start_index,
data_sampling_loc,
data_attn_weight,
batch_size,
spatial_size,
num_heads,
channels,
num_levels,
num_query,
num_point,
grad_value,
grad_sampling_loc,
grad_attn_weight);
}
}
}
cudaError_t err = cudaGetLastError();
if (err != cudaSuccess)
{
printf("error in ms_deformable_col2im_cuda: %s\n", cudaGetErrorString(err));
}
}
| 0
|
mavonic_private_repos/transformers/src/transformers
|
mavonic_private_repos/transformers/src/transformers/sagemaker/__init__.py
|
# Copyright 2021 The HuggingFace Team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
from .trainer_sm import SageMakerTrainer
from .training_args_sm import SageMakerTrainingArguments, is_sagemaker_dp_enabled
| 0
|
mavonic_private_repos/transformers/src/transformers
|
mavonic_private_repos/transformers/src/transformers/sagemaker/trainer_sm.py
|
# Copyright 2021 The HuggingFace Team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import warnings
from ..trainer import Trainer
from ..utils import logging
logger = logging.get_logger(__name__)
class SageMakerTrainer(Trainer):
def __init__(self, args=None, **kwargs):
warnings.warn(
"`SageMakerTrainer` is deprecated and will be removed in v5 of Transformers. You can use `Trainer` "
"instead.",
FutureWarning,
)
super().__init__(args=args, **kwargs)
| 0
|
mavonic_private_repos/transformers/src/transformers
|
mavonic_private_repos/transformers/src/transformers/sagemaker/training_args_sm.py
|
# Copyright 2021 The HuggingFace Team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import importlib.util
import json
import os
import warnings
from dataclasses import dataclass, field
import torch
from ..training_args import TrainingArguments
from ..utils import cached_property, is_sagemaker_dp_enabled, logging
logger = logging.get_logger(__name__)
# TODO: should be moved to `utils` after refactoring of SageMakerTrainer
def is_sagemaker_model_parallel_available():
# Get the sagemaker specific mp parameters from smp_options variable.
smp_options = os.getenv("SM_HP_MP_PARAMETERS", "{}")
try:
# Parse it and check the field "partitions" is included, it is required for model parallel.
smp_options = json.loads(smp_options)
if "partitions" not in smp_options:
return False
except json.JSONDecodeError:
return False
# Get the sagemaker specific framework parameters from mpi_options variable.
mpi_options = os.getenv("SM_FRAMEWORK_PARAMS", "{}")
try:
# Parse it and check the field "sagemaker_distributed_dataparallel_enabled".
mpi_options = json.loads(mpi_options)
if not mpi_options.get("sagemaker_mpi_enabled", False):
return False
except json.JSONDecodeError:
return False
# Lastly, check if the `smdistributed` module is present.
return importlib.util.find_spec("smdistributed") is not None
if is_sagemaker_model_parallel_available():
import smdistributed.modelparallel.torch as smp
smp.init()
@dataclass
class SageMakerTrainingArguments(TrainingArguments):
mp_parameters: str = field(
default="",
metadata={"help": "Used by the SageMaker launcher to send mp-specific args. Ignored in SageMakerTrainer"},
)
def __post_init__(self):
super().__post_init__()
warnings.warn(
"`SageMakerTrainingArguments` is deprecated and will be removed in v5 of Transformers. You can use "
"`TrainingArguments` instead.",
FutureWarning,
)
@cached_property
def _setup_devices(self) -> "torch.device":
logger.info("PyTorch: setting up devices")
if torch.distributed.is_available() and torch.distributed.is_initialized() and self.local_rank == -1:
logger.warning(
"torch.distributed process group is initialized, but local_rank == -1. "
"In order to use Torch DDP, launch your script with `python -m torch.distributed.launch"
)
if self.no_cuda:
device = torch.device("cpu")
self._n_gpu = 0
elif is_sagemaker_model_parallel_available():
local_rank = smp.local_rank()
device = torch.device("cuda", local_rank)
self._n_gpu = 1
elif is_sagemaker_dp_enabled():
import smdistributed.dataparallel.torch.torch_smddp # noqa: F401
torch.distributed.init_process_group(backend="smddp", timeout=self.ddp_timeout_delta)
self.local_rank = int(os.getenv("SMDATAPARALLEL_LOCAL_RANK"))
device = torch.device("cuda", self.local_rank)
self._n_gpu = 1
elif self.local_rank == -1:
# if n_gpu is > 1 we'll use nn.DataParallel.
# If you only want to use a specific subset of GPUs use `CUDA_VISIBLE_DEVICES=0`
# Explicitly set CUDA to the first (index 0) CUDA device, otherwise `set_device` will
# trigger an error that a device index is missing. Index 0 takes into account the
# GPUs available in the environment, so `CUDA_VISIBLE_DEVICES=1,2` with `cuda:0`
# will use the first GPU in that env, i.e. GPU#1
device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
# Sometimes the line in the postinit has not been run before we end up here, so just checking we're not at
# the default value.
self._n_gpu = torch.cuda.device_count()
else:
# Here, we'll use torch.distributed.
# Initializes the distributed backend which will take care of synchronizing nodes/GPUs
if not torch.distributed.is_initialized():
torch.distributed.init_process_group(backend="nccl", timeout=self.ddp_timeout_delta)
device = torch.device("cuda", self.local_rank)
self._n_gpu = 1
if device.type == "cuda":
torch.cuda.set_device(device)
return device
@property
def world_size(self):
if is_sagemaker_model_parallel_available():
return smp.dp_size()
return super().world_size
@property
def place_model_on_device(self):
return not is_sagemaker_model_parallel_available()
@property
def _no_sync_in_gradient_accumulation(self):
return False
| 0
|
mavonic_private_repos/transformers
|
mavonic_private_repos/transformers/notebooks/README.md
|
<!---
Copyright 2023 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
-->
# 🤗 Transformers Notebooks
You can find here a list of the official notebooks provided by Hugging Face.
Also, we would like to list here interesting content created by the community.
If you wrote some notebook(s) leveraging 🤗 Transformers and would like to be listed here, please open a
Pull Request so it can be included under the Community notebooks.
## Hugging Face's notebooks 🤗
### Documentation notebooks
You can open any page of the documentation as a notebook in Colab (there is a button directly on said pages) but they are also listed here if you need them:
| Notebook | Description | | |
|:----------|:-------------|:-------------|------:|
| [Quicktour of the library](https://github.com/huggingface/notebooks/blob/main/transformers_doc/en/quicktour.ipynb) | A presentation of the various APIs in Transformers |[](https://colab.research.google.com/github/huggingface/notebooks/blob/main/transformers_doc/en/quicktour.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/en/transformers_doc/quicktour.ipynb)|
| [Summary of the tasks](https://github.com/huggingface/notebooks/blob/main/transformers_doc/en/task_summary.ipynb) | How to run the models of the Transformers library task by task |[](https://colab.research.google.com/github/huggingface/notebooks/blob/main/transformers_doc/en/task_summary.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/transformers_doc/en/task_summary.ipynb)|
| [Preprocessing data](https://github.com/huggingface/notebooks/blob/main/transformers_doc/en/preprocessing.ipynb) | How to use a tokenizer to preprocess your data |[](https://colab.research.google.com/github/huggingface/notebooks/blob/main/transformers_doc/en/preprocessing.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/transformers_doc/en/preprocessing.ipynb)|
| [Fine-tuning a pretrained model](https://github.com/huggingface/notebooks/blob/main/transformers_doc/en/training.ipynb) | How to use the Trainer to fine-tune a pretrained model |[](https://colab.research.google.com/github/huggingface/notebooks/blob/main/transformers_doc/en/training.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/transformers_doc/en/training.ipynb)|
| [Summary of the tokenizers](https://github.com/huggingface/notebooks/blob/main/transformers_doc/en/tokenizer_summary.ipynb) | The differences between the tokenizers algorithm |[](https://colab.research.google.com/github/huggingface/notebooks/blob/main/transformers_doc/en/tokenizer_summary.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/transformers_doc/en/tokenizer_summary.ipynb)|
| [Multilingual models](https://github.com/huggingface/notebooks/blob/main/transformers_doc/en/multilingual.ipynb) | How to use the multilingual models of the library |[](https://colab.research.google.com/github/huggingface/notebooks/blob/main/transformers_doc/en/multilingual.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/transformers_doc/en/multilingual.ipynb)|
### PyTorch Examples
#### Natural Language Processing[[pytorch-nlp]]
| Notebook | Description | | |
|:----------|:-------------|:-------------|------:|
| [Train your tokenizer](https://github.com/huggingface/notebooks/blob/main/examples/tokenizer_training.ipynb) | How to train and use your very own tokenizer |[](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/tokenizer_training.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/tokenizer_training.ipynb)|
| [Train your language model](https://github.com/huggingface/notebooks/blob/main/examples/language_modeling_from_scratch.ipynb) | How to easily start using transformers |[](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/language_modeling_from_scratch.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/language_modeling_from_scratch.ipynb)|
| [How to fine-tune a model on text classification](https://github.com/huggingface/notebooks/blob/main/examples/text_classification.ipynb)| Show how to preprocess the data and fine-tune a pretrained model on any GLUE task. | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/text_classification.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/text_classification.ipynb)|
| [How to fine-tune a model on language modeling](https://github.com/huggingface/notebooks/blob/main/examples/language_modeling.ipynb)| Show how to preprocess the data and fine-tune a pretrained model on a causal or masked LM task. | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/language_modeling.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/language_modeling.ipynb)|
| [How to fine-tune a model on token classification](https://github.com/huggingface/notebooks/blob/main/examples/token_classification.ipynb)| Show how to preprocess the data and fine-tune a pretrained model on a token classification task (NER, PoS). | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/token_classification.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/token_classification.ipynb)|
| [How to fine-tune a model on question answering](https://github.com/huggingface/notebooks/blob/main/examples/question_answering.ipynb)| Show how to preprocess the data and fine-tune a pretrained model on SQUAD. | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/question_answering.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/question_answering.ipynb)|
| [How to fine-tune a model on multiple choice](https://github.com/huggingface/notebooks/blob/main/examples/multiple_choice.ipynb)| Show how to preprocess the data and fine-tune a pretrained model on SWAG. | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/multiple_choice.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/multiple_choice.ipynb)|
| [How to fine-tune a model on translation](https://github.com/huggingface/notebooks/blob/main/examples/translation.ipynb)| Show how to preprocess the data and fine-tune a pretrained model on WMT. | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/translation.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/translation.ipynb)|
| [How to fine-tune a model on summarization](https://github.com/huggingface/notebooks/blob/main/examples/summarization.ipynb)| Show how to preprocess the data and fine-tune a pretrained model on XSUM. | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/summarization.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/summarization.ipynb)|
| [How to train a language model from scratch](https://github.com/huggingface/blog/blob/main/notebooks/01_how_to_train.ipynb)| Highlight all the steps to effectively train Transformer model on custom data | [](https://colab.research.google.com/github/huggingface/blog/blob/main/notebooks/01_how_to_train.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/blog/blob/main/notebooks/01_how_to_train.ipynb)|
| [How to generate text](https://github.com/huggingface/blog/blob/main/notebooks/02_how_to_generate.ipynb)| How to use different decoding methods for language generation with transformers | [](https://colab.research.google.com/github/huggingface/blog/blob/main/notebooks/02_how_to_generate.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/blog/blob/main/notebooks/02_how_to_generate.ipynb)|
| [How to generate text (with constraints)](https://github.com/huggingface/blog/blob/main/notebooks/53_constrained_beam_search.ipynb)| How to guide language generation with user-provided constraints | [](https://colab.research.google.com/github/huggingface/blog/blob/main/notebooks/53_constrained_beam_search.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/blog/blob/main/notebooks/53_constrained_beam_search.ipynb)|
| [Reformer](https://github.com/huggingface/blog/blob/main/notebooks/03_reformer.ipynb)| How Reformer pushes the limits of language modeling | [](https://colab.research.google.com/github/patrickvonplaten/blog/blob/main/notebooks/03_reformer.ipynb)| [](https://studiolab.sagemaker.aws/import/github/patrickvonplaten/blog/blob/main/notebooks/03_reformer.ipynb)|
#### Computer Vision[[pytorch-cv]]
| Notebook | Description | | |
|:---------------------------------------------------------------------------------------------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------------------------------------------|:-----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|------:|
| [How to fine-tune a model on image classification (Torchvision)](https://github.com/huggingface/notebooks/blob/main/examples/image_classification.ipynb) | Show how to preprocess the data using Torchvision and fine-tune any pretrained Vision model on Image Classification | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/image_classification.ipynb) | [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/image_classification.ipynb)|
| [How to fine-tune a model on image classification (Albumentations)](https://github.com/huggingface/notebooks/blob/main/examples/image_classification_albumentations.ipynb) | Show how to preprocess the data using Albumentations and fine-tune any pretrained Vision model on Image Classification | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/image_classification_albumentations.ipynb) | [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/image_classification_albumentations.ipynb)|
| [How to fine-tune a model on image classification (Kornia)](https://github.com/huggingface/notebooks/blob/main/examples/image_classification_kornia.ipynb) | Show how to preprocess the data using Kornia and fine-tune any pretrained Vision model on Image Classification | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/image_classification_kornia.ipynb) | [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/image_classification_kornia.ipynb)|
| [How to perform zero-shot object detection with OWL-ViT](https://github.com/huggingface/notebooks/blob/main/examples/zeroshot_object_detection_with_owlvit.ipynb) | Show how to perform zero-shot object detection on images with text queries | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/zeroshot_object_detection_with_owlvit.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/zeroshot_object_detection_with_owlvit.ipynb)|
| [How to fine-tune an image captioning model](https://github.com/huggingface/notebooks/blob/main/examples/image_captioning_blip.ipynb) | Show how to fine-tune BLIP for image captioning on a custom dataset | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/image_captioning_blip.ipynb) | [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/image_captioning_blip.ipynb)|
| [How to build an image similarity system with Transformers](https://github.com/huggingface/notebooks/blob/main/examples/image_similarity.ipynb) | Show how to build an image similarity system | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/image_similarity.ipynb) | [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/image_similarity.ipynb)|
| [How to fine-tune a SegFormer model on semantic segmentation](https://github.com/huggingface/notebooks/blob/main/examples/semantic_segmentation.ipynb) | Show how to preprocess the data and fine-tune a pretrained SegFormer model on Semantic Segmentation | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/semantic_segmentation.ipynb) | [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/semantic_segmentation.ipynb)|
| [How to fine-tune a VideoMAE model on video classification](https://github.com/huggingface/notebooks/blob/main/examples/video_classification.ipynb) | Show how to preprocess the data and fine-tune a pretrained VideoMAE model on Video Classification | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/video_classification.ipynb) | [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/video_classification.ipynb)|
#### Audio[[pytorch-audio]]
| Notebook | Description | | |
|:----------|:-------------|:-------------|------:|
| [How to fine-tune a speech recognition model in English](https://github.com/huggingface/notebooks/blob/main/examples/speech_recognition.ipynb)| Show how to preprocess the data and fine-tune a pretrained Speech model on TIMIT | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/speech_recognition.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/speech_recognition.ipynb)|
| [How to fine-tune a speech recognition model in any language](https://github.com/huggingface/notebooks/blob/main/examples/multi_lingual_speech_recognition.ipynb)| Show how to preprocess the data and fine-tune a multi-lingually pretrained speech model on Common Voice | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/multi_lingual_speech_recognition.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/multi_lingual_speech_recognition.ipynb)|
| [How to fine-tune a model on audio classification](https://github.com/huggingface/notebooks/blob/main/examples/audio_classification.ipynb)| Show how to preprocess the data and fine-tune a pretrained Speech model on Keyword Spotting | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/audio_classification.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/audio_classification.ipynb)|
#### Biological Sequences[[pytorch-bio]]
| Notebook | Description | | |
|:----------|:----------------------------------------------------------------------------------------|:-------------|------:|
| [How to fine-tune a pre-trained protein model](https://github.com/huggingface/notebooks/blob/main/examples/protein_language_modeling.ipynb) | See how to tokenize proteins and fine-tune a large pre-trained protein "language" model | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/protein_language_modeling.ipynb) | [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/protein_language_modeling.ipynb) |
| [How to generate protein folds](https://github.com/huggingface/notebooks/blob/main/examples/protein_folding.ipynb) | See how to go from protein sequence to a full protein model and PDB file | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/protein_folding.ipynb) | [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/protein_folding.ipynb) |
| [How to fine-tune a Nucleotide Transformer model](https://github.com/huggingface/notebooks/blob/main/examples/nucleotide_transformer_dna_sequence_modelling.ipynb) | See how to tokenize DNA and fine-tune a large pre-trained DNA "language" model | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/nucleotide_transformer_dna_sequence_modelling.ipynb) | [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/nucleotide_transformer_dna_sequence_modelling.ipynb) |
| [Fine-tune a Nucleotide Transformer model with LoRA](https://github.com/huggingface/notebooks/blob/main/examples/nucleotide_transformer_dna_sequence_modelling_with_peft.ipynb) | Train even larger DNA models in a memory-efficient way | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/nucleotide_transformer_dna_sequence_modelling_with_peft.ipynb) | [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/nucleotide_transformer_dna_sequence_modelling_with_peft.ipynb) |
#### Other modalities[[pytorch-other]]
| Notebook | Description | | |
|:----------|:----------------------------------------------------------------------------------------|:-------------|------:|
| [Probabilistic Time Series Forecasting](https://github.com/huggingface/notebooks/blob/main/examples/time-series-transformers.ipynb) | See how to train Time Series Transformer on a custom dataset | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/time-series-transformers.ipynb) | [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/time-series-transformers.ipynb) |
#### Utility notebooks[[pytorch-utility]]
| Notebook | Description | | |
|:----------|:-------------|:-------------|------:|
| [How to export model to ONNX](https://github.com/huggingface/notebooks/blob/main/examples/onnx-export.ipynb)| Highlight how to export and run inference workloads through ONNX | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/onnx-export.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/onnx-export.ipynb)|
| [How to use Benchmarks](https://github.com/huggingface/notebooks/blob/main/examples/benchmark.ipynb)| How to benchmark models with transformers | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/benchmark.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/benchmark.ipynb)|
### TensorFlow Examples
#### Natural Language Processing[[tensorflow-nlp]]
| Notebook | Description | | |
|:----------|:-------------|:-------------|------:|
| [Train your tokenizer](https://github.com/huggingface/notebooks/blob/main/examples/tokenizer_training.ipynb) | How to train and use your very own tokenizer |[](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/tokenizer_training.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/tokenizer_training.ipynb)|
| [Train your language model](https://github.com/huggingface/notebooks/blob/main/examples/language_modeling_from_scratch-tf.ipynb) | How to easily start using transformers |[](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/language_modeling_from_scratch-tf.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/language_modeling_from_scratch-tf.ipynb)|
| [How to fine-tune a model on text classification](https://github.com/huggingface/notebooks/blob/main/examples/text_classification-tf.ipynb)| Show how to preprocess the data and fine-tune a pretrained model on any GLUE task. | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/text_classification-tf.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/text_classification-tf.ipynb)|
| [How to fine-tune a model on language modeling](https://github.com/huggingface/notebooks/blob/main/examples/language_modeling-tf.ipynb)| Show how to preprocess the data and fine-tune a pretrained model on a causal or masked LM task. | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/language_modeling-tf.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/language_modeling-tf.ipynb)|
| [How to fine-tune a model on token classification](https://github.com/huggingface/notebooks/blob/main/examples/token_classification-tf.ipynb)| Show how to preprocess the data and fine-tune a pretrained model on a token classification task (NER, PoS). | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/token_classification-tf.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/token_classification-tf.ipynb)|
| [How to fine-tune a model on question answering](https://github.com/huggingface/notebooks/blob/main/examples/question_answering-tf.ipynb)| Show how to preprocess the data and fine-tune a pretrained model on SQUAD. | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/question_answering-tf.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/question_answering-tf.ipynb)|
| [How to fine-tune a model on multiple choice](https://github.com/huggingface/notebooks/blob/main/examples/multiple_choice-tf.ipynb)| Show how to preprocess the data and fine-tune a pretrained model on SWAG. | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/multiple_choice-tf.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/multiple_choice-tf.ipynb)|
| [How to fine-tune a model on translation](https://github.com/huggingface/notebooks/blob/main/examples/translation-tf.ipynb)| Show how to preprocess the data and fine-tune a pretrained model on WMT. | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/translation-tf.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/translation-tf.ipynb)|
| [How to fine-tune a model on summarization](https://github.com/huggingface/notebooks/blob/main/examples/summarization-tf.ipynb)| Show how to preprocess the data and fine-tune a pretrained model on XSUM. | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/summarization-tf.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/summarization-tf.ipynb)|
#### Computer Vision[[tensorflow-cv]]
| Notebook | Description | | |
|:---------------------------------------------------------------------------------------------------------------------------------------------------------|:----------------------------------------------------------------------------------------------------|:-------------|------:|
| [How to fine-tune a model on image classification](https://github.com/huggingface/notebooks/blob/main/examples/image_classification-tf.ipynb) | Show how to preprocess the data and fine-tune any pretrained Vision model on Image Classification | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/image_classification-tf.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/image_classification-tf.ipynb)|
| [How to fine-tune a SegFormer model on semantic segmentation](https://github.com/huggingface/notebooks/blob/main/examples/semantic_segmentation-tf.ipynb) | Show how to preprocess the data and fine-tune a pretrained SegFormer model on Semantic Segmentation | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/semantic_segmentation-tf.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/semantic_segmentation-tf.ipynb)|
#### Biological Sequences[[tensorflow-bio]]
| Notebook | Description | | |
|:----------|:-------------|:-------------|------:|
| [How to fine-tune a pre-trained protein model](https://github.com/huggingface/notebooks/blob/main/examples/protein_language_modeling-tf.ipynb) | See how to tokenize proteins and fine-tune a large pre-trained protein "language" model | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/protein_language_modeling-tf.ipynb) | [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/protein_language_modeling-tf.ipynb) |
#### Utility notebooks[[tensorflow-utility]]
| Notebook | Description | | |
|:----------|:-------------|:-------------|------:|
| [How to train TF/Keras models on TPU](https://github.com/huggingface/notebooks/blob/main/examples/tpu_training-tf.ipynb) | See how to train at high speed on Google's TPU hardware | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/tpu_training-tf.ipynb) | [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/tpu_training-tf.ipynb) |
### Optimum notebooks
🤗 [Optimum](https://github.com/huggingface/optimum) is an extension of 🤗 Transformers, providing a set of performance optimization tools enabling maximum efficiency to train and run models on targeted hardwares.
| Notebook | Description | | |
|:----------|:-------------|:-------------|------:|
| [How to quantize a model with ONNX Runtime for text classification](https://github.com/huggingface/notebooks/blob/main/examples/text_classification_quantization_ort.ipynb)| Show how to apply static and dynamic quantization on a model using [ONNX Runtime](https://github.com/microsoft/onnxruntime) for any GLUE task. | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/text_classification_quantization_ort.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/text_classification_quantization_ort.ipynb)|
| [How to quantize a model with Intel Neural Compressor for text classification](https://github.com/huggingface/notebooks/blob/main/examples/text_classification_quantization_inc.ipynb)| Show how to apply static, dynamic and aware training quantization on a model using [Intel Neural Compressor (INC)](https://github.com/intel/neural-compressor) for any GLUE task. | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/text_classification_quantization_inc.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/text_classification_quantization_inc.ipynb)|
| [How to fine-tune a model on text classification with ONNX Runtime](https://github.com/huggingface/notebooks/blob/main/examples/text_classification_ort.ipynb)| Show how to preprocess the data and fine-tune a model on any GLUE task using [ONNX Runtime](https://github.com/microsoft/onnxruntime). | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/text_classification_ort.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/text_classification_ort.ipynb)|
| [How to fine-tune a model on summarization with ONNX Runtime](https://github.com/huggingface/notebooks/blob/main/examples/summarization_ort.ipynb)| Show how to preprocess the data and fine-tune a model on XSUM using [ONNX Runtime](https://github.com/microsoft/onnxruntime). | [](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/summarization_ort.ipynb)| [](https://studiolab.sagemaker.aws/import/github/huggingface/notebooks/blob/main/examples/summarization_ort.ipynb)|
## Community notebooks:
More notebooks developed by the community are available [here](https://hf.co/docs/transformers/community#community-notebooks).
| 0
|
mavonic_private_repos/transformers
|
mavonic_private_repos/transformers/model_cards/README.md
|
## 🔥 Model cards now live inside each huggingface.co model repo 🔥
For consistency, ease of use and scalability, `README.md` model cards now live directly inside each model repo on the HuggingFace model hub.
### How to update a model card
You can directly update a model card inside any model repo you have **write access** to, i.e.:
- a model under your username namespace
- a model under any organization you are a part of.
You can either:
- update it, commit and push using your usual git workflow (command line, GUI, etc.)
- or edit it directly from the website's UI.
**What if you want to create or update a model card for a model you don't have write access to?**
In that case, you can open a [Hub pull request](https://huggingface.co/docs/hub/repositories-pull-requests-discussions)! Check out the [announcement](https://huggingface.co/blog/community-update) of this feature for more details 🤗.
### What happened to the model cards here?
We migrated every model card from the repo to its corresponding huggingface.co model repo. Individual commits were preserved, and they link back to the original commit on GitHub.
| 0
|
mavonic_private_repos/transformers
|
mavonic_private_repos/transformers/scripts/stale.py
|
# Copyright 2021 The HuggingFace Team, the AllenNLP library authors. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""
Script to close stale issue. Taken in part from the AllenNLP repository.
https://github.com/allenai/allennlp.
"""
import os
from datetime import datetime as dt
import github.GithubException
from github import Github
LABELS_TO_EXEMPT = [
"good first issue",
"good second issue",
"good difficult issue",
"feature request",
"new model",
"wip",
]
def main():
g = Github(os.environ["GITHUB_TOKEN"])
repo = g.get_repo("huggingface/transformers")
open_issues = repo.get_issues(state="open")
for i, issue in enumerate(open_issues):
print(i, issue)
comments = sorted(list(issue.get_comments()), key=lambda i: i.created_at, reverse=True)
last_comment = comments[0] if len(comments) > 0 else None
if (
last_comment is not None and last_comment.user.login == "github-actions[bot]"
and (dt.utcnow() - issue.updated_at.replace(tzinfo=None)).days > 7
and (dt.utcnow() - issue.created_at.replace(tzinfo=None)).days >= 30
and not any(label.name.lower() in LABELS_TO_EXEMPT for label in issue.get_labels())
):
# print(f"Would close issue {issue.number} since it has been 7 days of inactivity since bot mention.")
try:
issue.edit(state="closed")
except github.GithubException as e:
print("Couldn't close the issue:", repr(e))
elif (
(dt.utcnow() - issue.updated_at.replace(tzinfo=None)).days > 23
and (dt.utcnow() - issue.created_at.replace(tzinfo=None)).days >= 30
and not any(label.name.lower() in LABELS_TO_EXEMPT for label in issue.get_labels())
):
# print(f"Would add stale comment to {issue.number}")
try:
issue.create_comment(
"This issue has been automatically marked as stale because it has not had "
"recent activity. If you think this still needs to be addressed "
"please comment on this thread.\n\nPlease note that issues that do not follow the "
"[contributing guidelines](https://github.com/huggingface/transformers/blob/main/CONTRIBUTING.md) "
"are likely to be ignored."
)
except github.GithubException as e:
print("Couldn't create comment:", repr(e))
if __name__ == "__main__":
main()
| 0
|
mavonic_private_repos/transformers
|
mavonic_private_repos/transformers/scripts/check_tokenizers.py
|
from collections import Counter
import datasets
import transformers
from transformers.convert_slow_tokenizer import SLOW_TO_FAST_CONVERTERS
from transformers.utils import logging
logging.set_verbosity_info()
TOKENIZER_CLASSES = {
name: (getattr(transformers, name), getattr(transformers, name + "Fast")) for name in SLOW_TO_FAST_CONVERTERS
}
dataset = datasets.load_dataset("xnli", split="test+validation")
total = 0
perfect = 0
imperfect = 0
wrong = 0
def check_diff(spm_diff, tok_diff, slow, fast):
if spm_diff == list(reversed(tok_diff)):
# AAA -> AA+A vs A+AA case.
return True
elif len(spm_diff) == len(tok_diff) and fast.decode(spm_diff) == fast.decode(tok_diff):
# Second order OK
# Barrich -> Barr + ich vs Bar + rich
return True
spm_reencoded = slow.encode(slow.decode(spm_diff))
tok_reencoded = fast.encode(fast.decode(spm_diff))
if spm_reencoded != spm_diff and spm_reencoded == tok_reencoded:
# Type 3 error.
# Snehagatha ->
# Sne, h, aga, th, a
# Sne, ha, gat, ha
# Encoding the wrong with sp does not even recover what spm gave us
# It fits tokenizer however...
return True
return False
def check_LTR_mark(line, idx, fast):
enc = fast.encode_plus(line)[0]
offsets = enc.offsets
curr, prev = offsets[idx], offsets[idx - 1]
if curr is not None and line[curr[0] : curr[1]] == "\u200f":
return True
if prev is not None and line[prev[0] : prev[1]] == "\u200f":
return True
def check_details(line, spm_ids, tok_ids, slow, fast):
# Encoding can be the same with same result AAA -> A + AA vs AA + A
# We can check that we use at least exactly the same number of tokens.
for i, (spm_id, tok_id) in enumerate(zip(spm_ids, tok_ids)):
if spm_id != tok_id:
break
first = i
for i, (spm_id, tok_id) in enumerate(zip(reversed(spm_ids), reversed(tok_ids))):
if spm_id != tok_id:
break
last = len(spm_ids) - i
spm_diff = spm_ids[first:last]
tok_diff = tok_ids[first:last]
if check_diff(spm_diff, tok_diff, slow, fast):
return True
if check_LTR_mark(line, first, fast):
return True
if last - first > 5:
# We might have twice a single problem, attempt to subdivide the disjointed tokens into smaller problems
spms = Counter(spm_ids[first:last])
toks = Counter(tok_ids[first:last])
removable_tokens = {spm_ for (spm_, si) in spms.items() if toks.get(spm_, 0) == si}
min_width = 3
for i in range(last - first - min_width):
if all(spm_ids[first + i + j] in removable_tokens for j in range(min_width)):
possible_matches = [
k
for k in range(last - first - min_width)
if tok_ids[first + k : first + k + min_width] == spm_ids[first + i : first + i + min_width]
]
for j in possible_matches:
if check_diff(spm_ids[first : first + i], tok_ids[first : first + j], sp, tok) and check_details(
line,
spm_ids[first + i : last],
tok_ids[first + j : last],
slow,
fast,
):
return True
print(f"Spm: {[fast.decode([spm_ids[i]]) for i in range(first, last)]}")
try:
print(f"Tok: {[fast.decode([tok_ids[i]]) for i in range(first, last)]}")
except Exception:
pass
fast.decode(spm_ids[:first])
fast.decode(spm_ids[last:])
wrong = fast.decode(spm_ids[first:last])
print()
print(wrong)
return False
def test_string(slow, fast, text):
global perfect
global imperfect
global wrong
global total
slow_ids = slow.encode(text)
fast_ids = fast.encode(text)
skip_assert = False
total += 1
if slow_ids != fast_ids:
if check_details(text, slow_ids, fast_ids, slow, fast):
skip_assert = True
imperfect += 1
else:
wrong += 1
else:
perfect += 1
if total % 10000 == 0:
print(f"({perfect} / {imperfect} / {wrong} ----- {perfect + imperfect + wrong})")
if skip_assert:
return
assert (
slow_ids == fast_ids
), f"line {text} : \n\n{slow_ids}\n{fast_ids}\n\n{slow.tokenize(text)}\n{fast.tokenize(text)}"
def test_tokenizer(slow, fast):
global batch_total
for i in range(len(dataset)):
# premise, all languages
for text in dataset[i]["premise"].values():
test_string(slow, fast, text)
# hypothesis, all languages
for text in dataset[i]["hypothesis"]["translation"]:
test_string(slow, fast, text)
if __name__ == "__main__":
for name, (slow_class, fast_class) in TOKENIZER_CLASSES.items():
checkpoint_names = list(slow_class.max_model_input_sizes.keys())
for checkpoint in checkpoint_names:
imperfect = 0
perfect = 0
wrong = 0
total = 0
print(f"========================== Checking {name}: {checkpoint} ==========================")
slow = slow_class.from_pretrained(checkpoint, force_download=True)
fast = fast_class.from_pretrained(checkpoint, force_download=True)
test_tokenizer(slow, fast)
print(f"Accuracy {perfect * 100 / total:.2f}")
| 0
|
mavonic_private_repos/transformers/scripts
|
mavonic_private_repos/transformers/scripts/benchmark/trainer-benchmark.py
|
#!/usr/bin/env python
# HF Trainer benchmarking tool
#
# This tool can be used to run and compare multiple dimensions of the HF Trainers args.
#
# It then prints a report once in github format with all the information that needs to be shared
# with others and second time in a console-friendly format, so it's easier to use for tuning things up.
#
# The main idea is:
#
# ./trainer-benchmark.py --base-cmd '<cmd args that don't change>' \
# --variations '--tf32 0|--tf32 1' '--fp16 0|--fp16 1|--bf16 1' \
# --target-metric-key train_samples_per_second
#
# The variations can be any command line argument that you want to compare and not just dtype as in
# the example.
#
# --variations allows you to compare variations in multiple dimensions.
#
# as the first dimention has 2 options and the second 3 in our example, this will run the trainer 6
# times adding one of:
#
# 1. --tf32 0 --fp16 0
# 2. --tf32 0 --fp16 1
# 3. --tf32 0 --bf16 1
# 4. --tf32 1 --fp16 0
# 5. --tf32 1 --fp16 1
# 6. --tf32 1 --bf16 1
#
# and print the results. This is just a cartesian product - and more than 2 dimensions can be used.
#
# If you want to rely on defaults, this:
# --variations '--tf32 0|--tf32 1' '--fp16 0|--fp16 1|--bf16 1'
# is identical to this:
# --variations '--tf32 0|--tf32 1' '|--fp16|--bf16'
#
# the leading empty variation in the 2nd dimension is a valid variation.
#
# So here we get the following 6 variations:
#
# 1. --tf32 0
# 2. --tf32 0 --fp16
# 3. --tf32 0 --bf16
# 4. --tf32 1
# 5. --tf32 1 --fp16
# 6. --tf32 1 --bf16
#
# In this particular case we don't know what the default tf32 setting is as it's normally
# pytorch-version dependent). That's why it's best to do an explicit setting of each variation:
# `--tf32 0|--tf32 1`
#
# Here is a full example of a train:
#
# CUDA_VISIBLE_DEVICES=0 python ./scripts/benchmark/trainer-benchmark.py \
# --base-cmd \
# ' examples/pytorch/translation/run_translation.py --model_name_or_path google-t5/t5-small \
# --output_dir output_dir --do_train --label_smoothing 0.1 --logging_strategy no \
# --save_strategy no --per_device_train_batch_size 32 --max_source_length 512 \
# --max_target_length 512 --num_train_epochs 1 --overwrite_output_dir \
# --source_lang en --target_lang ro --dataset_name wmt16 --dataset_config "ro-en" \
# --source_prefix "translate English to Romanian: " --warmup_steps 50 \
# --max_train_samples 20000 --dataloader_num_workers 2 ' \
# --target-metric-key train_samples_per_second --repeat-times 1 --variations \
# '|--fp16|--bf16' '--tf32 0|--tf32 1' --report-metric-keys train_loss \
# --repeat-times 1 --base-variation '--tf32 0'
#
# and here is a possible output:
#
#
# | Variation | Train | Diff | Train |
# | | samples | % | loss |
# | | per | | |
# | | second | | |
# |:----------------|----------:|-------:|--------:|
# | --tf32 0 | 285.11 | 0 | 2.51 |
# | --tf32 1 | 342.09 | 20 | 2.51 |
# | --fp16 --tf32 0 | 423.49 | 49 | 2.51 |
# | --fp16 --tf32 1 | 423.13 | 48 | 2.51 |
# | --bf16 --tf32 0 | 416.80 | 46 | 2.52 |
# | --bf16 --tf32 1 | 415.87 | 46 | 2.52 |
#
#
# So you can quickly compare the different outcomes.
#
# Typically running each experiment once is enough, but if the environment is unstable you can
# re-run each multiple times, e.g., 3 using --repeat-times 3 and it will report the averaged results.
#
# By default it'll use the lowest result as the base line to use as 100% and then compare the rest to
# it as can be seen from the table above, but you can also specify which combination is the one to use as
# the baseline, e.g., to change to another entry use: --base-variation '--tf32 1 --fp16 0'
#
# --target-metric-key is there to tell the program which metrics to compare - the different metric keys are
# inside output_dir/all_results.json. e.g., to measure eval performance instead of train use:
# --target-metric-key eval_samples_per_second
# but of course you will need to adjust the --base-cmd value in the example to perform evaluation as
# well (as currently it doesn't)
#
import argparse
import datetime
import io
import itertools
import json
import math
import os
import platform
import re
import shlex
import subprocess
import sys
from pathlib import Path
from statistics import fmean
import pandas as pd
import torch
from tqdm import tqdm
import transformers
nan = float("nan")
class Tee:
"""
A helper class to tee print's output into a file.
Usage:
sys.stdout = Tee(filename)
"""
def __init__(self, filename):
self.stdout = sys.stdout
self.file = open(filename, "a")
def __getattr__(self, attr):
return getattr(self.stdout, attr)
def write(self, msg):
self.stdout.write(msg)
# strip tqdm codes
self.file.write(re.sub(r"^.*\r", "", msg, 0, re.M))
def get_original_command(max_width=80, full_python_path=False):
"""
Return the original command line string that can be replayed nicely and wrapped for 80 char width.
Args:
max_width (`int`, `optional`, defaults to 80):
The width to wrap for.
full_python_path (`bool`, `optional`, defaults to `False`):
Whether to replicate the full path or just the last segment (i.e. `python`).
"""
cmd = []
# deal with critical env vars
env_keys = ["CUDA_VISIBLE_DEVICES"]
for key in env_keys:
val = os.environ.get(key, None)
if val is not None:
cmd.append(f"{key}={val}")
# python executable (not always needed if the script is executable)
python = sys.executable if full_python_path else sys.executable.split("/")[-1]
cmd.append(python)
# now the normal args
cmd += list(map(shlex.quote, sys.argv))
# split up into up to MAX_WIDTH lines with shell multi-line escapes
lines = []
current_line = ""
while len(cmd) > 0:
current_line += f"{cmd.pop(0)} "
if len(cmd) == 0 or len(current_line) + len(cmd[0]) + 1 > max_width - 1:
lines.append(current_line)
current_line = ""
return "\\\n".join(lines)
def get_base_command(args, output_dir):
# unwrap multi-line input
args.base_cmd = re.sub(r"[\\\n]+", " ", args.base_cmd)
# remove --output_dir if any and set our own
args.base_cmd = re.sub("--output_dir\s+[^\s]+", "", args.base_cmd)
args.base_cmd += f" --output_dir {output_dir}"
# ensure we have --overwrite_output_dir
args.base_cmd = re.sub("--overwrite_output_dir\s+", "", args.base_cmd)
args.base_cmd += " --overwrite_output_dir"
return [sys.executable] + shlex.split(args.base_cmd)
def process_run_single(id, cmd, variation, output_dir, target_metric_key, metric_keys, verbose):
# Enable to debug everything but the run itself, to do it fast and see the progress.
# This is useful for debugging the output formatting quickly - we can remove it later once
# everybody is happy with the output
if 0:
import random
from time import sleep
sleep(0)
return dict(
{k: random.uniform(0, 100) for k in metric_keys},
**{target_metric_key: random.choice([nan, 10.31, 100.2, 55.6666, 222.22222222])},
)
result = subprocess.run(cmd, capture_output=True, text=True)
if verbose:
print("STDOUT", result.stdout)
print("STDERR", result.stderr)
# save the streams
prefix = variation.replace(" ", "-")
with open(Path(output_dir) / f"log.{prefix}.stdout.txt", "w") as f:
f.write(result.stdout)
with open(Path(output_dir) / f"log.{prefix}.stderr.txt", "w") as f:
f.write(result.stderr)
if result.returncode != 0:
if verbose:
print("failed")
return {target_metric_key: nan}
with io.open(f"{output_dir}/all_results.json", "r", encoding="utf-8") as f:
metrics = json.load(f)
# filter out just the keys we want
return {k: v for k, v in metrics.items() if k in metric_keys}
def process_run(
id,
cmd,
variation_key,
variation,
longest_variation_len,
target_metric_key,
report_metric_keys,
repeat_times,
output_dir,
verbose,
):
results = []
metrics = []
preamble = f"{id}: {variation:<{longest_variation_len}}"
outcome = f"{preamble}: "
metric_keys = set(report_metric_keys + [target_metric_key])
for i in tqdm(range(repeat_times), desc=preamble, leave=False):
single_run_metrics = process_run_single(
id, cmd, variation, output_dir, target_metric_key, metric_keys, verbose
)
result = single_run_metrics[target_metric_key]
if not math.isnan(result):
metrics.append(single_run_metrics)
results.append(result)
outcome += "✓"
else:
outcome += "✘"
outcome = f"\33[2K\r{outcome}"
if len(metrics) > 0:
mean_metrics = {k: fmean([x[k] for x in metrics]) for k in metrics[0].keys()}
mean_target = round(mean_metrics[target_metric_key], 2)
results_str = f"{outcome} {mean_target}"
if len(metrics) > 1:
results_str += f" {tuple(round(x, 2) for x in results)}"
print(results_str)
mean_metrics[variation_key] = variation
return mean_metrics
else:
print(outcome)
return {variation_key: variation, target_metric_key: nan}
def get_versions():
properties = torch.cuda.get_device_properties(torch.device("cuda"))
return f"""
Datetime : {datetime.datetime.now().strftime('%Y-%m-%d %H:%M:%S')}
Software:
transformers: {transformers.__version__}
torch : {torch.__version__}
cuda : {torch.version.cuda}
python : {platform.python_version()}
Hardware:
{torch.cuda.device_count()} GPUs : {properties.name}, {properties.total_memory/2**30:0.2f}GB
"""
def process_results(results, target_metric_key, report_metric_keys, base_variation, output_dir):
df = pd.DataFrame(results)
variation_key = "variation"
diff_key = "diff_%"
sentinel_value = nan
if base_variation is not None and len(df[df[variation_key] == base_variation]):
# this may still return nan
sentinel_value = df.loc[df[variation_key] == base_variation][target_metric_key].item()
if math.isnan(sentinel_value):
# as a fallback, use the minimal value as the sentinel
sentinel_value = df.loc[df[target_metric_key] != nan][target_metric_key].min()
# create diff column if possible
if not math.isnan(sentinel_value):
df[diff_key] = df.apply(
lambda r: round(100 * (r[target_metric_key] - sentinel_value) / sentinel_value)
if not math.isnan(r[target_metric_key])
else 0,
axis="columns",
)
# re-order columns
cols = [variation_key, target_metric_key, diff_key, *report_metric_keys]
df = df.reindex(cols, axis="columns") # reorder cols
# capitalize
df = df.rename(str.capitalize, axis="columns")
# make the cols as narrow as possible
df_github = df.rename(lambda c: c.replace("_", "<br>"), axis="columns")
df_console = df.rename(lambda c: c.replace("_", "\n"), axis="columns")
report = ["", "Copy between the cut-here-lines and paste as is to github or a forum"]
report += ["----------8<-----------------8<--------"]
report += ["*** Results:", df_github.to_markdown(index=False, floatfmt=".2f")]
report += ["```"]
report += ["*** Setup:", get_versions()]
report += ["*** The benchmark command line was:", get_original_command()]
report += ["```"]
report += ["----------8<-----------------8<--------"]
report += ["*** Results (console):", df_console.to_markdown(index=False, floatfmt=".2f")]
print("\n\n".join(report))
def main():
parser = argparse.ArgumentParser()
parser.add_argument(
"--base-cmd",
default=None,
type=str,
required=True,
help="Base cmd",
)
parser.add_argument(
"--variations",
default=None,
type=str,
nargs="+",
required=True,
help="Multi-dimensional variations, example: '|--fp16|--bf16' '|--tf32'",
)
parser.add_argument(
"--base-variation",
default=None,
type=str,
help="Baseline variation to compare to. if None the minimal target value will be used to compare against",
)
parser.add_argument(
"--target-metric-key",
default=None,
type=str,
required=True,
help="Target metric key in output_dir/all_results.json, e.g., train_samples_per_second",
)
parser.add_argument(
"--report-metric-keys",
default="",
type=str,
help="Report metric keys - other metric keys from output_dir/all_results.json to report, e.g., train_loss. Use a single argument e.g., 'train_loss train_samples",
)
parser.add_argument(
"--repeat-times",
default=1,
type=int,
help="How many times to re-run each variation - an average will be reported",
)
parser.add_argument(
"--output_dir",
default="output_benchmark",
type=str,
help="The output directory where all the benchmark reports will go to and additionally this directory will be used to override --output_dir in the script that is being benchmarked",
)
parser.add_argument(
"--verbose",
default=False,
action="store_true",
help="Whether to show the outputs of each run or just the benchmark progress",
)
args = parser.parse_args()
output_dir = args.output_dir
Path(output_dir).mkdir(exist_ok=True)
base_cmd = get_base_command(args, output_dir)
# split each dimension into its --foo variations
dims = [list(map(str.strip, re.split(r"\|", x))) for x in args.variations]
# build a cartesian product of dimensions and convert those back into cmd-line arg strings,
# while stripping white space for inputs that were empty
variations = list(map(str.strip, map(" ".join, itertools.product(*dims))))
longest_variation_len = max(len(x) for x in variations)
# split wanted keys
report_metric_keys = args.report_metric_keys.split()
# capture prints into a log file for convenience
report_fn = f"benchmark-report-{datetime.datetime.now().strftime('%Y-%m-%d-%H-%M-%S')}.txt"
print(f"\nNote: each run's output is also logged under {output_dir}/log.*.std*.txt")
print(f"and this script's output is also piped into {report_fn}")
sys.stdout = Tee(report_fn)
print(f"\n*** Running {len(variations)} benchmarks:")
print(f"Base command: {' '.join(base_cmd)}")
variation_key = "variation"
results = []
for id, variation in enumerate(tqdm(variations, desc="Total completion: ", leave=False)):
cmd = base_cmd + variation.split()
results.append(
process_run(
id + 1,
cmd,
variation_key,
variation,
longest_variation_len,
args.target_metric_key,
report_metric_keys,
args.repeat_times,
output_dir,
args.verbose,
)
)
process_results(results, args.target_metric_key, report_metric_keys, args.base_variation, output_dir)
if __name__ == "__main__":
main()
| 0
|
mavonic_private_repos/transformers/scripts
|
mavonic_private_repos/transformers/scripts/fsmt/eval-facebook-wmt19.sh
|
#!/usr/bin/env bash
# Copyright 2020 The HuggingFace Team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# this script evals the following fsmt models
# it covers:
# - facebook/wmt19-ru-en
# - facebook/wmt19-en-ru
# - facebook/wmt19-de-en
# - facebook/wmt19-en-de
# this script needs to be run from the top level of the transformers repo
if [ ! -d "src/transformers" ]; then
echo "Error: This script needs to be run from the top of the transformers repo"
exit 1
fi
# In these scripts you may have to lower BS if you get CUDA OOM (or increase it if you have a large GPU)
### a short estimate version for quick testing ###
export PAIR=en-ru
export DATA_DIR=data/$PAIR
export SAVE_DIR=data/$PAIR
export BS=8
export NUM_BEAMS=8
mkdir -p $DATA_DIR
sacrebleu -t wmt19 -l $PAIR --echo src | head -10 > $DATA_DIR/val.source
sacrebleu -t wmt19 -l $PAIR --echo ref | head -10 > $DATA_DIR/val.target
echo $PAIR
PYTHONPATH="src:examples/seq2seq" python examples/seq2seq/run_eval.py facebook/wmt19-$PAIR $DATA_DIR/val.source $SAVE_DIR/test_translations.txt --reference_path $DATA_DIR/val.target --score_path $SAVE_DIR/test_bleu.json --bs $BS --task translation --num_beams $NUM_BEAMS
### Normal eval ###
# ru-en
export PAIR=ru-en
export DATA_DIR=data/$PAIR
export SAVE_DIR=data/$PAIR
export BS=8
export NUM_BEAMS=50
mkdir -p $DATA_DIR
sacrebleu -t wmt19 -l $PAIR --echo src > $DATA_DIR/val.source
sacrebleu -t wmt19 -l $PAIR --echo ref > $DATA_DIR/val.target
PYTHONPATH="src:examples/seq2seq" python examples/seq2seq/run_eval.py facebook/wmt19-$PAIR $DATA_DIR/val.source $SAVE_DIR/test_translations.txt --reference_path $DATA_DIR/val.target --score_path $SAVE_DIR/test_bleu.json --bs $BS --task translation --num_beams $NUM_BEAMS
# (target BLEU: 41.3 http://matrix.statmt.org/matrix/output/1907?run_id=6937)
# en-ru
export PAIR=en-ru
export DATA_DIR=data/$PAIR
export SAVE_DIR=data/$PAIR
export BS=8
export NUM_BEAMS=50
mkdir -p $DATA_DIR
sacrebleu -t wmt19 -l $PAIR --echo src > $DATA_DIR/val.source
sacrebleu -t wmt19 -l $PAIR --echo ref > $DATA_DIR/val.target
echo $PAIR
PYTHONPATH="src:examples/seq2seq" python examples/seq2seq/run_eval.py facebook/wmt19-$PAIR $DATA_DIR/val.source $SAVE_DIR/test_translations.txt --reference_path $DATA_DIR/val.target --score_path $SAVE_DIR/test_bleu.json --bs $BS --task translation --num_beams $NUM_BEAMS
# (target BLEU: 36.4 http://matrix.statmt.org/matrix/output/1914?score_id=37605)
# en-de
export PAIR=en-de
export DATA_DIR=data/$PAIR
export SAVE_DIR=data/$PAIR
export BS=8
mkdir -p $DATA_DIR
sacrebleu -t wmt19 -l $PAIR --echo src > $DATA_DIR/val.source
sacrebleu -t wmt19 -l $PAIR --echo ref > $DATA_DIR/val.target
echo $PAIR
PYTHONPATH="src:examples/seq2seq" python examples/seq2seq/run_eval.py facebook/wmt19-$PAIR $DATA_DIR/val.source $SAVE_DIR/test_translations.txt --reference_path $DATA_DIR/val.target --score_path $SAVE_DIR/test_bleu.json --bs $BS --task translation --num_beams $NUM_BEAMS
# (target BLEU: 43.1 http://matrix.statmt.org/matrix/output/1909?run_id=6862)
# de-en
export PAIR=de-en
export DATA_DIR=data/$PAIR
export SAVE_DIR=data/$PAIR
export BS=8
export NUM_BEAMS=50
mkdir -p $DATA_DIR
sacrebleu -t wmt19 -l $PAIR --echo src > $DATA_DIR/val.source
sacrebleu -t wmt19 -l $PAIR --echo ref > $DATA_DIR/val.target
echo $PAIR
PYTHONPATH="src:examples/seq2seq" python examples/seq2seq/run_eval.py facebook/wmt19-$PAIR $DATA_DIR/val.source $SAVE_DIR/test_translations.txt --reference_path $DATA_DIR/val.target --score_path $SAVE_DIR/test_bleu.json --bs $BS --task translation --num_beams $NUM_BEAMS
# (target BLEU: 42.3 http://matrix.statmt.org/matrix/output/1902?run_id=6750)
### Searching hparams eval ###
# en-ru
export PAIR=ru-en
export DATA_DIR=data/$PAIR
export SAVE_DIR=data/$PAIR
export BS=32
mkdir -p $DATA_DIR
sacrebleu -t wmt19 -l $PAIR --echo src > $DATA_DIR/val.source
sacrebleu -t wmt19 -l $PAIR --echo ref > $DATA_DIR/val.target
CUDA_VISIBLE_DEVICES="0" PYTHONPATH="src:examples/seq2seq" python examples/seq2seq/run_eval_search.py facebook/wmt19-$PAIR $DATA_DIR/val.source $SAVE_DIR/test_translations.txt --reference_path $DATA_DIR/val.target --score_path $SAVE_DIR/test_bleu.json --bs $BS --task translation --search="num_beams=5 length_penalty=0.6:0.7:0.8:0.9:1.0:1.1"
# en-ru
export PAIR=en-ru
export DATA_DIR=data/$PAIR
export SAVE_DIR=data/$PAIR
export BS=16
mkdir -p $DATA_DIR
mkdir -p $DATA_DIR
sacrebleu -t wmt19 -l $PAIR --echo src > $DATA_DIR/val.source
sacrebleu -t wmt19 -l $PAIR --echo ref > $DATA_DIR/val.target
CUDA_VISIBLE_DEVICES="0" PYTHONPATH="src:examples/seq2seq" python examples/seq2seq/run_eval_search.py facebook/wmt19-$PAIR $DATA_DIR/val.source $SAVE_DIR/test_translations.txt --reference_path $DATA_DIR/val.target --score_path $SAVE_DIR/test_bleu.json --bs $BS --task translation --search="num_beams=5:8:11:15 length_penalty=0.6:0.7:0.8:0.9:1.0:1.1 early_stopping=true:false"
# en-de
export PAIR=en-de
export DATA_DIR=data/$PAIR
export SAVE_DIR=data/$PAIR
export BS=16
mkdir -p $DATA_DIR
sacrebleu -t wmt19 -l $PAIR --echo src > $DATA_DIR/val.source
sacrebleu -t wmt19 -l $PAIR --echo ref > $DATA_DIR/val.target
CUDA_VISIBLE_DEVICES="1" PYTHONPATH="src:examples/seq2seq" python examples/seq2seq/run_eval_search.py facebook/wmt19-$PAIR $DATA_DIR/val.source $SAVE_DIR/test_translations.txt --reference_path $DATA_DIR/val.target --score_path $SAVE_DIR/test_bleu.json --bs $BS --task translation --search="num_beams=5:8:11:15 length_penalty=0.6:0.7:0.8:0.9:1.0:1.1 early_stopping=true:false"
# de-en
export PAIR=de-en
export DATA_DIR=data/$PAIR
export SAVE_DIR=data/$PAIR
export BS=16
mkdir -p $DATA_DIR
mkdir -p $DATA_DIR
sacrebleu -t wmt19 -l $PAIR --echo src > $DATA_DIR/val.source
sacrebleu -t wmt19 -l $PAIR --echo ref > $DATA_DIR/val.target
CUDA_VISIBLE_DEVICES="1" PYTHONPATH="src:examples/seq2seq" python examples/seq2seq/run_eval_search.py facebook/wmt19-$PAIR $DATA_DIR/val.source $SAVE_DIR/test_translations.txt --reference_path $DATA_DIR/val.target --score_path $SAVE_DIR/test_bleu.json --bs $BS --task translation --search="num_beams=5:8:11:15 length_penalty=0.6:0.7:0.8:0.9:1.0:1.1 early_stopping=true:false"
| 0
|
mavonic_private_repos/transformers/scripts
|
mavonic_private_repos/transformers/scripts/fsmt/fsmt-make-super-tiny-model.py
|
#!/usr/bin/env python
# coding: utf-8
# Copyright 2020 The HuggingFace Team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# This script creates a super tiny model that is useful inside tests, when we just want to test that
# the machinery works, without needing to the check the quality of the outcomes.
#
# This version creates a tiny vocab first, and then a tiny model - so the outcome is truly tiny -
# all files ~60KB. As compared to taking a full-size model, reducing to the minimum its layers and
# emb dimensions, but keeping the full vocab + merges files, leading to ~3MB in total for all files.
# The latter is done by `fsmt-make-super-tiny-model.py`.
#
# It will be used then as "stas/tiny-wmt19-en-ru"
import json
import tempfile
from pathlib import Path
from transformers import FSMTConfig, FSMTForConditionalGeneration, FSMTTokenizer
from transformers.models.fsmt.tokenization_fsmt import VOCAB_FILES_NAMES
mname_tiny = "tiny-wmt19-en-ru"
# Build
# borrowed from a test
vocab = [ "l", "o", "w", "e", "r", "s", "t", "i", "d", "n", "w</w>", "r</w>", "t</w>", "lo", "low", "er</w>", "low</w>", "lowest</w>", "newer</w>", "wider</w>", "<unk>", ]
vocab_tokens = dict(zip(vocab, range(len(vocab))))
merges = ["l o 123", "lo w 1456", "e r</w> 1789", ""]
with tempfile.TemporaryDirectory() as tmpdirname:
build_dir = Path(tmpdirname)
src_vocab_file = build_dir / VOCAB_FILES_NAMES["src_vocab_file"]
tgt_vocab_file = build_dir / VOCAB_FILES_NAMES["tgt_vocab_file"]
merges_file = build_dir / VOCAB_FILES_NAMES["merges_file"]
with open(src_vocab_file, "w") as fp: fp.write(json.dumps(vocab_tokens))
with open(tgt_vocab_file, "w") as fp: fp.write(json.dumps(vocab_tokens))
with open(merges_file, "w") as fp : fp.write("\n".join(merges))
tokenizer = FSMTTokenizer(
langs=["en", "ru"],
src_vocab_size = len(vocab),
tgt_vocab_size = len(vocab),
src_vocab_file=src_vocab_file,
tgt_vocab_file=tgt_vocab_file,
merges_file=merges_file,
)
config = FSMTConfig(
langs=['ru', 'en'],
src_vocab_size=1000, tgt_vocab_size=1000,
d_model=4,
encoder_layers=1, decoder_layers=1,
encoder_ffn_dim=4, decoder_ffn_dim=4,
encoder_attention_heads=1, decoder_attention_heads=1,
)
tiny_model = FSMTForConditionalGeneration(config)
print(f"num of params {tiny_model.num_parameters()}")
# Test
batch = tokenizer(["Making tiny model"], return_tensors="pt")
outputs = tiny_model(**batch)
print("test output:", len(outputs.logits[0]))
# Save
tiny_model.half() # makes it smaller
tiny_model.save_pretrained(mname_tiny)
tokenizer.save_pretrained(mname_tiny)
print(f"Generated {mname_tiny}")
# Upload
# transformers-cli upload tiny-wmt19-en-ru
| 0
|
mavonic_private_repos/transformers/scripts
|
mavonic_private_repos/transformers/scripts/fsmt/gen-card-allenai-wmt19.py
|
#!/usr/bin/env python
# Copyright 2020 The HuggingFace Team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# Usage:
# ./gen-card-allenai-wmt19.py
import os
from pathlib import Path
def write_model_card(model_card_dir, src_lang, tgt_lang, model_name):
texts = {
"en": "Machine learning is great, isn't it?",
"ru": "Машинное обучение - это здорово, не так ли?",
"de": "Maschinelles Lernen ist großartig, nicht wahr?",
}
# BLUE scores as follows:
# "pair": [fairseq, transformers]
scores = {
"wmt19-de-en-6-6-base": [0, 38.37],
"wmt19-de-en-6-6-big": [0, 39.90],
}
pair = f"{src_lang}-{tgt_lang}"
readme = f"""
---
language:
- {src_lang}
- {tgt_lang}
thumbnail:
tags:
- translation
- wmt19
- allenai
license: apache-2.0
datasets:
- wmt19
metrics:
- bleu
---
# FSMT
## Model description
This is a ported version of fairseq-based [wmt19 transformer](https://github.com/jungokasai/deep-shallow/) for {src_lang}-{tgt_lang}.
For more details, please, see [Deep Encoder, Shallow Decoder: Reevaluating the Speed-Quality Tradeoff in Machine Translation](https://arxiv.org/abs/2006.10369).
2 models are available:
* [wmt19-de-en-6-6-big](https://huggingface.co/allenai/wmt19-de-en-6-6-big)
* [wmt19-de-en-6-6-base](https://huggingface.co/allenai/wmt19-de-en-6-6-base)
## Intended uses & limitations
#### How to use
```python
from transformers import FSMTForConditionalGeneration, FSMTTokenizer
mname = "allenai/{model_name}"
tokenizer = FSMTTokenizer.from_pretrained(mname)
model = FSMTForConditionalGeneration.from_pretrained(mname)
input = "{texts[src_lang]}"
input_ids = tokenizer.encode(input, return_tensors="pt")
outputs = model.generate(input_ids)
decoded = tokenizer.decode(outputs[0], skip_special_tokens=True)
print(decoded) # {texts[tgt_lang]}
```
#### Limitations and bias
## Training data
Pretrained weights were left identical to the original model released by allenai. For more details, please, see the [paper](https://arxiv.org/abs/2006.10369).
## Eval results
Here are the BLEU scores:
model | transformers
-------|---------
{model_name} | {scores[model_name][1]}
The score was calculated using this code:
```bash
git clone https://github.com/huggingface/transformers
cd transformers
export PAIR={pair}
export DATA_DIR=data/$PAIR
export SAVE_DIR=data/$PAIR
export BS=8
export NUM_BEAMS=5
mkdir -p $DATA_DIR
sacrebleu -t wmt19 -l $PAIR --echo src > $DATA_DIR/val.source
sacrebleu -t wmt19 -l $PAIR --echo ref > $DATA_DIR/val.target
echo $PAIR
PYTHONPATH="src:examples/seq2seq" python examples/seq2seq/run_eval.py allenai/{model_name} $DATA_DIR/val.source $SAVE_DIR/test_translations.txt --reference_path $DATA_DIR/val.target --score_path $SAVE_DIR/test_bleu.json --bs $BS --task translation --num_beams $NUM_BEAMS
```
## Data Sources
- [training, etc.](http://www.statmt.org/wmt19/)
- [test set](http://matrix.statmt.org/test_sets/newstest2019.tgz?1556572561)
### BibTeX entry and citation info
```
@misc{{kasai2020deep,
title={{Deep Encoder, Shallow Decoder: Reevaluating the Speed-Quality Tradeoff in Machine Translation}},
author={{Jungo Kasai and Nikolaos Pappas and Hao Peng and James Cross and Noah A. Smith}},
year={{2020}},
eprint={{2006.10369}},
archivePrefix={{arXiv}},
primaryClass={{cs.CL}}
}}
```
"""
model_card_dir.mkdir(parents=True, exist_ok=True)
path = os.path.join(model_card_dir, "README.md")
print(f"Generating {path}")
with open(path, "w", encoding="utf-8") as f:
f.write(readme)
# make sure we are under the root of the project
repo_dir = Path(__file__).resolve().parent.parent.parent
model_cards_dir = repo_dir / "model_cards"
for model_name in ["wmt19-de-en-6-6-base", "wmt19-de-en-6-6-big"]:
model_card_dir = model_cards_dir / "allenai" / model_name
write_model_card(model_card_dir, src_lang="de", tgt_lang="en", model_name=model_name)
| 0
|
mavonic_private_repos/transformers/scripts
|
mavonic_private_repos/transformers/scripts/fsmt/s3-move.sh
|
# this is the process of uploading the updated models to s3. As I can't upload them directly to the correct orgs, this script shows how this is done
# Copyright 2020 The HuggingFace Team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
1. upload updated models to my account
transformers-cli upload -y wmt19-ru-en
transformers-cli upload -y wmt19-en-ru
transformers-cli upload -y wmt19-de-en
transformers-cli upload -y wmt19-en-de
transformers-cli upload -y wmt19-de-en-6-6-base
transformers-cli upload -y wmt19-de-en-6-6-big
transformers-cli upload -y wmt16-en-de-dist-12-1
transformers-cli upload -y wmt16-en-de-dist-6-1
transformers-cli upload -y wmt16-en-de-12-1
2. ask someone to move them to:
* to facebook: "wmt19-ru-en", "wmt19-en-ru", "wmt19-en-de", "wmt19-de-en"
* to allenai: "wmt16-en-de-dist-12-1", "wmt16-en-de-dist-6-1", "wmt16-en-de-12-1", "wmt19-de-en-6-6-base", "wmt19-de-en-6-6-big"
export b="s3://models.huggingface.co/bert"
stas_to_fb () {
src=$1
shift
aws s3 sync $b/stas/$src $b/facebook/$src $@
}
stas_to_allenai () {
src=$1
shift
aws s3 sync $b/stas/$src $b/allenai/$src $@
}
stas_to_fb wmt19-en-ru
stas_to_fb wmt19-ru-en
stas_to_fb wmt19-en-de
stas_to_fb wmt19-de-en
stas_to_allenai wmt16-en-de-dist-12-1
stas_to_allenai wmt16-en-de-dist-6-1
stas_to_allenai wmt16-en-de-6-1
stas_to_allenai wmt16-en-de-12-1
stas_to_allenai wmt19-de-en-6-6-base
stas_to_allenai wmt19-de-en-6-6-big
3. and then remove all these model files from my account
transformers-cli s3 rm wmt16-en-de-12-1/config.json
transformers-cli s3 rm wmt16-en-de-12-1/merges.txt
transformers-cli s3 rm wmt16-en-de-12-1/pytorch_model.bin
transformers-cli s3 rm wmt16-en-de-12-1/tokenizer_config.json
transformers-cli s3 rm wmt16-en-de-12-1/vocab-src.json
transformers-cli s3 rm wmt16-en-de-12-1/vocab-tgt.json
transformers-cli s3 rm wmt16-en-de-dist-12-1/config.json
transformers-cli s3 rm wmt16-en-de-dist-12-1/merges.txt
transformers-cli s3 rm wmt16-en-de-dist-12-1/pytorch_model.bin
transformers-cli s3 rm wmt16-en-de-dist-12-1/tokenizer_config.json
transformers-cli s3 rm wmt16-en-de-dist-12-1/vocab-src.json
transformers-cli s3 rm wmt16-en-de-dist-12-1/vocab-tgt.json
transformers-cli s3 rm wmt16-en-de-dist-6-1/config.json
transformers-cli s3 rm wmt16-en-de-dist-6-1/merges.txt
transformers-cli s3 rm wmt16-en-de-dist-6-1/pytorch_model.bin
transformers-cli s3 rm wmt16-en-de-dist-6-1/tokenizer_config.json
transformers-cli s3 rm wmt16-en-de-dist-6-1/vocab-src.json
transformers-cli s3 rm wmt16-en-de-dist-6-1/vocab-tgt.json
transformers-cli s3 rm wmt19-de-en-6-6-base/config.json
transformers-cli s3 rm wmt19-de-en-6-6-base/merges.txt
transformers-cli s3 rm wmt19-de-en-6-6-base/pytorch_model.bin
transformers-cli s3 rm wmt19-de-en-6-6-base/tokenizer_config.json
transformers-cli s3 rm wmt19-de-en-6-6-base/vocab-src.json
transformers-cli s3 rm wmt19-de-en-6-6-base/vocab-tgt.json
transformers-cli s3 rm wmt19-de-en-6-6-big/config.json
transformers-cli s3 rm wmt19-de-en-6-6-big/merges.txt
transformers-cli s3 rm wmt19-de-en-6-6-big/pytorch_model.bin
transformers-cli s3 rm wmt19-de-en-6-6-big/tokenizer_config.json
transformers-cli s3 rm wmt19-de-en-6-6-big/vocab-src.json
transformers-cli s3 rm wmt19-de-en-6-6-big/vocab-tgt.json
transformers-cli s3 rm wmt19-de-en/config.json
transformers-cli s3 rm wmt19-de-en/merges.txt
transformers-cli s3 rm wmt19-de-en/pytorch_model.bin
transformers-cli s3 rm wmt19-de-en/tokenizer_config.json
transformers-cli s3 rm wmt19-de-en/vocab-src.json
transformers-cli s3 rm wmt19-de-en/vocab-tgt.json
transformers-cli s3 rm wmt19-en-de/config.json
transformers-cli s3 rm wmt19-en-de/merges.txt
transformers-cli s3 rm wmt19-en-de/pytorch_model.bin
transformers-cli s3 rm wmt19-en-de/tokenizer_config.json
transformers-cli s3 rm wmt19-en-de/vocab-src.json
transformers-cli s3 rm wmt19-en-de/vocab-tgt.json
transformers-cli s3 rm wmt19-en-ru/config.json
transformers-cli s3 rm wmt19-en-ru/merges.txt
transformers-cli s3 rm wmt19-en-ru/pytorch_model.bin
transformers-cli s3 rm wmt19-en-ru/tokenizer_config.json
transformers-cli s3 rm wmt19-en-ru/vocab-src.json
transformers-cli s3 rm wmt19-en-ru/vocab-tgt.json
transformers-cli s3 rm wmt19-ru-en/config.json
transformers-cli s3 rm wmt19-ru-en/merges.txt
transformers-cli s3 rm wmt19-ru-en/pytorch_model.bin
transformers-cli s3 rm wmt19-ru-en/tokenizer_config.json
transformers-cli s3 rm wmt19-ru-en/vocab-src.json
transformers-cli s3 rm wmt19-ru-en/vocab-tgt.json
| 0
|
mavonic_private_repos/transformers/scripts
|
mavonic_private_repos/transformers/scripts/fsmt/gen-card-facebook-wmt19.py
|
#!/usr/bin/env python
# Copyright 2020 The HuggingFace Team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# Usage:
# ./gen-card-facebook-wmt19.py
import os
from pathlib import Path
def write_model_card(model_card_dir, src_lang, tgt_lang):
texts = {
"en": "Machine learning is great, isn't it?",
"ru": "Машинное обучение - это здорово, не так ли?",
"de": "Maschinelles Lernen ist großartig, oder?",
}
# BLUE scores as follows:
# "pair": [fairseq, transformers]
scores = {
"ru-en": ["[41.3](http://matrix.statmt.org/matrix/output/1907?run_id=6937)", "39.20"],
"en-ru": ["[36.4](http://matrix.statmt.org/matrix/output/1914?run_id=6724)", "33.47"],
"en-de": ["[43.1](http://matrix.statmt.org/matrix/output/1909?run_id=6862)", "42.83"],
"de-en": ["[42.3](http://matrix.statmt.org/matrix/output/1902?run_id=6750)", "41.35"],
}
pair = f"{src_lang}-{tgt_lang}"
readme = f"""
---
language:
- {src_lang}
- {tgt_lang}
thumbnail:
tags:
- translation
- wmt19
- facebook
license: apache-2.0
datasets:
- wmt19
metrics:
- bleu
---
# FSMT
## Model description
This is a ported version of [fairseq wmt19 transformer](https://github.com/pytorch/fairseq/blob/master/examples/wmt19/README.md) for {src_lang}-{tgt_lang}.
For more details, please see, [Facebook FAIR's WMT19 News Translation Task Submission](https://arxiv.org/abs/1907.06616).
The abbreviation FSMT stands for FairSeqMachineTranslation
All four models are available:
* [wmt19-en-ru](https://huggingface.co/facebook/wmt19-en-ru)
* [wmt19-ru-en](https://huggingface.co/facebook/wmt19-ru-en)
* [wmt19-en-de](https://huggingface.co/facebook/wmt19-en-de)
* [wmt19-de-en](https://huggingface.co/facebook/wmt19-de-en)
## Intended uses & limitations
#### How to use
```python
from transformers import FSMTForConditionalGeneration, FSMTTokenizer
mname = "facebook/wmt19-{src_lang}-{tgt_lang}"
tokenizer = FSMTTokenizer.from_pretrained(mname)
model = FSMTForConditionalGeneration.from_pretrained(mname)
input = "{texts[src_lang]}"
input_ids = tokenizer.encode(input, return_tensors="pt")
outputs = model.generate(input_ids)
decoded = tokenizer.decode(outputs[0], skip_special_tokens=True)
print(decoded) # {texts[tgt_lang]}
```
#### Limitations and bias
- The original (and this ported model) doesn't seem to handle well inputs with repeated sub-phrases, [content gets truncated](https://discuss.huggingface.co/t/issues-with-translating-inputs-containing-repeated-phrases/981)
## Training data
Pretrained weights were left identical to the original model released by fairseq. For more details, please, see the [paper](https://arxiv.org/abs/1907.06616).
## Eval results
pair | fairseq | transformers
-------|---------|----------
{pair} | {scores[pair][0]} | {scores[pair][1]}
The score is slightly below the score reported by `fairseq`, since `transformers`` currently doesn't support:
- model ensemble, therefore the best performing checkpoint was ported (``model4.pt``).
- re-ranking
The score was calculated using this code:
```bash
git clone https://github.com/huggingface/transformers
cd transformers
export PAIR={pair}
export DATA_DIR=data/$PAIR
export SAVE_DIR=data/$PAIR
export BS=8
export NUM_BEAMS=15
mkdir -p $DATA_DIR
sacrebleu -t wmt19 -l $PAIR --echo src > $DATA_DIR/val.source
sacrebleu -t wmt19 -l $PAIR --echo ref > $DATA_DIR/val.target
echo $PAIR
PYTHONPATH="src:examples/seq2seq" python examples/seq2seq/run_eval.py facebook/wmt19-$PAIR $DATA_DIR/val.source $SAVE_DIR/test_translations.txt --reference_path $DATA_DIR/val.target --score_path $SAVE_DIR/test_bleu.json --bs $BS --task translation --num_beams $NUM_BEAMS
```
note: fairseq reports using a beam of 50, so you should get a slightly higher score if re-run with `--num_beams 50`.
## Data Sources
- [training, etc.](http://www.statmt.org/wmt19/)
- [test set](http://matrix.statmt.org/test_sets/newstest2019.tgz?1556572561)
### BibTeX entry and citation info
```bibtex
@inproceedings{{...,
year={{2020}},
title={{Facebook FAIR's WMT19 News Translation Task Submission}},
author={{Ng, Nathan and Yee, Kyra and Baevski, Alexei and Ott, Myle and Auli, Michael and Edunov, Sergey}},
booktitle={{Proc. of WMT}},
}}
```
## TODO
- port model ensemble (fairseq uses 4 model checkpoints)
"""
os.makedirs(model_card_dir, exist_ok=True)
path = os.path.join(model_card_dir, "README.md")
print(f"Generating {path}")
with open(path, "w", encoding="utf-8") as f:
f.write(readme)
# make sure we are under the root of the project
repo_dir = Path(__file__).resolve().parent.parent.parent
model_cards_dir = repo_dir / "model_cards"
for model_name in ["wmt19-ru-en", "wmt19-en-ru", "wmt19-en-de", "wmt19-de-en"]:
base, src_lang, tgt_lang = model_name.split("-")
model_card_dir = model_cards_dir / "facebook" / model_name
write_model_card(model_card_dir, src_lang=src_lang, tgt_lang=tgt_lang)
| 0
|
mavonic_private_repos/transformers/scripts
|
mavonic_private_repos/transformers/scripts/fsmt/tests-to-run.sh
|
#!/usr/bin/env bash
# Copyright 2020 The HuggingFace Team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# these scripts need to be run before any changes to FSMT-related code - it should cover all bases
CUDA_VISIBLE_DEVICES="" RUN_SLOW=1 pytest --disable-warnings tests/test_tokenization_fsmt.py tests/test_configuration_auto.py tests/test_modeling_fsmt.py examples/seq2seq/test_fsmt_bleu_score.py
RUN_SLOW=1 pytest --disable-warnings tests/test_tokenization_fsmt.py tests/test_configuration_auto.py tests/test_modeling_fsmt.py examples/seq2seq/test_fsmt_bleu_score.py
| 0
|
mavonic_private_repos/transformers/scripts
|
mavonic_private_repos/transformers/scripts/fsmt/fsmt-make-tiny-model.py
|
#!/usr/bin/env python
# coding: utf-8
# Copyright 2020 The HuggingFace Team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# This script creates a super tiny model that is useful inside tests, when we just want to test that
# the machinery works, without needing to the check the quality of the outcomes.
#
# This version creates a tiny model through reduction of a normal pre-trained model, but keeping the
# full vocab, merges file, and thus also resulting in a larger model due to a large vocab size.
# This gives ~3MB in total for all files.
#
# If you want a 50 times smaller than this see `fsmt-make-super-tiny-model.py`, which is slightly more complicated
#
#
# It will be used then as "stas/tiny-wmt19-en-de"
# Build
from transformers import FSMTConfig, FSMTForConditionalGeneration, FSMTTokenizer
mname = "facebook/wmt19-en-de"
tokenizer = FSMTTokenizer.from_pretrained(mname)
# get the correct vocab sizes, etc. from the master model
config = FSMTConfig.from_pretrained(mname)
config.update({
"d_model": 4,
"encoder_layers": 1, "decoder_layers": 1,
"encoder_ffn_dim": 4, "decoder_ffn_dim": 4,
"encoder_attention_heads": 1, "decoder_attention_heads": 1})
tiny_model = FSMTForConditionalGeneration(config)
print(f"num of params {tiny_model.num_parameters()}")
# Test
batch = tokenizer(["Making tiny model"], return_tensors="pt")
outputs = tiny_model(**batch)
print("test output:", len(outputs.logits[0]))
# Save
mname_tiny = "tiny-wmt19-en-de"
tiny_model.half() # makes it smaller
tiny_model.save_pretrained(mname_tiny)
tokenizer.save_pretrained(mname_tiny)
print(f"Generated {mname_tiny}")
# Upload
# transformers-cli upload tiny-wmt19-en-de
| 0
|
mavonic_private_repos/transformers/scripts
|
mavonic_private_repos/transformers/scripts/fsmt/eval-allenai-wmt16.sh
|
#!/usr/bin/env bash
# Copyright 2020 The HuggingFace Team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# this script evals the following fsmt models
# it covers:
# - allenai/wmt16-en-de-dist-12-1
# - allenai/wmt16-en-de-dist-6-1
# - allenai/wmt16-en-de-12-1
# this script needs to be run from the top level of the transformers repo
if [ ! -d "src/transformers" ]; then
echo "Error: This script needs to be run from the top of the transformers repo"
exit 1
fi
# In these scripts you may have to lower BS if you get CUDA OOM (or increase it if you have a large GPU)
### Normal eval ###
export PAIR=en-de
export DATA_DIR=data/$PAIR
export SAVE_DIR=data/$PAIR
export BS=64
export NUM_BEAMS=5
mkdir -p $DATA_DIR
sacrebleu -t wmt19 -l $PAIR --echo src > $DATA_DIR/val.source
sacrebleu -t wmt19 -l $PAIR --echo ref > $DATA_DIR/val.target
MODEL_PATH=allenai/wmt16-en-de-dist-12-1
echo $PAIR $MODEL_PATH
PYTHONPATH="src:examples/seq2seq" python examples/seq2seq/run_eval.py $MODEL_PATH $DATA_DIR/val.source $SAVE_DIR/test_translations.txt --reference_path $DATA_DIR/val.target --score_path $SAVE_DIR/test_bleu.json --bs $BS --task translation --num_beams $NUM_BEAMS
MODEL_PATH=allenai/wmt16-en-de-dist-6-1
echo $PAIR $MODEL_PATH
PYTHONPATH="src:examples/seq2seq" python examples/seq2seq/run_eval.py $MODEL_PATH $DATA_DIR/val.source $SAVE_DIR/test_translations.txt --reference_path $DATA_DIR/val.target --score_path $SAVE_DIR/test_bleu.json --bs $BS --task translation --num_beams $NUM_BEAMS
MODEL_PATH=allenai/wmt16-en-de-12-1
echo $PAIR $MODEL_PATH
PYTHONPATH="src:examples/seq2seq" python examples/seq2seq/run_eval.py $MODEL_PATH $DATA_DIR/val.source $SAVE_DIR/test_translations.txt --reference_path $DATA_DIR/val.target --score_path $SAVE_DIR/test_bleu.json --bs $BS --task translation --num_beams $NUM_BEAMS
### Searching hparams eval ###
export PAIR=en-de
export DATA_DIR=data/$PAIR
export SAVE_DIR=data/$PAIR
export BS=32
export NUM_BEAMS=5
mkdir -p $DATA_DIR
sacrebleu -t wmt19 -l $PAIR --echo src > $DATA_DIR/val.source
sacrebleu -t wmt19 -l $PAIR --echo ref > $DATA_DIR/val.target
MODEL_PATH=allenai/wmt16-en-de-dist-12-1
echo $PAIR $MODEL_PATH
PYTHONPATH="src:examples/seq2seq" python examples/seq2seq/run_eval_search.py $MODEL_PATH $DATA_DIR/val.source $SAVE_DIR/test_translations.txt --reference_path $DATA_DIR/val.target --score_path $SAVE_DIR/test_bleu.json --bs $BS --task translation --search="num_beams=5:10:15 length_penalty=0.6:0.7:0.8:0.9:1.0:1.1"
MODEL_PATH=allenai/wmt16-en-de-dist-6-1
echo $PAIR $MODEL_PATH
PYTHONPATH="src:examples/seq2seq" python examples/seq2seq/run_eval_search.py $MODEL_PATH $DATA_DIR/val.source $SAVE_DIR/test_translations.txt --reference_path $DATA_DIR/val.target --score_path $SAVE_DIR/test_bleu.json --bs $BS --task translation --search="num_beams=5:10:15 length_penalty=0.6:0.7:0.8:0.9:1.0:1.1"
MODEL_PATH=allenai/wmt16-en-de-12-1
echo $PAIR $MODEL_PATH
PYTHONPATH="src:examples/seq2seq" python examples/seq2seq/run_eval_search.py $MODEL_PATH $DATA_DIR/val.source $SAVE_DIR/test_translations.txt --reference_path $DATA_DIR/val.target --score_path $SAVE_DIR/test_bleu.json --bs $BS --task translation --search="num_beams=5:10:15 length_penalty=0.6:0.7:0.8:0.9:1.0:1.1"
| 0
|
mavonic_private_repos/transformers/scripts
|
mavonic_private_repos/transformers/scripts/fsmt/convert-allenai-wmt19.sh
|
#!/usr/bin/env bash
# Copyright 2020 The HuggingFace Team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# this script acquires data and converts it to fsmt model
# it covers:
# - allenai/wmt19-de-en-6-6-base
# - allenai/wmt19-de-en-6-6-big
# this script needs to be run from the top level of the transformers repo
if [ ! -d "src/transformers" ]; then
echo "Error: This script needs to be run from the top of the transformers repo"
exit 1
fi
mkdir data
# get data (run once)
cd data
gdown 'https://drive.google.com/uc?id=1j6z9fYdlUyOYsh7KJoumRlr1yHczxR5T'
gdown 'https://drive.google.com/uc?id=1yT7ZjqfvUYOBXvMjeY8uGRHQFWoSo8Q5'
gdown 'https://drive.google.com/uc?id=15gAzHeRUCs-QV8vHeTReMPEh1j8excNE'
tar -xvzf wmt19.de-en.tar.gz
tar -xvzf wmt19_deen_base_dr0.1_1.tar.gz
tar -xvzf wmt19_deen_big_dr0.1_2.tar.gz
cp wmt19.de-en/data-bin/dict.*.txt wmt19_deen_base_dr0.1_1
cp wmt19.de-en/data-bin/dict.*.txt wmt19_deen_big_dr0.1_2
cd -
# run conversions and uploads
PYTHONPATH="src" python src/transformers/convert_fsmt_original_pytorch_checkpoint_to_pytorch.py --fsmt_checkpoint_path data/wmt19_deen_base_dr0.1_1/checkpoint_last3_avg.pt --pytorch_dump_folder_path data/wmt19-de-en-6-6-base
PYTHONPATH="src" python src/transformers/convert_fsmt_original_pytorch_checkpoint_to_pytorch.py --fsmt_checkpoint_path data/wmt19_deen_big_dr0.1_2/checkpoint_last3_avg.pt --pytorch_dump_folder_path data/wmt19-de-en-6-6-big
# upload
cd data
transformers-cli upload -y wmt19-de-en-6-6-base
transformers-cli upload -y wmt19-de-en-6-6-big
cd -
# if updating just small files and not the large models, here is a script to generate the right commands:
perl -le 'for $f (@ARGV) { print qq[transformers-cli upload -y $_/$f --filename $_/$f] for ("wmt19-de-en-6-6-base", "wmt19-de-en-6-6-big")}' vocab-src.json vocab-tgt.json tokenizer_config.json config.json
# add/remove files as needed
| 0
|
mavonic_private_repos/transformers/scripts
|
mavonic_private_repos/transformers/scripts/fsmt/gen-card-allenai-wmt16.py
|
#!/usr/bin/env python
# Copyright 2020 The HuggingFace Team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# Usage:
# ./gen-card-allenai-wmt16.py
import os
from pathlib import Path
def write_model_card(model_card_dir, src_lang, tgt_lang, model_name):
texts = {
"en": "Machine learning is great, isn't it?",
"ru": "Машинное обучение - это здорово, не так ли?",
"de": "Maschinelles Lernen ist großartig, nicht wahr?",
}
# BLUE scores as follows:
# "pair": [fairseq, transformers]
scores = {
"wmt16-en-de-dist-12-1": [28.3, 27.52],
"wmt16-en-de-dist-6-1": [27.4, 27.11],
"wmt16-en-de-12-1": [26.9, 25.75],
}
pair = f"{src_lang}-{tgt_lang}"
readme = f"""
---
language:
- {src_lang}
- {tgt_lang}
thumbnail:
tags:
- translation
- wmt16
- allenai
license: apache-2.0
datasets:
- wmt16
metrics:
- bleu
---
# FSMT
## Model description
This is a ported version of fairseq-based [wmt16 transformer](https://github.com/jungokasai/deep-shallow/) for {src_lang}-{tgt_lang}.
For more details, please, see [Deep Encoder, Shallow Decoder: Reevaluating the Speed-Quality Tradeoff in Machine Translation](https://arxiv.org/abs/2006.10369).
All 3 models are available:
* [wmt16-en-de-dist-12-1](https://huggingface.co/allenai/wmt16-en-de-dist-12-1)
* [wmt16-en-de-dist-6-1](https://huggingface.co/allenai/wmt16-en-de-dist-6-1)
* [wmt16-en-de-12-1](https://huggingface.co/allenai/wmt16-en-de-12-1)
## Intended uses & limitations
#### How to use
```python
from transformers import FSMTForConditionalGeneration, FSMTTokenizer
mname = "allenai/{model_name}"
tokenizer = FSMTTokenizer.from_pretrained(mname)
model = FSMTForConditionalGeneration.from_pretrained(mname)
input = "{texts[src_lang]}"
input_ids = tokenizer.encode(input, return_tensors="pt")
outputs = model.generate(input_ids)
decoded = tokenizer.decode(outputs[0], skip_special_tokens=True)
print(decoded) # {texts[tgt_lang]}
```
#### Limitations and bias
## Training data
Pretrained weights were left identical to the original model released by allenai. For more details, please, see the [paper](https://arxiv.org/abs/2006.10369).
## Eval results
Here are the BLEU scores:
model | fairseq | transformers
-------|---------|----------
{model_name} | {scores[model_name][0]} | {scores[model_name][1]}
The score is slightly below the score reported in the paper, as the researchers don't use `sacrebleu` and measure the score on tokenized outputs. `transformers` score was measured using `sacrebleu` on detokenized outputs.
The score was calculated using this code:
```bash
git clone https://github.com/huggingface/transformers
cd transformers
export PAIR={pair}
export DATA_DIR=data/$PAIR
export SAVE_DIR=data/$PAIR
export BS=8
export NUM_BEAMS=5
mkdir -p $DATA_DIR
sacrebleu -t wmt16 -l $PAIR --echo src > $DATA_DIR/val.source
sacrebleu -t wmt16 -l $PAIR --echo ref > $DATA_DIR/val.target
echo $PAIR
PYTHONPATH="src:examples/seq2seq" python examples/seq2seq/run_eval.py allenai/{model_name} $DATA_DIR/val.source $SAVE_DIR/test_translations.txt --reference_path $DATA_DIR/val.target --score_path $SAVE_DIR/test_bleu.json --bs $BS --task translation --num_beams $NUM_BEAMS
```
## Data Sources
- [training, etc.](http://www.statmt.org/wmt16/)
- [test set](http://matrix.statmt.org/test_sets/newstest2016.tgz?1504722372)
### BibTeX entry and citation info
```
@misc{{kasai2020deep,
title={{Deep Encoder, Shallow Decoder: Reevaluating the Speed-Quality Tradeoff in Machine Translation}},
author={{Jungo Kasai and Nikolaos Pappas and Hao Peng and James Cross and Noah A. Smith}},
year={{2020}},
eprint={{2006.10369}},
archivePrefix={{arXiv}},
primaryClass={{cs.CL}}
}}
```
"""
model_card_dir.mkdir(parents=True, exist_ok=True)
path = os.path.join(model_card_dir, "README.md")
print(f"Generating {path}")
with open(path, "w", encoding="utf-8") as f:
f.write(readme)
# make sure we are under the root of the project
repo_dir = Path(__file__).resolve().parent.parent.parent
model_cards_dir = repo_dir / "model_cards"
for model_name in ["wmt16-en-de-dist-12-1", "wmt16-en-de-dist-6-1", "wmt16-en-de-12-1"]:
model_card_dir = model_cards_dir / "allenai" / model_name
write_model_card(model_card_dir, src_lang="en", tgt_lang="de", model_name=model_name)
| 0
|
mavonic_private_repos/transformers/scripts
|
mavonic_private_repos/transformers/scripts/fsmt/convert-allenai-wmt16.sh
|
#!/usr/bin/env bash
# Copyright 2020 The HuggingFace Team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# this script acquires data and converts it to fsmt model
# it covers:
# - allenai/wmt16-en-de-dist-12-1
# - allenai/wmt16-en-de-dist-6-1
# - allenai/wmt16-en-de-12-1
# this script needs to be run from the top level of the transformers repo
if [ ! -d "src/transformers" ]; then
echo "Error: This script needs to be run from the top of the transformers repo"
exit 1
fi
mkdir data
# get data (run once)
cd data
gdown 'https://drive.google.com/uc?id=1x_G2cjvM1nW5hjAB8-vWxRqtQTlmIaQU'
gdown 'https://drive.google.com/uc?id=1oA2aqZlVNj5FarxBlNXEHpBS4lRetTzU'
gdown 'https://drive.google.com/uc?id=1Wup2D318QYBFPW_NKI1mfP_hXOfmUI9r'
tar -xvzf trans_ende_12-1_0.2.tar.gz
tar -xvzf trans_ende-dist_12-1_0.2.tar.gz
tar -xvzf trans_ende-dist_6-1_0.2.tar.gz
gdown 'https://drive.google.com/uc?id=1mNufoynJ9-Zy1kJh2TA_lHm2squji0i9'
gdown 'https://drive.google.com/uc?id=1iO7um-HWoNoRKDtw27YUSgyeubn9uXqj'
tar -xvzf wmt16.en-de.deep-shallow.dist.tar.gz
tar -xvzf wmt16.en-de.deep-shallow.tar.gz
cp wmt16.en-de.deep-shallow/data-bin/dict.*.txt trans_ende_12-1_0.2
cp wmt16.en-de.deep-shallow.dist/data-bin/dict.*.txt trans_ende-dist_12-1_0.2
cp wmt16.en-de.deep-shallow.dist/data-bin/dict.*.txt trans_ende-dist_6-1_0.2
cp wmt16.en-de.deep-shallow/bpecodes trans_ende_12-1_0.2
cp wmt16.en-de.deep-shallow.dist/bpecodes trans_ende-dist_12-1_0.2
cp wmt16.en-de.deep-shallow.dist/bpecodes trans_ende-dist_6-1_0.2
cd -
# run conversions and uploads
PYTHONPATH="src" python src/transformers/convert_fsmt_original_pytorch_checkpoint_to_pytorch.py --fsmt_checkpoint_path data/trans_ende-dist_12-1_0.2/checkpoint_top5_average.pt --pytorch_dump_folder_path data/wmt16-en-de-dist-12-1
PYTHONPATH="src" python src/transformers/convert_fsmt_original_pytorch_checkpoint_to_pytorch.py --fsmt_checkpoint_path data/trans_ende-dist_6-1_0.2/checkpoint_top5_average.pt --pytorch_dump_folder_path data/wmt16-en-de-dist-6-1
PYTHONPATH="src" python src/transformers/convert_fsmt_original_pytorch_checkpoint_to_pytorch.py --fsmt_checkpoint_path data/trans_ende_12-1_0.2/checkpoint_top5_average.pt --pytorch_dump_folder_path data/wmt16-en-de-12-1
# upload
cd data
transformers-cli upload -y wmt16-en-de-dist-12-1
transformers-cli upload -y wmt16-en-de-dist-6-1
transformers-cli upload -y wmt16-en-de-12-1
cd -
# if updating just small files and not the large models, here is a script to generate the right commands:
perl -le 'for $f (@ARGV) { print qq[transformers-cli upload -y $_/$f --filename $_/$f] for ("wmt16-en-de-dist-12-1", "wmt16-en-de-dist-6-1", "wmt16-en-de-12-1")}' vocab-src.json vocab-tgt.json tokenizer_config.json config.json
# add/remove files as needed
| 0
|
mavonic_private_repos/transformers/scripts
|
mavonic_private_repos/transformers/scripts/fsmt/eval-allenai-wmt19.sh
|
#!/usr/bin/env bash
# Copyright 2020 The HuggingFace Team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# this script evals the following fsmt models
# it covers:
# - allenai/wmt19-de-en-6-6-base
# - allenai/wmt19-de-en-6-6-big
# this script needs to be run from the top level of the transformers repo
if [ ! -d "src/transformers" ]; then
echo "Error: This script needs to be run from the top of the transformers repo"
exit 1
fi
# In these scripts you may have to lower BS if you get CUDA OOM (or increase it if you have a large GPU)
### Normal eval ###
export PAIR=de-en
export DATA_DIR=data/$PAIR
export SAVE_DIR=data/$PAIR
export BS=64
export NUM_BEAMS=5
mkdir -p $DATA_DIR
sacrebleu -t wmt19 -l $PAIR --echo src > $DATA_DIR/val.source
sacrebleu -t wmt19 -l $PAIR --echo ref > $DATA_DIR/val.target
MODEL_PATH=allenai/wmt19-de-en-6-6-base
echo $PAIR $MODEL_PATH
PYTHONPATH="src:examples/seq2seq" python examples/seq2seq/run_eval.py $MODEL_PATH $DATA_DIR/val.source $SAVE_DIR/test_translations.txt --reference_path $DATA_DIR/val.target --score_path $SAVE_DIR/test_bleu.json --bs $BS --task translation --num_beams $NUM_BEAMS
MODEL_PATH=allenai/wmt19-de-en-6-6-big
echo $PAIR $MODEL_PATH
PYTHONPATH="src:examples/seq2seq" python examples/seq2seq/run_eval.py $MODEL_PATH $DATA_DIR/val.source $SAVE_DIR/test_translations.txt --reference_path $DATA_DIR/val.target --score_path $SAVE_DIR/test_bleu.json --bs $BS --task translation --num_beams $NUM_BEAMS
### Searching hparams eval ###
export PAIR=de-en
export DATA_DIR=data/$PAIR
export SAVE_DIR=data/$PAIR
export BS=16
export NUM_BEAMS=5
mkdir -p $DATA_DIR
sacrebleu -t wmt19 -l $PAIR --echo src > $DATA_DIR/val.source
sacrebleu -t wmt19 -l $PAIR --echo ref > $DATA_DIR/val.target
MODEL_PATH=allenai/wmt19-de-en-6-6-base
echo $PAIR $MODEL_PATH
PYTHONPATH="src:examples/seq2seq" python examples/seq2seq/run_eval_search.py $MODEL_PATH $DATA_DIR/val.source $SAVE_DIR/test_translations.txt --reference_path $DATA_DIR/val.target --score_path $SAVE_DIR/test_bleu.json --bs $BS --task translation --search="num_beams=5:10:15 length_penalty=0.6:0.7:0.8:0.9:1.0:1.1"
MODEL_PATH=allenai/wmt19-de-en-6-6-big
echo $PAIR $MODEL_PATH
PYTHONPATH="src:examples/seq2seq" python examples/seq2seq/run_eval_search.py $MODEL_PATH $DATA_DIR/val.source $SAVE_DIR/test_translations.txt --reference_path $DATA_DIR/val.target --score_path $SAVE_DIR/test_bleu.json --bs $BS --task translation --search="num_beams=5:10:15 length_penalty=0.6:0.7:0.8:0.9:1.0:1.1"
| 0
|
mavonic_private_repos/transformers/scripts
|
mavonic_private_repos/transformers/scripts/fsmt/convert-facebook-wmt19.sh
|
#!/usr/bin/env bash
# Copyright 2020 The HuggingFace Team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# this script acquires data and converts it to fsmt model
# it covers:
# - facebook/wmt19-ru-en
# - facebook/wmt19-en-ru
# - facebook/wmt19-de-en
# - facebook/wmt19-en-de
# this script needs to be run from the top level of the transformers repo
if [ ! -d "src/transformers" ]; then
echo "Error: This script needs to be run from the top of the transformers repo"
exit 1
fi
mkdir data
# get data (run once)
cd data
wget https://dl.fbaipublicfiles.com/fairseq/models/wmt19.en-de.joined-dict.ensemble.tar.gz
wget https://dl.fbaipublicfiles.com/fairseq/models/wmt19.de-en.joined-dict.ensemble.tar.gz
wget https://dl.fbaipublicfiles.com/fairseq/models/wmt19.en-ru.ensemble.tar.gz
wget https://dl.fbaipublicfiles.com/fairseq/models/wmt19.ru-en.ensemble.tar.gz
tar -xvzf wmt19.en-de.joined-dict.ensemble.tar.gz
tar -xvzf wmt19.de-en.joined-dict.ensemble.tar.gz
tar -xvzf wmt19.en-ru.ensemble.tar.gz
tar -xvzf wmt19.ru-en.ensemble.tar.gz
cd -
# run conversions and uploads
export PAIR=ru-en
PYTHONPATH="src" python src/transformers/convert_fsmt_original_pytorch_checkpoint_to_pytorch.py --fsmt_checkpoint_path data/wmt19.$PAIR.ensemble/model4.pt --pytorch_dump_folder_path data/wmt19-$PAIR
export PAIR=en-ru
PYTHONPATH="src" python src/transformers/convert_fsmt_original_pytorch_checkpoint_to_pytorch.py --fsmt_checkpoint_path data/wmt19.$PAIR.ensemble/model4.pt --pytorch_dump_folder_path data/wmt19-$PAIR
export PAIR=de-en
PYTHONPATH="src" python src/transformers/convert_fsmt_original_pytorch_checkpoint_to_pytorch.py --fsmt_checkpoint_path data/wmt19.$PAIR.joined-dict.ensemble/model4.pt --pytorch_dump_folder_path data/wmt19-$PAIR
export PAIR=en-de
PYTHONPATH="src" python src/transformers/convert_fsmt_original_pytorch_checkpoint_to_pytorch.py --fsmt_checkpoint_path data/wmt19.$PAIR.joined-dict.ensemble/model4.pt --pytorch_dump_folder_path data/wmt19-$PAIR
# upload
cd data
transformers-cli upload -y wmt19-ru-en
transformers-cli upload -y wmt19-en-ru
transformers-cli upload -y wmt19-de-en
transformers-cli upload -y wmt19-en-de
cd -
# if updating just small files and not the large models, here is a script to generate the right commands:
perl -le 'for $f (@ARGV) { print qq[transformers-cli upload -y $_/$f --filename $_/$f] for map { "wmt19-$_" } ("en-ru", "ru-en", "de-en", "en-de")}' vocab-src.json vocab-tgt.json tokenizer_config.json config.json
# add/remove files as needed
| 0
|
mavonic_private_repos/transformers/scripts
|
mavonic_private_repos/transformers/scripts/pegasus/build_test_sample_spm_no_bos.py
|
#!/usr/bin/env python
# Copyright 2020 The HuggingFace Team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# this script builds a small sample spm file tests/fixtures/test_sentencepiece_no_bos.model, with features needed by pegasus
# 1. pip install sentencepiece
#
# 2. wget https://raw.githubusercontent.com/google/sentencepiece/master/data/botchan.txt
# 3. build
import sentencepiece as spm
# pegasus:
# 1. no bos
# 2. eos_id is 1
# 3. unk_id is 2
# build a sample spm file accordingly
spm.SentencePieceTrainer.train('--input=botchan.txt --model_prefix=test_sentencepiece_no_bos --bos_id=-1 --unk_id=2 --eos_id=1 --vocab_size=1000')
# 4. now update the fixture
# mv test_sentencepiece_no_bos.model ../../tests/fixtures/
| 0
|
mavonic_private_repos/transformers/scripts
|
mavonic_private_repos/transformers/scripts/tatoeba/upload_models.sh
|
#!/bin/bash
for FILE in converted/*; do
model_name=`basename $FILE`
huggingface-cli repo create $model_name -y
git clone https://huggingface.co/Helsinki-NLP/$model_name
mv $FILE/* $model_name/
cd $model_name
git add . && git commit -m "initial commit"
git push
cd ..
done
| 0
|
mavonic_private_repos/transformers/scripts
|
mavonic_private_repos/transformers/scripts/tatoeba/README.md
|
<!---
Copyright 2020 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
-->
Setup transformers following instructions in README.md, (I would fork first).
```bash
git clone git@github.com:huggingface/transformers.git
cd transformers
pip install -e .
pip install pandas GitPython wget
```
Get required metadata
```bash
curl https://cdn-datasets.huggingface.co/language_codes/language-codes-3b2.csv > language-codes-3b2.csv
curl https://cdn-datasets.huggingface.co/language_codes/iso-639-3.csv > iso-639-3.csv
```
Install Tatoeba-Challenge repo inside transformers
```bash
git clone git@github.com:Helsinki-NLP/Tatoeba-Challenge.git
```
To convert a few models, call the conversion script from command line:
```bash
python src/transformers/models/marian/convert_marian_tatoeba_to_pytorch.py --models heb-eng eng-heb --save_dir converted
```
To convert lots of models you can pass your list of Tatoeba model names to `resolver.convert_models` in a python client or script.
```python
from transformers.convert_marian_tatoeba_to_pytorch import TatoebaConverter
resolver = TatoebaConverter(save_dir='converted')
resolver.convert_models(['heb-eng', 'eng-heb'])
```
### Upload converted models
Since version v3.5.0, the model sharing workflow is switched to git-based system . Refer to [model sharing doc](https://huggingface.co/transformers/main/model_sharing.html#model-sharing-and-uploading) for more details.
To upload all converted models,
1. Install [git-lfs](https://git-lfs.github.com/).
2. Login to `huggingface-cli`
```bash
huggingface-cli login
```
3. Run the `upload_models` script
```bash
./scripts/tatoeba/upload_models.sh
```
### Modifications
- To change naming logic, change the code near `os.rename`. The model card creation code may also need to change.
- To change model card content, you must modify `TatoebaCodeResolver.write_model_card`
| 0
|
mavonic_private_repos/transformers/scripts
|
mavonic_private_repos/transformers/scripts/distributed/torch-distributed-gpu-test.py
|
#!/usr/bin/env python
#
# This a `torch.distributed` diagnostics script that checks that all GPUs in the cluster (one or
# many nodes) can talk to each other via nccl and allocate gpu memory.
#
# To run first adjust the number of processes and nodes:
#
# python -m torch.distributed.run --nproc_per_node 2 --nnodes 1 torch-distributed-gpu-test.py
#
# You may need to add --master_addr $MASTER_ADDR --master_port $MASTER_PORT if using a custom addr:port
#
# You can also use the rdzv API: --rdzv_endpoint $MASTER_ADDR:$MASTER_PORT --rdzv_backend c10d
#
# use torch.distributed.launch instead of torch.distributed.run for torch < 1.9
#
# If you get a hanging in `barrier` calls you have some network issues, you may try to debug this with:
#
# NCCL_DEBUG=INFO python -m torch.distributed.run --nproc_per_node 2 --nnodes 1 torch-distributed-gpu-test.py
#
# which should tell you what's going on behind the scenes.
#
#
# This script can be run via `srun` in the SLURM environment as well. Here is a SLURM script that
# runs on 2 nodes of 4 gpus per node:
#
# #SBATCH --job-name=test-nodes # name
# #SBATCH --nodes=2 # nodes
# #SBATCH --ntasks-per-node=1 # crucial - only 1 task per dist per node!
# #SBATCH --cpus-per-task=10 # number of cores per tasks
# #SBATCH --gres=gpu:4 # number of gpus
# #SBATCH --time 0:05:00 # maximum execution time (HH:MM:SS)
# #SBATCH --output=%x-%j.out # output file name
#
# GPUS_PER_NODE=4
# MASTER_ADDR=$(scontrol show hostnames $SLURM_JOB_NODELIST | head -n 1)
# MASTER_PORT=6000
#
# srun --jobid $SLURM_JOBID bash -c 'python -m torch.distributed.run \
# --nproc_per_node $GPUS_PER_NODE --nnodes $SLURM_NNODES --node_rank $SLURM_PROCID \
# --master_addr $MASTER_ADDR --master_port $MASTER_PORT \
# torch-distributed-gpu-test.py'
#
import fcntl
import os
import socket
import torch
import torch.distributed as dist
def printflock(*msgs):
"""solves multi-process interleaved print problem"""
with open(__file__, "r") as fh:
fcntl.flock(fh, fcntl.LOCK_EX)
try:
print(*msgs)
finally:
fcntl.flock(fh, fcntl.LOCK_UN)
local_rank = int(os.environ["LOCAL_RANK"])
torch.cuda.set_device(local_rank)
device = torch.device("cuda", local_rank)
hostname = socket.gethostname()
gpu = f"[{hostname}-{local_rank}]"
try:
# test distributed
dist.init_process_group("nccl")
dist.all_reduce(torch.ones(1).to(device), op=dist.ReduceOp.SUM)
dist.barrier()
# test cuda is available and can allocate memory
torch.cuda.is_available()
torch.ones(1).cuda(local_rank)
# global rank
rank = dist.get_rank()
world_size = dist.get_world_size()
printflock(f"{gpu} is OK (global rank: {rank}/{world_size})")
dist.barrier()
if rank == 0:
printflock(f"pt={torch.__version__}, cuda={torch.version.cuda}, nccl={torch.cuda.nccl.version()}")
except Exception:
printflock(f"{gpu} is broken")
raise
| 0
|
mavonic_private_repos/transformers/templates
|
mavonic_private_repos/transformers/templates/adding_a_new_example_script/cookiecutter.json
|
{
"example_name": "text classification",
"directory_name": "{{cookiecutter.example_name|lower|replace(' ', '-')}}",
"example_shortcut": "{{cookiecutter.directory_name}}",
"model_class": "AutoModel",
"authors": "The HuggingFace Team",
"can_train_from_scratch": ["True", "False"],
"with_trainer": ["True", "False"]
}
| 0
|
mavonic_private_repos/transformers/templates
|
mavonic_private_repos/transformers/templates/adding_a_new_example_script/README.md
|
<!---
Copyright 2020 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
-->
# How to add a new example script in 🤗 Transformers
This folder provide a template for adding a new example script implementing a training or inference task with the
models in the 🤗 Transformers library. To use it, you will need to install cookiecutter:
```bash
pip install cookiecutter
```
or refer to the installation page of the [cookiecutter documentation](https://cookiecutter.readthedocs.io/).
You can then run the following command inside the `examples` folder of the transformers repo:
```bash
cookiecutter ../templates/adding_a_new_example_script/
```
and answer the questions asked, which will generate a new folder where you will find a pre-filled template for your
example following the best practices we recommend for them.
Adjust the way the data is preprocessed, the model is loaded or the Trainer is instantiated then when you're happy, add
a `README.md` in the folder (or complete the existing one if you added a script to an existing folder) telling a user
how to run your script.
Make a PR to the 🤗 Transformers repo. Don't forget to tweet about your new example with a carbon screenshot of how to
run it and tag @huggingface!
| 0
|
mavonic_private_repos/transformers/templates/adding_a_new_example_script
|
mavonic_private_repos/transformers/templates/adding_a_new_example_script/{{cookiecutter.directory_name}}/run_{{cookiecutter.example_shortcut}}.py
|
#!/usr/bin/env python
# coding=utf-8
# Copyright 2022 {{cookiecutter.authors}} and The HuggingFace Inc. team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""
Fine-tuning a 🤗 Transformers model on {{cookiecutter.example_name}}.
"""
# You can also adapt this script on your own {{cookiecutter.example_name}} task. Pointers for this are left as comments.
{%- if cookiecutter.with_trainer == "True" %}
import logging
import math
import os
import sys
from dataclasses import dataclass, field
from typing import Optional, List
import datasets
import torch
from datasets import load_dataset
import transformers
from transformers import (
CONFIG_MAPPING,
MODEL_MAPPING,
AutoConfig,
{{cookiecutter.model_class}},
AutoTokenizer,
DataCollatorWithPadding,
HfArgumentParser,
Trainer,
TrainingArguments,
default_data_collator,
set_seed,
)
from transformers.trainer_utils import get_last_checkpoint
from transformers.utils import send_example_telemetry
logger = logging.getLogger(__name__)
{%- if cookiecutter.can_train_from_scratch == "True" %}
# You should update this to your particular problem to have better documentation of `model_type`
MODEL_CONFIG_CLASSES = list(MODEL_MAPPING.keys())
MODEL_TYPES = tuple(conf.model_type for conf in MODEL_CONFIG_CLASSES)
@dataclass
class ModelArguments:
"""
Arguments pertaining to which model/config/tokenizer we are going to fine-tune, or train from scratch.
"""
model_name_or_path: Optional[str] = field(
default=None,
metadata={
"help": "The model checkpoint for weights initialization. "
"Don't set if you want to train a model from scratch."
},
)
model_type: Optional[str] = field(
default=None,
metadata={"help": "If training from scratch, pass a model type from the list: " + ", ".join(MODEL_TYPES)},
)
config_name: Optional[str] = field(
default=None, metadata={"help": "Pretrained config name or path if not the same as model_name"}
)
tokenizer_name: Optional[str] = field(
default=None, metadata={"help": "Pretrained tokenizer name or path if not the same as model_name"}
)
cache_dir: Optional[str] = field(
default=None, metadata={"help": "Where do you want to store the pretrained models downloaded from huggingface.co"}
)
use_fast_tokenizer: bool = field(
default=True,
metadata={"help": "Whether to use one of the fast tokenizer (backed by the tokenizers library) or not."},
)
{%- elif cookiecutter.can_train_from_scratch == "False" %}
@dataclass
class ModelArguments:
"""
Arguments pertaining to which model/config/tokenizer we are going to fine-tune from.
"""
model_name_or_path: str = field(
metadata={"help": "Path to pretrained model or model identifier from huggingface.co/models"}
)
config_name: Optional[str] = field(
default=None, metadata={"help": "Pretrained config name or path if not the same as model_name"}
)
tokenizer_name: Optional[str] = field(
default=None, metadata={"help": "Pretrained tokenizer name or path if not the same as model_name"}
)
cache_dir: Optional[str] = field(
default=None, metadata={"help": "Where do you want to store the pretrained models downloaded from huggingface.co"}
)
use_fast_tokenizer: bool = field(
default=True,
metadata={"help": "Whether to use one of the fast tokenizer (backed by the tokenizers library) or not."},
)
model_revision: str = field(
default="main",
metadata={"help": "The specific model version to use (can be a branch name, tag name or commit id)."},
)
token: str = field(
default=None,
metadata={
"help": (
"The token to use as HTTP bearer authorization for remote files. If not specified, will use the token "
"generated when running `huggingface-cli login` (stored in `~/.huggingface`)."
)
},
)
trust_remote_code: bool = field(
default=False,
metadata={
"help": (
"Whether or not to allow for custom models defined on the Hub in their own modeling files. This option "
"should only be set to `True` for repositories you trust and in which you have read the code, as it will "
"execute code present on the Hub on your local machine."
)
},
)
{% endif %}
@dataclass
class DataTrainingArguments:
"""
Arguments pertaining to what data we are going to input our model for training and eval.
"""
dataset_name: Optional[str] = field(
default=None, metadata={"help": "The name of the dataset to use (via the datasets library)."}
)
dataset_config_name: Optional[str] = field(
default=None, metadata={"help": "The configuration name of the dataset to use (via the datasets library)."}
)
train_file: Optional[str] = field(default=None, metadata={"help": "The input training data file (a text file)."})
validation_file: Optional[str] = field(
default=None,
metadata={"help": "An optional input evaluation data file to evaluate the perplexity on (a text file)."},
)
test_file: Optional[str] = field(
default=None,
metadata={"help": "An optional input test data file to predict the label on (a text file)."},
)
overwrite_cache: bool = field(
default=False, metadata={"help": "Overwrite the cached training and evaluation sets"}
)
preprocessing_num_workers: Optional[int] = field(
default=None,
metadata={"help": "The number of processes to use for the preprocessing."},
)
max_train_samples: Optional[int] = field(
default=None,
metadata={
"help": "For debugging purposes or quicker training, truncate the number of training examples to this "
"value if set."
},
)
max_eval_samples: Optional[int] = field(
default=None,
metadata={
"help": "For debugging purposes or quicker training, truncate the number of evaluation examples to this "
"value if set."
},
)
max_predict_samples: Optional[int] = field(
default=None,
metadata={
"help": "For debugging purposes or quicker training, truncate the number of prediction examples to this "
"value if set."
},
)
def __post_init__(self):
if (
self.dataset_name is None
and self.train_file is None
and self.validation_file is None
and self.test_file is None
):
raise ValueError("Need either a dataset name or a training/validation/test file.")
else:
if self.train_file is not None:
extension = self.train_file.split(".")[-1]
assert extension in ["csv", "json", "txt"], "`train_file` should be a csv, a json or a txt file."
if self.validation_file is not None:
extension = self.validation_file.split(".")[-1]
assert extension in ["csv", "json", "txt"], "`validation_file` should be a csv, a json or a txt file."
if self.test_file is not None:
extension = self.test_file.split(".")[-1]
assert extension in ["csv", "json", "txt"], "`test_file` should be a csv, a json or a txt file."
def main():
# See all possible arguments in src/transformers/training_args.py
# or by passing the --help flag to this script.
# We now keep distinct sets of args, for a cleaner separation of concerns.
parser = HfArgumentParser((ModelArguments, DataTrainingArguments, TrainingArguments))
if len(sys.argv) == 2 and sys.argv[1].endswith(".json"):
# If we pass only one argument to the script and it's the path to a json file,
# let's parse it to get our arguments.
model_args, data_args, training_args = parser.parse_json_file(json_file=os.path.abspath(sys.argv[1]))
else:
model_args, data_args, training_args = parser.parse_args_into_dataclasses()
# Sending telemetry. Tracking the example usage helps us better allocate resources to maintain them. The
# information sent is the one passed as arguments along with your Python/PyTorch versions.
send_example_telemetry("run_{{cookiecutter.example_shortcut}}", model_args, data_args)
# Detecting last checkpoint.
last_checkpoint = None
if os.path.isdir(training_args.output_dir) and training_args.do_train and not training_args.overwrite_output_dir:
last_checkpoint = get_last_checkpoint(training_args.output_dir)
if last_checkpoint is None and len(os.listdir(training_args.output_dir)) > 0:
raise ValueError(
f"Output directory ({training_args.output_dir}) already exists and is not empty. "
"Use --overwrite_output_dir to overcome."
)
elif last_checkpoint is not None:
logger.info(
f"Checkpoint detected, resuming training at {last_checkpoint}. To avoid this behavior, change "
"the `--output_dir` or add `--overwrite_output_dir` to train from scratch."
)
# Setup logging
logging.basicConfig(
format="%(asctime)s - %(levelname)s - %(name)s - %(message)s",
datefmt="%m/%d/%Y %H:%M:%S",
handlers=[logging.StreamHandler(sys.stdout)],
)
log_level = training_args.get_process_log_level()
logger.setLevel(log_level)
datasets.utils.logging.set_verbosity(log_level)
transformers.utils.logging.set_verbosity(log_level)
transformers.utils.logging.enable_default_handler()
transformers.utils.logging.enable_explicit_format()
# Log on each process the small summary:
logger.warning(
f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}"
+ f"distributed training: {bool(training_args.local_rank != -1)}, 16-bits training: {training_args.fp16}"
)
logger.info(f"Training/evaluation parameters {training_args}")
# Set seed before initializing model.
set_seed(training_args.seed)
# Get the datasets: you can either provide your own CSV/JSON/TXT training and evaluation files (see below)
# or just provide the name of one of the public datasets available on the hub at https://huggingface.co/datasets/
# (the dataset will be downloaded automatically from the datasets Hub).
#
# For CSV/JSON files, this script will use the column called 'text' or the first column if no column called
# 'text' is found. You can easily tweak this behavior (see below).
#
# In distributed training, the load_dataset function guarantee that only one local process can concurrently
# download the dataset.
if data_args.dataset_name is not None:
# Downloading and loading a dataset from the hub.
raw_datasets = load_dataset(data_args.dataset_name, data_args.dataset_config_name)
else:
data_files = {}
if data_args.train_file is not None:
data_files["train"] = data_args.train_file
extension = data_args.train_file.split(".")[-1]
if data_args.validation_file is not None:
data_files["validation"] = data_args.validation_file
extension = data_args.validation_file.split(".")[-1]
if data_args.test_file is not None:
data_files["test"] = data_args.test_file
extension = data_args.test_file.split(".")[-1]
if extension == "txt":
extension = "text"
raw_datasets = load_dataset(extension, data_files=data_files)
# See more about loading any type of standard or custom dataset (from files, python dict, pandas DataFrame, etc) at
# https://huggingface.co/docs/datasets/loading_datasets.
# Load pretrained model and tokenizer
#
# Distributed training:
# The .from_pretrained methods guarantee that only one local process can concurrently
# download model & vocab.
{%- if cookiecutter.can_train_from_scratch == "True" %}
config_kwargs = {
"cache_dir": model_args.cache_dir,
"revision": model_args.model_revision,
"token": model_args.token,
"trust_remote_code": model_args.trust_remote_code,
}
if model_args.config_name:
config = AutoConfig.from_pretrained(model_args.config_name, **config_kwargs)
elif model_args.model_name_or_path:
config = AutoConfig.from_pretrained(model_args.model_name_or_path, **config_kwargs)
else:
config = CONFIG_MAPPING[model_args.model_type]()
logger.warning("You are instantiating a new config instance from scratch.")
tokenizer_kwargs = {
"cache_dir": model_args.cache_dir,
"use_fast": model_args.use_fast_tokenizer,
"revision": model_args.model_revision,
"token": model_args.token,
"trust_remote_code": model_args.trust_remote_code,
}
if model_args.tokenizer_name:
tokenizer = AutoTokenizer.from_pretrained(model_args.tokenizer_name, **tokenizer_kwargs)
elif model_args.model_name_or_path:
tokenizer = AutoTokenizer.from_pretrained(model_args.model_name_or_path, **tokenizer_kwargs)
else:
raise ValueError(
"You are instantiating a new tokenizer from scratch. This is not supported by this script. "
"You can do it from another script, save it, and load it from here, using --tokenizer_name."
)
if model_args.model_name_or_path:
model = {{cookiecutter.model_class}}.from_pretrained(
model_args.model_name_or_path,
from_tf=bool(".ckpt" in model_args.model_name_or_path),
config=config,
cache_dir=model_args.cache_dir,
revision=model_args.model_revision,
token=model_args.token,
trust_remote_code=model_args.trust_remote_code,
)
else:
logger.info("Training new model from scratch")
model = {{cookiecutter.model_class}}.from_config(config)
model.resize_token_embeddings(len(tokenizer))
{%- elif cookiecutter.can_train_from_scratch == "False" %}
config = AutoConfig.from_pretrained(
model_args.config_name if model_args.config_name else model_args.model_name_or_path,
# num_labels=num_labels, Uncomment if you have a certain number of labels
finetuning_task=data_args.task_name,
cache_dir=model_args.cache_dir,
revision=model_args.model_revision,
token=model_args.token,
trust_remote_code=model_args.trust_remote_code,
)
tokenizer = AutoTokenizer.from_pretrained(
model_args.tokenizer_name if model_args.tokenizer_name else model_args.model_name_or_path,
cache_dir=model_args.cache_dir,
use_fast=model_args.use_fast_tokenizer,
revision=model_args.model_revision,
token=model_args.token,
trust_remote_code=model_args.trust_remote_code,
)
model = AutoModelForSequenceClassification.from_pretrained(
model_args.model_name_or_path,
from_tf=bool(".ckpt" in model_args.model_name_or_path),
config=config,
cache_dir=model_args.cache_dir,
revision=model_args.model_revision,
token=model_args.token,
trust_remote_code=model_args.trust_remote_code,
)
{% endif %}
# Preprocessing the datasets.
# First we tokenize all the texts.
if training_args.do_train:
column_names = raw_datasets["train"].column_names
elif training_args.do_eval:
column_names = raw_datasets["validation"].column_names
elif training_args.do_predict:
column_names = raw_datasets["test"].column_names
text_column_name = "text" if "text" in column_names else column_names[0]
def tokenize_function(examples):
return tokenizer(examples[text_column_name], padding="max_length", truncation=True)
if training_args.do_train:
if "train" not in raw_datasets:
raise ValueError("--do_train requires a train dataset")
train_dataset = raw_datasets["train"]
if data_args.max_train_samples is not None:
# Select Sample from Dataset
train_dataset = train_dataset.select(range(data_args.max_train_samples))
# tokenize train dataset in batch
with training_args.main_process_first(desc="train dataset map tokenization"):
train_dataset = train_dataset.map(
tokenize_function,
batched=True,
num_proc=data_args.preprocessing_num_workers,
remove_columns=[text_column_name],
load_from_cache_file=not data_args.overwrite_cache,
)
if training_args.do_eval:
if "validation" not in raw_datasets:
raise ValueError("--do_eval requires a validation dataset")
eval_dataset = raw_datasets["validation"]
# Selecting samples from dataset
if data_args.max_eval_samples is not None:
eval_dataset = eval_dataset.select(range(data_args.max_eval_samples))
# tokenize validation dataset
with training_args.main_process_first(desc="validation dataset map tokenization"):
eval_dataset = eval_dataset.map(
tokenize_function,
batched=True,
num_proc=data_args.preprocessing_num_workers,
remove_columns=[text_column_name],
load_from_cache_file=not data_args.overwrite_cache,
)
if training_args.do_predict:
if "test" not in raw_datasets:
raise ValueError("--do_predict requires a test dataset")
predict_dataset = raw_datasets["test"]
# Selecting samples from dataset
if data_args.max_predict_samples is not None:
predict_dataset = predict_dataset.select(range(data_args.max_predict_samples))
# tokenize predict dataset
with training_args.main_process_first(desc="prediction dataset map tokenization"):
predict_dataset = predict_dataset.map(
tokenize_function,
batched=True,
num_proc=data_args.preprocessing_num_workers,
remove_columns=[text_column_name],
load_from_cache_file=not data_args.overwrite_cache,
)
# Data collator
data_collator=default_data_collator if not training_args.fp16 else DataCollatorWithPadding(tokenizer, pad_to_multiple_of=8)
# Initialize our Trainer
trainer = Trainer(
model=model,
args=training_args,
train_dataset=train_dataset if training_args.do_train else None,
eval_dataset=eval_dataset if training_args.do_eval else None,
tokenizer=tokenizer,
data_collator=data_collator,
)
# Training
if training_args.do_train:
{%- if cookiecutter.can_train_from_scratch == "False" %}
if last_checkpoint is not None:
checkpoint = last_checkpoint
elif os.path.isdir(model_args.model_name_or_path):
checkpoint = model_args.model_name_or_path
else:
checkpoint = None
{%- elif cookiecutter.can_train_from_scratch == "True" %}
if last_checkpoint is not None:
checkpoint = last_checkpoint
elif model_args.model_name_or_path is not None and os.path.isdir(model_args.model_name_or_path):
checkpoint = model_args.model_name_or_path
else:
checkpoint = None
{% endif %}
train_result = trainer.train(resume_from_checkpoint=checkpoint)
trainer.save_model() # Saves the tokenizer too for easy upload
metrics = train_result.metrics
max_train_samples = (
data_args.max_train_samples if data_args.max_train_samples is not None else len(train_dataset)
)
metrics["train_samples"] = min(max_train_samples, len(train_dataset))
trainer.log_metrics("train", metrics)
trainer.save_metrics("train", metrics)
trainer.save_state()
# Evaluation
if training_args.do_eval:
logger.info("*** Evaluate ***")
metrics = trainer.evaluate()
max_eval_samples = data_args.max_eval_samples if data_args.max_eval_samples is not None else len(eval_dataset)
metrics["eval_samples"] = min(max_eval_samples, len(eval_dataset))
trainer.log_metrics("eval", metrics)
trainer.save_metrics("eval", metrics)
# Prediction
if training_args.do_predict:
logger.info("*** Predict ***")
predictions, labels, metrics = trainer.predict(predict_dataset)
max_predict_samples = data_args.max_predict_samples if data_args.max_predict_samples is not None else len(predict_dataset)
metrics["predict_samples"] = min(max_predict_samples, len(predict_dataset))
trainer.log_metrics("predict", metrics)
trainer.save_metrics("predict", metrics)
# write custom code for saving predictions according to task
def _mp_fn(index):
# For xla_spawn (TPUs)
main()
if __name__ == "__main__":
main()
{%- elif cookiecutter.with_trainer == "False" %}
import argparse
import logging
import math
import os
import random
import datasets
from datasets import load_dataset, load_metric
from torch.utils.data import DataLoader
from tqdm.auto import tqdm
import transformers
from accelerate import Accelerator
from transformers import (
CONFIG_MAPPING,
MODEL_MAPPING,
AdamW,
AutoConfig,
{{cookiecutter.model_class}},
AutoTokenizer,
DataCollatorWithPadding,
PretrainedConfig,
SchedulerType,
default_data_collator,
get_scheduler,
set_seed,
)
from transformers.utils import send_example_telemetry
logger = logging.getLogger(__name__)
{%- if cookiecutter.can_train_from_scratch == "True" %}
# You should update this to your particular problem to have better documentation of `model_type`
MODEL_CONFIG_CLASSES = list(MODEL_MAPPING.keys())
MODEL_TYPES = tuple(conf.model_type for conf in MODEL_CONFIG_CLASSES)
{% endif %}
def parse_args():
parser = argparse.ArgumentParser(description="Finetune a transformers model on a text classification task")
parser.add_argument(
"--dataset_name",
type=str,
default=None,
help="The name of the dataset to use (via the datasets library).",
)
parser.add_argument(
"--dataset_config_name",
type=str,
default=None,
help= "The configuration name of the dataset to use (via the datasets library).",
)
parser.add_argument(
"--train_file", type=str, default=None, help="A csv or a json file containing the training data."
)
parser.add_argument(
"--validation_file", type=str, default=None, help="A csv or a json file containing the validation data."
)
parser.add_argument(
"--max_length",
type=int,
default=128,
help=(
"The maximum total input sequence length after tokenization. Sequences longer than this will be truncated,"
" sequences shorter will be padded if `--pad_to_max_length` is passed."
),
)
parser.add_argument(
"--pad_to_max_length",
action="store_true",
help="If passed, pad all samples to `max_length`. Otherwise, dynamic padding is used.",
)
parser.add_argument(
"--model_name_or_path",
type=str,
help="Path to pretrained model or model identifier from huggingface.co/models.",
required=True,
)
parser.add_argument(
"--config_name",
type=str,
default=None,
help="Pretrained config name or path if not the same as model_name",
)
parser.add_argument(
"--tokenizer_name",
type=str,
default=None,
help="Pretrained tokenizer name or path if not the same as model_name",
)
parser.add_argument(
"--use_slow_tokenizer",
action="store_true",
help="If passed, will use a slow tokenizer (not backed by the 🤗 Tokenizers library).",
)
parser.add_argument(
"--per_device_train_batch_size",
type=int,
default=8,
help="Batch size (per device) for the training dataloader.",
)
parser.add_argument(
"--per_device_eval_batch_size",
type=int,
default=8,
help="Batch size (per device) for the evaluation dataloader.",
)
parser.add_argument(
"--learning_rate",
type=float,
default=5e-5,
help="Initial learning rate (after the potential warmup period) to use.",
)
parser.add_argument("--weight_decay", type=float, default=0.0, help="Weight decay to use.")
parser.add_argument("--num_train_epochs", type=int, default=3, help="Total number of training epochs to perform.")
parser.add_argument(
"--max_train_steps",
type=int,
default=None,
help="Total number of training steps to perform. If provided, overrides num_train_epochs.",
)
parser.add_argument(
"--gradient_accumulation_steps",
type=int,
default=1,
help="Number of updates steps to accumulate before performing a backward/update pass.",
)
parser.add_argument(
"--lr_scheduler_type",
type=SchedulerType,
default="linear",
help="The scheduler type to use.",
choices=["linear", "cosine", "cosine_with_restarts", "polynomial", "constant", "constant_with_warmup"],
)
parser.add_argument(
"--num_warmup_steps", type=int, default=0, help="Number of steps for the warmup in the lr scheduler."
)
parser.add_argument("--output_dir", type=str, default=None, help="Where to store the final model.")
parser.add_argument("--seed", type=int, default=None, help="A seed for reproducible training.")
{%- if cookiecutter.can_train_from_scratch == "True" %}
parser.add_argument(
"--model_type",
type=str,
default=None,
help="Model type to use if training from scratch.",
choices=MODEL_TYPES,
)
{% endif %}
args = parser.parse_args()
# Sanity checks
if args.task_name is None and args.train_file is None and args.validation_file is None:
raise ValueError("Need either a task name or a training/validation file.")
else:
if args.train_file is not None:
extension = args.train_file.split(".")[-1]
assert extension in ["csv", "json"], "`train_file` should be a csv or a json file."
if args.validation_file is not None:
extension = args.validation_file.split(".")[-1]
assert extension in ["csv", "json"], "`validation_file` should be a csv or a json file."
if args.output_dir is not None:
os.makedirs(args.output_dir, exist_ok=True)
return args
def main():
args = parse_args()
# Sending telemetry. Tracking the example usage helps us better allocate resources to maintain them. The
# information sent is the one passed as arguments along with your Python/PyTorch versions.
send_example_telemetry("run_{{cookiecutter.example_shortcut}", args)
# Initialize the accelerator. We will let the accelerator handle device placement for us in this example.
accelerator = Accelerator()
# Make one log on every process with the configuration for debugging.
logging.basicConfig(
format="%(asctime)s - %(levelname)s - %(name)s - %(message)s",
datefmt="%m/%d/%Y %H:%M:%S",
level=logging.INFO,
)
logger.info(accelerator.state)
# Setup logging, we only want one process per machine to log things on the screen.
# accelerator.is_local_main_process is only True for one process per machine.
logger.setLevel(logging.INFO if accelerator.is_local_main_process else logging.ERROR)
if accelerator.is_local_main_process:
datasets.utils.logging.set_verbosity_warning()
transformers.utils.logging.set_verbosity_info()
else:
datasets.utils.logging.set_verbosity_error()
transformers.utils.logging.set_verbosity_error()
# If passed along, set the training seed now.
if args.seed is not None:
set_seed(args.seed)
# Get the datasets: you can either provide your own CSV/JSON/TXT training and evaluation files (see below)
# or just provide the name of one of the public datasets available on the hub at https://huggingface.co/datasets/
# (the dataset will be downloaded automatically from the datasets Hub).
#
# For CSV/JSON files, this script will use the column called 'text' or the first column if no column called
# 'text' is found. You can easily tweak this behavior (see below).
#
# In distributed training, the load_dataset function guarantee that only one local process can concurrently
# download the dataset.
if args.dataset_name is not None:
# Downloading and loading a dataset from the hub.
raw_datasets = load_dataset(args.dataset_name, args.dataset_config_name)
else:
data_files = {}
if args.train_file is not None:
data_files["train"] = args.train_file
extension = args.train_file.split(".")[-1]
if args.validation_file is not None:
data_files["validation"] = args.validation_file
extension = args.validation_file.split(".")[-1]
raw_datasets = load_dataset(extension, data_files=data_files)
# See more about loading any type of standard or custom dataset (from files, python dict, pandas DataFrame, etc) at
# https://huggingface.co/docs/datasets/loading_datasets.
# Load pretrained model and tokenizer
#
# In distributed training, the .from_pretrained methods guarantee that only one local process can concurrently
# download model & vocab.
{%- if cookiecutter.can_train_from_scratch == "True" %}
if model_args.config_name:
config = AutoConfig.from_pretrained(args.model_name_or_path)
elif model_args.model_name_or_path:
config = AutoConfig.from_pretrained(args.model_name_or_path)
else:
config = CONFIG_MAPPING[args.model_type]()
logger.warning("You are instantiating a new config instance from scratch.")
if model_args.tokenizer_name:
tokenizer = AutoTokenizer.from_pretrained(model_args.tokenizer_name, use_fast=not args.use_slow_tokenizer)
elif model_args.model_name_or_path:
tokenizer = AutoTokenizer.from_pretrained(model_args.model_name_or_path, use_fast=not args.use_slow_tokenizer)
else:
raise ValueError(
"You are instantiating a new tokenizer from scratch. This is not supported by this script. "
"You can do it from another script, save it, and load it from here, using --tokenizer_name."
)
if model_args.model_name_or_path:
model = {{cookiecutter.model_class}}.from_pretrained(
model_args.model_name_or_path,
from_tf=bool(".ckpt" in model_args.model_name_or_path),
config=config,
)
else:
logger.info("Training new model from scratch")
model = {{cookiecutter.model_class}}.from_config(config)
model.resize_token_embeddings(len(tokenizer))
{%- elif cookiecutter.can_train_from_scratch == "False" %}
config = AutoConfig.from_pretrained(
args.config_name if model_args.config_name else args.model_name_or_path,
# num_labels=num_labels, Uncomment if you have a certain number of labels
finetuning_task=data_args.task_name,
)
tokenizer = AutoTokenizer.from_pretrained(
args.tokenizer_name if model_args.tokenizer_name else args.model_name_or_path,
use_fast=not args.use_slow_tokenizer,
)
model = AutoModelForSequenceClassification.from_pretrained(
model_args.model_name_or_path,
from_tf=bool(".ckpt" in model_args.model_name_or_path),
config=config,
)
{% endif %}
# Preprocessing the datasets.
# First we tokenize all the texts.
column_names = raw_datasets["train"].column_names
text_column_name = "text" if "text" in column_names else column_names[0]
padding = "max_length" if args.pad_to_max_length else False
def tokenize_function(examples):
result = tokenizer(examples[text_column_name], padding=padding, max_length=args.max_length, truncation=True)
if "label" in examples:
result["labels"] = examples["label"]
return result
processed_datasets = raw_datasets.map(
preprocess_function, batched=True, remove_columns=raw_datasets["train"].column_names
)
train_dataset = processed_datasets["train"]
eval_dataset = processed_datasets["validation"]
# Log a few random samples from the training set:
for index in random.sample(range(len(train_dataset)), 3):
logger.info(f"Sample {index} of the training set: {train_dataset[index]}.")
# DataLoaders creation:
if args.pad_to_max_length:
# If padding was already done ot max length, we use the default data collator that will just convert everything
# to tensors.
data_collator = default_data_collator
else:
# Otherwise, `DataCollatorWithPadding` will apply dynamic padding for us (by padding to the maximum length of
# the samples passed). When using mixed precision, we add `pad_to_multiple_of=8` to pad all tensors to multiple
# of 8s, which will enable the use of Tensor Cores on NVIDIA hardware with compute capability >= 7.5 (Volta).
data_collator = DataCollatorWithPadding(tokenizer, pad_to_multiple_of=(8 if accelerator.use_fp16 else None))
train_dataloader = DataLoader(
train_dataset, shuffle=True, collate_fn=data_collator, batch_size=args.per_device_train_batch_size
)
eval_dataloader = DataLoader(eval_dataset, collate_fn=data_collator, batch_size=args.per_device_eval_batch_size)
# Optimizer
# Split weights in two groups, one with weight decay and the other not.
no_decay = ["bias", "LayerNorm.weight"]
optimizer_grouped_parameters = [
{
"params": [p for n, p in model.named_parameters() if not any(nd in n for nd in no_decay)],
"weight_decay": args.weight_decay,
},
{
"params": [p for n, p in model.named_parameters() if any(nd in n for nd in no_decay)],
"weight_decay": 0.0,
},
]
optimizer = AdamW(optimizer_grouped_parameters, lr=args.learning_rate)
# Prepare everything with our `accelerator`.
model, optimizer, train_dataloader, eval_dataloader = accelerator.prepare(
model, optimizer, train_dataloader, eval_dataloader
)
# Note -> the training dataloader needs to be prepared before we grab his length below (cause its length will be
# shorter in multiprocess)
# Scheduler and math around the number of training steps.
num_update_steps_per_epoch = math.ceil(len(train_dataloader) / args.gradient_accumulation_steps)
if args.max_train_steps is None:
args.max_train_steps = args.num_train_epochs * num_update_steps_per_epoch
else:
args.num_train_epochs = math.ceil(args.max_train_steps / num_update_steps_per_epoch)
lr_scheduler = get_scheduler(
name=args.lr_scheduler_type,
optimizer=optimizer,
num_warmup_steps=args.num_warmup_steps,
num_training_steps=args.max_train_steps,
)
# TODO Get the proper metric function
# metric = load_metric(xxx)
# Train!
total_batch_size = args.per_device_train_batch_size * accelerator.num_processes * args.gradient_accumulation_steps
logger.info("***** Running training *****")
logger.info(f" Num examples = {len(train_dataset)}")
logger.info(f" Num Epochs = {args.num_train_epochs}")
logger.info(f" Instantaneous batch size per device = {args.per_device_train_batch_size}")
logger.info(f" Total train batch size (w. parallel, distributed & accumulation) = {total_batch_size}")
logger.info(f" Gradient Accumulation steps = {args.gradient_accumulation_steps}")
logger.info(f" Total optimization steps = {args.max_train_steps}")
# Only show the progress bar once on each machine.
progress_bar = tqdm(range(args.max_train_steps), disable=not accelerator.is_local_main_process)
completed_steps = 0
for epoch in range(args.num_train_epochs):
model.train()
for step, batch in enumerate(train_dataloader):
outputs = model(**batch)
loss = outputs.loss
loss = loss / args.gradient_accumulation_steps
accelerator.backward(loss)
if step % args.gradient_accumulation_steps == 0 or step == len(train_dataloader) - 1:
optimizer.step()
lr_scheduler.step()
optimizer.zero_grad()
progress_bar.update(1)
completed_steps += 1
if completed_steps >= args.max_train_steps:
break
model.eval()
for step, batch in enumerate(eval_dataloader):
with torch.no_grad():
outputs = model(**batch)
predictions = outputs.logits.argmax(dim=-1)
metric.add_batch(
predictions=accelerator.gather(predictions),
references=accelerator.gather(batch["labels"]),
)
eval_metric = metric.compute()
logger.info(f"epoch {epoch}: {eval_metric}")
if args.output_dir is not None:
accelerator.wait_for_everyone()
unwrapped_model = accelerator.unwrap_model(model)
unwrapped_model.save_pretrained(args.output_dir, save_function=accelerator.save)
if __name__ == "__main__":
main()
{% endif %}
| 0
|
mavonic_private_repos/transformers/templates
|
mavonic_private_repos/transformers/templates/adding_a_missing_tokenization_test/cookiecutter.json
|
{
"modelname": "BrandNewBERT",
"uppercase_modelname": "BRAND_NEW_BERT",
"lowercase_modelname": "brand_new_bert",
"camelcase_modelname": "BrandNewBert",
"has_slow_class": ["True", "False"],
"has_fast_class": ["True", "False"],
"slow_tokenizer_use_sentencepiece": ["True", "False"],
"authors": "The HuggingFace Team"
}
| 0
|
mavonic_private_repos/transformers/templates
|
mavonic_private_repos/transformers/templates/adding_a_missing_tokenization_test/README.md
|
<!---
Copyright 2022 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
-->
This folder contains a template to add a tokenization test.
## Usage
Using the `cookiecutter` utility requires to have all the `dev` dependencies installed.
Let's first [fork](https://docs.github.com/en/get-started/quickstart/fork-a-repo) the `transformers` repo on github. Once it's done you can clone your fork and install `transformers` in our environment:
```shell script
git clone https://github.com/YOUR-USERNAME/transformers
cd transformers
pip install -e ".[dev]"
```
Once the installation is done, you can generate the template by running the following command. Be careful, the template will be generated inside a new folder in your current working directory.
```shell script
cookiecutter path-to-the folder/adding_a_missing_tokenization_test/
```
You will then have to answer some questions about the tokenizer for which you want to add tests. The `modelname` should be cased according to the plain text casing, i.e., BERT, RoBERTa, DeBERTa.
Once the command has finished, you should have a one new file inside the newly created folder named `test_tokenization_Xxx.py`. At this point the template is finished and you can move it to the sub-folder of the corresponding model in the test folder.
| 0
|
mavonic_private_repos/transformers/templates/adding_a_missing_tokenization_test
|
mavonic_private_repos/transformers/templates/adding_a_missing_tokenization_test/cookiecutter-template-{{cookiecutter.modelname}}/test_tokenization_{{cookiecutter.lowercase_modelname}}.py
|
# coding=utf-8
# Copyright 2022 {{cookiecutter.authors}}. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
""" Testing suite for the {{cookiecutter.modelname}} tokenizer. """
import unittest
{% if cookiecutter.has_slow_class == "True" and cookiecutter.has_fast_class == "True" -%}
from transformers import {{cookiecutter.camelcase_modelname}}Tokenizer, {{cookiecutter.camelcase_modelname}}TokenizerFast
{% elif cookiecutter.has_slow_class == "True" -%}
from transformers import {{cookiecutter.camelcase_modelname}}Tokenizer
{% elif cookiecutter.has_fast_class == "True" -%}
from transformers import {{cookiecutter.camelcase_modelname}}TokenizerFast
{% endif -%}
{% if cookiecutter.has_fast_class == "True" and cookiecutter.slow_tokenizer_use_sentencepiece == "True" -%}
from transformers.testing_utils import require_sentencepiece, require_tokenizers
from ...test_tokenization_common import TokenizerTesterMixin
@require_sentencepiece
@require_tokenizers
{% elif cookiecutter.slow_tokenizer_use_sentencepiece == "True" -%}
from transformers.testing_utils import require_sentencepiece
from ...test_tokenization_common import TokenizerTesterMixin
@require_sentencepiece
{% elif cookiecutter.has_fast_class == "True" -%}
from transformers.testing_utils import require_tokenizers
from ...test_tokenization_common import TokenizerTesterMixin
@require_tokenizers
{% else -%}
from ...test_tokenization_common import TokenizerTesterMixin
{% endif -%}
class {{cookiecutter.camelcase_modelname}}TokenizationTest(TokenizerTesterMixin, unittest.TestCase):
{% if cookiecutter.has_slow_class == "True" -%}
tokenizer_class = {{cookiecutter.camelcase_modelname}}Tokenizer
test_slow_tokenizer = True
{% else -%}
tokenizer_class = None
test_slow_tokenizer = False
{% endif -%}
{% if cookiecutter.has_fast_class == "True" -%}
rust_tokenizer_class = {{cookiecutter.camelcase_modelname}}TokenizerFast
test_rust_tokenizer = True
{% else -%}
rust_tokenizer_class = None
test_rust_tokenizer = False
{% endif -%}
{% if cookiecutter.slow_tokenizer_use_sentencepiece == "True" -%}
test_sentencepiece = True
{% endif -%}
# TODO: Check in `TokenizerTesterMixin` if other attributes need to be changed
def setUp(self):
super().setUp()
raise NotImplementedError(
"Here you have to implement the saving of a toy tokenizer in "
"`self.tmpdirname`."
)
# TODO: add tests with hard-coded target values
| 0
|
mavonic_private_repos/transformers/templates
|
mavonic_private_repos/transformers/templates/adding_a_new_model/ADD_NEW_MODEL_PROPOSAL_TEMPLATE.md
|
**TEMPLATE**
=====================================
*search & replace the following keywords, e.g.:*
`:%s/\[name of model\]/brand_new_bert/g`
-[lowercase name of model] # e.g. brand_new_bert
-[camelcase name of model] # e.g. BrandNewBert
-[name of mentor] # e.g. [Peter](https://github.com/peter)
-[link to original repo]
-[start date]
-[end date]
How to add [camelcase name of model] to 🤗 Transformers?
=====================================
Mentor: [name of mentor]
Begin: [start date]
Estimated End: [end date]
Adding a new model is often difficult and requires an in-depth knowledge
of the 🤗 Transformers library and ideally also of the model's original
repository. At Hugging Face, we are trying to empower the community more
and more to add models independently.
The following sections explain in detail how to add [camelcase name of model]
to Transformers. You will work closely with [name of mentor] to
integrate [camelcase name of model] into Transformers. By doing so, you will both gain a
theoretical and deep practical understanding of [camelcase name of model].
But more importantly, you will have made a major
open-source contribution to Transformers. Along the way, you will:
- get insights into open-source best practices
- understand the design principles of one of the most popular NLP
libraries
- learn how to do efficiently test large NLP models
- learn how to integrate Python utilities like `black`, `ruff`,
`make fix-copies` into a library to always ensure clean and readable
code
To start, let's try to get a general overview of the Transformers
library.
General overview of 🤗 Transformers
----------------------------------
First, you should get a general overview of 🤗 Transformers. Transformers
is a very opinionated library, so there is a chance that
you don't agree with some of the library's philosophies or design
choices. From our experience, however, we found that the fundamental
design choices and philosophies of the library are crucial to
efficiently scale Transformers while keeping maintenance costs at a
reasonable level.
A good first starting point to better understand the library is to read
the [documentation of our philosophy](https://huggingface.co/transformers/philosophy.html).
As a result of our way of working, there are some choices that we try to apply to all models:
- Composition is generally favored over abstraction
- Duplicating code is not always bad if it strongly improves the
readability or accessibility of a model
- Model files are as self-contained as possible so that when you read
the code of a specific model, you ideally only have to look into the
respective `modeling_....py` file.
In our opinion, the library's code is not just a means to provide a
product, *e.g.*, the ability to use BERT for inference, but also as the
very product that we want to improve. Hence, when adding a model, the
user is not only the person that will use your model, but also everybody
that will read, try to understand, and possibly tweak your code.
With this in mind, let's go a bit deeper into the general library
design.
### Overview of models
To successfully add a model, it is important to understand the
interaction between your model and its config,
`PreTrainedModel`, and `PretrainedConfig`. For
exemplary purposes, we will call the PyTorch model to be added to 🤗 Transformers
`BrandNewBert`.
Let's take a look:
As you can see, we do make use of inheritance in 🤗 Transformers, but we
keep the level of abstraction to an absolute minimum. There are never
more than two levels of abstraction for any model in the library.
`BrandNewBertModel` inherits from
`BrandNewBertPreTrainedModel` which in
turn inherits from `PreTrainedModel` and that's it.
As a general rule, we want to make sure
that a new model only depends on `PreTrainedModel`. The
important functionalities that are automatically provided to every new
model are
`PreTrainedModel.from_pretrained` and `PreTrainedModel.save_pretrained`, which are
used for serialization and deserialization. All
of the other important functionalities, such as
`BrandNewBertModel.forward` should be
completely defined in the new `modeling_brand_new_bert.py` module. Next,
we want to make sure that a model with a specific head layer, such as
`BrandNewBertForMaskedLM` does not inherit
from `BrandNewBertModel`, but rather uses
`BrandNewBertModel` as a component that
can be called in its forward pass to keep the level of abstraction low.
Every new model requires a configuration class, called
`BrandNewBertConfig`. This configuration
is always stored as an attribute in
`PreTrainedModel`, and
thus can be accessed via the `config` attribute for all classes
inheriting from `BrandNewBertPreTrainedModel`
```python
# assuming that `brand_new_bert` belongs to the organization `brandy`
model = BrandNewBertModel.from_pretrained("brandy/brand_new_bert")
model.config # model has access to its config
```
Similar to the model, the configuration inherits basic serialization and
deserialization functionalities from
`PretrainedConfig`. Note
that the configuration and the model are always serialized into two
different formats - the model to a `pytorch_model.bin` file
and the configuration to a `config.json` file. Calling
`PreTrainedModel.save_pretrained` will automatically call
`PretrainedConfig.save_pretrained`, so that both model and configuration are saved.
### Overview of tokenizers
Not quite ready yet :-( This section will be added soon!
Step-by-step recipe to add a model to 🤗 Transformers
----------------------------------------------------
Everyone has different preferences of how to port a model so it can be
very helpful for you to take a look at summaries of how other
contributors ported models to Hugging Face. Here is a list of community
blog posts on how to port a model:
1. [Porting GPT2
Model](https://medium.com/huggingface/from-tensorflow-to-pytorch-265f40ef2a28)
by [Thomas](https://huggingface.co/thomwolf)
2. [Porting WMT19 MT Model](https://huggingface.co/blog/porting-fsmt)
by [Stas](https://huggingface.co/stas)
From experience, we can tell you that the most important things to keep
in mind when adding a model are:
- Don't reinvent the wheel! Most parts of the code you will add for
the new 🤗 Transformers model already exist somewhere in 🤗
Transformers. Take some time to find similar, already existing
models and tokenizers you can copy from.
[grep](https://www.gnu.org/software/grep/) and
[rg](https://github.com/BurntSushi/ripgrep) are your friends. Note
that it might very well happen that your model's tokenizer is based
on one model implementation, and your model's modeling code on
another one. *E.g.*, FSMT's modeling code is based on BART, while
FSMT's tokenizer code is based on XLM.
- It's more of an engineering challenge than a scientific challenge.
You should spend more time on creating an efficient debugging
environment than trying to understand all theoretical aspects of the
model in the paper.
- Ask for help when you're stuck! Models are the core component of 🤗
Transformers so we, at Hugging Face, are more than happy to help
you at every step to add your model. Don't hesitate to ask if you
notice you are not making progress.
In the following, we try to give you a general recipe that we found most
useful when porting a model to 🤗 Transformers.
The following list is a summary of everything that has to be done to add
a model and can be used by you as a To-Do List:
1. [ ] (Optional) Understood theoretical aspects
2. [ ] Prepared transformers dev environment
3. [ ] Set up debugging environment of the original repository
4. [ ] Created script that successfully runs forward pass using
original repository and checkpoint
5. [ ] Successfully opened a PR and added the model skeleton to Transformers
6. [ ] Successfully converted original checkpoint to Transformers
checkpoint
7. [ ] Successfully ran forward pass in Transformers that gives
identical output to original checkpoint
8. [ ] Finished model tests in Transformers
9. [ ] Successfully added Tokenizer in Transformers
10. [ ] Run end-to-end integration tests
11. [ ] Finished docs
12. [ ] Uploaded model weights to the hub
13. [ ] Submitted the pull request for review
14. [ ] (Optional) Added a demo notebook
To begin with, we usually recommend to start by getting a good
theoretical understanding of `[camelcase name of model]`. However, if you prefer to
understand the theoretical aspects of the model *on-the-job*, then it is
totally fine to directly dive into the `[camelcase name of model]`'s code-base. This
option might suit you better, if your engineering skills are better than
your theoretical skill, if you have trouble understanding
`[camelcase name of model]`'s paper, or if you just enjoy programming much more than
reading scientific papers.
### 1. (Optional) Theoretical aspects of [camelcase name of model]
You should take some time to read *[camelcase name of model]'s* paper, if such
descriptive work exists. There might be large sections of the paper that
are difficult to understand. If this is the case, this is fine - don't
worry! The goal is not to get a deep theoretical understanding of the
paper, but to extract the necessary information required to effectively
re-implement the model in 🤗 Transformers. That being said, you don't
have to spend too much time on the theoretical aspects, but rather focus
on the practical ones, namely:
- What type of model is *[camelcase name of model]*? BERT-like encoder-only
model? GPT2-like decoder-only model? BART-like encoder-decoder
model? Look at the `model_summary` if
you're not familiar with the differences between those.
- What are the applications of *[camelcase name of model]*? Text
classification? Text generation? Seq2Seq tasks, *e.g.,*
summarization?
- What is the novel feature of the model making it different from
BERT/GPT-2/BART?
- Which of the already existing [🤗 Transformers
models](https://huggingface.co/transformers/#contents) is most
similar to *[camelcase name of model]*?
- What type of tokenizer is used? A sentencepiece tokenizer? Word
piece tokenizer? Is it the same tokenizer as used for BERT or BART?
After you feel like you have gotten a good overview of the architecture
of the model, you might want to write to [name of mentor] with any
questions you might have. This might include questions regarding the
model's architecture, its attention layer, etc. We will be more than
happy to help you.
#### Additional resources
Before diving into the code, here are some additional resources that might be worth taking a look at:
- [link 1]
- [link 2]
- [link 3]
- ...
#### Make sure you've understood the fundamental aspects of [camelcase name of model]
Alright, now you should be ready to take a closer look into the actual code of [camelcase name of model].
You should have understood the following aspects of [camelcase name of model] by now:
- [characteristic 1 of [camelcase name of model]]
- [characteristic 2 of [camelcase name of model]]
- ...
If any of the mentioned aspects above are **not** clear to you, now is a great time to talk to [name of mentor].
### 2. Next prepare your environment
1. Fork the [repository](https://github.com/huggingface/transformers)
by clicking on the 'Fork' button on the repository's page. This
creates a copy of the code under your GitHub user account.
2. Clone your `transformers` fork to your local disk, and add the base
repository as a remote:
```bash
git clone https://github.com/[your Github handle]/transformers.git
cd transformers
git remote add upstream https://github.com/huggingface/transformers.git
```
3. Set up a development environment, for instance by running the
following command:
```bash
python -m venv .env
source .env/bin/activate
pip install -e ".[dev]"
```
and return to the parent directory
```bash
cd ..
```
4. We recommend adding the PyTorch version of *[camelcase name of model]* to
Transformers. To install PyTorch, please follow the instructions [here](https://pytorch.org/get-started/locally/).
**Note:** You don't need to have CUDA installed. Making the new model
work on CPU is sufficient.
5. To port *[camelcase name of model]*, you will also need access to its
original repository:
```bash
git clone [link to original repo].git
cd [lowercase name of model]
pip install -e .
```
Now you have set up a development environment to port *[camelcase name of model]*
to 🤗 Transformers.
### Run a pretrained checkpoint using the original repository
**3. Set up debugging environment**
At first, you will work on the original *[camelcase name of model]* repository.
Often, the original implementation is very "researchy". Meaning that
documentation might be lacking and the code can be difficult to
understand. But this should be exactly your motivation to reimplement
*[camelcase name of model]*. At Hugging Face, one of our main goals is to *make
people stand on the shoulders of giants* which translates here very well
into taking a working model and rewriting it to make it as **accessible,
user-friendly, and beautiful** as possible. This is the number-one
motivation to re-implement models into 🤗 Transformers - trying to make
complex new NLP technology accessible to **everybody**.
You should start thereby by diving into the [original repository]([link to original repo]).
Successfully running the official pretrained model in the original
repository is often **the most difficult** step. From our experience, it
is very important to spend some time getting familiar with the original
code-base. You need to figure out the following:
- Where to find the pretrained weights?
- How to load the pretrained weights into the corresponding model?
- How to run the tokenizer independently from the model?
- Trace one forward pass so that you know which classes and functions
are required for a simple forward pass. Usually, you only have to
reimplement those functions.
- Be able to locate the important components of the model: Where is
the model's class? Are there model sub-classes, *e.g.*,
EncoderModel, DecoderModel? Where is the self-attention layer? Are
there multiple different attention layers, *e.g.*, *self-attention*,
*cross-attention*...?
- How can you debug the model in the original environment of the repo?
Do you have to add `print` statements, can you work with
an interactive debugger like [ipdb](https://pypi.org/project/ipdb/), or should you use
an efficient IDE to debug the model, like PyCharm?
It is very important that before you start the porting process, that you
can **efficiently** debug code in the original repository! Also,
remember that you are working with an open-source library, so do not
hesitate to open an issue, or even a pull request in the original
repository. The maintainers of this repository are most likely very
happy about someone looking into their code!
At this point, it is really up to you which debugging environment and
strategy you prefer to use to debug the original model. We strongly
advise against setting up a costly GPU environment, but simply work on a
CPU both when starting to dive into the original repository and also
when starting to write the 🤗 Transformers implementation of the model.
Only at the very end, when the model has already been successfully
ported to 🤗 Transformers, one should verify that the model also works as
expected on GPU.
In general, there are two possible debugging environments for running
the original model
- [Jupyter notebooks](https://jupyter.org/) / [google colab](https://colab.research.google.com/notebooks/intro.ipynb)
- Local python scripts.
Jupyter notebooks have the advantage that they allow for cell-by-cell
execution which can be helpful to better split logical components from
one another and to have faster debugging cycles as intermediate results
can be stored. Also, notebooks are often easier to share with other
contributors, which might be very helpful if you want to ask the Hugging
Face team for help. If you are familiar with Jupyter notebooks, we
strongly recommend you to work with them.
The obvious disadvantage of Jupyter notebooks is that if you are not
used to working with them you will have to spend some time adjusting to
the new programming environment and that you might not be able to use
your known debugging tools anymore, like `ipdb`.
**4. Successfully run forward pass**
For each code-base, a good first step is always to load a **small**
pretrained checkpoint and to be able to reproduce a single forward pass
using a dummy integer vector of input IDs as an input. Such a script
could look like this (in pseudocode):
```python
model = [camelcase name of model]Model.load_pretrained_checkpoint("/path/to/checkpoint/")
input_ids = [0, 4, 5, 2, 3, 7, 9] # vector of input ids
original_output = model.predict(input_ids)
```
Next, regarding the debugging strategy, there are generally a few from
which to choose from:
- Decompose the original model into many small testable components and
run a forward pass on each of those for verification
- Decompose the original model only into the original *tokenizer* and
the original *model*, run a forward pass on those, and use
intermediate print statements or breakpoints for verification
Again, it is up to you which strategy to choose. Often, one or the other
is advantageous depending on the original code base.
If the original code-base allows you to decompose the model into smaller
sub-components, *e.g.*, if the original code-base can easily be run in
eager mode, it is usually worth the effort to do so. There are some
important advantages to taking the more difficult road in the beginning:
- at a later stage when comparing the original model to the Hugging
Face implementation, you can verify automatically for each component
individually that the corresponding component of the 🤗 Transformers
implementation matches instead of relying on visual comparison via
print statements
- it can give you some rope to decompose the big problem of porting a
model into smaller problems of just porting individual components
and thus structure your work better
- separating the model into logical meaningful components will help
you to get a better overview of the model's design and thus to
better understand the model
- at a later stage those component-by-component tests help you to
ensure that no regression occurs as you continue changing your code
[Lysandre's](https://gist.github.com/LysandreJik/db4c948f6b4483960de5cbac598ad4ed)
integration checks for ELECTRA gives a nice example of how this can be
done.
However, if the original code-base is very complex or only allows
intermediate components to be run in a compiled mode, it might be too
time-consuming or even impossible to separate the model into smaller
testable sub-components. A good example is [T5's
MeshTensorFlow](https://github.com/tensorflow/mesh/tree/master/mesh_tensorflow)
library which is very complex and does not offer a simple way to
decompose the model into its sub-components. For such libraries, one
often relies on verifying print statements.
No matter which strategy you choose, the recommended procedure is often
the same in that you should start to debug the starting layers first and
the ending layers last.
It is recommended that you retrieve the output, either by print
statements or sub-component functions, of the following layers in the
following order:
1. Retrieve the input IDs passed to the model
2. Retrieve the word embeddings
3. Retrieve the input of the first Transformer layer
4. Retrieve the output of the first Transformer layer
5. Retrieve the output of the following n - 1 Transformer layers
6. Retrieve the output of the whole [camelcase name of model] Model
Input IDs should thereby consists of an array of integers, *e.g.*,
`input_ids = [0, 4, 4, 3, 2, 4, 1, 7, 19]`
The outputs of the following layers often consist of multi-dimensional
float arrays and can look like this:
```bash
[[
[-0.1465, -0.6501, 0.1993, ..., 0.1451, 0.3430, 0.6024],
[-0.4417, -0.5920, 0.3450, ..., -0.3062, 0.6182, 0.7132],
[-0.5009, -0.7122, 0.4548, ..., -0.3662, 0.6091, 0.7648],
...,
[-0.5613, -0.6332, 0.4324, ..., -0.3792, 0.7372, 0.9288],
[-0.5416, -0.6345, 0.4180, ..., -0.3564, 0.6992, 0.9191],
[-0.5334, -0.6403, 0.4271, ..., -0.3339, 0.6533, 0.8694]]],
```
We expect that every model added to 🤗 Transformers passes a couple of
integration tests, meaning that the original model and the reimplemented
version in 🤗 Transformers have to give the exact same output up to a
precision of 0.001! Since it is normal that the exact same model written
in different libraries can give a slightly different output depending on
the library framework, we accept an error tolerance of 1e-3 (0.001). It
is not enough if the model gives nearly the same output, they have to be
the almost identical. Therefore, you will certainly compare the
intermediate outputs of the 🤗 Transformers version multiple times
against the intermediate outputs of the original implementation of
*[camelcase name of model]* in which case an **efficient** debugging environment
of the original repository is absolutely important. Here is some advice
to make your debugging environment as efficient as possible.
- Find the best way of debugging intermediate results. Is the original
repository written in PyTorch? Then you should probably take the
time to write a longer script that decomposes the original model
into smaller sub-components to retrieve intermediate values. Is the
original repository written in Tensorflow 1? Then you might have to
rely on TensorFlow print operations like
[tf.print](https://www.tensorflow.org/api_docs/python/tf/print) to
output intermediate values. Is the original repository written in
Jax? Then make sure that the model is **not jitted** when running
the forward pass, *e.g.*, check-out [this
link](https://github.com/google/jax/issues/196).
- Use the smallest pretrained checkpoint you can find. The smaller the
checkpoint, the faster your debug cycle becomes. It is not efficient
if your pretrained model is so big that your forward pass takes more
than 10 seconds. In case only very large checkpoints are available,
it might make more sense to create a dummy model in the new
environment with randomly initialized weights and save those weights
for comparison with the 🤗 Transformers version of your model
- Make sure you are using the easiest way of calling a forward pass in
the original repository. Ideally, you want to find the function in
the original repository that **only** calls a single forward pass,
*i.e.* that is often called `predict`, `evaluate`, `forward` or
`__call__`. You don't want to debug a function that calls `forward`
multiple times, *e.g.*, to generate text, like
`autoregressive_sample`, `generate`.
- Try to separate the tokenization from the model's
forward pass. If the original repository shows
examples where you have to input a string, then try to find out
where in the forward call the string input is changed to input ids
and start from this point. This might mean that you have to possibly
write a small script yourself or change the original code so that
you can directly input the ids instead of an input string.
- Make sure that the model in your debugging setup is **not** in
training mode, which often causes the model to yield random outputs
due to multiple dropout layers in the model. Make sure that the
forward pass in your debugging environment is **deterministic** so
that the dropout layers are not used. Or use
`transformers.utils.set_seed` if the old and new
implementations are in the same framework.
#### More details on how to create a debugging environment for [camelcase name of model]
[TODO FILL: Here the mentor should add very specific information on what the student should do]
[to set up an efficient environment for the special requirements of this model]
### Port [camelcase name of model] to 🤗 Transformers
Next, you can finally start adding new code to 🤗 Transformers. Go into
the clone of your 🤗 Transformers' fork:
cd transformers
In the special case that you are adding a model whose architecture
exactly matches the model architecture of an existing model you only
have to add a conversion script as described in [this
section](#write-a-conversion-script). In this case, you can just re-use
the whole model architecture of the already existing model.
Otherwise, let's start generating a new model with the amazing
Cookiecutter!
**Use the Cookiecutter to automatically generate the model's code**
To begin with head over to the [🤗 Transformers
templates](https://github.com/huggingface/transformers/tree/main/templates/adding_a_new_model)
to make use of our `cookiecutter` implementation to automatically
generate all the relevant files for your model. Again, we recommend only
adding the PyTorch version of the model at first. Make sure you follow
the instructions of the `README.md` on the [🤗 Transformers
templates](https://github.com/huggingface/transformers/tree/main/templates/adding_a_new_model)
carefully.
**Open a Pull Request on the main huggingface/transformers repo**
Before starting to adapt the automatically generated code, now is the
time to open a "Work in progress (WIP)" pull request, *e.g.*, "\[WIP\]
Add *[camelcase name of model]*", in 🤗 Transformers so that you and the Hugging
Face team can work side-by-side on integrating the model into 🤗
Transformers.
You should do the following:
1. Create a branch with a descriptive name from your main branch
```bash
git checkout -b add_[lowercase name of model]
```
2. Commit the automatically generated code:
```bash
git add .
git commit
```
3. Fetch and rebase to current main
```bash
git fetch upstream
git rebase upstream/main
```
4. Push the changes to your account using:
```bash
git push -u origin a-descriptive-name-for-my-changes
```
5. Once you are satisfied, go to the webpage of your fork on GitHub.
Click on "Pull request". Make sure to add the GitHub handle of
[name of mentor] as a reviewer, so that the Hugging
Face team gets notified for future changes.
6. Change the PR into a draft by clicking on "Convert to draft" on the
right of the GitHub pull request web page.
In the following, whenever you have done some progress, don't forget to
commit your work and push it to your account so that it shows in the
pull request. Additionally, you should make sure to update your work
with the current main from time to time by doing:
git fetch upstream
git merge upstream/main
In general, all questions you might have regarding the model or your
implementation should be asked in your PR and discussed/solved in the
PR. This way, [name of mentor] will always be notified when you are
committing new code or if you have a question. It is often very helpful
to point [name of mentor] to your added code so that the Hugging
Face team can efficiently understand your problem or question.
To do so, you can go to the "Files changed" tab where you see all of
your changes, go to a line regarding which you want to ask a question,
and click on the "+" symbol to add a comment. Whenever a question or
problem has been solved, you can click on the "Resolve" button of the
created comment.
In the same way, [name of mentor] will open comments when reviewing
your code. We recommend asking most questions on GitHub on your PR. For
some very general questions that are not very useful for the public,
feel free to ping [name of mentor] by Slack or email.
**5. Adapt the generated models code for [camelcase name of model]**
At first, we will focus only on the model itself and not care about the
tokenizer. All the relevant code should be found in the generated files
`src/transformers/models/[lowercase name of model]/modeling_[lowercase name of model].py` and
`src/transformers/models/[lowercase name of model]/configuration_[lowercase name of model].py`.
Now you can finally start coding :). The generated code in
`src/transformers/models/[lowercase name of model]/modeling_[lowercase name of model].py` will
either have the same architecture as BERT if it's an encoder-only model
or BART if it's an encoder-decoder model. At this point, you should
remind yourself what you've learned in the beginning about the
theoretical aspects of the model: *How is the model different from BERT
or BART?*\". Implement those changes which often means to change the
*self-attention* layer, the order of the normalization layer, etc...
Again, it is often useful to look at the similar architecture of already
existing models in Transformers to get a better feeling of how your
model should be implemented.
**Note** that at this point, you don't have to be very sure that your
code is fully correct or clean. Rather, it is advised to add a first
*unclean*, copy-pasted version of the original code to
`src/transformers/models/[lowercase name of model]/modeling_[lowercase name of model].py`
until you feel like all the necessary code is added. From our
experience, it is much more efficient to quickly add a first version of
the required code and improve/correct the code iteratively with the
conversion script as described in the next section. The only thing that
has to work at this point is that you can instantiate the 🤗 Transformers
implementation of *[camelcase name of model]*, *i.e.* the following command
should work:
```python
from transformers import [camelcase name of model]Model, [camelcase name of model]Config
model = [camelcase name of model]Model([camelcase name of model]Config())
```
The above command will create a model according to the default
parameters as defined in `[camelcase name of model]Config()` with random weights,
thus making sure that the `init()` methods of all components works.
[TODO FILL: Here the mentor should add very specific information on what exactly has to be changed for this model]
[...]
[...]
**6. Write a conversion script**
Next, you should write a conversion script that lets you convert the
checkpoint you used to debug *[camelcase name of model]* in the original
repository to a checkpoint compatible with your just created 🤗
Transformers implementation of *[camelcase name of model]*. It is not advised to
write the conversion script from scratch, but rather to look through
already existing conversion scripts in 🤗 Transformers for one that has
been used to convert a similar model that was written in the same
framework as *[camelcase name of model]*. Usually, it is enough to copy an
already existing conversion script and slightly adapt it for your use
case. Don't hesitate to ask [name of mentor] to point you to a
similar already existing conversion script for your model.
- If you are porting a model from TensorFlow to PyTorch, a good
starting point might be BERT's conversion script
[here](https://github.com/huggingface/transformers/blob/7acfa95afb8194f8f9c1f4d2c6028224dbed35a2/src/transformers/models/bert/modeling_bert.py#L91)
- If you are porting a model from PyTorch to PyTorch, a good starting
point might be BART's conversion script
[here](https://github.com/huggingface/transformers/blob/main/src/transformers/models/bart/convert_bart_original_pytorch_checkpoint_to_pytorch.py)
In the following, we'll quickly explain how PyTorch models store layer
weights and define layer names. In PyTorch, the name of a layer is
defined by the name of the class attribute you give the layer. Let's
define a dummy model in PyTorch, called `SimpleModel` as follows:
```python
from torch import nn
class SimpleModel(nn.Module):
def __init__(self):
super().__init__()
self.dense = nn.Linear(10, 10)
self.intermediate = nn.Linear(10, 10)
self.layer_norm = nn.LayerNorm(10)
```
Now we can create an instance of this model definition which will fill
all weights: `dense`, `intermediate`, `layer_norm` with random weights.
We can print the model to see its architecture
```python
model = SimpleModel()
print(model)
```
This will print out the following:
```bash
SimpleModel(
(dense): Linear(in_features=10, out_features=10, bias=True)
(intermediate): Linear(in_features=10, out_features=10, bias=True)
(layer_norm): LayerNorm((10,), eps=1e-05, elementwise_affine=True)
)
```
We can see that the layer names are defined by the name of the class
attribute in PyTorch. You can print out the weight values of a specific
layer:
```python
print(model.dense.weight.data)
```
to see that the weights were randomly initialized
```bash
tensor([[-0.0818, 0.2207, -0.0749, -0.0030, 0.0045, -0.1569, -0.1598, 0.0212,
-0.2077, 0.2157],
[ 0.1044, 0.0201, 0.0990, 0.2482, 0.3116, 0.2509, 0.2866, -0.2190,
0.2166, -0.0212],
[-0.2000, 0.1107, -0.1999, -0.3119, 0.1559, 0.0993, 0.1776, -0.1950,
-0.1023, -0.0447],
[-0.0888, -0.1092, 0.2281, 0.0336, 0.1817, -0.0115, 0.2096, 0.1415,
-0.1876, -0.2467],
[ 0.2208, -0.2352, -0.1426, -0.2636, -0.2889, -0.2061, -0.2849, -0.0465,
0.2577, 0.0402],
[ 0.1502, 0.2465, 0.2566, 0.0693, 0.2352, -0.0530, 0.1859, -0.0604,
0.2132, 0.1680],
[ 0.1733, -0.2407, -0.1721, 0.1484, 0.0358, -0.0633, -0.0721, -0.0090,
0.2707, -0.2509],
[-0.1173, 0.1561, 0.2945, 0.0595, -0.1996, 0.2988, -0.0802, 0.0407,
0.1829, -0.1568],
[-0.1164, -0.2228, -0.0403, 0.0428, 0.1339, 0.0047, 0.1967, 0.2923,
0.0333, -0.0536],
[-0.1492, -0.1616, 0.1057, 0.1950, -0.2807, -0.2710, -0.1586, 0.0739,
0.2220, 0.2358]]).
```
In the conversion script, you should fill those randomly initialized
weights with the exact weights of the corresponding layer in the
checkpoint. *E.g.*,
```python
# retrieve matching layer weights, e.g. by
# recursive algorithm
layer_name = "dense"
pretrained_weight = array_of_dense_layer
model_pointer = getattr(model, "dense")
model_pointer.weight.data = torch.from_numpy(pretrained_weight)
```
While doing so, you must verify that each randomly initialized weight of
your PyTorch model and its corresponding pretrained checkpoint weight
exactly match in both **shape and name**. To do so, it is **necessary**
to add assert statements for the shape and print out the names of the
checkpoints weights. *E.g.*, you should add statements like:
```python
assert (
model_pointer.weight.shape == pretrained_weight.shape
), f"Pointer shape of random weight {model_pointer.shape} and array shape of checkpoint weight {pretrained_weight.shape} mismatched"
```
Besides, you should also print out the names of both weights to make
sure they match, *e.g.*,
```python
logger.info(f"Initialize PyTorch weight {layer_name} from {pretrained_weight.name}")
```
If either the shape or the name doesn't match, you probably assigned
the wrong checkpoint weight to a randomly initialized layer of the 🤗
Transformers implementation.
An incorrect shape is most likely due to an incorrect setting of the
config parameters in `[camelcase name of model]Config()` that do not exactly match
those that were used for the checkpoint you want to convert. However, it
could also be that PyTorch's implementation of a layer requires the
weight to be transposed beforehand.
Finally, you should also check that **all** required weights are
initialized and print out all checkpoint weights that were not used for
initialization to make sure the model is correctly converted. It is
completely normal, that the conversion trials fail with either a wrong
shape statement or wrong name assignment. This is most likely because
either you used incorrect parameters in `[camelcase name of model]Config()`, have a
wrong architecture in the 🤗 Transformers implementation, you have a bug
in the `init()` functions of one of the components of the 🤗 Transformers
implementation or you need to transpose one of the checkpoint weights.
This step should be iterated with the previous step until all weights of
the checkpoint are correctly loaded in the Transformers model. Having
correctly loaded the checkpoint into the 🤗 Transformers implementation,
you can then save the model under a folder of your choice
`/path/to/converted/checkpoint/folder` that should then contain both a
`pytorch_model.bin` file and a `config.json` file:
```python
model.save_pretrained("/path/to/converted/checkpoint/folder")
```
[TODO FILL: Here the mentor should add very specific information on what exactly has to be done for the conversion of this model]
[...]
[...]
**7. Implement the forward pass**
Having managed to correctly load the pretrained weights into the 🤗
Transformers implementation, you should now make sure that the forward
pass is correctly implemented. In [Get familiar with the original
repository](#34-run-a-pretrained-checkpoint-using-the-original-repository),
you have already created a script that runs a forward pass of the model
using the original repository. Now you should write an analogous script
using the 🤗 Transformers implementation instead of the original one. It
should look as follows:
[TODO FILL: Here the model name might have to be adapted, *e.g.*, maybe [camelcase name of model]ForConditionalGeneration instead of [camelcase name of model]Model]
```python
model = [camelcase name of model]Model.from_pretrained("/path/to/converted/checkpoint/folder")
input_ids = [0, 4, 4, 3, 2, 4, 1, 7, 19]
output = model(input_ids).last_hidden_states
```
It is very likely that the 🤗 Transformers implementation and the
original model implementation don't give the exact same output the very
first time or that the forward pass throws an error. Don't be
disappointed - it's expected! First, you should make sure that the
forward pass doesn't throw any errors. It often happens that the wrong
dimensions are used leading to a `"Dimensionality mismatch"`
error or that the wrong data type object is used, *e.g.*, `torch.long`
instead of `torch.float32`. Don't hesitate to ask [name of mentor]
for help, if you don't manage to solve certain errors.
The final part to make sure the 🤗 Transformers implementation works
correctly is to ensure that the outputs are equivalent to a precision of
`1e-3`. First, you should ensure that the output shapes are identical,
*i.e.* `outputs.shape` should yield the same value for the script of the
🤗 Transformers implementation and the original implementation. Next, you
should make sure that the output values are identical as well. This one
of the most difficult parts of adding a new model. Common mistakes why
the outputs are not identical are:
- Some layers were not added, *i.e.* an activation layer
was not added, or the residual connection was forgotten
- The word embedding matrix was not tied
- The wrong positional embeddings are used because the original
implementation uses on offset
- Dropout is applied during the forward pass. To fix this make sure
`model.training is False` and that no dropout layer is
falsely activated during the forward pass, *i.e.* pass
`self.training` to [PyTorch's functional
dropout](https://pytorch.org/docs/stable/nn.functional.html?highlight=dropout#torch.nn.functional.dropout)
The best way to fix the problem is usually to look at the forward pass
of the original implementation and the 🤗 Transformers implementation
side-by-side and check if there are any differences. Ideally, you should
debug/print out intermediate outputs of both implementations of the
forward pass to find the exact position in the network where the 🤗
Transformers implementation shows a different output than the original
implementation. First, make sure that the hard-coded `input_ids` in both
scripts are identical. Next, verify that the outputs of the first
transformation of the `input_ids` (usually the word embeddings) are
identical. And then work your way up to the very last layer of the
network. At some point, you will notice a difference between the two
implementations, which should point you to the bug in the 🤗 Transformers
implementation. From our experience, a simple and efficient way is to
add many print statements in both the original implementation and 🤗
Transformers implementation, at the same positions in the network
respectively, and to successively remove print statements showing the
same values for intermediate presentions.
When you're confident that both implementations yield the same output,
verifying the outputs with
`torch.allclose(original_output, output, atol=1e-3)`, you're done with
the most difficult part! Congratulations - the work left to be done
should be a cakewalk 😊.
**8. Adding all necessary model tests**
At this point, you have successfully added a new model. However, it is
very much possible that the model does not yet fully comply with the
required design. To make sure, the implementation is fully compatible
with 🤗 Transformers, all common tests should pass. The Cookiecutter
should have automatically added a test file for your model, probably
under the same `tests/test_modeling_[lowercase name of model].py`. Run this test
file to verify that all common tests pass:
```python
pytest tests/test_modeling_[lowercase name of model].py
```
[TODO FILL: Here the mentor should add very specific information on what tests are likely to fail after having implemented the model
, e.g. given the model, it might be very likely that `test_attention_output` fails]
[...]
[...]
Having fixed all common tests, it is now crucial to ensure that all the
nice work you have done is well tested, so that
- a) The community can easily understand your work by looking at
specific tests of *[camelcase name of model]*
- b) Future changes to your model will not break any important
feature of the model.
At first, integration tests should be added. Those integration tests
essentially do the same as the debugging scripts you used earlier to
implement the model to 🤗 Transformers. A template of those model tests
is already added by the Cookiecutter, called
`[camelcase name of model]ModelIntegrationTests` and only has to be filled out by
you. To ensure that those tests are passing, run
```python
RUN_SLOW=1 pytest -sv tests/test_modeling_[lowercase name of model].py::[camelcase name of model]ModelIntegrationTests
```
**Note:** In case you are using Windows, you should replace `RUN_SLOW=1` with `SET RUN_SLOW=1`
Second, all features that are special to *[camelcase name of model]* should be
tested additionally in a separate test under
`[camelcase name of model]ModelTester`/`[camelcase name of model]ModelTest`. This part is often
forgotten but is extremely useful in two ways:
- It helps to transfer the knowledge you have acquired during the
model addition to the community by showing how the special features
of *[camelcase name of model]* should work.
- Future contributors can quickly test changes to the model by running
those special tests.
[TODO FILL: Here the mentor should add very specific information on what special features of the model should be tested additionally]
[...]
[...]
**9. Implement the tokenizer**
Next, we should add the tokenizer of *[camelcase name of model]*. Usually, the
tokenizer is equivalent or very similar to an already existing tokenizer
of 🤗 Transformers.
[TODO FILL: Here the mentor should add a comment whether a new tokenizer is required or if this is not the case which existing tokenizer closest resembles
[camelcase name of model]'s tokenizer and how the tokenizer should be implemented]
[...]
[...]
It is very important to find/extract the original tokenizer file and to
manage to load this file into the 🤗 Transformers' implementation of the
tokenizer.
For [camelcase name of model], the tokenizer files can be found here:
- [To be filled out by mentor]
and having implemented the 🤗 Transformers' version of the tokenizer can be loaded as follows:
[To be filled out by mentor]
To ensure that the tokenizer works correctly, it is recommended to first
create a script in the original repository that inputs a string and
returns the `input_ids`. It could look similar to this (in pseudo-code):
```bash
input_str = "This is a long example input string containing special characters .$?-, numbers 2872 234 12 and words."
model = [camelcase name of model]Model.load_pretrained_checkpoint("/path/to/checkpoint/")
input_ids = model.tokenize(input_str)
```
You might have to take a deeper look again into the original repository
to find the correct tokenizer function or you might even have to do
changes to your clone of the original repository to only output the
`input_ids`. Having written a functional tokenization script that uses
the original repository, an analogous script for 🤗 Transformers should
be created. It should look similar to this:
```python
from transformers import [camelcase name of model]Tokenizer
input_str = "This is a long example input string containing special characters .$?-, numbers 2872 234 12 and words."
tokenizer = [camelcase name of model]Tokenizer.from_pretrained("/path/to/tokenizer/folder/")
input_ids = tokenizer(input_str).input_ids
```
When both `input_ids` yield the same values, as a final step a tokenizer
test file should also be added.
[TODO FILL: Here mentor should point the student to test files of similar tokenizers]
Analogous to the modeling test files of *[camelcase name of model]*, the
tokenization test files of *[camelcase name of model]* should contain a couple of
hard-coded integration tests.
[TODO FILL: Here mentor should again point to an existing similar test of another model that the student can copy & adapt]
**10. Run End-to-end integration tests**
Having added the tokenizer, you should also add a couple of end-to-end
integration tests using both the model and the tokenizer to
`tests/test_modeling_[lowercase name of model].py` in 🤗 Transformers. Such a test
should show on a meaningful text-to-text sample that the 🤗 Transformers
implementation works as expected. A meaningful text-to-text sample can
include *e.g.* a source-to-target-translation pair, an
article-to-summary pair, a question-to-answer pair, etc... If none of
the ported checkpoints has been fine-tuned on a downstream task it is
enough to simply rely on the model tests. In a final step to ensure that
the model is fully functional, it is advised that you also run all tests
on GPU. It can happen that you forgot to add some `.to(self.device)`
statements to internal tensors of the model, which in such a test would
show in an error. In case you have no access to a GPU, the Hugging Face
team can take care of running those tests for you.
**11. Add Docstring**
Now, all the necessary functionality for *[camelcase name of model]* is added -
you're almost done! The only thing left to add is a nice docstring and
a doc page. The Cookiecutter should have added a template file called
`docs/source/model_doc/[lowercase name of model].rst` that you should fill out.
Users of your model will usually first look at this page before using
your model. Hence, the documentation must be understandable and concise.
It is very useful for the community to add some *Tips* to show how the
model should be used. Don't hesitate to ping [name of mentor]
regarding the docstrings.
Next, make sure that the docstring added to
`src/transformers/models/[lowercase name of model]/modeling_[lowercase name of model].py` is
correct and included all necessary inputs and outputs. It is always to
good to remind oneself that documentation should be treated at least as
carefully as the code in 🤗 Transformers since the documentation is
usually the first contact point of the community with the model.
**Code refactor**
Great, now you have added all the necessary code for *[camelcase name of model]*.
At this point, you should correct some potential incorrect code style by
running:
```bash
make style
```
and verify that your coding style passes the quality check:
```bash
make quality
```
There are a couple of other very strict design tests in 🤗 Transformers
that might still be failing, which shows up in the tests of your pull
request. This is often because of some missing information in the
docstring or some incorrect naming. [name of mentor] will surely
help you if you're stuck here.
Lastly, it is always a good idea to refactor one's code after having
ensured that the code works correctly. With all tests passing, now it's
a good time to go over the added code again and do some refactoring.
You have now finished the coding part, congratulation! 🎉 You are
Awesome! 😎
**12. Upload the models to the model hub**
In this final part, you should convert and upload all checkpoints to the
model hub and add a model card for each uploaded model checkpoint. You
should work alongside [name of mentor] here to decide on a fitting
name for each checkpoint and to get the required access rights to be
able to upload the model under the author's organization of
*[camelcase name of model]*.
It is worth spending some time to create fitting model cards for each
checkpoint. The model cards should highlight the specific
characteristics of this particular checkpoint, *e.g.*, On which dataset
was the checkpoint pretrained/fine-tuned on? On what down-stream task
should the model be used? And also include some code on how to correctly
use the model.
**13. (Optional) Add notebook**
It is very helpful to add a notebook that showcases in-detail how
*[camelcase name of model]* can be used for inference and/or fine-tuned on a
downstream task. This is not mandatory to merge your PR, but very useful
for the community.
**14. Submit your finished PR**
You're done programming now and can move to the last step, which is
getting your PR merged into main. Usually, [name of mentor]
should have helped you already at this point, but it is worth taking
some time to give your finished PR a nice description and eventually add
comments to your code, if you want to point out certain design choices
to your reviewer.
### Share your work!!
Now, it's time to get some credit from the community for your work!
Having completed a model addition is a major contribution to
Transformers and the whole NLP community. Your code and the ported
pre-trained models will certainly be used by hundreds and possibly even
thousands of developers and researchers. You should be proud of your
work and share your achievement with the community.
**You have made another model that is super easy to access for everyone
in the community! 🤯**
| 0
|
mavonic_private_repos/transformers/templates
|
mavonic_private_repos/transformers/templates/adding_a_new_model/README.md
|
<!---
Copyright 2020 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
-->
# Adding a new model
This page has been updated in light of the removal of the `add_new_model` script in favor of the more complete
`add_new_model_like` script.
We recommend you checkout the documentation of [How to add a model](https://huggingface.co/docs/transformers/main/en/add_new_model)
in the Hugging Face Transformers documentation for complete and up-to-date instructions.
| 0
|
mavonic_private_repos/transformers/templates/adding_a_new_model
|
mavonic_private_repos/transformers/templates/adding_a_new_model/open_model_proposals/ADD_BIG_BIRD.md
|
How to add BigBird to 🤗 Transformers?
=====================================
Mentor: [Patrick](https://github.com/patrickvonplaten)
Begin: 12.02.2020
Estimated End: 19.03.2020
Contributor: [Vasudev](https://github.com/thevasudevgupta)
Adding a new model is often difficult and requires an in-depth knowledge
of the 🤗 Transformers library and ideally also of the model's original
repository. At Hugging Face, we are trying to empower the community more
and more to add models independently.
The following sections explain in detail how to add BigBird
to Transformers. You will work closely with Patrick to
integrate BigBird into Transformers. By doing so, you will both gain a
theoretical and deep practical understanding of BigBird.
But more importantly, you will have made a major
open-source contribution to Transformers. Along the way, you will:
- get insights into open-source best practices
- understand the design principles of one of the most popular NLP
libraries
- learn how to do efficiently test large NLP models
- learn how to integrate Python utilities like `black`, `ruff`,
`make fix-copies` into a library to always ensure clean and readable
code
To start, let's try to get a general overview of the Transformers
library.
General overview of 🤗 Transformers
----------------------------------
First, you should get a general overview of 🤗 Transformers. Transformers
is a very opinionated library, so there is a chance that
you don't agree with some of the library's philosophies or design
choices. From our experience, however, we found that the fundamental
design choices and philosophies of the library are crucial to
efficiently scale Transformers while keeping maintenance costs at a
reasonable level.
A good first starting point to better understand the library is to read
the [documentation of our philosophy](https://huggingface.co/transformers/philosophy.html).
As a result of our way of working, there are some choices that we try to apply to all models:
- Composition is generally favored over abstraction
- Duplicating code is not always bad if it strongly improves the
readability or accessibility of a model
- Model files are as self-contained as possible so that when you read
the code of a specific model, you ideally only have to look into the
respective `modeling_....py` file.
In our opinion, the library's code is not just a means to provide a
product, *e.g.*, the ability to use BERT for inference, but also as the
very product that we want to improve. Hence, when adding a model, the
user is not only the person that will use your model, but also everybody
that will read, try to understand, and possibly tweak your code.
With this in mind, let's go a bit deeper into the general library
design.
### Overview of models
To successfully add a model, it is important to understand the
interaction between your model and its config,
`PreTrainedModel`, and `PretrainedConfig`. For
exemplary purposes, we will call the PyTorch model to be added to 🤗 Transformers
`BrandNewBert`.
Let's take a look:
As you can see, we do make use of inheritance in 🤗 Transformers, but we
keep the level of abstraction to an absolute minimum. There are never
more than two levels of abstraction for any model in the library.
`BrandNewBertModel` inherits from
`BrandNewBertPreTrainedModel` which in
turn inherits from `PreTrainedModel` and that's it.
As a general rule, we want to make sure
that a new model only depends on `PreTrainedModel`. The
important functionalities that are automatically provided to every new
model are
`PreTrainedModel.from_pretrained` and `PreTrainedModel.save_pretrained`, which are
used for serialization and deserialization. All
of the other important functionalities, such as
`BrandNewBertModel.forward` should be
completely defined in the new `modeling_brand_new_bert.py` module. Next,
we want to make sure that a model with a specific head layer, such as
`BrandNewBertForMaskedLM` does not inherit
from `BrandNewBertModel`, but rather uses
`BrandNewBertModel` as a component that
can be called in its forward pass to keep the level of abstraction low.
Every new model requires a configuration class, called
`BrandNewBertConfig`. This configuration
is always stored as an attribute in
`PreTrainedModel`, and
thus can be accessed via the `config` attribute for all classes
inheriting from `BrandNewBertPreTrainedModel`
```python
# assuming that `brand_new_bert` belongs to the organization `brandy`
model = BrandNewBertModel.from_pretrained("brandy/brand_new_bert")
model.config # model has access to its config
```
Similar to the model, the configuration inherits basic serialization and
deserialization functionalities from
`PretrainedConfig`. Note
that the configuration and the model are always serialized into two
different formats - the model to a `pytorch_model.bin` file
and the configuration to a `config.json` file. Calling
`PreTrainedModel.save_pretrained` will automatically call
`PretrainedConfig.save_pretrained`, so that both model and configuration are saved.
### Overview of tokenizers
Not quite ready yet :-( This section will be added soon!
Step-by-step recipe to add a model to 🤗 Transformers
----------------------------------------------------
Everyone has different preferences of how to port a model so it can be
very helpful for you to take a look at summaries of how other
contributors ported models to Hugging Face. Here is a list of community
blog posts on how to port a model:
1. [Porting GPT2
Model](https://medium.com/huggingface/from-tensorflow-to-pytorch-265f40ef2a28)
by [Thomas](https://huggingface.co/thomwolf)
2. [Porting WMT19 MT Model](https://huggingface.co/blog/porting-fsmt)
by [Stas](https://huggingface.co/stas)
From experience, we can tell you that the most important things to keep
in mind when adding a model are:
- Don't reinvent the wheel! Most parts of the code you will add for
the new 🤗 Transformers model already exist somewhere in 🤗
Transformers. Take some time to find similar, already existing
models and tokenizers you can copy from.
[grep](https://www.gnu.org/software/grep/) and
[rg](https://github.com/BurntSushi/ripgrep) are your friends. Note
that it might very well happen that your model's tokenizer is based
on one model implementation, and your model's modeling code on
another one. *E.g.*, FSMT's modeling code is based on BART, while
FSMT's tokenizer code is based on XLM.
- It's more of an engineering challenge than a scientific challenge.
You should spend more time on creating an efficient debugging
environment than trying to understand all theoretical aspects of the
model in the paper.
- Ask for help when you're stuck! Models are the core component of 🤗
Transformers so we, at Hugging Face, are more than happy to help
you at every step to add your model. Don't hesitate to ask if you
notice you are not making progress.
In the following, we try to give you a general recipe that we found most
useful when porting a model to 🤗 Transformers.
The following list is a summary of everything that has to be done to add
a model and can be used by you as a To-Do List:
1. [ ] (Optional) Understood theoretical aspects
2. [ ] Prepared transformers dev environment
3. [ ] Set up debugging environment of the original repository
4. [ ] Created script that successfully runs forward pass using
original repository and checkpoint
5. [ ] Successfully opened a PR and added the model skeleton to Transformers
6. [ ] Successfully converted original checkpoint to Transformers
checkpoint
7. [ ] Successfully ran forward pass in Transformers that gives
identical output to original checkpoint
8. [ ] Finished model tests in Transformers
9. [ ] Successfully added Tokenizer in Transformers
10. [ ] Run end-to-end integration tests
11. [ ] Finished docs
12. [ ] Uploaded model weights to the hub
13. [ ] Submitted the pull request for review
14. [ ] (Optional) Added a demo notebook
To begin with, we usually recommend to start by getting a good
theoretical understanding of `BigBird`. However, if you prefer to
understand the theoretical aspects of the model *on-the-job*, then it is
totally fine to directly dive into the `BigBird`'s code-base. This
option might suit you better, if your engineering skills are better than
your theoretical skill, if you have trouble understanding
`BigBird`'s paper, or if you just enjoy programming much more than
reading scientific papers.
### 1. (Optional) Theoretical aspects of BigBird
You should take some time to read *BigBird's* paper, if such
descriptive work exists. There might be large sections of the paper that
are difficult to understand. If this is the case, this is fine - don't
worry! The goal is not to get a deep theoretical understanding of the
paper, but to extract the necessary information required to effectively
re-implement the model in 🤗 Transformers. That being said, you don't
have to spend too much time on the theoretical aspects, but rather focus
on the practical ones, namely:
- What type of model is *BigBird*? BERT-like encoder-only
model? GPT2-like decoder-only model? BART-like encoder-decoder
model? Look at the `model_summary` if
you're not familiar with the differences between those.
- What are the applications of *BigBird*? Text
classification? Text generation? Seq2Seq tasks, *e.g.,*
summarization?
- What is the novel feature of the model making it different from
BERT/GPT-2/BART?
- Which of the already existing [🤗 Transformers
models](https://huggingface.co/transformers/#contents) is most
similar to *BigBird*?
- What type of tokenizer is used? A sentencepiece tokenizer? Word
piece tokenizer? Is it the same tokenizer as used for BERT or BART?
After you feel like you have gotten a good overview of the architecture
of the model, you might want to write to Patrick with any
questions you might have. This might include questions regarding the
model's architecture, its attention layer, etc. We will be more than
happy to help you.
#### Additional resources
Before diving into the code, here are some additional resources that might be worth taking a look at:
- [Yannic Kilcher's paper summary](https://www.youtube.com/watch?v=WVPE62Gk3EM&ab_channel=YannicKilcher)
- [Yannic Kilcher's summary of Longformer](https://www.youtube.com/watch?v=_8KNb5iqblE&ab_channel=YannicKilcher) - Longformer and BigBird are **very** similar models. Since Longformer has already been ported to 🤗 Transformers, it is useful to understand the differences between the two models
- [Blog post](https://medium.com/dsc-msit/is-google-bigbird-gonna-be-the-new-leader-in-nlp-domain-8c95cecc30f8) - A relatively superficial blog post about BigBird. Might be a good starting point to understand BigBird
#### Make sure you've understood the fundamental aspects of BigBird
Alright, now you should be ready to take a closer look into the actual code of BigBird.
You should have understood the following aspects of BigBird by now:
- BigBird provides a new attention layer for long-range sequence modelling that can be used
as a drop-in replacement for already existing architectures. This means that every transformer-based model architecture can replace its [Self-attention layer](https://towardsdatascience.com/illustrated-self-attention-2d627e33b20a) with BigBird's self-attention layer.
- BigBird's self-attention layer is composed of three mechanisms: block sparse (local) self-attention, global self-attention, random self-attention
- BigBird's block sparse (local) self-attention is different from Longformer's local self-attention. How so? Why does that matter? => Can be deployed on TPU much easier this way
- BigBird can be implemented for both an encoder-only model **and**
for an encoder-decoder model, which means that we can reuse lots of [code from RoBERTa](https://github.com/huggingface/transformers/blob/main/src/transformers/models/roberta/modeling_roberta.py) and [from PEGASUS](https://github.com/huggingface/transformers/blob/main/src/transformers/models/pegasus/modeling_pegasus.py) at a later stage.
If any of the mentioned aspects above are **not** clear to you, now is a great time to talk to Patrick.
### 2. Next prepare your environment
1. Fork the [repository](https://github.com/huggingface/transformers)
by clicking on the 'Fork' button on the repository's page. This
creates a copy of the code under your GitHub user account.
2. Clone your `transformers` fork to your local disk, and add the base
repository as a remote:
```bash
git clone https://github.com/[your Github handle]/transformers.git
cd transformers
git remote add upstream https://github.com/huggingface/transformers.git
```
3. Set up a development environment, for instance by running the
following command:
```bash
python -m venv .env
source .env/bin/activate
pip install -e ".[dev]"
```
and return to the parent directory
```bash
cd ..
```
4. We recommend adding the PyTorch version of *BigBird* to
Transformers. To install PyTorch, please follow the instructions [here](https://pytorch.org/get-started/locally/).
**Note:** You don't need to have CUDA installed. Making the new model
work on CPU is sufficient.
5. To port *BigBird*, you will also need access to its
original repository:
```bash
git clone https://github.com/google-research/bigbird.git
cd big_bird
pip install -e .
```
Now you have set up a development environment to port *BigBird*
to 🤗 Transformers.
### Run a pretrained checkpoint using the original repository
**3. Set up debugging environment**
At first, you will work on the original *BigBird* repository.
Often, the original implementation is very "researchy". Meaning that
documentation might be lacking and the code can be difficult to
understand. But this should be exactly your motivation to reimplement
*BigBird*. At Hugging Face, one of our main goals is to *make
people stand on the shoulders of giants* which translates here very well
into taking a working model and rewriting it to make it as **accessible,
user-friendly, and beautiful** as possible. This is the number-one
motivation to re-implement models into 🤗 Transformers - trying to make
complex new NLP technology accessible to **everybody**.
You should start thereby by diving into the [original repository](https://github.com/google-research/bigbird).
Successfully running the official pretrained model in the original
repository is often **the most difficult** step. From our experience, it
is very important to spend some time getting familiar with the original
code-base. You need to figure out the following:
- Where to find the pretrained weights?
- How to load the pretrained weights into the corresponding model?
- How to run the tokenizer independently from the model?
- Trace one forward pass so that you know which classes and functions
are required for a simple forward pass. Usually, you only have to
reimplement those functions.
- Be able to locate the important components of the model: Where is
the model's class? Are there model sub-classes, *e.g.*,
EncoderModel, DecoderModel? Where is the self-attention layer? Are
there multiple different attention layers, *e.g.*, *self-attention*,
*cross-attention*...?
- How can you debug the model in the original environment of the repo?
Do you have to add `print` statements, can you work with
an interactive debugger like [ipdb](https://pypi.org/project/ipdb/), or should you use
an efficient IDE to debug the model, like PyCharm?
It is very important that before you start the porting process, that you
can **efficiently** debug code in the original repository! Also,
remember that you are working with an open-source library, so do not
hesitate to open an issue, or even a pull request in the original
repository. The maintainers of this repository are most likely very
happy about someone looking into their code!
At this point, it is really up to you which debugging environment and
strategy you prefer to use to debug the original model. We strongly
advise against setting up a costly GPU environment, but simply work on a
CPU both when starting to dive into the original repository and also
when starting to write the 🤗 Transformers implementation of the model.
Only at the very end, when the model has already been successfully
ported to 🤗 Transformers, one should verify that the model also works as
expected on GPU.
In general, there are two possible debugging environments for running
the original model
- [Jupyter notebooks](https://jupyter.org/) / [google colab](https://colab.research.google.com/notebooks/intro.ipynb)
- Local python scripts.
Jupyter notebooks have the advantage that they allow for cell-by-cell
execution which can be helpful to better split logical components from
one another and to have faster debugging cycles as intermediate results
can be stored. Also, notebooks are often easier to share with other
contributors, which might be very helpful if you want to ask the Hugging
Face team for help. If you are familiar with Jupyter notebooks, we
strongly recommend you to work with them.
The obvious disadvantage of Jupyter notebooks is that if you are not
used to working with them you will have to spend some time adjusting to
the new programming environment and that you might not be able to use
your known debugging tools anymore, like `ipdb`.
**4. Successfully run forward pass**
For each code-base, a good first step is always to load a **small**
pretrained checkpoint and to be able to reproduce a single forward pass
using a dummy integer vector of input IDs as an input. Such a script
could look something like this:
```python
from bigbird.core import modeling
model = modeling.BertModel(bert_config)
from bigbird.core import utils
params = utils.BigBirdConfig(vocab_size=32000, hidden_size=512,
num_hidden_layers=8, num_attention_heads=6, intermediate_size=1024)
ckpt_path = 'gs://bigbird-transformer/pretrain/bigbr_base/model.ckpt-0'
ckpt_reader = tf.compat.v1.train.NewCheckpointReader(ckpt_path)
model.set_weights([ckpt_reader.get_tensor(v.name[:-2]) for v in tqdm(model.trainable_weights, position=0)])
input_ids = tf.constant([[31, 51, 99], [15, 5, 0]])
_, pooled_output = model(input_ids=input_ids, token_type_ids=token_type_ids)
...
```
Next, regarding the debugging strategy, there are generally a few from
which to choose from:
- Decompose the original model into many small testable components and
run a forward pass on each of those for verification
- Decompose the original model only into the original *tokenizer* and
the original *model*, run a forward pass on those, and use
intermediate print statements or breakpoints for verification
Again, it is up to you which strategy to choose. Often, one or the other
is advantageous depending on the original code base.
If the original code-base allows you to decompose the model into smaller
sub-components, *e.g.*, if the original code-base can easily be run in
eager mode, it is usually worth the effort to do so. There are some
important advantages to taking the more difficult road in the beginning:
- at a later stage when comparing the original model to the Hugging
Face implementation, you can verify automatically for each component
individually that the corresponding component of the 🤗 Transformers
implementation matches instead of relying on visual comparison via
print statements
- it can give you some rope to decompose the big problem of porting a
model into smaller problems of just porting individual components
and thus structure your work better
- separating the model into logical meaningful components will help
you to get a better overview of the model's design and thus to
better understand the model
- at a later stage those component-by-component tests help you to
ensure that no regression occurs as you continue changing your code
[Lysandre's](https://gist.github.com/LysandreJik/db4c948f6b4483960de5cbac598ad4ed)
integration checks for ELECTRA gives a nice example of how this can be
done.
However, if the original code-base is very complex or only allows
intermediate components to be run in a compiled mode, it might be too
time-consuming or even impossible to separate the model into smaller
testable sub-components. A good example is [T5's
MeshTensorFlow](https://github.com/tensorflow/mesh/tree/master/mesh_tensorflow)
library which is very complex and does not offer a simple way to
decompose the model into its sub-components. For such libraries, one
often relies on verifying print statements.
No matter which strategy you choose, the recommended procedure is often
the same in that you should start to debug the starting layers first and
the ending layers last.
It is recommended that you retrieve the output, either by print
statements or sub-component functions, of the following layers in the
following order:
1. Retrieve the input IDs passed to the model
2. Retrieve the word embeddings
3. Retrieve the input of the first Transformer layer
4. Retrieve the output of the first Transformer layer
5. Retrieve the output of the following n - 1 Transformer layers
6. Retrieve the output of the whole BigBird Model
Input IDs should thereby consists of an array of integers, *e.g.*,
`input_ids = [0, 4, 4, 3, 2, 4, 1, 7, 19]`
The outputs of the following layers often consist of multi-dimensional
float arrays and can look like this:
```bash
[[
[-0.1465, -0.6501, 0.1993, ..., 0.1451, 0.3430, 0.6024],
[-0.4417, -0.5920, 0.3450, ..., -0.3062, 0.6182, 0.7132],
[-0.5009, -0.7122, 0.4548, ..., -0.3662, 0.6091, 0.7648],
...,
[-0.5613, -0.6332, 0.4324, ..., -0.3792, 0.7372, 0.9288],
[-0.5416, -0.6345, 0.4180, ..., -0.3564, 0.6992, 0.9191],
[-0.5334, -0.6403, 0.4271, ..., -0.3339, 0.6533, 0.8694]]],
```
We expect that every model added to 🤗 Transformers passes a couple of
integration tests, meaning that the original model and the reimplemented
version in 🤗 Transformers have to give the exact same output up to a
precision of 0.001! Since it is normal that the exact same model written
in different libraries can give a slightly different output depending on
the library framework, we accept an error tolerance of 1e-3 (0.001). It
is not enough if the model gives nearly the same output, they have to be
the almost identical. Therefore, you will certainly compare the
intermediate outputs of the 🤗 Transformers version multiple times
against the intermediate outputs of the original implementation of
*BigBird* in which case an **efficient** debugging environment
of the original repository is absolutely important. Here is some advice
to make your debugging environment as efficient as possible.
- Find the best way of debugging intermediate results. Is the original
repository written in PyTorch? Then you should probably take the
time to write a longer script that decomposes the original model
into smaller sub-components to retrieve intermediate values. Is the
original repository written in Tensorflow 1? Then you might have to
rely on TensorFlow print operations like
[tf.print](https://www.tensorflow.org/api_docs/python/tf/print) to
output intermediate values. Is the original repository written in
Jax? Then make sure that the model is **not jitted** when running
the forward pass, *e.g.*, check-out [this
link](https://github.com/google/jax/issues/196).
- Use the smallest pretrained checkpoint you can find. The smaller the
checkpoint, the faster your debug cycle becomes. It is not efficient
if your pretrained model is so big that your forward pass takes more
than 10 seconds. In case only very large checkpoints are available,
it might make more sense to create a dummy model in the new
environment with randomly initialized weights and save those weights
for comparison with the 🤗 Transformers version of your model
- Make sure you are using the easiest way of calling a forward pass in
the original repository. Ideally, you want to find the function in
the original repository that **only** calls a single forward pass,
*i.e.* that is often called `predict`, `evaluate`, `forward` or
`__call__`. You don't want to debug a function that calls `forward`
multiple times, *e.g.*, to generate text, like
`autoregressive_sample`, `generate`.
- Try to separate the tokenization from the model's
forward pass. If the original repository shows
examples where you have to input a string, then try to find out
where in the forward call the string input is changed to input ids
and start from this point. This might mean that you have to possibly
write a small script yourself or change the original code so that
you can directly input the ids instead of an input string.
- Make sure that the model in your debugging setup is **not** in
training mode, which often causes the model to yield random outputs
due to multiple dropout layers in the model. Make sure that the
forward pass in your debugging environment is **deterministic** so
that the dropout layers are not used. Or use
`transformers.utils.set_seed` if the old and new
implementations are in the same framework.
#### (Important) More details on how to create a debugging environment for BigBird
- BigBird has multiple pretrained checkpoints that should eventually all be ported to
🤗 Transformers. The pretrained checkpoints can be found [here](https://console.cloud.google.com/storage/browser/bigbird-transformer/pretrain;tab=objects?prefix=&forceOnObjectsSortingFiltering=false).
Those checkpoints include both pretrained weights for encoder-only (BERT/RoBERTa) under the folder `bigbr_base` and encoder-decoder (PEGASUS) under the folder `bigbp_large`.
You should start by porting the `bigbr_base` model. The encoder-decoder model
can be ported afterward.
for an encoder-decoder architecture as well as an encoder-only architecture.
- BigBird was written in tf.compat meaning that a mixture of a TensorFlow 1 and
TensorFlow 2 API was used.
- The most important part of the BigBird code-base is [bigbird.bigbird.core](https://github.com/google-research/bigbird/tree/master/bigbird/core) which includes all logic necessary
to implement BigBird.
- The first goal should be to successfully run a forward pass using the RoBERTa checkpoint `bigbr_base/model.ckpt-0.data-00000-of-00001` and `bigbr_base/model.ckpt-0.index`.
### Port BigBird to 🤗 Transformers
Next, you can finally start adding new code to 🤗 Transformers. Go into
the clone of your 🤗 Transformers' fork:
cd transformers
In the special case that you are adding a model whose architecture
exactly matches the model architecture of an existing model you only
have to add a conversion script as described in [this
section](#write-a-conversion-script). In this case, you can just re-use
the whole model architecture of the already existing model.
Otherwise, let's start generating a new model with the amazing
Cookiecutter!
**Use the Cookiecutter to automatically generate the model's code**
To begin with head over to the [🤗 Transformers
templates](https://github.com/huggingface/transformers/tree/main/templates/adding_a_new_model)
to make use of our `cookiecutter` implementation to automatically
generate all the relevant files for your model. Again, we recommend only
adding the PyTorch version of the model at first. Make sure you follow
the instructions of the `README.md` on the [🤗 Transformers
templates](https://github.com/huggingface/transformers/tree/main/templates/adding_a_new_model)
carefully.
Since you will first implement the Encoder-only/RoBERTa-like version of BigBird you should
select the `is_encoder_decoder_model = False` option in the cookiecutter. Also, it is recommended
that you implement the model only in PyTorch in the beginning and select "Standalone" as the
tokenizer type for now.
**Open a Pull Request on the main huggingface/transformers repo**
Before starting to adapt the automatically generated code, now is the
time to open a "Work in progress (WIP)" pull request, *e.g.*, "\[WIP\]
Add *BigBird*", in 🤗 Transformers so that you and the Hugging
Face team can work side-by-side on integrating the model into 🤗
Transformers.
You should do the following:
1. Create a branch with a descriptive name from your main branch
```bash
git checkout -b add_big_bird
```
2. Commit the automatically generated code:
```bash
git add .
git commit
```
3. Fetch and rebase to current main
```bash
git fetch upstream
git rebase upstream/main
```
4. Push the changes to your account using:
```bash
git push -u origin a-descriptive-name-for-my-changes
```
5. Once you are satisfied, go to the webpage of your fork on GitHub.
Click on "Pull request". Make sure to add the GitHub handle of Patrick
as one reviewer, so that the Hugging Face team gets notified for future changes.
6. Change the PR into a draft by clicking on "Convert to draft" on the
right of the GitHub pull request web page.
In the following, whenever you have done some progress, don't forget to
commit your work and push it to your account so that it shows in the
pull request. Additionally, you should make sure to update your work
with the current main from time to time by doing:
git fetch upstream
git merge upstream/main
In general, all questions you might have regarding the model or your
implementation should be asked in your PR and discussed/solved in the
PR. This way, Patrick will always be notified when you are
committing new code or if you have a question. It is often very helpful
to point Patrick to your added code so that the Hugging
Face team can efficiently understand your problem or question.
To do so, you can go to the "Files changed" tab where you see all of
your changes, go to a line regarding which you want to ask a question,
and click on the "+" symbol to add a comment. Whenever a question or
problem has been solved, you can click on the "Resolve" button of the
created comment.
In the same way, Patrick will open comments when reviewing
your code. We recommend asking most questions on GitHub on your PR. For
some very general questions that are not very useful for the public,
feel free to ping Patrick by Slack or email.
**5. Adapt the generated models code for BigBird**
At first, we will focus only on the model itself and not care about the
tokenizer. All the relevant code should be found in the generated files
`src/transformers/models/big_bird/modeling_big_bird.py` and
`src/transformers/models/big_bird/configuration_big_bird.py`.
Now you can finally start coding :). The generated code in
`src/transformers/models/big_bird/modeling_big_bird.py` will
either have the same architecture as BERT if it's an encoder-only model
or BART if it's an encoder-decoder model. At this point, you should
remind yourself what you've learned in the beginning about the
theoretical aspects of the model: *How is the model different from BERT
or BART?*\". Implement those changes which often means to change the
*self-attention* layer, the order of the normalization layer, etc...
Again, it is often useful to look at the similar architecture of already
existing models in Transformers to get a better feeling of how your
model should be implemented.
**Note** that at this point, you don't have to be very sure that your
code is fully correct or clean. Rather, it is advised to add a first
*unclean*, copy-pasted version of the original code to
`src/transformers/models/big_bird/modeling_big_bird.py`
until you feel like all the necessary code is added. From our
experience, it is much more efficient to quickly add a first version of
the required code and improve/correct the code iteratively with the
conversion script as described in the next section. The only thing that
has to work at this point is that you can instantiate the 🤗 Transformers
implementation of *BigBird*, *i.e.* the following command
should work:
```python
from transformers import BigBirdModel, BigBirdConfig
model = BigBirdModel(BigBirdConfig())
```
The above command will create a model according to the default
parameters as defined in `BigBirdConfig()` with random weights,
thus making sure that the `init()` methods of all components works.
Note that for BigBird you have to change the attention layer. BigBird's attention
layer is quite complex as you can see [here](https://github.com/google-research/bigbird/blob/103a3345f94bf6364749b51189ed93024ca5ef26/bigbird/core/attention.py#L560). Don't
feel discouraged by this! In a first step you should simply make sure that
the layer `BigBirdAttention` has the correct weights as can be found in the
pretrained checkpoints. This means that you have to make sure that in the
`__init__(self, ...)` function of `BigBirdAttention`, all submodules include all
necessary `nn.Module` layers. Only at a later stage do we need to fully rewrite
the complex attention function.
**6. Write a conversion script**
Next, you should write a conversion script that lets you convert the
checkpoint you used to debug *BigBird* in the original
repository to a checkpoint compatible with your just created 🤗
Transformers implementation of *BigBird*. It is not advised to
write the conversion script from scratch, but rather to look through
already existing conversion scripts in 🤗 Transformers for one that has
been used to convert a similar model that was written in the same
framework as *BigBird*. Usually, it is enough to copy an
already existing conversion script and slightly adapt it for your use
case. Don't hesitate to ask Patrick to point you to a
similar already existing conversion script for your model.
- A good starting point to convert the original TF BigBird implementation to the PT Hugging Face implementation is probably BERT's conversion script
[here](https://github.com/huggingface/transformers/blob/7acfa95afb8194f8f9c1f4d2c6028224dbed35a2/src/transformers/models/bert/modeling_bert.py#L91)
You can copy paste the conversion function into `modeling_big_bird.py` and then adapt it
to your needs.
In the following, we'll quickly explain how PyTorch models store layer
weights and define layer names. In PyTorch, the name of a layer is
defined by the name of the class attribute you give the layer. Let's
define a dummy model in PyTorch, called `SimpleModel` as follows:
```python
from torch import nn
class SimpleModel(nn.Module):
def __init__(self):
super().__init__()
self.dense = nn.Linear(10, 10)
self.intermediate = nn.Linear(10, 10)
self.layer_norm = nn.LayerNorm(10)
```
Now we can create an instance of this model definition which will fill
all weights: `dense`, `intermediate`, `layer_norm` with random weights.
We can print the model to see its architecture
```python
model = SimpleModel()
print(model)
```
This will print out the following:
```bash
SimpleModel(
(dense): Linear(in_features=10, out_features=10, bias=True)
(intermediate): Linear(in_features=10, out_features=10, bias=True)
(layer_norm): LayerNorm((10,), eps=1e-05, elementwise_affine=True)
)
```
We can see that the layer names are defined by the name of the class
attribute in PyTorch. You can print out the weight values of a specific
layer:
```python
print(model.dense.weight.data)
```
to see that the weights were randomly initialized
```bash
tensor([[-0.0818, 0.2207, -0.0749, -0.0030, 0.0045, -0.1569, -0.1598, 0.0212,
-0.2077, 0.2157],
[ 0.1044, 0.0201, 0.0990, 0.2482, 0.3116, 0.2509, 0.2866, -0.2190,
0.2166, -0.0212],
[-0.2000, 0.1107, -0.1999, -0.3119, 0.1559, 0.0993, 0.1776, -0.1950,
-0.1023, -0.0447],
[-0.0888, -0.1092, 0.2281, 0.0336, 0.1817, -0.0115, 0.2096, 0.1415,
-0.1876, -0.2467],
[ 0.2208, -0.2352, -0.1426, -0.2636, -0.2889, -0.2061, -0.2849, -0.0465,
0.2577, 0.0402],
[ 0.1502, 0.2465, 0.2566, 0.0693, 0.2352, -0.0530, 0.1859, -0.0604,
0.2132, 0.1680],
[ 0.1733, -0.2407, -0.1721, 0.1484, 0.0358, -0.0633, -0.0721, -0.0090,
0.2707, -0.2509],
[-0.1173, 0.1561, 0.2945, 0.0595, -0.1996, 0.2988, -0.0802, 0.0407,
0.1829, -0.1568],
[-0.1164, -0.2228, -0.0403, 0.0428, 0.1339, 0.0047, 0.1967, 0.2923,
0.0333, -0.0536],
[-0.1492, -0.1616, 0.1057, 0.1950, -0.2807, -0.2710, -0.1586, 0.0739,
0.2220, 0.2358]]).
```
In the conversion script, you should fill those randomly initialized
weights with the exact weights of the corresponding layer in the
checkpoint. *E.g.*,
```python
# retrieve matching layer weights, e.g. by
# recursive algorithm
layer_name = "dense"
pretrained_weight = array_of_dense_layer
model_pointer = getattr(model, "dense")
model_pointer.weight.data = torch.from_numpy(pretrained_weight)
```
While doing so, you must verify that each randomly initialized weight of
your PyTorch model and its corresponding pretrained checkpoint weight
exactly match in both **shape and name**. To do so, it is **necessary**
to add assert statements for the shape and print out the names of the
checkpoints weights. *E.g.*, you should add statements like:
```python
assert (
model_pointer.weight.shape == pretrained_weight.shape
), f"Pointer shape of random weight {model_pointer.shape} and array shape of checkpoint weight {pretrained_weight.shape} mismatched"
```
Besides, you should also print out the names of both weights to make
sure they match, *e.g.*,
```python
logger.info(f"Initialize PyTorch weight {layer_name} from {pretrained_weight.name}")
```
If either the shape or the name doesn't match, you probably assigned
the wrong checkpoint weight to a randomly initialized layer of the 🤗
Transformers implementation.
An incorrect shape is most likely due to an incorrect setting of the
config parameters in `BigBirdConfig()` that do not exactly match
those that were used for the checkpoint you want to convert. However, it
could also be that PyTorch's implementation of a layer requires the
weight to be transposed beforehand.
Finally, you should also check that **all** required weights are
initialized and print out all checkpoint weights that were not used for
initialization to make sure the model is correctly converted. It is
completely normal, that the conversion trials fail with either a wrong
shape statement or wrong name assignment. This is most likely because
either you used incorrect parameters in `BigBirdConfig()`, have a
wrong architecture in the 🤗 Transformers implementation, you have a bug
in the `init()` functions of one of the components of the 🤗 Transformers
implementation or you need to transpose one of the checkpoint weights.
This step should be iterated with the previous step until all weights of
the checkpoint are correctly loaded in the Transformers model. Having
correctly loaded the checkpoint into the 🤗 Transformers implementation,
you can then save the model under a folder of your choice
`/path/to/converted/checkpoint/folder` that should then contain both a
`pytorch_model.bin` file and a `config.json` file:
```python
model.save_pretrained("/path/to/converted/checkpoint/folder")
```
**7. Implement the forward pass**
Having managed to correctly load the pretrained weights into the 🤗
Transformers implementation, you should now make sure that the forward
pass is correctly implemented. In [Get familiar with the original
repository](#run-a-pretrained-checkpoint-using-the-original-repository),
you have already created a script that runs a forward pass of the model
using the original repository. Now you should write an analogous script
using the 🤗 Transformers implementation instead of the original one. It
should look as follows:
[Here the model name might have to be adapted, *e.g.*, maybe BigBirdForConditionalGeneration instead of BigBirdModel]
```python
model = BigBirdModel.from_pretrained("/path/to/converted/checkpoint/folder")
input_ids = [0, 4, 4, 3, 2, 4, 1, 7, 19]
output = model(input_ids).last_hidden_states
```
It is very likely that the 🤗 Transformers implementation and the
original model implementation don't give the exact same output the very
first time or that the forward pass throws an error. Don't be
disappointed - it's expected! First, you should make sure that the
forward pass doesn't throw any errors. It often happens that the wrong
dimensions are used leading to a `"Dimensionality mismatch"`
error or that the wrong data type object is used, *e.g.*, `torch.long`
instead of `torch.float32`. Don't hesitate to ask Patrick
for help, if you don't manage to solve certain errors.
The final part to make sure the 🤗 Transformers implementation works
correctly is to ensure that the outputs are equivalent to a precision of
`1e-3`. First, you should ensure that the output shapes are identical,
*i.e.* `outputs.shape` should yield the same value for the script of the
🤗 Transformers implementation and the original implementation. Next, you
should make sure that the output values are identical as well. This one
of the most difficult parts of adding a new model. Common mistakes why
the outputs are not identical are:
- Some layers were not added, *i.e.* an activation layer
was not added, or the residual connection was forgotten
- The word embedding matrix was not tied
- The wrong positional embeddings are used because the original
implementation uses on offset
- Dropout is applied during the forward pass. To fix this make sure
`model.training is False` and that no dropout layer is
falsely activated during the forward pass, *i.e.* pass
`self.training` to [PyTorch's functional
dropout](https://pytorch.org/docs/stable/nn.functional.html?highlight=dropout#torch.nn.functional.dropout)
The best way to fix the problem is usually to look at the forward pass
of the original implementation and the 🤗 Transformers implementation
side-by-side and check if there are any differences. Ideally, you should
debug/print out intermediate outputs of both implementations of the
forward pass to find the exact position in the network where the 🤗
Transformers implementation shows a different output than the original
implementation. First, make sure that the hard-coded `input_ids` in both
scripts are identical. Next, verify that the outputs of the first
transformation of the `input_ids` (usually the word embeddings) are
identical. And then work your way up to the very last layer of the
network. At some point, you will notice a difference between the two
implementations, which should point you to the bug in the 🤗 Transformers
implementation. From our experience, a simple and efficient way is to
add many print statements in both the original implementation and 🤗
Transformers implementation, at the same positions in the network
respectively, and to successively remove print statements showing the
same values for intermediate presentions.
When you're confident that both implementations yield the same output,
verifying the outputs with
`torch.allclose(original_output, output, atol=1e-3)`, you're done with
the most difficult part! Congratulations - the work left to be done
should be a cakewalk 😊.
**8. Adding all necessary model tests**
At this point, you have successfully added a new model. However, it is
very much possible that the model does not yet fully comply with the
required design. To make sure, the implementation is fully compatible
with 🤗 Transformers, all common tests should pass. The Cookiecutter
should have automatically added a test file for your model, probably
under the same `tests/test_modeling_big_bird.py`. Run this test
file to verify that all common tests pass:
```python
pytest tests/test_modeling_big_bird.py
```
Having fixed all common tests, it is now crucial to ensure that all the
nice work you have done is well tested, so that
- a) The community can easily understand your work by looking at
specific tests of *BigBird*
- b) Future changes to your model will not break any important
feature of the model.
At first, integration tests should be added. Those integration tests
essentially do the same as the debugging scripts you used earlier to
implement the model to 🤗 Transformers. A template of those model tests
is already added by the Cookiecutter, called
`BigBirdModelIntegrationTests` and only has to be filled out by
you. To ensure that those tests are passing, run
```python
RUN_SLOW=1 pytest -sv tests/test_modeling_big_bird.py::BigBirdModelIntegrationTests
```
**Note**: In case you are using Windows, you should replace `RUN_SLOW=1` with
`SET RUN_SLOW=1`
Second, all features that are special to *BigBird* should be
tested additionally in a separate test under
`BigBirdModelTester`/`BigBirdModelTest`. This part is often
forgotten but is extremely useful in two ways:
- It helps to transfer the knowledge you have acquired during the
model addition to the community by showing how the special features
of *BigBird* should work.
- Future contributors can quickly test changes to the model by running
those special tests.
BigBird has quite a complex attention layer, so it is very important
to add more tests verifying the all parts of BigBird's self-attention layer
works as expected. This means that there should be at least 3 additional tests:
- 1. Verify that the sparse attention works correctly
- 2. Verify that the global attention works correctly
- 3. Verify that the random attention works correctly
**9. Implement the tokenizer**
Next, we should add the tokenizer of *BigBird*. Usually, the
tokenizer is equivalent or very similar to an already existing tokenizer
of 🤗 Transformers.
In the case of BigBird you should be able to just rely on an already existing tokenizer.
If not mistaken, BigBird uses the same tokenizer that was used for `BertGenerationTokenizer`,
which is based on `sentencepiece`. So you should be able to just set the config parameter
`tokenizer_class` to `BertGenerationTokenizer` without having to implement any new tokenizer.
It is very important to find/extract the original tokenizer file and to
manage to load this file into the 🤗 Transformers' implementation of the
tokenizer.
For BigBird, the tokenizer (sentencepiece) files can be found [here](https://github.com/google-research/bigbird/blob/master/bigbird/vocab/gpt2.model), which you should be able to load
as easily as:
```python
from transformers import BertGenerationTokenizer
tokenizer = BertGenerationTokenizer("/path/to/gpt2.model/file")
```
To ensure that the tokenizer works correctly, it is recommended to first
create a script in the original repository that inputs a string and
returns the `input_ids`. It could look similar to this (in pseudo-code):
```bash
input_str = "This is a long example input string containing special characters .$?-, numbers 2872 234 12 and words."
model = BigBirdModel.load_pretrained_checkpoint("/path/to/checkpoint/")
input_ids = model.tokenize(input_str)
```
You might have to take a deeper look again into the original repository
to find the correct tokenizer function or you might even have to do
changes to your clone of the original repository to only output the
`input_ids`. Having written a functional tokenization script that uses
the original repository, an analogous script for 🤗 Transformers should
be created. It should look similar to this:
```python
from transformers import BertGenerationTokenizer
input_str = "This is a long example input string containing special characters .$?-, numbers 2872 234 12 and words."
tokenizer = BertGenerationTokenizer.from_pretrained("/path/big/bird/folder")
input_ids = tokenizer(input_str).input_ids
```
When both `input_ids` yield the same values, as a final step a tokenizer
test file should also be added.
Since BigBird is most likely fully based on `BertGenerationTokenizer`,
you should only add a couple of "slow" integration tests. However, in this
case you do **not** need to add any `BigBirdTokenizationTest`.
**10. Run End-to-end integration tests**
Having added the tokenizer, you should also add a couple of end-to-end
integration tests using both the model and the tokenizer to
`tests/test_modeling_big_bird.py` in 🤗 Transformers. Such a test
should show on a meaningful text-to-text sample that the 🤗 Transformers
implementation works as expected. A meaningful text-to-text sample can
include, *e.g.*, a source-to-target-translation pair, an
article-to-summary pair, a question-to-answer pair, etc... If none of
the ported checkpoints has been fine-tuned on a downstream task it is
enough to simply rely on the model tests. In a final step to ensure that
the model is fully functional, it is advised that you also run all tests
on GPU. It can happen that you forgot to add some `.to(self.device)`
statements to internal tensors of the model, which in such a test would
show in an error. In case you have no access to a GPU, the Hugging Face
team can take care of running those tests for you.
**11. Add Docstring**
Now, all the necessary functionality for *BigBird* is added -
you're almost done! The only thing left to add is a nice docstring and
a doc page. The Cookiecutter should have added a template file called
`docs/source/model_doc/big_bird.rst` that you should fill out.
Users of your model will usually first look at this page before using
your model. Hence, the documentation must be understandable and concise.
It is very useful for the community to add some *Tips* to show how the
model should be used. Don't hesitate to ping Patrick
regarding the docstrings.
Next, make sure that the docstring added to
`src/transformers/models/big_bird/modeling_big_bird.py` is
correct and included all necessary inputs and outputs. It is always to
good to remind oneself that documentation should be treated at least as
carefully as the code in 🤗 Transformers since the documentation is
usually the first contact point of the community with the model.
**Code refactor**
Great, now you have added all the necessary code for *BigBird*.
At this point, you should correct some potential incorrect code style by
running:
```bash
make style
```
and verify that your coding style passes the quality check:
```bash
make quality
```
There are a couple of other very strict design tests in 🤗 Transformers
that might still be failing, which shows up in the tests of your pull
request. This is often because of some missing information in the
docstring or some incorrect naming. Patrick will surely
help you if you're stuck here.
Lastly, it is always a good idea to refactor one's code after having
ensured that the code works correctly. With all tests passing, now it's
a good time to go over the added code again and do some refactoring.
You have now finished the coding part, congratulation! 🎉 You are
Awesome! 😎
**12. Upload the models to the model hub**
In this final part, you should convert and upload all checkpoints to the
model hub and add a model card for each uploaded model checkpoint. You
should work alongside Patrick here to decide on a fitting
name for each checkpoint and to get the required access rights to be
able to upload the model under the author's organization of
*BigBird*.
It is worth spending some time to create fitting model cards for each
checkpoint. The model cards should highlight the specific
characteristics of this particular checkpoint, *e.g.*, On which dataset
was the checkpoint pretrained/fine-tuned on? On what down-stream task
should the model be used? And also include some code on how to correctly
use the model.
**13. (Optional) Add notebook**
It is very helpful to add a notebook that showcases in-detail how
*BigBird* can be used for inference and/or fine-tuned on a
downstream task. This is not mandatory to merge your PR, but very useful
for the community.
**14. Submit your finished PR**
You're done programming now and can move to the last step, which is
getting your PR merged into main. Usually, Patrick
should have helped you already at this point, but it is worth taking
some time to give your finished PR a nice description and eventually add
comments to your code, if you want to point out certain design choices
to your reviewer.
### Share your work!!
Now, it's time to get some credit from the community for your work!
Having completed a model addition is a major contribution to
Transformers and the whole NLP community. Your code and the ported
pre-trained models will certainly be used by hundreds and possibly even
thousands of developers and researchers. You should be proud of your
work and share your achievement with the community.
**You have made another model that is super easy to access for everyone
in the community! 🤯**
| 0
|
mavonic_private_repos/transformers/templates/adding_a_new_model
|
mavonic_private_repos/transformers/templates/adding_a_new_model/open_model_proposals/README.md
|
Currently the following model proposals are available:
- <s>[BigBird (Google)](./ADD_BIG_BIRD.md)</s>
| 0
|
mavonic_private_repos/transformers
|
mavonic_private_repos/transformers/.circleci/parse_test_outputs.py
|
import re
import argparse
def parse_pytest_output(file_path):
skipped_tests = {}
skipped_count = 0
with open(file_path, 'r') as file:
for line in file:
match = re.match(r'^SKIPPED \[(\d+)\] (tests/.*): (.*)$', line)
if match:
skipped_count += 1
test_file, test_line, reason = match.groups()
skipped_tests[reason] = skipped_tests.get(reason, []) + [(test_file, test_line)]
for k,v in sorted(skipped_tests.items(), key=lambda x:len(x[1])):
print(f"{len(v):4} skipped because: {k}")
print("Number of skipped tests:", skipped_count)
def parse_pytest_failure_output(file_path):
failed_tests = {}
failed_count = 0
with open(file_path, 'r') as file:
for line in file:
match = re.match(r'^FAILED (tests/.*) - (.*): (.*)$', line)
if match:
failed_count += 1
_, error, reason = match.groups()
failed_tests[reason] = failed_tests.get(reason, []) + [error]
for k,v in sorted(failed_tests.items(), key=lambda x:len(x[1])):
print(f"{len(v):4} failed because `{v[0]}` -> {k}")
print("Number of failed tests:", failed_count)
if failed_count>0:
exit(1)
def parse_pytest_errors_output(file_path):
print(file_path)
error_tests = {}
error_count = 0
with open(file_path, 'r') as file:
for line in file:
match = re.match(r'^ERROR (tests/.*) - (.*): (.*)$', line)
if match:
error_count += 1
_, test_error, reason = match.groups()
error_tests[reason] = error_tests.get(reason, []) + [test_error]
for k,v in sorted(error_tests.items(), key=lambda x:len(x[1])):
print(f"{len(v):4} errored out because of `{v[0]}` -> {k}")
print("Number of errors:", error_count)
if error_count>0:
exit(1)
def main():
parser = argparse.ArgumentParser()
parser.add_argument("--file", help="file to parse")
parser.add_argument("--skip", action="store_true", help="show skipped reasons")
parser.add_argument("--fail", action="store_true", help="show failed tests")
parser.add_argument("--errors", action="store_true", help="show failed tests")
args = parser.parse_args()
if args.skip:
parse_pytest_output(args.file)
if args.fail:
parse_pytest_failure_output(args.file)
if args.errors:
parse_pytest_errors_output(args.file)
if __name__ == "__main__":
main()
| 0
|
mavonic_private_repos/transformers
|
mavonic_private_repos/transformers/.circleci/config.yml
|
version: 2.1
setup: true
orbs:
continuation: circleci/continuation@0.1.0
parameters:
nightly:
type: boolean
default: false
jobs:
# Ensure running with CircleCI/huggingface
check_circleci_user:
docker:
- image: python:3.10-slim
parallelism: 1
steps:
- run: echo $CIRCLE_PROJECT_USERNAME
- run: |
if [ "$CIRCLE_PROJECT_USERNAME" = "huggingface" ]; then
exit 0
else
echo "The CI is running under $CIRCLE_PROJECT_USERNAME personal account. Please follow https://support.circleci.com/hc/en-us/articles/360008097173-Troubleshooting-why-pull-requests-are-not-triggering-jobs-on-my-organization- to fix it."; exit -1
fi
# Fetch the tests to run
fetch_tests:
working_directory: ~/transformers
docker:
- image: huggingface/transformers-quality
parallelism: 1
steps:
- checkout
- run: uv pip install -U -e .
- run: mkdir -p test_preparation
- run: python utils/tests_fetcher.py | tee tests_fetched_summary.txt
- store_artifacts:
path: ~/transformers/tests_fetched_summary.txt
- run: |
if [ -f test_list.txt ]; then
cp test_list.txt test_preparation/test_list.txt
else
touch test_preparation/test_list.txt
fi
- run: |
if [ -f examples_test_list.txt ]; then
mv examples_test_list.txt test_preparation/examples_test_list.txt
else
touch test_preparation/examples_test_list.txt
fi
- run: |
if [ -f filtered_test_list_cross_tests.txt ]; then
mv filtered_test_list_cross_tests.txt test_preparation/filtered_test_list_cross_tests.txt
else
touch test_preparation/filtered_test_list_cross_tests.txt
fi
- run: |
if [ -f doctest_list.txt ]; then
cp doctest_list.txt test_preparation/doctest_list.txt
else
touch test_preparation/doctest_list.txt
fi
- run: |
if [ -f test_repo_utils.txt ]; then
mv test_repo_utils.txt test_preparation/test_repo_utils.txt
else
touch test_preparation/test_repo_utils.txt
fi
- run: python utils/tests_fetcher.py --filter_tests
- run: |
if [ -f test_list.txt ]; then
mv test_list.txt test_preparation/filtered_test_list.txt
else
touch test_preparation/filtered_test_list.txt
fi
- store_artifacts:
path: test_preparation/test_list.txt
- store_artifacts:
path: test_preparation/doctest_list.txt
- store_artifacts:
path: ~/transformers/test_preparation/filtered_test_list.txt
- store_artifacts:
path: test_preparation/examples_test_list.txt
- run: python .circleci/create_circleci_config.py --fetcher_folder test_preparation
- run: |
if [ ! -s test_preparation/generated_config.yml ]; then
echo "No tests to run, exiting early!"
circleci-agent step halt
fi
- store_artifacts:
path: test_preparation/generated_config.yml
- store_artifacts:
path: test_preparation/filtered_test_list_cross_tests.txt
- continuation/continue:
configuration_path: test_preparation/generated_config.yml
# To run all tests for the nightly build
fetch_all_tests:
working_directory: ~/transformers
docker:
- image: huggingface/transformers-consistency
parallelism: 1
steps:
- checkout
- run: uv pip install -e .
- run: |
mkdir test_preparation
echo -n "tests" > test_preparation/test_list.txt
echo -n "all" > test_preparation/examples_test_list.txt
echo -n "tests/repo_utils" > test_preparation/test_repo_utils.txt
- run: |
echo -n "tests" > test_list.txt
python utils/tests_fetcher.py --filter_tests
mv test_list.txt test_preparation/filtered_test_list.txt
- run: python .circleci/create_circleci_config.py --fetcher_folder test_preparation
- run: cp test_preparation/generated_config.yml test_preparation/generated_config.txt
- store_artifacts:
path: test_preparation/generated_config.txt
- continuation/continue:
configuration_path: test_preparation/generated_config.yml
check_code_quality:
working_directory: ~/transformers
docker:
- image: huggingface/transformers-quality
resource_class: large
environment:
TRANSFORMERS_IS_CI: yes
PYTEST_TIMEOUT: 120
parallelism: 1
steps:
- checkout
- run: uv pip install -e .
- run:
name: Show installed libraries and their versions
command: pip freeze | tee installed.txt
- store_artifacts:
path: ~/transformers/installed.txt
- run: python -c "from transformers import *" || (echo '🚨 import failed, this means you introduced unprotected imports! 🚨'; exit 1)
- run: ruff check examples tests src utils
- run: ruff format tests src utils --check
- run: python utils/custom_init_isort.py --check_only
- run: python utils/sort_auto_mappings.py --check_only
- run: python utils/check_doc_toc.py
check_repository_consistency:
working_directory: ~/transformers
docker:
- image: huggingface/transformers-consistency
resource_class: large
environment:
TRANSFORMERS_IS_CI: yes
PYTEST_TIMEOUT: 120
parallelism: 1
steps:
- checkout
- run: uv pip install -e .
- run:
name: Show installed libraries and their versions
command: pip freeze | tee installed.txt
- store_artifacts:
path: ~/transformers/installed.txt
- run: python utils/check_copies.py
- run: python utils/check_table.py
- run: python utils/check_dummies.py
- run: python utils/check_repo.py
- run: python utils/check_inits.py
- run: python utils/check_config_docstrings.py
- run: python utils/check_config_attributes.py
- run: python utils/check_doctest_list.py
- run: make deps_table_check_updated
- run: python utils/update_metadata.py --check-only
- run: python utils/check_docstrings.py
- run: python utils/check_support_list.py
workflows:
version: 2
setup_and_quality:
when:
not: <<pipeline.parameters.nightly>>
jobs:
- check_circleci_user
- check_code_quality
- check_repository_consistency
- fetch_tests
nightly:
when: <<pipeline.parameters.nightly>>
jobs:
- check_circleci_user
- check_code_quality
- check_repository_consistency
- fetch_all_tests
| 0
|
mavonic_private_repos/transformers
|
mavonic_private_repos/transformers/.circleci/TROUBLESHOOT.md
|
# Troubleshooting
This is a document explaining how to deal with various issues on Circle-CI. The entries may include actual solutions or pointers to Issues that cover those.
## Circle CI
* pytest worker runs out of resident RAM and gets killed by `cgroups`: https://github.com/huggingface/transformers/issues/11408
| 0
|
mavonic_private_repos/transformers
|
mavonic_private_repos/transformers/.circleci/create_circleci_config.py
|
# coding=utf-8
# Copyright 2022 The HuggingFace Inc. team.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import argparse
import copy
import os
import random
from dataclasses import dataclass
from typing import Any, Dict, List, Optional
import glob
import yaml
COMMON_ENV_VARIABLES = {
"OMP_NUM_THREADS": 1,
"TRANSFORMERS_IS_CI": True,
"PYTEST_TIMEOUT": 120,
"RUN_PIPELINE_TESTS": False,
"RUN_PT_TF_CROSS_TESTS": False,
"RUN_PT_FLAX_CROSS_TESTS": False,
}
# Disable the use of {"s": None} as the output is way too long, causing the navigation on CircleCI impractical
COMMON_PYTEST_OPTIONS = {"max-worker-restart": 0, "dist": "loadfile", "v": None}
DEFAULT_DOCKER_IMAGE = [{"image": "cimg/python:3.8.12"}]
class EmptyJob:
job_name = "empty"
def to_dict(self):
return {
"docker": copy.deepcopy(DEFAULT_DOCKER_IMAGE),
"steps":["checkout"],
}
@dataclass
class CircleCIJob:
name: str
additional_env: Dict[str, Any] = None
cache_name: str = None
cache_version: str = "0.8.2"
docker_image: List[Dict[str, str]] = None
install_steps: List[str] = None
marker: Optional[str] = None
parallelism: Optional[int] = 1
pytest_num_workers: int = 12
pytest_options: Dict[str, Any] = None
resource_class: Optional[str] = "2xlarge"
tests_to_run: Optional[List[str]] = None
# This should be only used for doctest job!
command_timeout: Optional[int] = None
def __post_init__(self):
# Deal with defaults for mutable attributes.
if self.additional_env is None:
self.additional_env = {}
if self.cache_name is None:
self.cache_name = self.name
if self.docker_image is None:
# Let's avoid changing the default list and make a copy.
self.docker_image = copy.deepcopy(DEFAULT_DOCKER_IMAGE)
if self.install_steps is None:
self.install_steps = []
if self.pytest_options is None:
self.pytest_options = {}
if isinstance(self.tests_to_run, str):
self.tests_to_run = [self.tests_to_run]
if self.parallelism is None:
self.parallelism = 1
def to_dict(self):
env = COMMON_ENV_VARIABLES.copy()
env.update(self.additional_env)
cache_branch_prefix = os.environ.get("CIRCLE_BRANCH", "pull")
if cache_branch_prefix != "main":
cache_branch_prefix = "pull"
job = {
"docker": self.docker_image,
"environment": env,
}
if self.resource_class is not None:
job["resource_class"] = self.resource_class
if self.parallelism is not None:
job["parallelism"] = self.parallelism
steps = [
"checkout",
{"attach_workspace": {"at": "test_preparation"}},
]
steps.extend([{"run": l} for l in self.install_steps])
steps.append({"run": {"name": "Show installed libraries and their size", "command": """du -h -d 1 "$(pip -V | cut -d ' ' -f 4 | sed 's/pip//g')" | grep -vE "dist-info|_distutils_hack|__pycache__" | sort -h | tee installed.txt || true"""}})
steps.append({"run": {"name": "Show installed libraries and their versions", "command": """pip list --format=freeze | tee installed.txt || true"""}})
steps.append({"run":{"name":"Show biggest libraries","command":"""dpkg-query --show --showformat='${Installed-Size}\t${Package}\n' | sort -rh | head -25 | sort -h | awk '{ package=$2; sub(".*/", "", package); printf("%.5f GB %s\n", $1/1024/1024, package)}' || true"""}})
steps.append({"store_artifacts": {"path": "installed.txt"}})
all_options = {**COMMON_PYTEST_OPTIONS, **self.pytest_options}
pytest_flags = [f"--{key}={value}" if (value is not None or key in ["doctest-modules"]) else f"-{key}" for key, value in all_options.items()]
pytest_flags.append(
f"--make-reports={self.name}" if "examples" in self.name else f"--make-reports=tests_{self.name}"
)
steps.append({"run": {"name": "Create `test-results` directory", "command": "mkdir test-results"}})
test_command = ""
if self.command_timeout:
test_command = f"timeout {self.command_timeout} "
# junit familiy xunit1 is necessary to support splitting on test name or class name with circleci split
test_command += f"python3 -m pytest -rsfE -p no:warnings -o junit_family=xunit1 --tb=short --junitxml=test-results/junit.xml -n {self.pytest_num_workers} " + " ".join(pytest_flags)
if self.parallelism == 1:
if self.tests_to_run is None:
test_command += " << pipeline.parameters.tests_to_run >>"
else:
test_command += " " + " ".join(self.tests_to_run)
else:
# We need explicit list instead of `pipeline.parameters.tests_to_run` (only available at job runtime)
tests = self.tests_to_run
if tests is None:
folder = os.environ["test_preparation_dir"]
test_file = os.path.join(folder, "filtered_test_list.txt")
if os.path.exists(test_file):
with open(test_file) as f:
tests = f.read().split(" ")
# expand the test list
if tests == ["tests"]:
tests = [os.path.join("tests", x) for x in os.listdir("tests")]
expanded_tests = []
for test in tests:
if test.endswith(".py"):
expanded_tests.append(test)
elif test == "tests/models":
expanded_tests.extend(glob.glob("tests/models/**/test*.py", recursive=True))
elif test == "tests/pipelines":
expanded_tests.extend([os.path.join(test, x) for x in os.listdir(test)])
else:
expanded_tests.append(test)
tests = " ".join(expanded_tests)
# Each executor to run ~10 tests
n_executors = max(len(expanded_tests) // 10, 1)
# Avoid empty test list on some executor(s) or launching too many executors
if n_executors > self.parallelism:
n_executors = self.parallelism
job["parallelism"] = n_executors
# Need to be newline separated for the command `circleci tests split` below
command = f'echo {tests} | tr " " "\\n" >> tests.txt'
steps.append({"run": {"name": "Get tests", "command": command}})
command = 'TESTS=$(circleci tests split tests.txt) && echo $TESTS > splitted_tests.txt'
steps.append({"run": {"name": "Split tests", "command": command}})
steps.append({"store_artifacts": {"path": "tests.txt"}})
steps.append({"store_artifacts": {"path": "splitted_tests.txt"}})
test_command = ""
if self.command_timeout:
test_command = f"timeout {self.command_timeout} "
test_command += f"python3 -m pytest -rsfE -p no:warnings --tb=short -o junit_family=xunit1 --junitxml=test-results/junit.xml -n {self.pytest_num_workers} " + " ".join(pytest_flags)
test_command += " $(cat splitted_tests.txt)"
if self.marker is not None:
test_command += f" -m {self.marker}"
if self.name == "pr_documentation_tests":
# can't use ` | tee tee tests_output.txt` as usual
test_command += " > tests_output.txt"
# Save the return code, so we can check if it is timeout in the next step.
test_command += '; touch "$?".txt'
# Never fail the test step for the doctest job. We will check the results in the next step, and fail that
# step instead if the actual test failures are found. This is to avoid the timeout being reported as test
# failure.
test_command = f"({test_command}) || true"
else:
test_command = f"({test_command} | tee tests_output.txt)"
steps.append({"run": {"name": "Run tests", "command": test_command}})
steps.append({"run": {"name": "Skipped tests", "when": "always", "command": f"python3 .circleci/parse_test_outputs.py --file tests_output.txt --skip"}})
steps.append({"run": {"name": "Failed tests", "when": "always", "command": f"python3 .circleci/parse_test_outputs.py --file tests_output.txt --fail"}})
steps.append({"run": {"name": "Errors", "when": "always", "command": f"python3 .circleci/parse_test_outputs.py --file tests_output.txt --errors"}})
steps.append({"store_test_results": {"path": "test-results"}})
steps.append({"store_artifacts": {"path": "tests_output.txt"}})
steps.append({"store_artifacts": {"path": "test-results/junit.xml"}})
steps.append({"store_artifacts": {"path": "reports"}})
job["steps"] = steps
return job
@property
def job_name(self):
return self.name if "examples" in self.name else f"tests_{self.name}"
# JOBS
torch_and_tf_job = CircleCIJob(
"torch_and_tf",
docker_image=[{"image":"huggingface/transformers-torch-tf-light"}],
install_steps=["uv venv && uv pip install ."],
additional_env={"RUN_PT_TF_CROSS_TESTS": True},
marker="is_pt_tf_cross_test",
pytest_options={"rA": None, "durations": 0},
)
torch_and_flax_job = CircleCIJob(
"torch_and_flax",
additional_env={"RUN_PT_FLAX_CROSS_TESTS": True},
docker_image=[{"image":"huggingface/transformers-torch-jax-light"}],
install_steps=["uv venv && uv pip install ."],
marker="is_pt_flax_cross_test",
pytest_options={"rA": None, "durations": 0},
)
torch_job = CircleCIJob(
"torch",
docker_image=[{"image": "huggingface/transformers-torch-light"}],
install_steps=["uv venv && uv pip install ."],
parallelism=6,
pytest_num_workers=16
)
tf_job = CircleCIJob(
"tf",
docker_image=[{"image":"huggingface/transformers-tf-light"}],
install_steps=["uv venv", "uv pip install -e."],
parallelism=6,
pytest_num_workers=16,
)
flax_job = CircleCIJob(
"flax",
docker_image=[{"image":"huggingface/transformers-jax-light"}],
install_steps=["uv venv && uv pip install ."],
parallelism=6,
pytest_num_workers=16
)
pipelines_torch_job = CircleCIJob(
"pipelines_torch",
additional_env={"RUN_PIPELINE_TESTS": True},
docker_image=[{"image":"huggingface/transformers-torch-light"}],
install_steps=["uv venv && uv pip install ."],
marker="is_pipeline_test",
)
pipelines_tf_job = CircleCIJob(
"pipelines_tf",
additional_env={"RUN_PIPELINE_TESTS": True},
docker_image=[{"image":"huggingface/transformers-tf-light"}],
install_steps=["uv venv && uv pip install ."],
marker="is_pipeline_test",
)
custom_tokenizers_job = CircleCIJob(
"custom_tokenizers",
additional_env={"RUN_CUSTOM_TOKENIZERS": True},
docker_image=[{"image": "huggingface/transformers-custom-tokenizers"}],
install_steps=["uv venv","uv pip install -e ."],
parallelism=None,
resource_class=None,
tests_to_run=[
"./tests/models/bert_japanese/test_tokenization_bert_japanese.py",
"./tests/models/openai/test_tokenization_openai.py",
"./tests/models/clip/test_tokenization_clip.py",
],
)
examples_torch_job = CircleCIJob(
"examples_torch",
additional_env={"OMP_NUM_THREADS": 8},
cache_name="torch_examples",
docker_image=[{"image":"huggingface/transformers-examples-torch"}],
install_steps=["uv venv && uv pip install ."],
pytest_num_workers=1,
)
examples_tensorflow_job = CircleCIJob(
"examples_tensorflow",
cache_name="tensorflow_examples",
docker_image=[{"image":"huggingface/transformers-examples-tf"}],
install_steps=["uv venv && uv pip install ."],
parallelism=8
)
hub_job = CircleCIJob(
"hub",
additional_env={"HUGGINGFACE_CO_STAGING": True},
docker_image=[{"image":"huggingface/transformers-torch-light"}],
install_steps=[
"uv venv && uv pip install .",
'git config --global user.email "ci@dummy.com"',
'git config --global user.name "ci"',
],
marker="is_staging_test",
pytest_num_workers=1,
)
onnx_job = CircleCIJob(
"onnx",
docker_image=[{"image":"huggingface/transformers-torch-tf-light"}],
install_steps=[
"uv venv && uv pip install .",
"uv pip install --upgrade eager pip",
"uv pip install .[torch,tf,testing,sentencepiece,onnxruntime,vision,rjieba]",
],
pytest_options={"k onnx": None},
pytest_num_workers=1,
)
exotic_models_job = CircleCIJob(
"exotic_models",
install_steps=["uv venv && uv pip install ."],
docker_image=[{"image":"huggingface/transformers-exotic-models"}],
tests_to_run=[
"tests/models/*layoutlmv*",
"tests/models/*nat",
"tests/models/deta",
"tests/models/udop",
"tests/models/nougat",
],
pytest_num_workers=12,
parallelism=4,
pytest_options={"durations": 100},
)
repo_utils_job = CircleCIJob(
"repo_utils",
docker_image=[{"image":"huggingface/transformers-consistency"}],
install_steps=["uv venv && uv pip install ."],
parallelism=None,
pytest_num_workers=1,
resource_class="large",
tests_to_run="tests/repo_utils",
)
# We also include a `dummy.py` file in the files to be doc-tested to prevent edge case failure. Otherwise, the pytest
# hangs forever during test collection while showing `collecting 0 items / 21 errors`. (To see this, we have to remove
# the bash output redirection.)
py_command = 'from utils.tests_fetcher import get_doctest_files; to_test = get_doctest_files() + ["dummy.py"]; to_test = " ".join(to_test); print(to_test)'
py_command = f"$(python3 -c '{py_command}')"
command = f'echo "{py_command}" > pr_documentation_tests_temp.txt'
doc_test_job = CircleCIJob(
"pr_documentation_tests",
docker_image=[{"image":"huggingface/transformers-consistency"}],
additional_env={"TRANSFORMERS_VERBOSITY": "error", "DATASETS_VERBOSITY": "error", "SKIP_CUDA_DOCTEST": "1"},
install_steps=[
# Add an empty file to keep the test step running correctly even no file is selected to be tested.
"touch dummy.py",
{
"name": "Get files to test",
"command": command,
},
{
"name": "Show information in `Get files to test`",
"command":
"cat pr_documentation_tests_temp.txt"
},
{
"name": "Get the last line in `pr_documentation_tests.txt`",
"command":
"tail -n1 pr_documentation_tests_temp.txt | tee pr_documentation_tests.txt"
},
],
tests_to_run="$(cat pr_documentation_tests.txt)", # noqa
pytest_options={"-doctest-modules": None, "doctest-glob": "*.md", "dist": "loadfile", "rvsA": None},
command_timeout=1200, # test cannot run longer than 1200 seconds
pytest_num_workers=1,
)
REGULAR_TESTS = [
torch_and_tf_job,
torch_and_flax_job,
torch_job,
tf_job,
flax_job,
custom_tokenizers_job,
hub_job,
onnx_job,
exotic_models_job,
]
EXAMPLES_TESTS = [
examples_torch_job,
examples_tensorflow_job,
]
PIPELINE_TESTS = [
pipelines_torch_job,
pipelines_tf_job,
]
REPO_UTIL_TESTS = [repo_utils_job]
DOC_TESTS = [doc_test_job]
def create_circleci_config(folder=None):
if folder is None:
folder = os.getcwd()
# Used in CircleCIJob.to_dict() to expand the test list (for using parallelism)
os.environ["test_preparation_dir"] = folder
jobs = []
all_test_file = os.path.join(folder, "test_list.txt")
if os.path.exists(all_test_file):
with open(all_test_file) as f:
all_test_list = f.read()
else:
all_test_list = []
if len(all_test_list) > 0:
jobs.extend(PIPELINE_TESTS)
test_file = os.path.join(folder, "filtered_test_list.txt")
if os.path.exists(test_file):
with open(test_file) as f:
test_list = f.read()
else:
test_list = []
if len(test_list) > 0:
jobs.extend(REGULAR_TESTS)
extended_tests_to_run = set(test_list.split())
# Extend the test files for cross test jobs
for job in jobs:
if job.job_name in ["tests_torch_and_tf", "tests_torch_and_flax"]:
for test_path in copy.copy(extended_tests_to_run):
dir_path, fn = os.path.split(test_path)
if fn.startswith("test_modeling_tf_"):
fn = fn.replace("test_modeling_tf_", "test_modeling_")
elif fn.startswith("test_modeling_flax_"):
fn = fn.replace("test_modeling_flax_", "test_modeling_")
else:
if job.job_name == "test_torch_and_tf":
fn = fn.replace("test_modeling_", "test_modeling_tf_")
elif job.job_name == "test_torch_and_flax":
fn = fn.replace("test_modeling_", "test_modeling_flax_")
new_test_file = str(os.path.join(dir_path, fn))
if os.path.isfile(new_test_file):
if new_test_file not in extended_tests_to_run:
extended_tests_to_run.add(new_test_file)
extended_tests_to_run = sorted(extended_tests_to_run)
for job in jobs:
if job.job_name in ["tests_torch_and_tf", "tests_torch_and_flax"]:
job.tests_to_run = extended_tests_to_run
fn = "filtered_test_list_cross_tests.txt"
f_path = os.path.join(folder, fn)
with open(f_path, "w") as fp:
fp.write(" ".join(extended_tests_to_run))
example_file = os.path.join(folder, "examples_test_list.txt")
if os.path.exists(example_file) and os.path.getsize(example_file) > 0:
with open(example_file, "r", encoding="utf-8") as f:
example_tests = f.read()
for job in EXAMPLES_TESTS:
framework = job.name.replace("examples_", "").replace("torch", "pytorch")
if example_tests == "all":
job.tests_to_run = [f"examples/{framework}"]
else:
job.tests_to_run = [f for f in example_tests.split(" ") if f.startswith(f"examples/{framework}")]
if len(job.tests_to_run) > 0:
jobs.append(job)
doctest_file = os.path.join(folder, "doctest_list.txt")
if os.path.exists(doctest_file):
with open(doctest_file) as f:
doctest_list = f.read()
else:
doctest_list = []
if len(doctest_list) > 0:
jobs.extend(DOC_TESTS)
repo_util_file = os.path.join(folder, "test_repo_utils.txt")
if os.path.exists(repo_util_file) and os.path.getsize(repo_util_file) > 0:
jobs.extend(REPO_UTIL_TESTS)
if len(jobs) == 0:
jobs = [EmptyJob()]
config = {"version": "2.1"}
config["parameters"] = {
# Only used to accept the parameters from the trigger
"nightly": {"type": "boolean", "default": False},
"tests_to_run": {"type": "string", "default": test_list},
}
config["jobs"] = {j.job_name: j.to_dict() for j in jobs}
config["workflows"] = {"version": 2, "run_tests": {"jobs": [j.job_name for j in jobs]}}
with open(os.path.join(folder, "generated_config.yml"), "w") as f:
f.write(yaml.dump(config, indent=2, width=1000000, sort_keys=False))
if __name__ == "__main__":
parser = argparse.ArgumentParser()
parser.add_argument(
"--fetcher_folder", type=str, default=None, help="Only test that all tests and modules are accounted for."
)
args = parser.parse_args()
create_circleci_config(args.fetcher_folder)
| 0
|
mavonic_private_repos/transformers
|
mavonic_private_repos/transformers/docs/TRANSLATING.md
|
### Translating the Transformers documentation into your language
As part of our mission to democratize machine learning, we'd love to make the Transformers library available in many more languages! Follow the steps below if you want to help translate the documentation into your language 🙏.
**🗞️ Open an issue**
To get started, navigate to the [Issues](https://github.com/huggingface/transformers/issues) page of this repo and check if anyone else has opened an issue for your language. If not, open a new issue by selecting the "Translation template" from the "New issue" button.
Once an issue exists, post a comment to indicate which chapters you'd like to work on, and we'll add your name to the list.
**🍴 Fork the repository**
First, you'll need to [fork the Transformers repo](https://docs.github.com/en/get-started/quickstart/fork-a-repo). You can do this by clicking on the **Fork** button on the top-right corner of this repo's page.
Once you've forked the repo, you'll want to get the files on your local machine for editing. You can do that by cloning the fork with Git as follows:
```bash
git clone https://github.com/YOUR-USERNAME/transformers.git
```
**📋 Copy-paste the English version with a new language code**
The documentation files are in one leading directory:
- [`docs/source`](https://github.com/huggingface/transformers/tree/main/docs/source): All the documentation materials are organized here by language.
You'll only need to copy the files in the [`docs/source/en`](https://github.com/huggingface/transformers/tree/main/docs/source/en) directory, so first navigate to your fork of the repo and run the following:
```bash
cd ~/path/to/transformers/docs
cp -r source/en source/LANG-ID
```
Here, `LANG-ID` should be one of the ISO 639-1 or ISO 639-2 language codes -- see [here](https://www.loc.gov/standards/iso639-2/php/code_list.php) for a handy table.
**✍️ Start translating**
The fun part comes - translating the text!
The first thing we recommend is translating the part of the `_toctree.yml` file that corresponds to your doc chapter. This file is used to render the table of contents on the website.
> 🙋 If the `_toctree.yml` file doesn't yet exist for your language, you can create one by copy-pasting from the English version and deleting the sections unrelated to your chapter. Just make sure it exists in the `docs/source/LANG-ID/` directory!
The fields you should add are `local` (with the name of the file containing the translation; e.g. `autoclass_tutorial`), and `title` (with the title of the doc in your language; e.g. `Load pretrained instances with an AutoClass`) -- as a reference, here is the `_toctree.yml` for [English](https://github.com/huggingface/transformers/blob/main/docs/source/en/_toctree.yml):
```yaml
- sections:
- local: pipeline_tutorial # Do not change this! Use the same name for your .md file
title: Pipelines for inference # Translate this!
...
title: Tutorials # Translate this!
```
Once you have translated the `_toctree.yml` file, you can start translating the [MDX](https://mdxjs.com/) files associated with your docs chapter.
> 🙋 If you'd like others to help you with the translation, you should [open an issue](https://github.com/huggingface/transformers/issues) and tag @stevhliu and @MKhalusova.
| 0
|
mavonic_private_repos/transformers
|
mavonic_private_repos/transformers/docs/README.md
|
<!---
Copyright 2020 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
-->
# Generating the documentation
To generate the documentation, you first have to build it. Several packages are necessary to build the doc,
you can install them with the following command, at the root of the code repository:
```bash
pip install -e ".[docs]"
```
Then you need to install our special tool that builds the documentation:
```bash
pip install git+https://github.com/huggingface/doc-builder
```
---
**NOTE**
You only need to generate the documentation to inspect it locally (if you're planning changes and want to
check how they look before committing for instance). You don't have to commit the built documentation.
---
## Building the documentation
Once you have setup the `doc-builder` and additional packages, you can generate the documentation by
typing the following command:
```bash
doc-builder build transformers docs/source/en/ --build_dir ~/tmp/test-build
```
You can adapt the `--build_dir` to set any temporary folder that you prefer. This command will create it and generate
the MDX files that will be rendered as the documentation on the main website. You can inspect them in your favorite
Markdown editor.
## Previewing the documentation
To preview the docs, first install the `watchdog` module with:
```bash
pip install watchdog
```
Then run the following command:
```bash
doc-builder preview {package_name} {path_to_docs}
```
For example:
```bash
doc-builder preview transformers docs/source/en/
```
The docs will be viewable at [http://localhost:3000](http://localhost:3000). You can also preview the docs once you have opened a PR. You will see a bot add a comment to a link where the documentation with your changes lives.
---
**NOTE**
The `preview` command only works with existing doc files. When you add a completely new file, you need to update `_toctree.yml` & restart `preview` command (`ctrl-c` to stop it & call `doc-builder preview ...` again).
---
## Adding a new element to the navigation bar
Accepted files are Markdown (.md).
Create a file with its extension and put it in the source directory. You can then link it to the toc-tree by putting
the filename without the extension in the [`_toctree.yml`](https://github.com/huggingface/transformers/blob/main/docs/source/en/_toctree.yml) file.
## Renaming section headers and moving sections
It helps to keep the old links working when renaming the section header and/or moving sections from one document to another. This is because the old links are likely to be used in Issues, Forums, and Social media and it'd make for a much more superior user experience if users reading those months later could still easily navigate to the originally intended information.
Therefore, we simply keep a little map of moved sections at the end of the document where the original section was. The key is to preserve the original anchor.
So if you renamed a section from: "Section A" to "Section B", then you can add at the end of the file:
```
Sections that were moved:
[ <a href="#section-b">Section A</a><a id="section-a"></a> ]
```
and of course, if you moved it to another file, then:
```
Sections that were moved:
[ <a href="../new-file#section-b">Section A</a><a id="section-a"></a> ]
```
Use the relative style to link to the new file so that the versioned docs continue to work.
For an example of a rich moved section set please see the very end of [the Trainer doc](https://github.com/huggingface/transformers/blob/main/docs/source/en/main_classes/trainer.md).
## Writing Documentation - Specification
The `huggingface/transformers` documentation follows the
[Google documentation](https://sphinxcontrib-napoleon.readthedocs.io/en/latest/example_google.html) style for docstrings,
although we can write them directly in Markdown.
### Adding a new tutorial
Adding a new tutorial or section is done in two steps:
- Add a new file under `./source`. This file can either be ReStructuredText (.rst) or Markdown (.md).
- Link that file in `./source/_toctree.yml` on the correct toc-tree.
Make sure to put your new file under the proper section. It's unlikely to go in the first section (*Get Started*), so
depending on the intended targets (beginners, more advanced users, or researchers) it should go in sections two, three, or
four.
### Translating
When translating, refer to the guide at [./TRANSLATING.md](https://github.com/huggingface/transformers/blob/main/docs/TRANSLATING.md).
### Adding a new model
When adding a new model:
- Create a file `xxx.md` or under `./source/model_doc` (don't hesitate to copy an existing file as template).
- Link that file in `./source/_toctree.yml`.
- Write a short overview of the model:
- Overview with paper & authors
- Paper abstract
- Tips and tricks and how to use it best
- Add the classes that should be linked in the model. This generally includes the configuration, the tokenizer, and
every model of that class (the base model, alongside models with additional heads), both in PyTorch and TensorFlow.
The order is generally:
- Configuration
- Tokenizer
- PyTorch base model
- PyTorch head models
- TensorFlow base model
- TensorFlow head models
- Flax base model
- Flax head models
These classes should be added using our Markdown syntax. Usually as follows:
```
## XXXConfig
[[autodoc]] XXXConfig
```
This will include every public method of the configuration that is documented. If for some reason you wish for a method
not to be displayed in the documentation, you can do so by specifying which methods should be in the docs:
```
## XXXTokenizer
[[autodoc]] XXXTokenizer
- build_inputs_with_special_tokens
- get_special_tokens_mask
- create_token_type_ids_from_sequences
- save_vocabulary
```
If you just want to add a method that is not documented (for instance magic methods like `__call__` are not documented
by default) you can put the list of methods to add in a list that contains `all`:
```
## XXXTokenizer
[[autodoc]] XXXTokenizer
- all
- __call__
```
### Writing source documentation
Values that should be put in `code` should either be surrounded by backticks: \`like so\`. Note that argument names
and objects like True, None, or any strings should usually be put in `code`.
When mentioning a class, function, or method, it is recommended to use our syntax for internal links so that our tool
adds a link to its documentation with this syntax: \[\`XXXClass\`\] or \[\`function\`\]. This requires the class or
function to be in the main package.
If you want to create a link to some internal class or function, you need to
provide its path. For instance: \[\`utils.ModelOutput\`\]. This will be converted into a link with
`utils.ModelOutput` in the description. To get rid of the path and only keep the name of the object you are
linking to in the description, add a ~: \[\`~utils.ModelOutput\`\] will generate a link with `ModelOutput` in the description.
The same works for methods so you can either use \[\`XXXClass.method\`\] or \[\`~XXXClass.method\`\].
#### Defining arguments in a method
Arguments should be defined with the `Args:` (or `Arguments:` or `Parameters:`) prefix, followed by a line return and
an indentation. The argument should be followed by its type, with its shape if it is a tensor, a colon, and its
description:
```
Args:
n_layers (`int`): The number of layers of the model.
```
If the description is too long to fit in one line, another indentation is necessary before writing the description
after the argument.
Here's an example showcasing everything so far:
```
Args:
input_ids (`torch.LongTensor` of shape `(batch_size, sequence_length)`):
Indices of input sequence tokens in the vocabulary.
Indices can be obtained using [`AlbertTokenizer`]. See [`~PreTrainedTokenizer.encode`] and
[`~PreTrainedTokenizer.__call__`] for details.
[What are input IDs?](../glossary#input-ids)
```
For optional arguments or arguments with defaults we follow the following syntax: imagine we have a function with the
following signature:
```
def my_function(x: str = None, a: float = 1):
```
then its documentation should look like this:
```
Args:
x (`str`, *optional*):
This argument controls ...
a (`float`, *optional*, defaults to 1):
This argument is used to ...
```
Note that we always omit the "defaults to \`None\`" when None is the default for any argument. Also note that even
if the first line describing your argument type and its default gets long, you can't break it on several lines. You can
however, write as many lines as you want in the indented description (see the example above with `input_ids`).
#### Writing a multi-line code block
Multi-line code blocks can be useful for displaying examples. They are done between two lines of three backticks as usual in Markdown:
````
```
# first line of code
# second line
# etc
```
````
We follow the [doctest](https://docs.python.org/3/library/doctest.html) syntax for the examples to automatically test
the results to stay consistent with the library.
#### Writing a return block
The return block should be introduced with the `Returns:` prefix, followed by a line return and an indentation.
The first line should be the type of the return, followed by a line return. No need to indent further for the elements
building the return.
Here's an example of a single value return:
```
Returns:
`List[int]`: A list of integers in the range [0, 1] --- 1 for a special token, 0 for a sequence token.
```
Here's an example of a tuple return, comprising several objects:
```
Returns:
`tuple(torch.FloatTensor)` comprising various elements depending on the configuration ([`BertConfig`]) and inputs:
- ** loss** (*optional*, returned when `masked_lm_labels` is provided) `torch.FloatTensor` of shape `(1,)` --
Total loss is the sum of the masked language modeling loss and the next sequence prediction (classification) loss.
- **prediction_scores** (`torch.FloatTensor` of shape `(batch_size, sequence_length, config.vocab_size)`) --
Prediction scores of the language modeling head (scores for each vocabulary token before SoftMax).
```
#### Adding an image
Due to the rapidly growing repository, it is important to make sure that no files that would significantly weigh down the repository are added. This includes images, videos, and other non-text files. We prefer to leverage a hf.co hosted `dataset` like
the ones hosted on [`hf-internal-testing`](https://huggingface.co/hf-internal-testing) in which to place these files and reference
them by URL. We recommend putting them in the following dataset: [huggingface/documentation-images](https://huggingface.co/datasets/huggingface/documentation-images).
If an external contribution, feel free to add the images to your PR and ask a Hugging Face member to migrate your images
to this dataset.
## Styling the docstring
We have an automatic script running with the `make style` comment that will make sure that:
- the docstrings fully take advantage of the line width
- all code examples are formatted using black, like the code of the Transformers library
This script may have some weird failures if you made a syntax mistake or if you uncover a bug. Therefore, it's
recommended to commit your changes before running `make style`, so you can revert the changes done by that script
easily.
# Testing documentation examples
Good documentation often comes with an example of how a specific function or class should be used.
Each model class should contain at least one example showcasing
how to use this model class in inference. *E.g.* the class [Wav2Vec2ForCTC](https://huggingface.co/docs/transformers/model_doc/wav2vec2#transformers.Wav2Vec2ForCTC)
includes an example of how to transcribe speech to text in the
[docstring of its forward function](https://huggingface.co/docs/transformers/model_doc/wav2vec2#transformers.Wav2Vec2ForCTC.forward).
## Writing documentation examples
The syntax for Example docstrings can look as follows:
```
Example:
```python
>>> from transformers import Wav2Vec2Processor, Wav2Vec2ForCTC
>>> from datasets import load_dataset
>>> import torch
>>> dataset = load_dataset("hf-internal-testing/librispeech_asr_demo", "clean", split="validation")
>>> dataset = dataset.sort("id")
>>> sampling_rate = dataset.features["audio"].sampling_rate
>>> processor = Wav2Vec2Processor.from_pretrained("facebook/wav2vec2-base-960h")
>>> model = Wav2Vec2ForCTC.from_pretrained("facebook/wav2vec2-base-960h")
>>> # audio file is decoded on the fly
>>> inputs = processor(dataset[0]["audio"]["array"], sampling_rate=sampling_rate, return_tensors="pt")
>>> with torch.no_grad():
... logits = model(**inputs).logits
>>> predicted_ids = torch.argmax(logits, dim=-1)
>>> # transcribe speech
>>> transcription = processor.batch_decode(predicted_ids)
>>> transcription[0]
'MISTER QUILTER IS THE APOSTLE OF THE MIDDLE CLASSES AND WE ARE GLAD TO WELCOME HIS GOSPEL'
```
```
The docstring should give a minimal, clear example of how the respective model
is to be used in inference and also include the expected (ideally sensible)
output.
Often, readers will try out the example before even going through the function
or class definitions. Therefore, it is of utmost importance that the example
works as expected.
## Docstring testing
To do so each example should be included in the doctests.
We use pytests' [doctest integration](https://docs.pytest.org/doctest.html) to verify that all of our examples run correctly.
For Transformers, the doctests are run on a daily basis via GitHub Actions as can be
seen [here](https://github.com/huggingface/transformers/actions/workflows/doctests.yml).
### For Python files
Run all the tests in the docstrings of a given file with the following command, here is how we test the modeling file of Wav2Vec2 for instance:
```bash
pytest --doctest-modules src/transformers/models/wav2vec2/modeling_wav2vec2.py -sv --doctest-continue-on-failure
```
If you want to isolate a specific docstring, just add `::` after the file name then type the whole path of the function/class/method whose docstring you want to test. For instance, here is how to just test the forward method of `Wav2Vec2ForCTC`:
```bash
pytest --doctest-modules src/transformers/models/wav2vec2/modeling_wav2vec2.py::transformers.models.wav2vec2.modeling_wav2vec2.Wav2Vec2ForCTC.forward -sv --doctest-continue-on-failure
```
### For Markdown files
You can test locally a given file with this command (here testing the quicktour):
```bash
pytest --doctest-modules docs/source/quicktour.md -sv --doctest-continue-on-failure --doctest-glob="*.md"
```
### Writing doctests
Here are a few tips to help you debug the doctests and make them pass:
- The outputs of the code need to match the expected output **exactly**, so make sure you have the same outputs. In particular doctest will see a difference between single quotes and double quotes, or a missing parenthesis. The only exceptions to that rule are:
* whitespace: one give whitespace (space, tabulation, new line) is equivalent to any number of whitespace, so you can add new lines where there are spaces to make your output more readable.
* numerical values: you should never put more than 4 or 5 digits to expected results as different setups or library versions might get you slightly different results. `doctest` is configured to ignore any difference lower than the precision to which you wrote (so 1e-4 if you write 4 digits).
- Don't leave a block of code that is very long to execute. If you can't make it fast, you can either not use the doctest syntax on it (so that it's ignored), or if you want to use the doctest syntax to show the results, you can add a comment `# doctest: +SKIP` at the end of the lines of code too long to execute
- Each line of code that produces a result needs to have that result written below. You can ignore an output if you don't want to show it in your code example by adding a comment ` # doctest: +IGNORE_RESULT` at the end of the line of code producing it.
| 0
|
mavonic_private_repos/transformers/docs
|
mavonic_private_repos/transformers/docs/source/_config.py
|
# docstyle-ignore
INSTALL_CONTENT = """
# Transformers installation
! pip install transformers datasets evaluate accelerate
# To install from source instead of the last release, comment the command above and uncomment the following one.
# ! pip install git+https://github.com/huggingface/transformers.git
"""
notebook_first_cells = [{"type": "code", "content": INSTALL_CONTENT}]
black_avoid_patterns = {
"{processor_class}": "FakeProcessorClass",
"{model_class}": "FakeModelClass",
"{object_class}": "FakeObjectClass",
}
| 0
|
mavonic_private_repos/transformers/docs/source
|
mavonic_private_repos/transformers/docs/source/pt/fast_tokenizers.md
|
<!--Copyright 2020 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
specific language governing permissions and limitations under the License.
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer.
-->
# Usando os Tokenizers do 🤗 Tokenizers
O [`PreTrainedTokenizerFast`] depende da biblioteca [🤗 Tokenizers](https://huggingface.co/docs/tokenizers). O Tokenizer obtido da biblioteca 🤗 Tokenizers pode ser carregado facilmente pelo 🤗 Transformers.
Antes de entrar nos detalhes, vamos começar criando um tokenizer fictício em algumas linhas:
```python
>>> from tokenizers import Tokenizer
>>> from tokenizers.models import BPE
>>> from tokenizers.trainers import BpeTrainer
>>> from tokenizers.pre_tokenizers import Whitespace
>>> tokenizer = Tokenizer(BPE(unk_token="[UNK]"))
>>> trainer = BpeTrainer(special_tokens=["[UNK]", "[CLS]", "[SEP]", "[PAD]", "[MASK]"])
>>> tokenizer.pre_tokenizer = Whitespace()
>>> files = [...]
>>> tokenizer.train(files, trainer)
```
Agora temos um tokenizer treinado nos arquivos que foram definidos. Nós podemos continuar usando nessa execução ou salvar em um arquivo JSON para re-utilizar no futuro.
## Carregando diretamente de um objeto tokenizer
Vamos ver como aproveitar esse objeto tokenizer na biblioteca 🤗 Transformers. A classe [`PreTrainedTokenizerFast`] permite uma instanciação fácil, aceitando o objeto *tokenizer* instanciado como um argumento:
```python
>>> from transformers import PreTrainedTokenizerFast
>>> fast_tokenizer = PreTrainedTokenizerFast(tokenizer_object=tokenizer)
```
Esse objeto pode ser utilizado com todos os métodos compartilhados pelos tokenizers dos 🤗 Transformers! Vá para [a página do tokenizer](main_classes/tokenizer) para mais informações.
## Carregando de um arquivo JSON
Para carregar um tokenizer de um arquivo JSON vamos primeiro começar salvando nosso tokenizer:
```python
>>> tokenizer.save("tokenizer.json")
```
A pasta para qual salvamos esse arquivo pode ser passada para o método de inicialização do [`PreTrainedTokenizerFast`] usando o `tokenizer_file` parâmetro:
```python
>>> from transformers import PreTrainedTokenizerFast
>>> fast_tokenizer = PreTrainedTokenizerFast(tokenizer_file="tokenizer.json")
```
Esse objeto pode ser utilizado com todos os métodos compartilhados pelos tokenizers dos 🤗 Transformers! Vá para [a página do tokenizer](main_classes/tokenizer) para mais informações.
| 0
|
mavonic_private_repos/transformers/docs/source
|
mavonic_private_repos/transformers/docs/source/pt/run_scripts.md
|
<!--Copyright 2022 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
specific language governing permissions and limitations under the License.
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer.
-->
# Treinamento a partir de um script
Junto com os 🤗 Transformers [notebooks](./notebooks), também há scripts de exemplo demonstrando como treinar um modelo para uma tarefa com [PyTorch](https://github.com/huggingface/transformers/tree/main/examples/pytorch), [TensorFlow](https://github.com/huggingface/transformers/tree/main/examples/tensorflow) ou [JAX/Flax](https://github.com/huggingface/transformers/tree/main/examples/flax).
Você também encontrará scripts que usamos em nossos [projetos de pesquisa](https://github.com/huggingface/transformers/tree/main/examples/research_projects) e [exemplos legados](https://github.com/huggingface/transformers/tree/main/examples/legacy) que são principalmente contribuições da comunidade. Esses scripts não são mantidos ativamente e exigem uma versão específica de 🤗 Transformers que provavelmente será incompatível com a versão mais recente da biblioteca.
Não se espera que os scripts de exemplo funcionem imediatamente em todos os problemas, você pode precisar adaptar o script ao problema que está tentando resolver. Para ajudá-lo com isso, a maioria dos scripts expõe totalmente como os dados são pré-processados, permitindo que você os edite conforme necessário para seu caso de uso.
Para qualquer recurso que você gostaria de implementar em um script de exemplo, discuta-o no [fórum](https://discuss.huggingface.co/) ou em uma [issue](https://github.com/huggingface/transformers/issues) antes de enviar um Pull Request. Embora recebamos correções de bugs, é improvável que mesclaremos um Pull Request que adicione mais funcionalidades ao custo de legibilidade.
Este guia mostrará como executar um exemplo de script de treinamento de sumarização em [PyTorch](https://github.com/huggingface/transformers/tree/main/examples/pytorch/summarization) e [TensorFlow](https://github.com/huggingface/transformers/tree/main/examples/tensorflow/summarization). Espera-se que todos os exemplos funcionem com ambas as estruturas, a menos que especificado de outra forma.
## Configuração
Para executar com êxito a versão mais recente dos scripts de exemplo, você precisa **instalar o 🤗 Transformers da fonte** em um novo ambiente virtual:
```bash
git clone https://github.com/huggingface/transformers
cd transformers
pip install .
```
Para versões mais antigas dos scripts de exemplo, clique no botão abaixo:
<details>
<summary>Exemplos para versões antigas dos 🤗 Transformers</summary>
<ul>
<li><a href="https://github.com/huggingface/transformers/tree/v4.5.1/examples">v4.5.1</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v4.4.2/examples">v4.4.2</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v4.3.3/examples">v4.3.3</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v4.2.2/examples">v4.2.2</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v4.1.1/examples">v4.1.1</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v4.0.1/examples">v4.0.1</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v3.5.1/examples">v3.5.1</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v3.4.0/examples">v3.4.0</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v3.3.1/examples">v3.3.1</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v3.2.0/examples">v3.2.0</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v3.1.0/examples">v3.1.0</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v3.0.2/examples">v3.0.2</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v2.11.0/examples">v2.11.0</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v2.10.0/examples">v2.10.0</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v2.9.1/examples">v2.9.1</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v2.8.0/examples">v2.8.0</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v2.7.0/examples">v2.7.0</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v2.6.0/examples">v2.6.0</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v2.5.1/examples">v2.5.1</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v2.4.0/examples">v2.4.0</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v2.3.0/examples">v2.3.0</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v2.2.0/examples">v2.2.0</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v2.1.0/examples">v2.1.1</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v2.0.0/examples">v2.0.0</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v1.2.0/examples">v1.2.0</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v1.1.0/examples">v1.1.0</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v1.0.0/examples">v1.0.0</a></li>
</ul>
</details>
Em seguida, mude seu clone atual dos 🤗 Transformers para uma versão específica, como v3.5.1, por exemplo:
```bash
git checkout tags/v3.5.1
```
Depois de configurar a versão correta da biblioteca, navegue até a pasta de exemplo de sua escolha e instale os requisitos específicos do exemplo:
```bash
pip install -r requirements.txt
```
## Executando um script
<frameworkcontent>
<pt>
O script de exemplo baixa e pré-processa um conjunto de dados da biblioteca 🤗 [Datasets](https://huggingface.co/docs/datasets/). Em seguida, o script ajusta um conjunto de dados com o [Trainer](https://huggingface.co/docs/transformers/main_classes/trainer) em uma arquitetura que oferece suporte à sumarização. O exemplo a seguir mostra como ajustar [T5-small](https://huggingface.co/google-t5/t5-small) no conjunto de dados [CNN/DailyMail](https://huggingface.co/datasets/cnn_dailymail). O modelo T5 requer um argumento `source_prefix` adicional devido à forma como foi treinado. Este prompt informa ao T5 que esta é uma tarefa de sumarização.
```bash
python examples/pytorch/summarization/run_summarization.py \
--model_name_or_path google-t5/t5-small \
--do_train \
--do_eval \
--dataset_name cnn_dailymail \
--dataset_config "3.0.0" \
--source_prefix "summarize: " \
--output_dir /tmp/tst-summarization \
--per_device_train_batch_size=4 \
--per_device_eval_batch_size=4 \
--overwrite_output_dir \
--predict_with_generate
```
</pt>
<tf>
Este outro script de exemplo baixa e pré-processa um conjunto de dados da biblioteca 🤗 [Datasets](https://huggingface.co/docs/datasets/). Em seguida, o script ajusta um conjunto de dados usando Keras em uma arquitetura que oferece suporte à sumarização. O exemplo a seguir mostra como ajustar [T5-small](https://huggingface.co/google-t5/t5-small) no conjunto de dados [CNN/DailyMail](https://huggingface.co/datasets/cnn_dailymail). O modelo T5 requer um argumento `source_prefix` adicional devido à forma como foi treinado. Este prompt informa ao T5 que esta é uma tarefa de sumarização.
```bash
python examples/tensorflow/summarization/run_summarization.py \
--model_name_or_path google-t5/t5-small \
--dataset_name cnn_dailymail \
--dataset_config "3.0.0" \
--output_dir /tmp/tst-summarization \
--per_device_train_batch_size 8 \
--per_device_eval_batch_size 16 \
--num_train_epochs 3 \
--do_train \
--do_eval
```
</tf>
</frameworkcontent>
## Treinamento distribuído e precisão mista
O [Trainer](https://huggingface.co/docs/transformers/main_classes/trainer) oferece suporte a treinamento distribuído e precisão mista, o que significa que você também pode usá-lo em um script. Para habilitar esses dois recursos:
- Adicione o argumento `fp16` para habilitar a precisão mista.
- Defina o número de GPUs a serem usadas com o argumento `nproc_per_node`.
```bash
torchrun \
--nproc_per_node 8 pytorch/summarization/run_summarization.py \
--fp16 \
--model_name_or_path google-t5/t5-small \
--do_train \
--do_eval \
--dataset_name cnn_dailymail \
--dataset_config "3.0.0" \
--source_prefix "summarize: " \
--output_dir /tmp/tst-summarization \
--per_device_train_batch_size=4 \
--per_device_eval_batch_size=4 \
--overwrite_output_dir \
--predict_with_generate
```
Os scripts do TensorFlow utilizam um [`MirroredStrategy`](https://www.tensorflow.org/guide/distributed_training#mirroredstrategy) para treinamento distribuído, e você não precisa adicionar argumentos adicionais ao script de treinamento. O script do TensorFlow usará várias GPUs por padrão, se estiverem disponíveis.
## Executando um script em uma TPU
<frameworkcontent>
<pt>
As Unidades de Processamento de Tensor (TPUs) são projetadas especificamente para acelerar o desempenho. O PyTorch oferece suporte a TPUs com o compilador de aprendizado profundo [XLA](https://www.tensorflow.org/xla) (consulte [aqui](https://github.com/pytorch/xla/blob/master/README.md) para mais detalhes). Para usar uma TPU, inicie o script `xla_spawn.py` e use o argumento `num_cores` para definir o número de núcleos de TPU que você deseja usar.
```bash
python xla_spawn.py --num_cores 8 \
summarization/run_summarization.py \
--model_name_or_path google-t5/t5-small \
--do_train \
--do_eval \
--dataset_name cnn_dailymail \
--dataset_config "3.0.0" \
--source_prefix "summarize: " \
--output_dir /tmp/tst-summarization \
--per_device_train_batch_size=4 \
--per_device_eval_batch_size=4 \
--overwrite_output_dir \
--predict_with_generate
```
</pt>
<tf>
As Unidades de Processamento de Tensor (TPUs) são projetadas especificamente para acelerar o desempenho. Os scripts do TensorFlow utilizam uma [`TPUStrategy`](https://www.tensorflow.org/guide/distributed_training#tpustrategy) para treinamento em TPUs. Para usar uma TPU, passe o nome do recurso TPU para o argumento `tpu`.
```bash
python run_summarization.py \
--tpu name_of_tpu_resource \
--model_name_or_path google-t5/t5-small \
--dataset_name cnn_dailymail \
--dataset_config "3.0.0" \
--output_dir /tmp/tst-summarization \
--per_device_train_batch_size 8 \
--per_device_eval_batch_size 16 \
--num_train_epochs 3 \
--do_train \
--do_eval
```
</tf>
</frameworkcontent>
## Execute um script com 🤗 Accelerate
🤗 [Accelerate](https://huggingface.co/docs/accelerate) é uma biblioteca somente do PyTorch que oferece um método unificado para treinar um modelo em vários tipos de configurações (CPU, multiplas GPUs, TPUs), mantendo visibilidade no loop de treinamento do PyTorch. Certifique-se de ter o 🤗 Accelerate instalado se ainda não o tiver:
> Nota: Como o Accelerate está se desenvolvendo rapidamente, a versão git do Accelerate deve ser instalada para executar os scripts
```bash
pip install git+https://github.com/huggingface/accelerate
```
Em vez do script `run_summarization.py`, você precisa usar o script `run_summarization_no_trainer.py`. Os scripts suportados pelo 🤗 Accelerate terão um arquivo `task_no_trainer.py` na pasta. Comece executando o seguinte comando para criar e salvar um arquivo de configuração:
```bash
accelerate config
```
Teste sua configuração para garantir que ela esteja corretamente configurada :
```bash
accelerate test
```
Agora você está pronto para iniciar o treinamento:
```bash
accelerate launch run_summarization_no_trainer.py \
--model_name_or_path google-t5/t5-small \
--dataset_name cnn_dailymail \
--dataset_config "3.0.0" \
--source_prefix "summarize: " \
--output_dir ~/tmp/tst-summarization
```
## Usando um conjunto de dados personalizado
O script de resumo oferece suporte a conjuntos de dados personalizados, desde que sejam um arquivo CSV ou JSON. Ao usar seu próprio conjunto de dados, você precisa especificar vários argumentos adicionais:
- `train_file` e `validation_file` especificam o caminho para seus arquivos de treinamento e validação respectivamente.
- `text_column` é o texto de entrada para sumarização.
- `summary_column` é o texto de destino para saída.
Um script para sumarização usando um conjunto de dados customizado ficaria assim:
```bash
python examples/pytorch/summarization/run_summarization.py \
--model_name_or_path google-t5/t5-small \
--do_train \
--do_eval \
--train_file path_to_csv_or_jsonlines_file \
--validation_file path_to_csv_or_jsonlines_file \
--text_column text_column_name \
--summary_column summary_column_name \
--source_prefix "summarize: " \
--output_dir /tmp/tst-summarization \
--overwrite_output_dir \
--per_device_train_batch_size=4 \
--per_device_eval_batch_size=4 \
--predict_with_generate
```
## Testando um script
Geralmente, é uma boa ideia executar seu script em um número menor de exemplos de conjuntos de dados para garantir que tudo funcione conforme o esperado antes de se comprometer com um conjunto de dados inteiro, que pode levar horas para ser concluído. Use os seguintes argumentos para truncar o conjunto de dados para um número máximo de amostras:
- `max_train_samples`
- `max_eval_samples`
- `max_predict_samples`
```bash
python examples/pytorch/summarization/run_summarization.py \
--model_name_or_path google-t5/t5-small \
--max_train_samples 50 \
--max_eval_samples 50 \
--max_predict_samples 50 \
--do_train \
--do_eval \
--dataset_name cnn_dailymail \
--dataset_config "3.0.0" \
--source_prefix "summarize: " \
--output_dir /tmp/tst-summarization \
--per_device_train_batch_size=4 \
--per_device_eval_batch_size=4 \
--overwrite_output_dir \
--predict_with_generate
```
Nem todos os scripts de exemplo suportam o argumento `max_predict_samples`. Se você não tiver certeza se seu script suporta este argumento, adicione o argumento `-h` para verificar:
```bash
examples/pytorch/summarization/run_summarization.py -h
```
## Retomar o treinamento a partir de um checkpoint
Outra opção útil para habilitar é retomar o treinamento de um checkpoint anterior. Isso garantirá que você possa continuar de onde parou sem recomeçar se o seu treinamento for interrompido. Existem dois métodos para retomar o treinamento a partir de um checkpoint.
O primeiro método usa o argumento `output_dir previous_output_dir` para retomar o treinamento do último checkpoint armazenado em `output_dir`. Neste caso, você deve remover `overwrite_output_dir`:
```bash
python examples/pytorch/summarization/run_summarization.py
--model_name_or_path google-t5/t5-small \
--do_train \
--do_eval \
--dataset_name cnn_dailymail \
--dataset_config "3.0.0" \
--source_prefix "summarize: " \
--output_dir /tmp/tst-summarization \
--per_device_train_batch_size=4 \
--per_device_eval_batch_size=4 \
--output_dir previous_output_dir \
--predict_with_generate
```
O segundo método usa o argumento `resume_from_checkpoint path_to_specific_checkpoint` para retomar o treinamento de uma pasta de checkpoint específica.
```bash
python examples/pytorch/summarization/run_summarization.py
--model_name_or_path google-t5/t5-small \
--do_train \
--do_eval \
--dataset_name cnn_dailymail \
--dataset_config "3.0.0" \
--source_prefix "summarize: " \
--output_dir /tmp/tst-summarization \
--per_device_train_batch_size=4 \
--per_device_eval_batch_size=4 \
--overwrite_output_dir \
--resume_from_checkpoint path_to_specific_checkpoint \
--predict_with_generate
```
## Compartilhando seu modelo
Todos os scripts podem enviar seu modelo final para o [Model Hub](https://huggingface.co/models). Certifique-se de estar conectado ao Hugging Face antes de começar:
```bash
huggingface-cli login
```
Em seguida, adicione o argumento `push_to_hub` ao script. Este argumento criará um repositório com seu nome de usuário do Hugging Face e o nome da pasta especificado em `output_dir`.
Para dar um nome específico ao seu repositório, use o argumento `push_to_hub_model_id` para adicioná-lo. O repositório será listado automaticamente em seu namespace.
O exemplo a seguir mostra como fazer upload de um modelo com um nome de repositório específico:
```bash
python examples/pytorch/summarization/run_summarization.py
--model_name_or_path google-t5/t5-small \
--do_train \
--do_eval \
--dataset_name cnn_dailymail \
--dataset_config "3.0.0" \
--source_prefix "summarize: " \
--push_to_hub \
--push_to_hub_model_id finetuned-t5-cnn_dailymail \
--output_dir /tmp/tst-summarization \
--per_device_train_batch_size=4 \
--per_device_eval_batch_size=4 \
--overwrite_output_dir \
--predict_with_generate
```
| 0
|
mavonic_private_repos/transformers/docs/source
|
mavonic_private_repos/transformers/docs/source/pt/create_a_model.md
|
<!--Copyright 2022 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
specific language governing permissions and limitations under the License.
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer.
-->
# Criar uma arquitetura customizada
Uma [`AutoClass`](model_doc/auto) automaticamente infere a arquitetura do modelo e baixa configurações e pesos pré-treinados. Geralmente, nós recomendamos usar uma `AutoClass` para produzir um código independente de checkpoints. Mas usuários que querem mais contole sobre parâmetros específicos do modelo pode criar um modelo customizado 🤗 Transformers a partir de algumas classes bases. Isso pode ser particulamente útil para alguém que está interessado em estudar, treinar ou fazer experimentos com um modelo 🤗 Transformers. Nesse tutorial, será explicado como criar um modelo customizado sem uma `AutoClass`. Aprenda como:
- Carregar e customizar a configuração de um modelo.
- Criar a arquitetura de um modelo.
- Criar um tokenizer rápido e devagar para textos.
- Criar extrator de features para tarefas envolvendo audio e imagem.
- Criar um processador para tarefas multimodais.
## configuration
A [configuration](main_classes/configuration) refere-se a atributos específicos de um modelo. Cada configuração de modelo tem atributos diferentes; por exemplo, todos modelo de PLN possuem os atributos `hidden_size`, `num_attention_heads`, `num_hidden_layers` e `vocab_size` em comum. Esse atributos especificam o numero de 'attention heads' ou 'hidden layers' para construir um modelo.
Dê uma olhada a mais em [DistilBERT](model_doc/distilbert) acessando [`DistilBertConfig`] para observar esses atributos:
```py
>>> from transformers import DistilBertConfig
>>> config = DistilBertConfig()
>>> print(config)
DistilBertConfig {
"activation": "gelu",
"attention_dropout": 0.1,
"dim": 768,
"dropout": 0.1,
"hidden_dim": 3072,
"initializer_range": 0.02,
"max_position_embeddings": 512,
"model_type": "distilbert",
"n_heads": 12,
"n_layers": 6,
"pad_token_id": 0,
"qa_dropout": 0.1,
"seq_classif_dropout": 0.2,
"sinusoidal_pos_embds": false,
"transformers_version": "4.16.2",
"vocab_size": 30522
}
```
[`DistilBertConfig`] mostra todos os atributos padrões usados para construir um [`DistilBertModel`] base. Todos atributos são customizáveis, o que cria espaço para experimentos. Por exemplo, você pode customizar um modelo padrão para:
- Tentar uma função de ativação diferente com o parâmetro `activation`.
- Usar uma taxa de desistência maior para as probabilidades de 'attention' com o parâmetro `attention_dropout`.
```py
>>> my_config = DistilBertConfig(activation="relu", attention_dropout=0.4)
>>> print(my_config)
DistilBertConfig {
"activation": "relu",
"attention_dropout": 0.4,
"dim": 768,
"dropout": 0.1,
"hidden_dim": 3072,
"initializer_range": 0.02,
"max_position_embeddings": 512,
"model_type": "distilbert",
"n_heads": 12,
"n_layers": 6,
"pad_token_id": 0,
"qa_dropout": 0.1,
"seq_classif_dropout": 0.2,
"sinusoidal_pos_embds": false,
"transformers_version": "4.16.2",
"vocab_size": 30522
}
```
Atributos de um modelo pré-treinado podem ser modificados na função [`~PretrainedConfig.from_pretrained`]:
```py
>>> my_config = DistilBertConfig.from_pretrained("distilbert/distilbert-base-uncased", activation="relu", attention_dropout=0.4)
```
Uma vez que você está satisfeito com as configurações do seu modelo, você consegue salvar elas com [`~PretrainedConfig.save_pretrained`]. Seu arquivo de configurações está salvo como um arquivo JSON no diretório especificado:
```py
>>> my_config.save_pretrained(save_directory="./your_model_save_path")
```
Para reusar o arquivo de configurações, carregue com [`~PretrainedConfig.from_pretrained`]:
```py
>>> my_config = DistilBertConfig.from_pretrained("./your_model_save_path/my_config.json")
```
<Tip>
Você pode também salvar seu arquivo de configurações como um dicionário ou até mesmo com a diferença entre as seus atributos de configuração customizados e os atributos de configuração padrões! Olhe a documentação [configuration](main_classes/configuration) para mais detalhes.
</Tip>
## Modelo
O próximo passo é criar um [model](main_classes/models). O modelo - também vagamente referido como arquitetura - define o que cada camada está fazendo e quais operações estão acontecendo. Atributos como `num_hidden_layers` das configurações são utilizados para definir a arquitetura. Todo modelo compartilha a classe base [`PreTrainedModel`] e alguns métodos em comum como redimensionar o tamanho dos embeddings de entrada e podar as 'self-attention heads'. Além disso, todos os modelos também são subclasses de [`torch.nn.Module`](https://pytorch.org/docs/stable/generated/torch.nn.Module.html), [`tf.keras.Model`](https://www.tensorflow.org/api_docs/python/tf/keras/Model) ou [`flax.linen.Module`](https://flax.readthedocs.io/en/latest/api_reference/flax.linen/module.html). Isso significa que os modelos são compatíveis com cada respectivo uso de framework.
<frameworkcontent>
<pt>
Carregar seus atributos de configuração customizados em um modelo:
```py
>>> from transformers import DistilBertModel
>>> my_config = DistilBertConfig.from_pretrained("./your_model_save_path/my_config.json")
>>> model = DistilBertModel(my_config)
```
Isso cria um modelo com valores aleatórios ao invés de pré-treinar os pesos. Você não irá conseguir usar usar esse modelo para nada útil ainda, até você treinar ele. Treino é um processo caro e demorado. Geralmente é melhor utilizar um modelo pré-treinado para obter melhores resultados mais rápido, enquanto usa apenas uma fração dos recursos necessários para treinar.
Criar um modelo pré-treinado com [`~PreTrainedModel.from_pretrained`]:
```py
>>> model = DistilBertModel.from_pretrained("distilbert/distilbert-base-uncased")
```
Quando você carregar os pesos pré-treinados, a configuração padrão do modelo é automaticamente carregada se o modelo é provido pelo 🤗 Transformers. No entanto, você ainda consegue mudar - alguns ou todos - os atributos padrões de configuração do modelo com os seus próprio atributos, se você preferir:
```py
>>> model = DistilBertModel.from_pretrained("distilbert/distilbert-base-uncased", config=my_config)
```
</pt>
<tf>
Carregar os seus próprios atributos padrões de contiguração no modelo:
```py
>>> from transformers import TFDistilBertModel
>>> my_config = DistilBertConfig.from_pretrained("./your_model_save_path/my_config.json")
>>> tf_model = TFDistilBertModel(my_config)
```
Isso cria um modelo com valores aleatórios ao invés de pré-treinar os pesos. Você não irá conseguir usar usar esse modelo para nada útil ainda, até você treinar ele. Treino é um processo caro e demorado. Geralmente é melhor utilizar um modelo pré-treinado para obter melhores resultados mais rápido, enquanto usa apenas uma fração dos recursos necessários para treinar.
Criar um modelo pré-treinado com [`~TFPreTrainedModel.from_pretrained`]:
```py
>>> tf_model = TFDistilBertModel.from_pretrained("distilbert/distilbert-base-uncased")
```
Quando você carregar os pesos pré-treinados, a configuração padrão do modelo é automaticamente carregada se o modelo é provido pelo 🤗 Transformers. No entanto, você ainda consegue mudar - alguns ou todos - os atributos padrões de configuração do modelo com os seus próprio atributos, se você preferir:
```py
>>> tf_model = TFDistilBertModel.from_pretrained("distilbert/distilbert-base-uncased", config=my_config)
```
</tf>
</frameworkcontent>
### Heads do modelo
Neste ponto, você tem um modelo básico do DistilBERT que gera os *estados ocultos*. Os estados ocultos são passados como entrada para a head do moelo para produzir a saída final. 🤗 Transformers fornece uma head de modelo diferente para cada tarefa desde que o modelo suporte essa tarefa (por exemplo, você não consegue utilizar o modelo DistilBERT para uma tarefa de 'sequence-to-sequence' como tradução).
<frameworkcontent>
<pt>
Por exemplo, [`DistilBertForSequenceClassification`] é um modelo DistilBERT base com uma head de classificação de sequência. A head de calssificação de sequência é uma camada linear no topo das saídas agrupadas.
```py
>>> from transformers import DistilBertForSequenceClassification
>>> model = DistilBertForSequenceClassification.from_pretrained("distilbert/distilbert-base-uncased")
```
Reutilize facilmente esse ponto de parada para outra tarefe mudando para uma head de modelo diferente. Para uma tarefe de responder questões, você usaria a head do modelo [`DistilBertForQuestionAnswering`]. A head de responder questões é similar com a de classificação de sequências exceto o fato de que ela é uma camada no topo dos estados das saídas ocultas.
```py
>>> from transformers import DistilBertForQuestionAnswering
>>> model = DistilBertForQuestionAnswering.from_pretrained("distilbert/distilbert-base-uncased")
```
</pt>
<tf>
Por exemplo, [`TFDistilBertForSequenceClassification`] é um modelo DistilBERT base com uma head de classificação de sequência. A head de calssificação de sequência é uma camada linear no topo das saídas agrupadas.
```py
>>> from transformers import TFDistilBertForSequenceClassification
>>> tf_model = TFDistilBertForSequenceClassification.from_pretrained("distilbert/distilbert-base-uncased")
```
Reutilize facilmente esse ponto de parada para outra tarefe mudando para uma head de modelo diferente. Para uma tarefe de responder questões, você usaria a head do modelo [`TFDistilBertForQuestionAnswering`]. A head de responder questões é similar com a de classificação de sequências exceto o fato de que ela é uma camada no topo dos estados das saídas ocultas.
```py
>>> from transformers import TFDistilBertForQuestionAnswering
>>> tf_model = TFDistilBertForQuestionAnswering.from_pretrained("distilbert/distilbert-base-uncased")
```
</tf>
</frameworkcontent>
## Tokenizer
A útlima classe base que você precisa antes de usar um modelo para dados textuais é a [tokenizer](main_classes/tokenizer) para converter textos originais para tensores. Existem dois tipos de tokenizers que você pode usar com 🤗 Transformers:
- [`PreTrainedTokenizer`]: uma implementação em Python de um tokenizer.
- [`PreTrainedTokenizerFast`]: um tokenizer da nossa biblioteca [🤗 Tokenizer](https://huggingface.co/docs/tokenizers/python/latest/) baseada em Rust. Esse tipo de tokenizer é significantemente mais rapido - especialmente durante tokenization de codificação - devido a implementação em Rust. O tokenizer rápido tambem oferece métodos adicionais como *offset mapping* que mapeia tokens para suar palavras ou caracteres originais.
Os dois tokenizers suporta métodos comuns como os de codificar e decodificar, adicionar novos tokens, e gerenciar tokens especiais.
<Tip warning={true}>
Nem todo modelo suporta um 'fast tokenizer'. De uma olhada aqui [table](index#supported-frameworks) pra checar se um modelo suporta 'fast tokenizer'.
</Tip>
Se você treinou seu prórpio tokenizer, você pode criar um a partir do seu arquivo *vocabulary*:
```py
>>> from transformers import DistilBertTokenizer
>>> my_tokenizer = DistilBertTokenizer(vocab_file="my_vocab_file.txt", do_lower_case=False, padding_side="left")
```
É importante lembrar que o vocabulário de um tokenizer customizado será diferente de um vocabulário gerado pelo tokenizer de um modelo pré treinado. Você precisa usar o vocabulário de um modelo pré treinado se você estiver usando um modelo pré treinado, caso contrário as entradas não farão sentido. Criando um tokenizer com um vocabulário de um modelo pré treinado com a classe [`DistilBertTokenizer`]:
```py
>>> from transformers import DistilBertTokenizer
>>> slow_tokenizer = DistilBertTokenizer.from_pretrained("distilbert/distilbert-base-uncased")
```
Criando um 'fast tokenizer' com a classe [`DistilBertTokenizerFast`]:
```py
>>> from transformers import DistilBertTokenizerFast
>>> fast_tokenizer = DistilBertTokenizerFast.from_pretrained("distilbert/distilbert-base-uncased")
```
<Tip>
Pos padrão, [`AutoTokenizer`] tentará carregar um 'fast tokenizer'. Você pode disabilitar esse comportamento colocando `use_fast=False` no `from_pretrained`.
</Tip>
## Extrator de features
Um extrator de features processa entradas de imagem ou áudio. Ele herda da classe base [`~feature_extraction_utils.FeatureExtractionMixin`], e pode também herdar da classe [`ImageFeatureExtractionMixin`] para processamento de features de imagem ou da classe [`SequenceFeatureExtractor`] para processamento de entradas de áudio.
Dependendo do que você está trabalhando em um audio ou uma tarefa de visão, crie um estrator de features associado com o modelo que você está usando. Por exemplo, crie um [`ViTFeatureExtractor`] padrão se você estiver usando [ViT](model_doc/vit) para classificação de imagens:
```py
>>> from transformers import ViTFeatureExtractor
>>> vit_extractor = ViTFeatureExtractor()
>>> print(vit_extractor)
ViTFeatureExtractor {
"do_normalize": true,
"do_resize": true,
"feature_extractor_type": "ViTFeatureExtractor",
"image_mean": [
0.5,
0.5,
0.5
],
"image_std": [
0.5,
0.5,
0.5
],
"resample": 2,
"size": 224
}
```
<Tip>
Se você não estiver procurando por nenhuma customização, apenas use o método `from_pretrained` para carregar parâmetros do modelo de extrator de features padrão.
</Tip>
Modifique qualquer parâmetro dentre os [`ViTFeatureExtractor`] para criar seu extrator de features customizado.
```py
>>> from transformers import ViTFeatureExtractor
>>> my_vit_extractor = ViTFeatureExtractor(resample="PIL.Image.BOX", do_normalize=False, image_mean=[0.3, 0.3, 0.3])
>>> print(my_vit_extractor)
ViTFeatureExtractor {
"do_normalize": false,
"do_resize": true,
"feature_extractor_type": "ViTFeatureExtractor",
"image_mean": [
0.3,
0.3,
0.3
],
"image_std": [
0.5,
0.5,
0.5
],
"resample": "PIL.Image.BOX",
"size": 224
}
```
Para entradas de áutio, você pode criar um [`Wav2Vec2FeatureExtractor`] e customizar os parâmetros de uma forma similar:
```py
>>> from transformers import Wav2Vec2FeatureExtractor
>>> w2v2_extractor = Wav2Vec2FeatureExtractor()
>>> print(w2v2_extractor)
Wav2Vec2FeatureExtractor {
"do_normalize": true,
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
"feature_size": 1,
"padding_side": "right",
"padding_value": 0.0,
"return_attention_mask": false,
"sampling_rate": 16000
}
```
## Processor
Para modelos que suportam tarefas multimodais, 🤗 Transformers oferece uma classe processadora que convenientemente cobre um extrator de features e tokenizer dentro de um único objeto. Por exemplo, vamos usar o [`Wav2Vec2Processor`] para uma tarefa de reconhecimento de fala automática (ASR). ASR transcreve áudio para texto, então você irá precisar de um extrator de um features e um tokenizer.
Crie um extrator de features para lidar com as entradas de áudio.
```py
>>> from transformers import Wav2Vec2FeatureExtractor
>>> feature_extractor = Wav2Vec2FeatureExtractor(padding_value=1.0, do_normalize=True)
```
Crie um tokenizer para lidar com a entrada de textos:
```py
>>> from transformers import Wav2Vec2CTCTokenizer
>>> tokenizer = Wav2Vec2CTCTokenizer(vocab_file="my_vocab_file.txt")
```
Combine o extrator de features e o tokenizer no [`Wav2Vec2Processor`]:
```py
>>> from transformers import Wav2Vec2Processor
>>> processor = Wav2Vec2Processor(feature_extractor=feature_extractor, tokenizer=tokenizer)
```
Com duas classes básicas - configuração e modelo - e um preprocessamento de classe adicional (tokenizer, extrator de features, ou processador), você pode criar qualquer modelo que suportado por 🤗 Transformers. Qualquer uma dessas classes base são configuráveis, te permitindo usar os atributos específicos que você queira. Você pode facilmente preparar um modelo para treinamento ou modificar um modelo pré-treinado com poucas mudanças.
| 0
|
mavonic_private_repos/transformers/docs/source
|
mavonic_private_repos/transformers/docs/source/pt/_config.py
|
# docstyle-ignore
INSTALL_CONTENT = """
# Transformers installation
! pip install transformers datasets evaluate accelerate
# To install from source instead of the last release, comment the command above and uncomment the following one.
# ! pip install git+https://github.com/huggingface/transformers.git
"""
notebook_first_cells = [{"type": "code", "content": INSTALL_CONTENT}]
black_avoid_patterns = {
"{processor_class}": "FakeProcessorClass",
"{model_class}": "FakeModelClass",
"{object_class}": "FakeObjectClass",
}
| 0
|
mavonic_private_repos/transformers/docs/source
|
mavonic_private_repos/transformers/docs/source/pt/multilingual.md
|
<!--Copyright 2022 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
specific language governing permissions and limitations under the License.
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer.
-->
# Modelos multilinguísticos para inferência
[[open-in-colab]]
Existem vários modelos multilinguísticos no 🤗 Transformers e seus usos para inferência diferem dos modelos monolíngues.
No entanto, nem *todos* os usos dos modelos multilíngues são tão diferentes.
Alguns modelos, como o [google-bert/bert-base-multilingual-uncased](https://huggingface.co/google-bert/bert-base-multilingual-uncased),
podem ser usados como se fossem monolíngues. Este guia irá te ajudar a usar modelos multilíngues cujo uso difere
para o propósito de inferência.
## XLM
O XLM tem dez checkpoints diferentes dos quais apenas um é monolíngue.
Os nove checkpoints restantes do modelo são subdivididos em duas categorias:
checkpoints que usam de language embeddings e os que não.
### XLM com language embeddings
Os seguintes modelos de XLM usam language embeddings para especificar a linguagem utilizada para a inferência.
- `FacebookAI/xlm-mlm-ende-1024` (Masked language modeling, English-German)
- `FacebookAI/xlm-mlm-enfr-1024` (Masked language modeling, English-French)
- `FacebookAI/xlm-mlm-enro-1024` (Masked language modeling, English-Romanian)
- `FacebookAI/xlm-mlm-xnli15-1024` (Masked language modeling, XNLI languages)
- `FacebookAI/xlm-mlm-tlm-xnli15-1024` (Masked language modeling + translation, XNLI languages)
- `FacebookAI/xlm-clm-enfr-1024` (Causal language modeling, English-French)
- `FacebookAI/xlm-clm-ende-1024` (Causal language modeling, English-German)
Os language embeddings são representados por um tensor de mesma dimensão que os `input_ids` passados ao modelo.
Os valores destes tensores dependem do idioma utilizado e se identificam pelos atributos `lang2id` e `id2lang` do tokenizador.
Neste exemplo, carregamos o checkpoint `FacebookAI/xlm-clm-enfr-1024`(Causal language modeling, English-French):
```py
>>> import torch
>>> from transformers import XLMTokenizer, XLMWithLMHeadModel
>>> tokenizer = XLMTokenizer.from_pretrained("FacebookAI/xlm-clm-enfr-1024")
>>> model = XLMWithLMHeadModel.from_pretrained("FacebookAI/xlm-clm-enfr-1024")
```
O atributo `lang2id` do tokenizador mostra os idiomas deste modelo e seus ids:
```py
>>> print(tokenizer.lang2id)
{'en': 0, 'fr': 1}
```
Em seguida, cria-se um input de exemplo:
```py
>>> input_ids = torch.tensor([tokenizer.encode("Wikipedia was used to")]) # batch size of 1
```
Estabelece-se o id do idioma, por exemplo `"en"`, e utiliza-se o mesmo para definir a language embedding.
A language embedding é um tensor preenchido com `0`, que é o id de idioma para o inglês.
Este tensor deve ser do mesmo tamanho que os `input_ids`.
```py
>>> language_id = tokenizer.lang2id["en"] # 0
>>> langs = torch.tensor([language_id] * input_ids.shape[1]) # torch.tensor([0, 0, 0, ..., 0])
>>> # We reshape it to be of size (batch_size, sequence_length)
>>> langs = langs.view(1, -1) # is now of shape [1, sequence_length] (we have a batch size of 1)
```
Agora você pode passar os `input_ids` e a language embedding ao modelo:
```py
>>> outputs = model(input_ids, langs=langs)
```
O script [run_generation.py](https://github.com/huggingface/transformers/tree/master/examples/pytorch/text-generation/run_generation.py) pode gerar um texto com language embeddings utilizando os checkpoints `xlm-clm`.
### XLM sem language embeddings
Os seguintes modelos XLM não requerem o uso de language embeddings durante a inferência:
- `FacebookAI/xlm-mlm-17-1280` (Modelagem de linguagem com máscara, 17 idiomas)
- `FacebookAI/xlm-mlm-100-1280` (Modelagem de linguagem com máscara, 100 idiomas)
Estes modelos são utilizados para representações genéricas de frase diferentemente dos checkpoints XLM anteriores.
## BERT
Os seguintes modelos do BERT podem ser utilizados para tarefas multilinguísticas:
- `google-bert/bert-base-multilingual-uncased` (Modelagem de linguagem com máscara + Previsão de frases, 102 idiomas)
- `google-bert/bert-base-multilingual-cased` (Modelagem de linguagem com máscara + Previsão de frases, 104 idiomas)
Estes modelos não requerem language embeddings durante a inferência. Devem identificar a linguagem a partir
do contexto e realizar a inferência em sequência.
## XLM-RoBERTa
Os seguintes modelos do XLM-RoBERTa podem ser utilizados para tarefas multilinguísticas:
- `FacebookAI/xlm-roberta-base` (Modelagem de linguagem com máscara, 100 idiomas)
- `FacebookAI/xlm-roberta-large` Modelagem de linguagem com máscara, 100 idiomas)
O XLM-RoBERTa foi treinado com 2,5 TB de dados do CommonCrawl recém-criados e testados em 100 idiomas.
Proporciona fortes vantagens sobre os modelos multilinguísticos publicados anteriormente como o mBERT e o XLM em tarefas
subsequentes como a classificação, a rotulagem de sequências e à respostas a perguntas.
## M2M100
Os seguintes modelos de M2M100 podem ser utilizados para traduções multilinguísticas:
- `facebook/m2m100_418M` (Tradução)
- `facebook/m2m100_1.2B` (Tradução)
Neste exemplo, o checkpoint `facebook/m2m100_418M` é carregado para traduzir do mandarim ao inglês. É possível
estabelecer o idioma de origem no tokenizador:
```py
>>> from transformers import M2M100ForConditionalGeneration, M2M100Tokenizer
>>> en_text = "Do not meddle in the affairs of wizards, for they are subtle and quick to anger."
>>> chinese_text = "不要插手巫師的事務, 因為他們是微妙的, 很快就會發怒."
>>> tokenizer = M2M100Tokenizer.from_pretrained("facebook/m2m100_418M", src_lang="zh")
>>> model = M2M100ForConditionalGeneration.from_pretrained("facebook/m2m100_418M")
```
Tokenização do texto:
```py
>>> encoded_zh = tokenizer(chinese_text, return_tensors="pt")
```
O M2M100 força o id do idioma de destino como o primeiro token gerado para traduzir ao idioma de destino.
É definido o `forced_bos_token_id` como `en` no método `generate` para traduzir ao inglês.
```py
>>> generated_tokens = model.generate(**encoded_zh, forced_bos_token_id=tokenizer.get_lang_id("en"))
>>> tokenizer.batch_decode(generated_tokens, skip_special_tokens=True)
'Do not interfere with the matters of the witches, because they are delicate and will soon be angry.'
```
## MBart
Os seguintes modelos do MBart podem ser utilizados para tradução multilinguística:
- `facebook/mbart-large-50-one-to-many-mmt` (Tradução automática multilinguística de um a vários, 50 idiomas)
- `facebook/mbart-large-50-many-to-many-mmt` (Tradução automática multilinguística de vários a vários, 50 idiomas)
- `facebook/mbart-large-50-many-to-one-mmt` (Tradução automática multilinguística vários a um, 50 idiomas)
- `facebook/mbart-large-50` (Tradução multilinguística, 50 idiomas)
- `facebook/mbart-large-cc25`
Neste exemplo, carrega-se o checkpoint `facebook/mbart-large-50-many-to-many-mmt` para traduzir do finlandês ao inglês.
Pode-se definir o idioma de origem no tokenizador:
```py
>>> from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
>>> en_text = "Do not meddle in the affairs of wizards, for they are subtle and quick to anger."
>>> fi_text = "Älä sekaannu velhojen asioihin, sillä ne ovat hienovaraisia ja nopeasti vihaisia."
>>> tokenizer = AutoTokenizer.from_pretrained("facebook/mbart-large-50-many-to-many-mmt", src_lang="fi_FI")
>>> model = AutoModelForSeq2SeqLM.from_pretrained("facebook/mbart-large-50-many-to-many-mmt")
```
Tokenizando o texto:
```py
>>> encoded_en = tokenizer(en_text, return_tensors="pt")
```
O MBart força o id do idioma de destino como o primeiro token gerado para traduzir ao idioma de destino.
É definido o `forced_bos_token_id` como `en` no método `generate` para traduzir ao inglês.
```py
>>> generated_tokens = model.generate(**encoded_en, forced_bos_token_id=tokenizer.lang_code_to_id("en_XX"))
>>> tokenizer.batch_decode(generated_tokens, skip_special_tokens=True)
"Don't interfere with the wizard's affairs, because they are subtle, will soon get angry."
```
Se estiver usando o checkpoint `facebook/mbart-large-50-many-to-one-mmt` não será necessário forçar o id do idioma de destino
como sendo o primeiro token generado, caso contrário a usagem é a mesma.
| 0
|
mavonic_private_repos/transformers/docs/source
|
mavonic_private_repos/transformers/docs/source/pt/custom_models.md
|
<!--Copyright 2020 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
specific language governing permissions and limitations under the License.
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer.
-->
# Compartilhando modelos customizados
A biblioteca 🤗 Transformers foi projetada para ser facilmente extensível. Cada modelo é totalmente codificado em uma determinada subpasta
do repositório sem abstração, para que você possa copiar facilmente um arquivo de modelagem e ajustá-lo às suas necessidades.
Se você estiver escrevendo um modelo totalmente novo, pode ser mais fácil começar do zero. Neste tutorial, mostraremos
como escrever um modelo customizado e sua configuração para que possa ser usado com Transformers, e como você pode compartilhá-lo
com a comunidade (com o código em que se baseia) para que qualquer pessoa possa usá-lo, mesmo se não estiver presente na biblioteca 🤗 Transformers.
Ilustraremos tudo isso em um modelo ResNet, envolvendo a classe ResNet do
[biblioteca timm](https://github.com/rwightman/pytorch-image-models) em um [`PreTrainedModel`].
## Escrevendo uma configuração customizada
Antes de mergulharmos no modelo, vamos primeiro escrever sua configuração. A configuração de um modelo é um objeto que
terá todas as informações necessárias para construir o modelo. Como veremos na próxima seção, o modelo só pode
ter um `config` para ser inicializado, então realmente precisamos que esse objeto seja o mais completo possível.
Em nosso exemplo, pegaremos alguns argumentos da classe ResNet que podemos querer ajustar. Diferentes
configurações nos dará os diferentes tipos de ResNets que são possíveis. Em seguida, apenas armazenamos esses argumentos,
após verificar a validade de alguns deles.
```python
from transformers import PretrainedConfig
from typing import List
class ResnetConfig(PretrainedConfig):
model_type = "resnet"
def __init__(
self,
block_type="bottleneck",
layers: List[int] = [3, 4, 6, 3],
num_classes: int = 1000,
input_channels: int = 3,
cardinality: int = 1,
base_width: int = 64,
stem_width: int = 64,
stem_type: str = "",
avg_down: bool = False,
**kwargs,
):
if block_type not in ["basic", "bottleneck"]:
raise ValueError(f"`block_type` must be 'basic' or bottleneck', got {block_type}.")
if stem_type not in ["", "deep", "deep-tiered"]:
raise ValueError(f"`stem_type` must be '', 'deep' or 'deep-tiered', got {stem_type}.")
self.block_type = block_type
self.layers = layers
self.num_classes = num_classes
self.input_channels = input_channels
self.cardinality = cardinality
self.base_width = base_width
self.stem_width = stem_width
self.stem_type = stem_type
self.avg_down = avg_down
super().__init__(**kwargs)
```
As três coisas importantes a serem lembradas ao escrever sua própria configuração são:
- você tem que herdar de `PretrainedConfig`,
- o `__init__` do seu `PretrainedConfig` deve aceitar quaisquer kwargs,
- esses `kwargs` precisam ser passados para a superclasse `__init__`.
A herança é para garantir que você obtenha todas as funcionalidades da biblioteca 🤗 Transformers, enquanto as outras duas
restrições vêm do fato de um `PretrainedConfig` ter mais campos do que os que você está configurando. Ao recarregar um
config com o método `from_pretrained`, esses campos precisam ser aceitos pelo seu config e então enviados para a
superclasse.
Definir um `model_type` para sua configuração (aqui `model_type="resnet"`) não é obrigatório, a menos que você queira
registrar seu modelo com as classes automáticas (veja a última seção).
Com isso feito, você pode facilmente criar e salvar sua configuração como faria com qualquer outra configuração de modelo da
biblioteca. Aqui está como podemos criar uma configuração resnet50d e salvá-la:
```py
resnet50d_config = ResnetConfig(block_type="bottleneck", stem_width=32, stem_type="deep", avg_down=True)
resnet50d_config.save_pretrained("custom-resnet")
```
Isso salvará um arquivo chamado `config.json` dentro da pasta `custom-resnet`. Você pode então recarregar sua configuração com o
método `from_pretrained`:
```py
resnet50d_config = ResnetConfig.from_pretrained("custom-resnet")
```
Você também pode usar qualquer outro método da classe [`PretrainedConfig`], como [`~PretrainedConfig.push_to_hub`] para
carregar diretamente sua configuração para o Hub.
## Escrevendo um modelo customizado
Agora que temos nossa configuração ResNet, podemos continuar escrevendo o modelo. Na verdade, escreveremos dois: um que
extrai os recursos ocultos de um lote de imagens (como [`BertModel`]) e um que é adequado para classificação de imagem
(como [`BertForSequenceClassification`]).
Como mencionamos antes, escreveremos apenas um wrapper solto do modelo para mantê-lo simples para este exemplo. A única
coisa que precisamos fazer antes de escrever esta classe é um mapa entre os tipos de bloco e as classes de bloco reais. Então o
modelo é definido a partir da configuração passando tudo para a classe `ResNet`:
```py
from transformers import PreTrainedModel
from timm.models.resnet import BasicBlock, Bottleneck, ResNet
from .configuration_resnet import ResnetConfig
BLOCK_MAPPING = {"basic": BasicBlock, "bottleneck": Bottleneck}
class ResnetModel(PreTrainedModel):
config_class = ResnetConfig
def __init__(self, config):
super().__init__(config)
block_layer = BLOCK_MAPPING[config.block_type]
self.model = ResNet(
block_layer,
config.layers,
num_classes=config.num_classes,
in_chans=config.input_channels,
cardinality=config.cardinality,
base_width=config.base_width,
stem_width=config.stem_width,
stem_type=config.stem_type,
avg_down=config.avg_down,
)
def forward(self, tensor):
return self.model.forward_features(tensor)
```
Para o modelo que irá classificar as imagens, vamos apenas alterar o método forward:
```py
import torch
class ResnetModelForImageClassification(PreTrainedModel):
config_class = ResnetConfig
def __init__(self, config):
super().__init__(config)
block_layer = BLOCK_MAPPING[config.block_type]
self.model = ResNet(
block_layer,
config.layers,
num_classes=config.num_classes,
in_chans=config.input_channels,
cardinality=config.cardinality,
base_width=config.base_width,
stem_width=config.stem_width,
stem_type=config.stem_type,
avg_down=config.avg_down,
)
def forward(self, tensor, labels=None):
logits = self.model(tensor)
if labels is not None:
loss = torch.nn.cross_entropy(logits, labels)
return {"loss": loss, "logits": logits}
return {"logits": logits}
```
Em ambos os casos, observe como herdamos de `PreTrainedModel` e chamamos a inicialização da superclasse com o `config`
(um pouco parecido quando você escreve um `torch.nn.Module`). A linha que define o `config_class` não é obrigatória, a menos que
você deseje registrar seu modelo com as classes automáticas (consulte a última seção).
<Tip>
Se o seu modelo for muito semelhante a um modelo dentro da biblioteca, você poderá reutilizar a mesma configuração desse modelo.
</Tip>
Você pode fazer com que seu modelo retorne o que você quiser,porém retornando um dicionário como fizemos para
`ResnetModelForImageClassification`, com a função de perda incluída quando os rótulos são passados, vai tornar seu modelo diretamente
utilizável dentro da classe [`Trainer`]. Você pode usar outro formato de saída, desde que esteja planejando usar seu próprio
laço de treinamento ou outra biblioteca para treinamento.
Agora que temos nossa classe do modelo, vamos criar uma:
```py
resnet50d = ResnetModelForImageClassification(resnet50d_config)
```
Novamente, você pode usar qualquer um dos métodos do [`PreTrainedModel`], como [`~PreTrainedModel.save_pretrained`] ou
[`~PreTrainedModel.push_to_hub`]. Usaremos o segundo na próxima seção e veremos como enviar os pesos e
o código do nosso modelo. Mas primeiro, vamos carregar alguns pesos pré-treinados dentro do nosso modelo.
Em seu próprio caso de uso, você provavelmente estará treinando seu modelo customizado em seus próprios dados. Para este tutorial ser rápido,
usaremos a versão pré-treinada do resnet50d. Como nosso modelo é apenas um wrapper em torno dele, será
fácil de transferir esses pesos:
```py
import timm
pretrained_model = timm.create_model("resnet50d", pretrained=True)
resnet50d.model.load_state_dict(pretrained_model.state_dict())
```
Agora vamos ver como ter certeza de que quando fazemos [`~PreTrainedModel.save_pretrained`] ou [`~PreTrainedModel.push_to_hub`], o
código do modelo é salvo.
## Enviando o código para o Hub
<Tip warning={true}>
Esta API é experimental e pode ter algumas pequenas alterações nas próximas versões.
</Tip>
Primeiro, certifique-se de que seu modelo esteja totalmente definido em um arquivo `.py`. Ele pode contar com importações relativas para alguns outros arquivos
desde que todos os arquivos estejam no mesmo diretório (ainda não suportamos submódulos para este recurso). Para o nosso exemplo,
vamos definir um arquivo `modeling_resnet.py` e um arquivo `configuration_resnet.py` em uma pasta no
diretório de trabalho atual chamado `resnet_model`. O arquivo de configuração contém o código para `ResnetConfig` e o arquivo de modelagem
contém o código do `ResnetModel` e `ResnetModelForImageClassification`.
```
.
└── resnet_model
├── __init__.py
├── configuration_resnet.py
└── modeling_resnet.py
```
O `__init__.py` pode estar vazio, apenas está lá para que o Python detecte que o `resnet_model` possa ser usado como um módulo.
<Tip warning={true}>
Se estiver copiando arquivos de modelagem da biblioteca, você precisará substituir todas as importações relativas na parte superior do arquivo
para importar do pacote `transformers`.
</Tip>
Observe que você pode reutilizar (ou subclasse) uma configuração/modelo existente.
Para compartilhar seu modelo com a comunidade, siga estas etapas: primeiro importe o modelo ResNet e a configuração do
arquivos criados:
```py
from resnet_model.configuration_resnet import ResnetConfig
from resnet_model.modeling_resnet import ResnetModel, ResnetModelForImageClassification
```
Então você tem que dizer à biblioteca que deseja copiar os arquivos de código desses objetos ao usar o `save_pretrained`
e registrá-los corretamente com uma determinada classe automáticas (especialmente para modelos), basta executar:
```py
ResnetConfig.register_for_auto_class()
ResnetModel.register_for_auto_class("AutoModel")
ResnetModelForImageClassification.register_for_auto_class("AutoModelForImageClassification")
```
Observe que não há necessidade de especificar uma classe automática para a configuração (há apenas uma classe automática,
[`AutoConfig`]), mas é diferente para os modelos. Seu modelo customizado pode ser adequado para muitas tarefas diferentes, então você
tem que especificar qual das classes automáticas é a correta para o seu modelo.
Em seguida, vamos criar a configuração e os modelos como fizemos antes:
```py
resnet50d_config = ResnetConfig(block_type="bottleneck", stem_width=32, stem_type="deep", avg_down=True)
resnet50d = ResnetModelForImageClassification(resnet50d_config)
pretrained_model = timm.create_model("resnet50d", pretrained=True)
resnet50d.model.load_state_dict(pretrained_model.state_dict())
```
Agora para enviar o modelo para o Hub, certifique-se de estar logado. Ou execute no seu terminal:
```bash
huggingface-cli login
```
ou a partir do notebook:
```py
from huggingface_hub import notebook_login
notebook_login()
```
Você pode então enviar para seu próprio namespace (ou uma organização da qual você é membro) assim:
```py
resnet50d.push_to_hub("custom-resnet50d")
```
Além dos pesos do modelo e da configuração no formato json, isso também copiou o modelo e
configuração `.py` na pasta `custom-resnet50d` e carregou o resultado para o Hub. Você pode conferir o resultado
neste [repositório de modelos](https://huggingface.co/sgugger/custom-resnet50d).
Consulte o [tutorial de compartilhamento](model_sharing) para obter mais informações sobre o método push_to_hub.
## Usando um modelo com código customizado
Você pode usar qualquer configuração, modelo ou tokenizador com arquivos de código customizados em seu repositório com as classes automáticas e
o método `from_pretrained`. Todos os arquivos e códigos carregados no Hub são verificados quanto a malware (consulte a documentação de [Segurança do Hub](https://huggingface.co/docs/hub/security#malware-scanning) para obter mais informações), mas você ainda deve
revisar o código do modelo e o autor para evitar a execução de código malicioso em sua máquina. Defina `trust_remote_code=True` para usar
um modelo com código customizado:
```py
from transformers import AutoModelForImageClassification
model = AutoModelForImageClassification.from_pretrained("sgugger/custom-resnet50d", trust_remote_code=True)
```
Também é fortemente recomendado passar um hash de confirmação como uma `revisão` para garantir que o autor dos modelos não
atualize o código com novas linhas maliciosas (a menos que você confie totalmente nos autores dos modelos).
```py
commit_hash = "ed94a7c6247d8aedce4647f00f20de6875b5b292"
model = AutoModelForImageClassification.from_pretrained(
"sgugger/custom-resnet50d", trust_remote_code=True, revision=commit_hash
)
```
Observe que ao navegar no histórico de commits do repositório do modelo no Hub, há um botão para copiar facilmente o commit
hash de qualquer commit.
## Registrando um modelo com código customizado para as classes automáticas
Se você estiver escrevendo uma biblioteca que estende 🤗 Transformers, talvez queira estender as classes automáticas para incluir seus próprios
modelos. Isso é diferente de enviar o código para o Hub no sentido de que os usuários precisarão importar sua biblioteca para
obter os modelos customizados (ao contrário de baixar automaticamente o código do modelo do Hub).
Desde que sua configuração tenha um atributo `model_type` diferente dos tipos de modelo existentes e que as classes do seu modelo
tenha os atributos `config_class` corretos, você pode simplesmente adicioná-los às classes automáticas assim:
```py
from transformers import AutoConfig, AutoModel, AutoModelForImageClassification
AutoConfig.register("resnet", ResnetConfig)
AutoModel.register(ResnetConfig, ResnetModel)
AutoModelForImageClassification.register(ResnetConfig, ResnetModelForImageClassification)
```
Observe que o primeiro argumento usado ao registrar sua configuração customizada para [`AutoConfig`] precisa corresponder ao `model_type`
de sua configuração customizada. E o primeiro argumento usado ao registrar seus modelos customizados, para qualquer necessidade de classe de modelo automático
deve corresponder ao `config_class` desses modelos.
| 0
|
mavonic_private_repos/transformers/docs/source
|
mavonic_private_repos/transformers/docs/source/pt/pipeline_tutorial.md
|
<!--Copyright 2022 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
specific language governing permissions and limitations under the License.
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer.
-->
# Pipelines para inferência
Um [pipeline] simplifica o uso dos modelos no [Model Hub](https://huggingface.co/models) para a inferência de uma diversidade de tarefas,
como a geração de texto, a segmentação de imagens e a classificação de áudio.
Inclusive, se não tem experiência com alguma modalidade específica ou não compreende o código que forma os modelos,
pode usar eles mesmo assim com o [pipeline]! Este tutorial te ensinará a:
* Utilizar um [`pipeline`] para inferência.
* Utilizar um tokenizador ou model específico.
* Utilizar um [`pipeline`] para tarefas de áudio e visão computacional.
<Tip>
Acesse a documentação do [`pipeline`] para obter uma lista completa de tarefas possíveis.
</Tip>
## Uso do pipeline
Mesmo que cada tarefa tenha um [`pipeline`] associado, é mais simples usar a abstração geral do [`pipeline`] que
contém todos os pipelines das tarefas mais específicas.
O [`pipeline`] carrega automaticamenta um modelo predeterminado e um tokenizador com capacidade de inferência para sua
tarefa.
1. Comece carregando um [`pipeline`] e especifique uma tarefa de inferência:
```py
>>> from transformers import pipeline
>>> generator = pipeline(task="text-generation")
```
2. Passe seu dado de entrada, no caso um texto, ao [`pipeline`]:
```py
>>> generator("Three Rings for the Elven-kings under the sky, Seven for the Dwarf-lords in their halls of stone")
[{'generated_text': 'Three Rings for the Elven-kings under the sky, Seven for the Dwarf-lords in their halls of stone, Seven for the Iron-priests at the door to the east, and thirteen for the Lord Kings at the end of the mountain'}]
```
Se tiver mais de uma entrada, passe-a como uma lista:
```py
>>> generator(
... [
... "Three Rings for the Elven-kings under the sky, Seven for the Dwarf-lords in their halls of stone",
... "Nine for Mortal Men, doomed to die, One for the Dark Lord on his dark throne",
... ]
... )
```
Qualquer parâmetro adicional para a sua tarefa também pode ser incluído no [`pipeline`]. A tarefa `text-generation` tem um método
[`~generation.GenerationMixin.generate`] com vários parâmetros para controlar a saída.
Por exemplo, se quiser gerar mais de uma saída, defina-a no parâmetro `num_return_sequences`:
```py
>>> generator(
... "Three Rings for the Elven-kings under the sky, Seven for the Dwarf-lords in their halls of stone",
... num_return_sequences=2,
... )
```
### Selecionando um modelo e um tokenizador
O [`pipeline`] aceita qualquer modelo do [Model Hub](https://huggingface.co/models). Há rótulos adicionais no Model Hub
que te permitem filtrar pelo modelo que gostaria de usar para sua tarefa. Uma vez que tiver escolhido o modelo apropriado,
carregue-o com as classes `AutoModelFor` e [`AutoTokenizer`] correspondentes. Por exemplo, carregue a classe [`AutoModelForCausalLM`]
para uma tarefa de modelagem de linguagem causal:
```py
>>> from transformers import AutoTokenizer, AutoModelForCausalLM
>>> tokenizer = AutoTokenizer.from_pretrained("distilbert/distilgpt2")
>>> model = AutoModelForCausalLM.from_pretrained("distilbert/distilgpt2")
```
Crie uma [`pipeline`] para a sua tarefa e especifíque o modelo e o tokenizador que foram carregados:
```py
>>> from transformers import pipeline
>>> generator = pipeline(task="text-generation", model=model, tokenizer=tokenizer)
```
Passe seu texto de entrada ao [`pipeline`] para gerar algum texto:
```py
>>> generator("Three Rings for the Elven-kings under the sky, Seven for the Dwarf-lords in their halls of stone")
[{'generated_text': 'Three Rings for the Elven-kings under the sky, Seven for the Dwarf-lords in their halls of stone, Seven for the Dragon-lords (for them to rule in a world ruled by their rulers, and all who live within the realm'}]
```
## Pipeline de audio
A flexibilidade do [`pipeline`] significa que também pode-se extender às tarefas de áudio.
La flexibilidad de [`pipeline`] significa que también se puede extender a tareas de audio.
Por exemplo, classifiquemos a emoção de um breve fragmento do famoso discurso de John F. Kennedy /home/rzimmerdev/dev/transformers/docs/source/pt/pipeline_tutorial.md
Encontre um modelo de [audio classification](https://huggingface.co/models?pipeline_tag=audio-classification) para
reconhecimento de emoções no Model Hub e carregue-o usando o [`pipeline`]:
```py
>>> from transformers import pipeline
>>> audio_classifier = pipeline(
... task="audio-classification", model="ehcalabres/wav2vec2-lg-xlsr-en-speech-emotion-recognition"
... )
```
Passe o arquivo de áudio ao [`pipeline`]:
```py
>>> audio_classifier("jfk_moon_speech.wav")
[{'label': 'calm', 'score': 0.13856211304664612},
{'label': 'disgust', 'score': 0.13148026168346405},
{'label': 'happy', 'score': 0.12635163962841034},
{'label': 'angry', 'score': 0.12439591437578201},
{'label': 'fearful', 'score': 0.12404385954141617}]
```
## Pipeline de visão computacional
Finalmente, utilizar um [`pipeline`] para tarefas de visão é praticamente a mesma coisa.
Especifique a sua tarefa de visão e passe a sua imagem ao classificador.
A imagem pode ser um link ou uma rota local à imagem. Por exemplo, que espécie de gato está presente na imagem?
```py
>>> from transformers import pipeline
>>> vision_classifier = pipeline(task="image-classification")
>>> vision_classifier(
... images="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/pipeline-cat-chonk.jpeg"
... )
[{'label': 'lynx, catamount', 'score': 0.4403027892112732},
{'label': 'cougar, puma, catamount, mountain lion, painter, panther, Felis concolor',
'score': 0.03433405980467796},
{'label': 'snow leopard, ounce, Panthera uncia',
'score': 0.032148055732250214},
{'label': 'Egyptian cat', 'score': 0.02353910356760025},
{'label': 'tiger cat', 'score': 0.023034192621707916}]
```
| 0
|
mavonic_private_repos/transformers/docs/source
|
mavonic_private_repos/transformers/docs/source/pt/_toctree.yml
|
- sections:
- local: index
title: 🤗 Transformers
- local: quicktour
title: Tour rápido
- local: installation
title: Instalação
title: Início
- sections:
- local: pipeline_tutorial
title: Pipelines para inferência
- local: training
title: Fine-tuning de um modelo pré-treinado
- local: accelerate
title: Treinamento distribuído com 🤗 Accelerate
title: Tutoriais
- sections:
- local: fast_tokenizers
title: Usando os Tokenizers do 🤗 Tokenizers
- local: create_a_model
title: Criando uma arquitetura customizada
- local: custom_models
title: Compartilhando modelos customizados
- local: run_scripts
title: Treinamento a partir de um script
- local: converting_tensorflow_models
title: Convertendo checkpoints do TensorFlow para Pytorch
- local: serialization
title: Exportando modelos para ONNX
- sections:
- local: tasks/sequence_classification
title: Classificação de texto
- local: tasks/token_classification
title: Classificação de tokens
title: Fine-tuning para tarefas específicas
- local: multilingual
title: Modelos multilinguísticos para inferência
title: Guias práticos
| 0
|
mavonic_private_repos/transformers/docs/source
|
mavonic_private_repos/transformers/docs/source/pt/index.md
|
<!--Copyright 2020 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
specific language governing permissions and limitations under the License.
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer.
-->
# 🤗 Transformers
Estado da Arte para Aprendizado de Máquina em PyTorch, TensorFlow e JAX.
O 🤗 Transformers disponibiliza APIs para facilmente baixar e treinar modelos pré-treinados de última geração.
O uso de modelos pré-treinados pode diminuir os seus custos de computação, a sua pegada de carbono, além de economizar o
tempo necessário para se treinar um modelo do zero. Os modelos podem ser usados para diversas tarefas:
* 📝 Textos: classificação, extração de informações, perguntas e respostas, resumir, traduzir e gerar textos em mais de 100 idiomas.
* 🖼 Imagens: classificação, deteção de objetos, e segmentação.
* 🗣 Audio: reconhecimento de fala e classificação de áudio.
* 🐙 Multimodal: perguntas tabeladas e respsostas, reconhecimento ótico de charactéres, extração de informação de
documentos escaneados, classificação de vídeo, perguntas e respostas visuais.
Nossa biblioteca aceita integração contínua entre três das bibliotecas mais populares de aprendizado profundo:
Our library supports seamless integration between three of the most popular deep learning libraries:
[PyTorch](https://pytorch.org/), [TensorFlow](https://www.tensorflow.org/) e [JAX](https://jax.readthedocs.io/en/latest/).
Treine seu modelo em três linhas de código em um framework, e carregue-o para execução em outro.
Cada arquitetura 🤗 Transformers é definida em um módulo individual do Python, para que seja facilmente customizável para pesquisa e experimentos.
## Se você estiver procurando suporte do time da Hugging Face, acesse
<a target="_blank" href="https://huggingface.co/support">
<img alt="HuggingFace Expert Acceleration Program" src="https://huggingface.co/front/thumbnails/support.png" style="width: 100%; max-width: 600px; border: 1px solid #eee; border-radius: 4px; box-shadow: 0 1px 2px 0 rgba(0, 0, 0, 0.05);">
</a>
## Conteúdo
A documentação é dividida em cinco partes:
- **INÍCIO** contém um tour rápido de instalação e instruções para te dar um empurrão inicial com os 🤗 Transformers.
- **TUTORIAIS** são perfeitos para começar a aprender sobre a nossa biblioteca. Essa seção irá te ajudar a desenvolver
habilidades básicas necessárias para usar o 🤗 Transformers.
- **GUIAS PRÁTICOS** irão te mostrar como alcançar um certo objetivo, como o fine-tuning de um modelo pré-treinado
para modelamento de idioma, ou como criar um cabeçalho personalizado para um modelo.
- **GUIAS CONCEITUAIS** te darão mais discussões e explicações dos conceitos fundamentais e idéias por trás dos modelos,
tarefas e da filosofia de design por trás do 🤗 Transformers.
- **API** descreve o funcionamento de cada classe e função, agrupada em:
- **CLASSES PRINCIPAIS** para as classes que expõe as APIs importantes da biblioteca.
- **MODELOS** para as classes e funções relacionadas à cada modelo implementado na biblioteca.
- **AUXILIARES INTERNOS** para as classes e funções usadas internamente.
Atualmente a biblioteca contém implementações do PyTorch, TensorFlow e JAX, pesos para modelos pré-treinados e scripts de uso e conversão de utilidades para os seguintes modelos:
### Modelos atuais
<!--This list is updated automatically from the README with _make fix-copies_. Do not update manually! -->
1. **[ALBERT](model_doc/albert)** (from Google Research and the Toyota Technological Institute at Chicago) released with the paper [ALBERT: A Lite BERT for Self-supervised Learning of Language Representations](https://arxiv.org/abs/1909.11942), by Zhenzhong Lan, Mingda Chen, Sebastian Goodman, Kevin Gimpel, Piyush Sharma, Radu Soricut.
1. **[BART](model_doc/bart)** (from Facebook) released with the paper [BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension](https://arxiv.org/abs/1910.13461) by Mike Lewis, Yinhan Liu, Naman Goyal, Marjan Ghazvininejad, Abdelrahman Mohamed, Omer Levy, Ves Stoyanov and Luke Zettlemoyer.
1. **[BARThez](model_doc/barthez)** (from École polytechnique) released with the paper [BARThez: a Skilled Pretrained French Sequence-to-Sequence Model](https://arxiv.org/abs/2010.12321) by Moussa Kamal Eddine, Antoine J.-P. Tixier, Michalis Vazirgiannis.
1. **[BARTpho](model_doc/bartpho)** (from VinAI Research) released with the paper [BARTpho: Pre-trained Sequence-to-Sequence Models for Vietnamese](https://arxiv.org/abs/2109.09701) by Nguyen Luong Tran, Duong Minh Le and Dat Quoc Nguyen.
1. **[BEiT](model_doc/beit)** (from Microsoft) released with the paper [BEiT: BERT Pre-Training of Image Transformers](https://arxiv.org/abs/2106.08254) by Hangbo Bao, Li Dong, Furu Wei.
1. **[BERT](model_doc/bert)** (from Google) released with the paper [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://arxiv.org/abs/1810.04805) by Jacob Devlin, Ming-Wei Chang, Kenton Lee and Kristina Toutanova.
1. **[BERTweet](model_doc/bertweet)** (from VinAI Research) released with the paper [BERTweet: A pre-trained language model for English Tweets](https://aclanthology.org/2020.emnlp-demos.2/) by Dat Quoc Nguyen, Thanh Vu and Anh Tuan Nguyen.
1. **[BERT For Sequence Generation](model_doc/bert-generation)** (from Google) released with the paper [Leveraging Pre-trained Checkpoints for Sequence Generation Tasks](https://arxiv.org/abs/1907.12461) by Sascha Rothe, Shashi Narayan, Aliaksei Severyn.
1. **[BigBird-RoBERTa](model_doc/big_bird)** (from Google Research) released with the paper [Big Bird: Transformers for Longer Sequences](https://arxiv.org/abs/2007.14062) by Manzil Zaheer, Guru Guruganesh, Avinava Dubey, Joshua Ainslie, Chris Alberti, Santiago Ontanon, Philip Pham, Anirudh Ravula, Qifan Wang, Li Yang, Amr Ahmed.
1. **[BigBird-Pegasus](model_doc/bigbird_pegasus)** (from Google Research) released with the paper [Big Bird: Transformers for Longer Sequences](https://arxiv.org/abs/2007.14062) by Manzil Zaheer, Guru Guruganesh, Avinava Dubey, Joshua Ainslie, Chris Alberti, Santiago Ontanon, Philip Pham, Anirudh Ravula, Qifan Wang, Li Yang, Amr Ahmed.
1. **[Blenderbot](model_doc/blenderbot)** (from Facebook) released with the paper [Recipes for building an open-domain chatbot](https://arxiv.org/abs/2004.13637) by Stephen Roller, Emily Dinan, Naman Goyal, Da Ju, Mary Williamson, Yinhan Liu, Jing Xu, Myle Ott, Kurt Shuster, Eric M. Smith, Y-Lan Boureau, Jason Weston.
1. **[BlenderbotSmall](model_doc/blenderbot-small)** (from Facebook) released with the paper [Recipes for building an open-domain chatbot](https://arxiv.org/abs/2004.13637) by Stephen Roller, Emily Dinan, Naman Goyal, Da Ju, Mary Williamson, Yinhan Liu, Jing Xu, Myle Ott, Kurt Shuster, Eric M. Smith, Y-Lan Boureau, Jason Weston.
1. **[BORT](model_doc/bort)** (from Alexa) released with the paper [Optimal Subarchitecture Extraction For BERT](https://arxiv.org/abs/2010.10499) by Adrian de Wynter and Daniel J. Perry.
1. **[ByT5](model_doc/byt5)** (from Google Research) released with the paper [ByT5: Towards a token-free future with pre-trained byte-to-byte models](https://arxiv.org/abs/2105.13626) by Linting Xue, Aditya Barua, Noah Constant, Rami Al-Rfou, Sharan Narang, Mihir Kale, Adam Roberts, Colin Raffel.
1. **[CamemBERT](model_doc/camembert)** (from Inria/Facebook/Sorbonne) released with the paper [CamemBERT: a Tasty French Language Model](https://arxiv.org/abs/1911.03894) by Louis Martin*, Benjamin Muller*, Pedro Javier Ortiz Suárez*, Yoann Dupont, Laurent Romary, Éric Villemonte de la Clergerie, Djamé Seddah and Benoît Sagot.
1. **[CANINE](model_doc/canine)** (from Google Research) released with the paper [CANINE: Pre-training an Efficient Tokenization-Free Encoder for Language Representation](https://arxiv.org/abs/2103.06874) by Jonathan H. Clark, Dan Garrette, Iulia Turc, John Wieting.
1. **[ConvNeXT](model_doc/convnext)** (from Facebook AI) released with the paper [A ConvNet for the 2020s](https://arxiv.org/abs/2201.03545) by Zhuang Liu, Hanzi Mao, Chao-Yuan Wu, Christoph Feichtenhofer, Trevor Darrell, Saining Xie.
1. **[ConvNeXTV2](model_doc/convnextv2)** (from Facebook AI) released with the paper [ConvNeXt V2: Co-designing and Scaling ConvNets with Masked Autoencoders](https://arxiv.org/abs/2301.00808) by Sanghyun Woo, Shoubhik Debnath, Ronghang Hu, Xinlei Chen, Zhuang Liu, In So Kweon, Saining Xie.
1. **[CLIP](model_doc/clip)** (from OpenAI) released with the paper [Learning Transferable Visual Models From Natural Language Supervision](https://arxiv.org/abs/2103.00020) by Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, Gretchen Krueger, Ilya Sutskever.
1. **[ConvBERT](model_doc/convbert)** (from YituTech) released with the paper [ConvBERT: Improving BERT with Span-based Dynamic Convolution](https://arxiv.org/abs/2008.02496) by Zihang Jiang, Weihao Yu, Daquan Zhou, Yunpeng Chen, Jiashi Feng, Shuicheng Yan.
1. **[CPM](model_doc/cpm)** (from Tsinghua University) released with the paper [CPM: A Large-scale Generative Chinese Pre-trained Language Model](https://arxiv.org/abs/2012.00413) by Zhengyan Zhang, Xu Han, Hao Zhou, Pei Ke, Yuxian Gu, Deming Ye, Yujia Qin, Yusheng Su, Haozhe Ji, Jian Guan, Fanchao Qi, Xiaozhi Wang, Yanan Zheng, Guoyang Zeng, Huanqi Cao, Shengqi Chen, Daixuan Li, Zhenbo Sun, Zhiyuan Liu, Minlie Huang, Wentao Han, Jie Tang, Juanzi Li, Xiaoyan Zhu, Maosong Sun.
1. **[CTRL](model_doc/ctrl)** (from Salesforce) released with the paper [CTRL: A Conditional Transformer Language Model for Controllable Generation](https://arxiv.org/abs/1909.05858) by Nitish Shirish Keskar*, Bryan McCann*, Lav R. Varshney, Caiming Xiong and Richard Socher.
1. **[Data2Vec](model_doc/data2vec)** (from Facebook) released with the paper [Data2Vec: A General Framework for Self-supervised Learning in Speech, Vision and Language](https://arxiv.org/abs/2202.03555) by Alexei Baevski, Wei-Ning Hsu, Qiantong Xu, Arun Babu, Jiatao Gu, Michael Auli.
1. **[DeBERTa](model_doc/deberta)** (from Microsoft) released with the paper [DeBERTa: Decoding-enhanced BERT with Disentangled Attention](https://arxiv.org/abs/2006.03654) by Pengcheng He, Xiaodong Liu, Jianfeng Gao, Weizhu Chen.
1. **[DeBERTa-v2](model_doc/deberta-v2)** (from Microsoft) released with the paper [DeBERTa: Decoding-enhanced BERT with Disentangled Attention](https://arxiv.org/abs/2006.03654) by Pengcheng He, Xiaodong Liu, Jianfeng Gao, Weizhu Chen.
1. **[Decision Transformer](model_doc/decision_transformer)** (from Berkeley/Facebook/Google) released with the paper [Decision Transformer: Reinforcement Learning via Sequence Modeling](https://arxiv.org/abs/2106.01345) by Lili Chen, Kevin Lu, Aravind Rajeswaran, Kimin Lee, Aditya Grover, Michael Laskin, Pieter Abbeel, Aravind Srinivas, Igor Mordatch.
1. **[DiT](model_doc/dit)** (from Microsoft Research) released with the paper [DiT: Self-supervised Pre-training for Document Image Transformer](https://arxiv.org/abs/2203.02378) by Junlong Li, Yiheng Xu, Tengchao Lv, Lei Cui, Cha Zhang, Furu Wei.
1. **[DeiT](model_doc/deit)** (from Facebook) released with the paper [Training data-efficient image transformers & distillation through attention](https://arxiv.org/abs/2012.12877) by Hugo Touvron, Matthieu Cord, Matthijs Douze, Francisco Massa, Alexandre Sablayrolles, Hervé Jégou.
1. **[DETR](model_doc/detr)** (from Facebook) released with the paper [End-to-End Object Detection with Transformers](https://arxiv.org/abs/2005.12872) by Nicolas Carion, Francisco Massa, Gabriel Synnaeve, Nicolas Usunier, Alexander Kirillov, Sergey Zagoruyko.
1. **[DialoGPT](model_doc/dialogpt)** (from Microsoft Research) released with the paper [DialoGPT: Large-Scale Generative Pre-training for Conversational Response Generation](https://arxiv.org/abs/1911.00536) by Yizhe Zhang, Siqi Sun, Michel Galley, Yen-Chun Chen, Chris Brockett, Xiang Gao, Jianfeng Gao, Jingjing Liu, Bill Dolan.
1. **[DistilBERT](model_doc/distilbert)** (from HuggingFace), released together with the paper [DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter](https://arxiv.org/abs/1910.01108) by Victor Sanh, Lysandre Debut and Thomas Wolf. The same method has been applied to compress GPT2 into [DistilGPT2](https://github.com/huggingface/transformers/tree/main/examples/research_projects/distillation), RoBERTa into [DistilRoBERTa](https://github.com/huggingface/transformers/tree/main/examples/research_projects/distillation), Multilingual BERT into [DistilmBERT](https://github.com/huggingface/transformers/tree/main/examples/research_projects/distillation) and a German version of DistilBERT.
1. **[DPR](model_doc/dpr)** (from Facebook) released with the paper [Dense Passage Retrieval for Open-Domain Question Answering](https://arxiv.org/abs/2004.04906) by Vladimir Karpukhin, Barlas Oğuz, Sewon Min, Patrick Lewis, Ledell Wu, Sergey Edunov, Danqi Chen, and Wen-tau Yih.
1. **[DPT](master/model_doc/dpt)** (from Intel Labs) released with the paper [Vision Transformers for Dense Prediction](https://arxiv.org/abs/2103.13413) by René Ranftl, Alexey Bochkovskiy, Vladlen Koltun.
1. **[EfficientNet](model_doc/efficientnet)** (from Google Research) released with the paper [EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks](https://arxiv.org/abs/1905.11946) by Mingxing Tan and Quoc V. Le.
1. **[EncoderDecoder](model_doc/encoder-decoder)** (from Google Research) released with the paper [Leveraging Pre-trained Checkpoints for Sequence Generation Tasks](https://arxiv.org/abs/1907.12461) by Sascha Rothe, Shashi Narayan, Aliaksei Severyn.
1. **[ELECTRA](model_doc/electra)** (from Google Research/Stanford University) released with the paper [ELECTRA: Pre-training text encoders as discriminators rather than generators](https://arxiv.org/abs/2003.10555) by Kevin Clark, Minh-Thang Luong, Quoc V. Le, Christopher D. Manning.
1. **[FlauBERT](model_doc/flaubert)** (from CNRS) released with the paper [FlauBERT: Unsupervised Language Model Pre-training for French](https://arxiv.org/abs/1912.05372) by Hang Le, Loïc Vial, Jibril Frej, Vincent Segonne, Maximin Coavoux, Benjamin Lecouteux, Alexandre Allauzen, Benoît Crabbé, Laurent Besacier, Didier Schwab.
1. **[FNet](model_doc/fnet)** (from Google Research) released with the paper [FNet: Mixing Tokens with Fourier Transforms](https://arxiv.org/abs/2105.03824) by James Lee-Thorp, Joshua Ainslie, Ilya Eckstein, Santiago Ontanon.
1. **[Funnel Transformer](model_doc/funnel)** (from CMU/Google Brain) released with the paper [Funnel-Transformer: Filtering out Sequential Redundancy for Efficient Language Processing](https://arxiv.org/abs/2006.03236) by Zihang Dai, Guokun Lai, Yiming Yang, Quoc V. Le.
1. **[GLPN](model_doc/glpn)** (from KAIST) released with the paper [Global-Local Path Networks for Monocular Depth Estimation with Vertical CutDepth](https://arxiv.org/abs/2201.07436) by Doyeon Kim, Woonghyun Ga, Pyungwhan Ahn, Donggyu Joo, Sehwan Chun, Junmo Kim.
1. **[GPT](model_doc/openai-gpt)** (from OpenAI) released with the paper [Improving Language Understanding by Generative Pre-Training](https://openai.com/research/language-unsupervised/) by Alec Radford, Karthik Narasimhan, Tim Salimans and Ilya Sutskever.
1. **[GPT-2](model_doc/gpt2)** (from OpenAI) released with the paper [Language Models are Unsupervised Multitask Learners](https://openai.com/research/better-language-models/) by Alec Radford, Jeffrey Wu, Rewon Child, David Luan, Dario Amodei and Ilya Sutskever.
1. **[GPT-J](model_doc/gptj)** (from EleutherAI) released in the repository [kingoflolz/mesh-transformer-jax](https://github.com/kingoflolz/mesh-transformer-jax/) by Ben Wang and Aran Komatsuzaki.
1. **[GPT Neo](model_doc/gpt_neo)** (from EleutherAI) released in the repository [EleutherAI/gpt-neo](https://github.com/EleutherAI/gpt-neo) by Sid Black, Stella Biderman, Leo Gao, Phil Wang and Connor Leahy.
1. **[GPTSAN-japanese](model_doc/gptsan-japanese)** released in the repository [tanreinama/GPTSAN](https://github.com/tanreinama/GPTSAN/blob/main/report/model.md) by Toshiyuki Sakamoto(tanreinama).
1. **[Hubert](model_doc/hubert)** (from Facebook) released with the paper [HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Units](https://arxiv.org/abs/2106.07447) by Wei-Ning Hsu, Benjamin Bolte, Yao-Hung Hubert Tsai, Kushal Lakhotia, Ruslan Salakhutdinov, Abdelrahman Mohamed.
1. **[I-BERT](model_doc/ibert)** (from Berkeley) released with the paper [I-BERT: Integer-only BERT Quantization](https://arxiv.org/abs/2101.01321) by Sehoon Kim, Amir Gholami, Zhewei Yao, Michael W. Mahoney, Kurt Keutzer.
1. **[ImageGPT](model_doc/imagegpt)** (from OpenAI) released with the paper [Generative Pretraining from Pixels](https://openai.com/blog/image-gpt/) by Mark Chen, Alec Radford, Rewon Child, Jeffrey Wu, Heewoo Jun, David Luan, Ilya Sutskever.
1. **[LayoutLM](model_doc/layoutlm)** (from Microsoft Research Asia) released with the paper [LayoutLM: Pre-training of Text and Layout for Document Image Understanding](https://arxiv.org/abs/1912.13318) by Yiheng Xu, Minghao Li, Lei Cui, Shaohan Huang, Furu Wei, Ming Zhou.
1. **[LayoutLMv2](model_doc/layoutlmv2)** (from Microsoft Research Asia) released with the paper [LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding](https://arxiv.org/abs/2012.14740) by Yang Xu, Yiheng Xu, Tengchao Lv, Lei Cui, Furu Wei, Guoxin Wang, Yijuan Lu, Dinei Florencio, Cha Zhang, Wanxiang Che, Min Zhang, Lidong Zhou.
1. **[LayoutXLM](model_doc/layoutxlm)** (from Microsoft Research Asia) released with the paper [LayoutXLM: Multimodal Pre-training for Multilingual Visually-rich Document Understanding](https://arxiv.org/abs/2104.08836) by Yiheng Xu, Tengchao Lv, Lei Cui, Guoxin Wang, Yijuan Lu, Dinei Florencio, Cha Zhang, Furu Wei.
1. **[LED](model_doc/led)** (from AllenAI) released with the paper [Longformer: The Long-Document Transformer](https://arxiv.org/abs/2004.05150) by Iz Beltagy, Matthew E. Peters, Arman Cohan.
1. **[Longformer](model_doc/longformer)** (from AllenAI) released with the paper [Longformer: The Long-Document Transformer](https://arxiv.org/abs/2004.05150) by Iz Beltagy, Matthew E. Peters, Arman Cohan.
1. **[LUKE](model_doc/luke)** (from Studio Ousia) released with the paper [LUKE: Deep Contextualized Entity Representations with Entity-aware Self-attention](https://arxiv.org/abs/2010.01057) by Ikuya Yamada, Akari Asai, Hiroyuki Shindo, Hideaki Takeda, Yuji Matsumoto.
1. **[mLUKE](model_doc/mluke)** (from Studio Ousia) released with the paper [mLUKE: The Power of Entity Representations in Multilingual Pretrained Language Models](https://arxiv.org/abs/2110.08151) by Ryokan Ri, Ikuya Yamada, and Yoshimasa Tsuruoka.
1. **[LXMERT](model_doc/lxmert)** (from UNC Chapel Hill) released with the paper [LXMERT: Learning Cross-Modality Encoder Representations from Transformers for Open-Domain Question Answering](https://arxiv.org/abs/1908.07490) by Hao Tan and Mohit Bansal.
1. **[M2M100](model_doc/m2m_100)** (from Facebook) released with the paper [Beyond English-Centric Multilingual Machine Translation](https://arxiv.org/abs/2010.11125) by Angela Fan, Shruti Bhosale, Holger Schwenk, Zhiyi Ma, Ahmed El-Kishky, Siddharth Goyal, Mandeep Baines, Onur Celebi, Guillaume Wenzek, Vishrav Chaudhary, Naman Goyal, Tom Birch, Vitaliy Liptchinsky, Sergey Edunov, Edouard Grave, Michael Auli, Armand Joulin.
1. **[MarianMT](model_doc/marian)** Machine translation models trained using [OPUS](http://opus.nlpl.eu/) data by Jörg Tiedemann. The [Marian Framework](https://marian-nmt.github.io/) is being developed by the Microsoft Translator Team.
1. **[Mask2Former](model_doc/mask2former)** (from FAIR and UIUC) released with the paper [Masked-attention Mask Transformer for Universal Image Segmentation](https://arxiv.org/abs/2112.01527) by Bowen Cheng, Ishan Misra, Alexander G. Schwing, Alexander Kirillov, Rohit Girdhar.
1. **[MaskFormer](model_doc/maskformer)** (from Meta and UIUC) released with the paper [Per-Pixel Classification is Not All You Need for Semantic Segmentation](https://arxiv.org/abs/2107.06278) by Bowen Cheng, Alexander G. Schwing, Alexander Kirillov.
1. **[MBart](model_doc/mbart)** (from Facebook) released with the paper [Multilingual Denoising Pre-training for Neural Machine Translation](https://arxiv.org/abs/2001.08210) by Yinhan Liu, Jiatao Gu, Naman Goyal, Xian Li, Sergey Edunov, Marjan Ghazvininejad, Mike Lewis, Luke Zettlemoyer.
1. **[MBart-50](model_doc/mbart)** (from Facebook) released with the paper [Multilingual Translation with Extensible Multilingual Pretraining and Finetuning](https://arxiv.org/abs/2008.00401) by Yuqing Tang, Chau Tran, Xian Li, Peng-Jen Chen, Naman Goyal, Vishrav Chaudhary, Jiatao Gu, Angela Fan.
1. **[Megatron-BERT](model_doc/megatron-bert)** (from NVIDIA) released with the paper [Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism](https://arxiv.org/abs/1909.08053) by Mohammad Shoeybi, Mostofa Patwary, Raul Puri, Patrick LeGresley, Jared Casper and Bryan Catanzaro.
1. **[Megatron-GPT2](model_doc/megatron_gpt2)** (from NVIDIA) released with the paper [Megatron-LM: Training Multi-Billion Parameter Language Models Using Model Parallelism](https://arxiv.org/abs/1909.08053) by Mohammad Shoeybi, Mostofa Patwary, Raul Puri, Patrick LeGresley, Jared Casper and Bryan Catanzaro.
1. **[MPNet](model_doc/mpnet)** (from Microsoft Research) released with the paper [MPNet: Masked and Permuted Pre-training for Language Understanding](https://arxiv.org/abs/2004.09297) by Kaitao Song, Xu Tan, Tao Qin, Jianfeng Lu, Tie-Yan Liu.
1. **[MT5](model_doc/mt5)** (from Google AI) released with the paper [mT5: A massively multilingual pre-trained text-to-text transformer](https://arxiv.org/abs/2010.11934) by Linting Xue, Noah Constant, Adam Roberts, Mihir Kale, Rami Al-Rfou, Aditya Siddhant, Aditya Barua, Colin Raffel.
1. **[Nyströmformer](model_doc/nystromformer)** (from the University of Wisconsin - Madison) released with the paper [Nyströmformer: A Nyström-Based Algorithm for Approximating Self-Attention](https://arxiv.org/abs/2102.03902) by Yunyang Xiong, Zhanpeng Zeng, Rudrasis Chakraborty, Mingxing Tan, Glenn Fung, Yin Li, Vikas Singh.
1. **[OneFormer](model_doc/oneformer)** (from SHI Labs) released with the paper [OneFormer: One Transformer to Rule Universal Image Segmentation](https://arxiv.org/abs/2211.06220) by Jitesh Jain, Jiachen Li, MangTik Chiu, Ali Hassani, Nikita Orlov, Humphrey Shi.
1. **[Pegasus](model_doc/pegasus)** (from Google) released with the paper [PEGASUS: Pre-training with Extracted Gap-sentences for Abstractive Summarization](https://arxiv.org/abs/1912.08777) by Jingqing Zhang, Yao Zhao, Mohammad Saleh and Peter J. Liu.
1. **[Perceiver IO](model_doc/perceiver)** (from Deepmind) released with the paper [Perceiver IO: A General Architecture for Structured Inputs & Outputs](https://arxiv.org/abs/2107.14795) by Andrew Jaegle, Sebastian Borgeaud, Jean-Baptiste Alayrac, Carl Doersch, Catalin Ionescu, David Ding, Skanda Koppula, Daniel Zoran, Andrew Brock, Evan Shelhamer, Olivier Hénaff, Matthew M. Botvinick, Andrew Zisserman, Oriol Vinyals, João Carreira.
1. **[PhoBERT](model_doc/phobert)** (from VinAI Research) released with the paper [PhoBERT: Pre-trained language models for Vietnamese](https://www.aclweb.org/anthology/2020.findings-emnlp.92/) by Dat Quoc Nguyen and Anh Tuan Nguyen.
1. **[PLBart](model_doc/plbart)** (from UCLA NLP) released with the paper [Unified Pre-training for Program Understanding and Generation](https://arxiv.org/abs/2103.06333) by Wasi Uddin Ahmad, Saikat Chakraborty, Baishakhi Ray, Kai-Wei Chang.
1. **[PoolFormer](model_doc/poolformer)** (from Sea AI Labs) released with the paper [MetaFormer is Actually What You Need for Vision](https://arxiv.org/abs/2111.11418) by Yu, Weihao and Luo, Mi and Zhou, Pan and Si, Chenyang and Zhou, Yichen and Wang, Xinchao and Feng, Jiashi and Yan, Shuicheng.
1. **[ProphetNet](model_doc/prophetnet)** (from Microsoft Research) released with the paper [ProphetNet: Predicting Future N-gram for Sequence-to-Sequence Pre-training](https://arxiv.org/abs/2001.04063) by Yu Yan, Weizhen Qi, Yeyun Gong, Dayiheng Liu, Nan Duan, Jiusheng Chen, Ruofei Zhang and Ming Zhou.
1. **[QDQBert](model_doc/qdqbert)** (from NVIDIA) released with the paper [Integer Quantization for Deep Learning Inference: Principles and Empirical Evaluation](https://arxiv.org/abs/2004.09602) by Hao Wu, Patrick Judd, Xiaojie Zhang, Mikhail Isaev and Paulius Micikevicius.
1. **[REALM](model_doc/realm.html)** (from Google Research) released with the paper [REALM: Retrieval-Augmented Language Model Pre-Training](https://arxiv.org/abs/2002.08909) by Kelvin Guu, Kenton Lee, Zora Tung, Panupong Pasupat and Ming-Wei Chang.
1. **[Reformer](model_doc/reformer)** (from Google Research) released with the paper [Reformer: The Efficient Transformer](https://arxiv.org/abs/2001.04451) by Nikita Kitaev, Łukasz Kaiser, Anselm Levskaya.
1. **[RemBERT](model_doc/rembert)** (from Google Research) released with the paper [Rethinking embedding coupling in pre-trained language models](https://arxiv.org/abs/2010.12821) by Hyung Won Chung, Thibault Févry, Henry Tsai, M. Johnson, Sebastian Ruder.
1. **[RegNet](model_doc/regnet)** (from META Platforms) released with the paper [Designing Network Design Space](https://arxiv.org/abs/2003.13678) by Ilija Radosavovic, Raj Prateek Kosaraju, Ross Girshick, Kaiming He, Piotr Dollár.
1. **[ResNet](model_doc/resnet)** (from Microsoft Research) released with the paper [Deep Residual Learning for Image Recognition](https://arxiv.org/abs/1512.03385) by Kaiming He, Xiangyu Zhang, Shaoqing Ren, Jian Sun.
1. **[RoBERTa](model_doc/roberta)** (from Facebook), released together with the paper [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://arxiv.org/abs/1907.11692) by Yinhan Liu, Myle Ott, Naman Goyal, Jingfei Du, Mandar Joshi, Danqi Chen, Omer Levy, Mike Lewis, Luke Zettlemoyer, Veselin Stoyanov.
1. **[RoFormer](model_doc/roformer)** (from ZhuiyiTechnology), released together with the paper [RoFormer: Enhanced Transformer with Rotary Position Embedding](https://arxiv.org/abs/2104.09864) by Jianlin Su and Yu Lu and Shengfeng Pan and Bo Wen and Yunfeng Liu.
1. **[SegFormer](model_doc/segformer)** (from NVIDIA) released with the paper [SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers](https://arxiv.org/abs/2105.15203) by Enze Xie, Wenhai Wang, Zhiding Yu, Anima Anandkumar, Jose M. Alvarez, Ping Luo.
1. **[SEW](model_doc/sew)** (from ASAPP) released with the paper [Performance-Efficiency Trade-offs in Unsupervised Pre-training for Speech Recognition](https://arxiv.org/abs/2109.06870) by Felix Wu, Kwangyoun Kim, Jing Pan, Kyu Han, Kilian Q. Weinberger, Yoav Artzi.
1. **[SEW-D](model_doc/sew_d)** (from ASAPP) released with the paper [Performance-Efficiency Trade-offs in Unsupervised Pre-training for Speech Recognition](https://arxiv.org/abs/2109.06870) by Felix Wu, Kwangyoun Kim, Jing Pan, Kyu Han, Kilian Q. Weinberger, Yoav Artzi.
1. **[SpeechToTextTransformer](model_doc/speech_to_text)** (from Facebook), released together with the paper [fairseq S2T: Fast Speech-to-Text Modeling with fairseq](https://arxiv.org/abs/2010.05171) by Changhan Wang, Yun Tang, Xutai Ma, Anne Wu, Dmytro Okhonko, Juan Pino.
1. **[SpeechToTextTransformer2](model_doc/speech_to_text_2)** (from Facebook), released together with the paper [Large-Scale Self- and Semi-Supervised Learning for Speech Translation](https://arxiv.org/abs/2104.06678) by Changhan Wang, Anne Wu, Juan Pino, Alexei Baevski, Michael Auli, Alexis Conneau.
1. **[Splinter](model_doc/splinter)** (from Tel Aviv University), released together with the paper [Few-Shot Question Answering by Pretraining Span Selection](https://arxiv.org/abs/2101.00438) by Ori Ram, Yuval Kirstain, Jonathan Berant, Amir Globerson, Omer Levy.
1. **[SqueezeBert](model_doc/squeezebert)** (from Berkeley) released with the paper [SqueezeBERT: What can computer vision teach NLP about efficient neural networks?](https://arxiv.org/abs/2006.11316) by Forrest N. Iandola, Albert E. Shaw, Ravi Krishna, and Kurt W. Keutzer.
1. **[Swin Transformer](model_doc/swin)** (from Microsoft) released with the paper [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://arxiv.org/abs/2103.14030) by Ze Liu, Yutong Lin, Yue Cao, Han Hu, Yixuan Wei, Zheng Zhang, Stephen Lin, Baining Guo.
1. **[T5](model_doc/t5)** (from Google AI) released with the paper [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://arxiv.org/abs/1910.10683) by Colin Raffel and Noam Shazeer and Adam Roberts and Katherine Lee and Sharan Narang and Michael Matena and Yanqi Zhou and Wei Li and Peter J. Liu.
1. **[T5v1.1](model_doc/t5v1.1)** (from Google AI) released in the repository [google-research/text-to-text-transfer-transformer](https://github.com/google-research/text-to-text-transfer-transformer/blob/main/released_checkpoints.md#t511) by Colin Raffel and Noam Shazeer and Adam Roberts and Katherine Lee and Sharan Narang and Michael Matena and Yanqi Zhou and Wei Li and Peter J. Liu.
1. **[TAPAS](model_doc/tapas)** (from Google AI) released with the paper [TAPAS: Weakly Supervised Table Parsing via Pre-training](https://arxiv.org/abs/2004.02349) by Jonathan Herzig, Paweł Krzysztof Nowak, Thomas Müller, Francesco Piccinno and Julian Martin Eisenschlos.
1. **[TAPEX](model_doc/tapex)** (from Microsoft Research) released with the paper [TAPEX: Table Pre-training via Learning a Neural SQL Executor](https://arxiv.org/abs/2107.07653) by Qian Liu, Bei Chen, Jiaqi Guo, Morteza Ziyadi, Zeqi Lin, Weizhu Chen, Jian-Guang Lou.
1. **[Transformer-XL](model_doc/transfo-xl)** (from Google/CMU) released with the paper [Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context](https://arxiv.org/abs/1901.02860) by Zihang Dai*, Zhilin Yang*, Yiming Yang, Jaime Carbonell, Quoc V. Le, Ruslan Salakhutdinov.
1. **[TrOCR](model_doc/trocr)** (from Microsoft), released together with the paper [TrOCR: Transformer-based Optical Character Recognition with Pre-trained Models](https://arxiv.org/abs/2109.10282) by Minghao Li, Tengchao Lv, Lei Cui, Yijuan Lu, Dinei Florencio, Cha Zhang, Zhoujun Li, Furu Wei.
1. **[UniSpeech](model_doc/unispeech)** (from Microsoft Research) released with the paper [UniSpeech: Unified Speech Representation Learning with Labeled and Unlabeled Data](https://arxiv.org/abs/2101.07597) by Chengyi Wang, Yu Wu, Yao Qian, Kenichi Kumatani, Shujie Liu, Furu Wei, Michael Zeng, Xuedong Huang.
1. **[UniSpeechSat](model_doc/unispeech-sat)** (from Microsoft Research) released with the paper [UNISPEECH-SAT: UNIVERSAL SPEECH REPRESENTATION LEARNING WITH SPEAKER AWARE PRE-TRAINING](https://arxiv.org/abs/2110.05752) by Sanyuan Chen, Yu Wu, Chengyi Wang, Zhengyang Chen, Zhuo Chen, Shujie Liu, Jian Wu, Yao Qian, Furu Wei, Jinyu Li, Xiangzhan Yu.
1. **[VAN](model_doc/van)** (from Tsinghua University and Nankai University) released with the paper [Visual Attention Network](https://arxiv.org/abs/2202.09741) by Meng-Hao Guo, Cheng-Ze Lu, Zheng-Ning Liu, Ming-Ming Cheng, Shi-Min Hu.
1. **[ViLT](model_doc/vilt)** (from NAVER AI Lab/Kakao Enterprise/Kakao Brain) released with the paper [ViLT: Vision-and-Language Transformer Without Convolution or Region Supervision](https://arxiv.org/abs/2102.03334) by Wonjae Kim, Bokyung Son, Ildoo Kim.
1. **[Vision Transformer (ViT)](model_doc/vit)** (from Google AI) released with the paper [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://arxiv.org/abs/2010.11929) by Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Sylvain Gelly, Jakob Uszkoreit, Neil Houlsby.
1. **[ViTMAE](model_doc/vit_mae)** (from Meta AI) released with the paper [Masked Autoencoders Are Scalable Vision Learners](https://arxiv.org/abs/2111.06377) by Kaiming He, Xinlei Chen, Saining Xie, Yanghao Li, Piotr Dollár, Ross Girshick.
1. **[VisualBERT](model_doc/visual_bert)** (from UCLA NLP) released with the paper [VisualBERT: A Simple and Performant Baseline for Vision and Language](https://arxiv.org/pdf/1908.03557) by Liunian Harold Li, Mark Yatskar, Da Yin, Cho-Jui Hsieh, Kai-Wei Chang.
1. **[WavLM](model_doc/wavlm)** (from Microsoft Research) released with the paper [WavLM: Large-Scale Self-Supervised Pre-Training for Full Stack Speech Processing](https://arxiv.org/abs/2110.13900) by Sanyuan Chen, Chengyi Wang, Zhengyang Chen, Yu Wu, Shujie Liu, Zhuo Chen, Jinyu Li, Naoyuki Kanda, Takuya Yoshioka, Xiong Xiao, Jian Wu, Long Zhou, Shuo Ren, Yanmin Qian, Yao Qian, Jian Wu, Michael Zeng, Furu Wei.
1. **[Wav2Vec2](model_doc/wav2vec2)** (from Facebook AI) released with the paper [wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations](https://arxiv.org/abs/2006.11477) by Alexei Baevski, Henry Zhou, Abdelrahman Mohamed, Michael Auli.
1. **[Wav2Vec2Phoneme](model_doc/wav2vec2_phoneme)** (from Facebook AI) released with the paper [Simple and Effective Zero-shot Cross-lingual Phoneme Recognition](https://arxiv.org/abs/2109.11680) by Qiantong Xu, Alexei Baevski, Michael Auli.
1. **[XGLM](model_doc/xglm)** (From Facebook AI) released with the paper [Few-shot Learning with Multilingual Language Models](https://arxiv.org/abs/2112.10668) by Xi Victoria Lin, Todor Mihaylov, Mikel Artetxe, Tianlu Wang, Shuohui Chen, Daniel Simig, Myle Ott, Naman Goyal, Shruti Bhosale, Jingfei Du, Ramakanth Pasunuru, Sam Shleifer, Punit Singh Koura, Vishrav Chaudhary, Brian O'Horo, Jeff Wang, Luke Zettlemoyer, Zornitsa Kozareva, Mona Diab, Veselin Stoyanov, Xian Li.
1. **[XLM](model_doc/xlm)** (from Facebook) released together with the paper [Cross-lingual Language Model Pretraining](https://arxiv.org/abs/1901.07291) by Guillaume Lample and Alexis Conneau.
1. **[XLM-ProphetNet](model_doc/xlm-prophetnet)** (from Microsoft Research) released with the paper [ProphetNet: Predicting Future N-gram for Sequence-to-Sequence Pre-training](https://arxiv.org/abs/2001.04063) by Yu Yan, Weizhen Qi, Yeyun Gong, Dayiheng Liu, Nan Duan, Jiusheng Chen, Ruofei Zhang and Ming Zhou.
1. **[XLM-RoBERTa](model_doc/xlm-roberta)** (from Facebook AI), released together with the paper [Unsupervised Cross-lingual Representation Learning at Scale](https://arxiv.org/abs/1911.02116) by Alexis Conneau*, Kartikay Khandelwal*, Naman Goyal, Vishrav Chaudhary, Guillaume Wenzek, Francisco Guzmán, Edouard Grave, Myle Ott, Luke Zettlemoyer and Veselin Stoyanov.
1. **[XLM-RoBERTa-XL](model_doc/xlm-roberta-xl)** (from Facebook AI), released together with the paper [Larger-Scale Transformers for Multilingual Masked Language Modeling](https://arxiv.org/abs/2105.00572) by Naman Goyal, Jingfei Du, Myle Ott, Giri Anantharaman, Alexis Conneau.
1. **[XLNet](model_doc/xlnet)** (from Google/CMU) released with the paper [XLNet: Generalized Autoregressive Pretraining for Language Understanding](https://arxiv.org/abs/1906.08237) by Zhilin Yang*, Zihang Dai*, Yiming Yang, Jaime Carbonell, Ruslan Salakhutdinov, Quoc V. Le.
1. **[XLSR-Wav2Vec2](model_doc/xlsr_wav2vec2)** (from Facebook AI) released with the paper [Unsupervised Cross-Lingual Representation Learning For Speech Recognition](https://arxiv.org/abs/2006.13979) by Alexis Conneau, Alexei Baevski, Ronan Collobert, Abdelrahman Mohamed, Michael Auli.
1. **[XLS-R](model_doc/xls_r)** (from Facebook AI) released with the paper [XLS-R: Self-supervised Cross-lingual Speech Representation Learning at Scale](https://arxiv.org/abs/2111.09296) by Arun Babu, Changhan Wang, Andros Tjandra, Kushal Lakhotia, Qiantong Xu, Naman Goyal, Kritika Singh, Patrick von Platen, Yatharth Saraf, Juan Pino, Alexei Baevski, Alexis Conneau, Michael Auli.
1. **[YOSO](model_doc/yoso)** (from the University of Wisconsin - Madison) released with the paper [You Only Sample (Almost) Once: Linear Cost Self-Attention Via Bernoulli Sampling](https://arxiv.org/abs/2111.09714) by Zhanpeng Zeng, Yunyang Xiong, Sathya N. Ravi, Shailesh Acharya, Glenn Fung, Vikas Singh.
### Frameworks aceitos
A tabela abaixo representa a lista de suporte na biblioteca para cada um dos seguintes modelos, caso tenham um tokenizer
do Python (chamado de "slow"), ou um tokenizer construído em cima da biblioteca 🤗 Tokenizers (chamado de "fast"). Além
disso, são diferenciados pelo suporte em diferentes frameworks: JAX (por meio do Flax); PyTorch; e/ou Tensorflow.
<!--This table is updated automatically from the auto modules with _make fix-copies_. Do not update manually!-->
| Model | Tokenizer slow | Tokenizer fast | PyTorch support | TensorFlow support | Flax Support |
|:---------------------------:|:--------------:|:--------------:|:---------------:|:------------------:|:------------:|
| ALBERT | ✅ | ✅ | ✅ | ✅ | ✅ |
| BART | ✅ | ✅ | ✅ | ✅ | ✅ |
| BEiT | ❌ | ❌ | ✅ | ❌ | ✅ |
| BERT | ✅ | ✅ | ✅ | ✅ | ✅ |
| Bert Generation | ✅ | ❌ | ✅ | ❌ | ❌ |
| BigBird | ✅ | ✅ | ✅ | ❌ | ✅ |
| BigBirdPegasus | ❌ | ❌ | ✅ | ❌ | ❌ |
| Blenderbot | ✅ | ✅ | ✅ | ✅ | ✅ |
| BlenderbotSmall | ✅ | ✅ | ✅ | ✅ | ✅ |
| CamemBERT | ✅ | ✅ | ✅ | ✅ | ❌ |
| Canine | ✅ | ❌ | ✅ | ❌ | ❌ |
| CLIP | ✅ | ✅ | ✅ | ✅ | ✅ |
| ConvBERT | ✅ | ✅ | ✅ | ✅ | ❌ |
| ConvNext | ❌ | ❌ | ✅ | ✅ | ❌ |
| CTRL | ✅ | ❌ | ✅ | ✅ | ❌ |
| Data2VecAudio | ❌ | ❌ | ✅ | ❌ | ❌ |
| Data2VecText | ❌ | ❌ | ✅ | ❌ | ❌ |
| Data2VecVision | ❌ | ❌ | ✅ | ❌ | ❌ |
| DeBERTa | ✅ | ✅ | ✅ | ✅ | ❌ |
| DeBERTa-v2 | ✅ | ✅ | ✅ | ✅ | ❌ |
| Decision Transformer | ❌ | ❌ | ✅ | ❌ | ❌ |
| DeiT | ❌ | ❌ | ✅ | ❌ | ❌ |
| DETR | ❌ | ❌ | ✅ | ❌ | ❌ |
| DistilBERT | ✅ | ✅ | ✅ | ✅ | ✅ |
| DPR | ✅ | ✅ | ✅ | ✅ | ❌ |
| DPT | ❌ | ❌ | ✅ | ❌ | ❌ |
| ELECTRA | ✅ | ✅ | ✅ | ✅ | ✅ |
| Encoder decoder | ❌ | ❌ | ✅ | ✅ | ✅ |
| FairSeq Machine-Translation | ✅ | ❌ | ✅ | ❌ | ❌ |
| FlauBERT | ✅ | ❌ | ✅ | ✅ | ❌ |
| FNet | ✅ | ✅ | ✅ | ❌ | ❌ |
| Funnel Transformer | ✅ | ✅ | ✅ | ✅ | ❌ |
| GLPN | ❌ | ❌ | ✅ | ❌ | ❌ |
| GPT Neo | ❌ | ❌ | ✅ | ❌ | ✅ |
| GPT-J | ❌ | ❌ | ✅ | ✅ | ✅ |
| Hubert | ❌ | ❌ | ✅ | ✅ | ❌ |
| I-BERT | ❌ | ❌ | ✅ | ❌ | ❌ |
| ImageGPT | ❌ | ❌ | ✅ | ❌ | ❌ |
| LayoutLM | ✅ | ✅ | ✅ | ✅ | ❌ |
| LayoutLMv2 | ✅ | ✅ | ✅ | ❌ | ❌ |
| LED | ✅ | ✅ | ✅ | ✅ | ❌ |
| Longformer | ✅ | ✅ | ✅ | ✅ | ❌ |
| LUKE | ✅ | ❌ | ✅ | ❌ | ❌ |
| LXMERT | ✅ | ✅ | ✅ | ✅ | ❌ |
| M2M100 | ✅ | ❌ | ✅ | ❌ | ❌ |
| Marian | ✅ | ❌ | ✅ | ✅ | ✅ |
| MaskFormer | ❌ | ❌ | ✅ | ❌ | ❌ |
| mBART | ✅ | ✅ | ✅ | ✅ | ✅ |
| MegatronBert | ❌ | ❌ | ✅ | ❌ | ❌ |
| MobileBERT | ✅ | ✅ | ✅ | ✅ | ❌ |
| MPNet | ✅ | ✅ | ✅ | ✅ | ❌ |
| mT5 | ✅ | ✅ | ✅ | ✅ | ✅ |
| Nystromformer | ❌ | ❌ | ✅ | ❌ | ❌ |
| OpenAI GPT | ✅ | ✅ | ✅ | ✅ | ❌ |
| OpenAI GPT-2 | ✅ | ✅ | ✅ | ✅ | ✅ |
| Pegasus | ✅ | ✅ | ✅ | ✅ | ✅ |
| Perceiver | ✅ | ❌ | ✅ | ❌ | ❌ |
| PLBart | ✅ | ❌ | ✅ | ❌ | ❌ |
| PoolFormer | ❌ | ❌ | ✅ | ❌ | ❌ |
| ProphetNet | ✅ | ❌ | ✅ | ❌ | ❌ |
| QDQBert | ❌ | ❌ | ✅ | ❌ | ❌ |
| RAG | ✅ | ❌ | ✅ | ✅ | ❌ |
| Realm | ✅ | ✅ | ✅ | ❌ | ❌ |
| Reformer | ✅ | ✅ | ✅ | ❌ | ❌ |
| RegNet | ❌ | ❌ | ✅ | ✅ | ✅ |
| RemBERT | ✅ | ✅ | ✅ | ✅ | ❌ |
| ResNet | ❌ | ❌ | ✅ | ❌ | ✅ |
| RetriBERT | ✅ | ✅ | ✅ | ❌ | ❌ |
| RoBERTa | ✅ | ✅ | ✅ | ✅ | ✅ |
| RoFormer | ✅ | ✅ | ✅ | ✅ | ✅ |
| SegFormer | ❌ | ❌ | ✅ | ❌ | ❌ |
| SEW | ❌ | ❌ | ✅ | ❌ | ❌ |
| SEW-D | ❌ | ❌ | ✅ | ❌ | ❌ |
| Speech Encoder decoder | ❌ | ❌ | ✅ | ❌ | ✅ |
| Speech2Text | ✅ | ❌ | ✅ | ✅ | ❌ |
| Speech2Text2 | ✅ | ❌ | ❌ | ❌ | ❌ |
| Splinter | ✅ | ✅ | ✅ | ❌ | ❌ |
| SqueezeBERT | ✅ | ✅ | ✅ | ❌ | ❌ |
| Swin | ❌ | ❌ | ✅ | ❌ | ❌ |
| T5 | ✅ | ✅ | ✅ | ✅ | ✅ |
| TAPAS | ✅ | ❌ | ✅ | ✅ | ❌ |
| TAPEX | ✅ | ✅ | ✅ | ✅ | ✅ |
| Transformer-XL | ✅ | ❌ | ✅ | ✅ | ❌ |
| TrOCR | ❌ | ❌ | ✅ | ❌ | ❌ |
| UniSpeech | ❌ | ❌ | ✅ | ❌ | ❌ |
| UniSpeechSat | ❌ | ❌ | ✅ | ❌ | ❌ |
| VAN | ❌ | ❌ | ✅ | ❌ | ❌ |
| ViLT | ❌ | ❌ | ✅ | ❌ | ❌ |
| Vision Encoder decoder | ❌ | ❌ | ✅ | ✅ | ✅ |
| VisionTextDualEncoder | ❌ | ❌ | ✅ | ❌ | ✅ |
| VisualBert | ❌ | ❌ | ✅ | ❌ | ❌ |
| ViT | ❌ | ❌ | ✅ | ✅ | ✅ |
| ViTMAE | ❌ | ❌ | ✅ | ✅ | ❌ |
| Wav2Vec2 | ✅ | ❌ | ✅ | ✅ | ✅ |
| WavLM | ❌ | ❌ | ✅ | ❌ | ❌ |
| XGLM | ✅ | ✅ | ✅ | ❌ | ✅ |
| XLM | ✅ | ❌ | ✅ | ✅ | ❌ |
| XLM-RoBERTa | ✅ | ✅ | ✅ | ✅ | ✅ |
| XLM-RoBERTa-XL | ❌ | ❌ | ✅ | ❌ | ❌ |
| XLMProphetNet | ✅ | ❌ | ✅ | ❌ | ❌ |
| XLNet | ✅ | ✅ | ✅ | ✅ | ❌ |
| YOSO | ❌ | ❌ | ✅ | ❌ | ❌ |
<!-- End table-->
| 0
|
mavonic_private_repos/transformers/docs/source
|
mavonic_private_repos/transformers/docs/source/pt/training.md
|
<!--Copyright 2022 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
specific language governing permissions and limitations under the License.
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer.
-->
# Fine-tuning de um modelo pré-treinado
[[open-in-colab]]
O uso de um modelo pré-treinado tem importantes vantagens. Redução do custo computacional, a pegada de carbono, e te
permite utilizar modelos de última geração sem ter que treinar um novo desde o início.
O 🤗 Transformers proporciona acesso a milhares de modelos pré-treinados numa ampla gama de tarefas.
Quando utilizar um modelo pré-treinado, treine-o com um dataset específico para a sua tarefa.
Isto é chamado de fine-tuning, uma técnica de treinamento incrivelmente poderosa. Neste tutorial faremos o fine-tuning
de um modelo pré-treinado com um framework de Deep Learning da sua escolha:
* Fine-tuning de um modelo pré-treinado com o 🤗 Transformers [`Trainer`].
* Fine-tuning de um modelo pré-treinado no TensorFlow com o Keras.
* Fine-tuning de um modelo pré-treinado em PyTorch nativo.
<a id='data-processing'></a>
## Preparando um dataset
<Youtube id="_BZearw7f0w"/>
Antes de aplicar o fine-tuning a um modelo pré-treinado, baixe um dataset e prepare-o para o treinamento.
O tutorial anterior ensinará a processar os dados para o treinamento, e então poderá ter a oportunidade de testar
esse novo conhecimento em algo prático.
Comece carregando o dataset [Yelp Reviews](https://huggingface.co/datasets/yelp_review_full):
```py
>>> from datasets import load_dataset
>>> dataset = load_dataset("yelp_review_full")
>>> dataset[100]
{'label': 0,
'text': 'My expectations for McDonalds are t rarely high. But for one to still fail so spectacularly...that takes something special!\\nThe cashier took my friends\'s order, then promptly ignored me. I had to force myself in front of a cashier who opened his register to wait on the person BEHIND me. I waited over five minutes for a gigantic order that included precisely one kid\'s meal. After watching two people who ordered after me be handed their food, I asked where mine was. The manager started yelling at the cashiers for \\"serving off their orders\\" when they didn\'t have their food. But neither cashier was anywhere near those controls, and the manager was the one serving food to customers and clearing the boards.\\nThe manager was rude when giving me my order. She didn\'t make sure that I had everything ON MY RECEIPT, and never even had the decency to apologize that I felt I was getting poor service.\\nI\'ve eaten at various McDonalds restaurants for over 30 years. I\'ve worked at more than one location. I expect bad days, bad moods, and the occasional mistake. But I have yet to have a decent experience at this store. It will remain a place I avoid unless someone in my party needs to avoid illness from low blood sugar. Perhaps I should go back to the racially biased service of Steak n Shake instead!'}
```
Como já sabe, é necessário ter um tokenizador para processar o texto e incluir uma estratégia de padding e truncamento,
para manejar qualquer tamanho varíavel de sequência. Para processar o seu dataset em apenas um passo, utilize o método de
🤗 Datasets [`map`](https://huggingface.co/docs/datasets/process#map) para aplicar uma função de preprocessamento sobre
todo o dataset.
```py
>>> from transformers import AutoTokenizer
>>> tokenizer = AutoTokenizer.from_pretrained("google-bert/bert-base-cased")
>>> def tokenize_function(examples):
... return tokenizer(examples["text"], padding="max_length", truncation=True)
>>> tokenized_datasets = dataset.map(tokenize_function, batched=True)
```
Se desejar, é possível criar um subconjunto menor do dataset completo para aplicar o fine-tuning e assim reduzir o tempo necessário.
```py
>>> small_train_dataset = tokenized_datasets["train"].shuffle(seed=42).select(range(1000))
>>> small_eval_dataset = tokenized_datasets["test"].shuffle(seed=42).select(range(1000))
```
<a id='trainer'></a>
## Fine-tuning com o `Trainer`
<Youtube id="nvBXf7s7vTI"/>
O 🤗 Transformers proporciona uma classe [`Trainer`] otimizada para o treinamento de modelos de 🤗 Transformers,
facilitando os primeiros passos do treinamento sem a necessidade de escrever manualmente o seu próprio ciclo.
A API do [`Trainer`] suporta um grande conjunto de opções de treinamento e funcionalidades, como o logging,
o gradient accumulation e o mixed precision.
Comece carregando seu modelo e especifique o número de labels de previsão.
A partir do [Card Dataset](https://huggingface.co/datasets/yelp_review_full#data-fields) do Yelp Reveiw, que ja
sabemos ter 5 labels usamos o seguinte código:
```py
>>> from transformers import AutoModelForSequenceClassification
>>> model = AutoModelForSequenceClassification.from_pretrained("google-bert/bert-base-cased", num_labels=5)
```
<Tip>
Você verá um alerta sobre alguns pesos pré-treinados que não estão sendo utilizados e que alguns pesos estão
sendo inicializados aleatoriamente. Não se preocupe, essa mensagem é completamente normal.
O header/cabeçário pré-treinado do modelo BERT é descartado e substitui-se por um header de classificação
inicializado aleatoriamente. Assim, pode aplicar o fine-tuning a este novo header do modelo em sua tarefa
de classificação de sequências fazendo um transfer learning do modelo pré-treinado.
</Tip>
### Hiperparâmetros de treinamento
Em seguida, crie uma classe [`TrainingArguments`] que contenha todos os hiperparâmetros que possam ser ajustados, assim
como os indicadores para ativar as diferentes opções de treinamento. Para este tutorial, você pode começar o treinamento
usando os [hiperparámetros](https://huggingface.co/docs/transformers/main_classes/trainer#transformers.TrainingArguments) padrão,
porém, sinta-se livre para experimentar com eles e encontrar uma configuração ótima.
Especifique onde salvar os checkpoints do treinamento:
```py
>>> from transformers import TrainingArguments
>>> training_args = TrainingArguments(output_dir="test_trainer")
```
### Métricas
O [`Trainer`] não avalia automaticamente o rendimento do modelo durante o treinamento. Será necessário passar ao
[`Trainer`] uma função para calcular e fazer um diagnóstico sobre as métricas. A biblioteca 🤗 Datasets proporciona
uma função de [`accuracy`](https://huggingface.co/metrics/accuracy) simples que pode ser carregada com a função
`load_metric` (ver este [tutorial](https://huggingface.co/docs/datasets/metrics) para mais informações):
```py
>>> import numpy as np
>>> from datasets import load_metric
>>> metric = load_metric("accuracy")
```
Defina a função `compute` dentro de `metric` para calcular a precisão das suas predições.
Antes de passar as suas predições ao `compute`, é necessário converter as predições à logits (lembre-se que
todos os modelos de 🤗 Transformers retornam logits).
```py
>>> def compute_metrics(eval_pred):
... logits, labels = eval_pred
... predictions = np.argmax(logits, axis=-1)
... return metric.compute(predictions=predictions, references=labels)
```
Se quiser controlar as suas métricas de avaliação durante o fine-tuning, especifique o parâmetro `eval_strategy`
nos seus argumentos de treinamento para que o modelo considere a métrica de avaliação ao final de cada época:
```py
>>> from transformers import TrainingArguments
>>> training_args = TrainingArguments(output_dir="test_trainer", eval_strategy="epoch")
```
### Trainer
Crie um objeto [`Trainer`] com o seu modelo, argumentos de treinamento, conjuntos de dados de treinamento e de teste, e a sua função de avaliação:
```py
>>> trainer = Trainer(
... model=model,
... args=training_args,
... train_dataset=small_train_dataset,
... eval_dataset=small_eval_dataset,
... compute_metrics=compute_metrics,
... )
```
Em seguida, aplique o fine-tuning a seu modelo chamado [`~transformers.Trainer.train`]:
```py
>>> trainer.train()
```
<a id='keras'></a>
## Fine-tuning com Keras
<Youtube id="rnTGBy2ax1c"/>
Os modelos de 🤗 Transformers também permitem realizar o treinamento com o TensorFlow com a API do Keras.
Contudo, será necessário fazer algumas mudanças antes de realizar o fine-tuning.
### Conversão do dataset ao formato do TensorFlow
O [`DefaultDataCollator`] junta os tensores em um batch para que o modelo possa ser treinado em cima deles.
Assegure-se de especificar os `return_tensors` para retornar os tensores do TensorFlow:
```py
>>> from transformers import DefaultDataCollator
>>> data_collator = DefaultDataCollator(return_tensors="tf")
```
<Tip>
O [`Trainer`] utiliza [`DataCollatorWithPadding`] por padrão, então você não precisa especificar explicitamente um
colador de dados (data collator).
</Tip>
Em seguida, converta os datasets tokenizados em datasets do TensorFlow com o método
[`to_tf_dataset`](https://huggingface.co/docs/datasets/package_reference/main_classes#datasets.Dataset.to_tf_dataset).
Especifique suas entradas em `columns` e seu rótulo em `label_cols`:
```py
>>> tf_train_dataset = small_train_dataset.to_tf_dataset(
... columns=["attention_mask", "input_ids", "token_type_ids"],
... label_cols="labels",
... shuffle=True,
... collate_fn=data_collator,
... batch_size=8,
... )
>>> tf_validation_dataset = small_eval_dataset.to_tf_dataset(
... columns=["attention_mask", "input_ids", "token_type_ids"],
... label_cols="labels",
... shuffle=False,
... collate_fn=data_collator,
... batch_size=8,
... )
```
### Compilação e ajustes
Carregue um modelo do TensorFlow com o número esperado de rótulos:
```py
>>> import tensorflow as tf
>>> from transformers import TFAutoModelForSequenceClassification
>>> model = TFAutoModelForSequenceClassification.from_pretrained("google-bert/bert-base-cased", num_labels=5)
```
A seguir, compile e ajuste o fine-tuning a seu modelo com [`fit`](https://keras.io/api/models/model_training_apis/) como
faria com qualquer outro modelo do Keras:
```py
>>> model.compile(
... optimizer=tf.keras.optimizers.Adam(learning_rate=5e-5),
... loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True),
... metrics=tf.metrics.SparseCategoricalAccuracy(),
... )
>>> model.fit(tf_train_dataset, validation_data=tf_validation_dataset, epochs=3)
```
<a id='pytorch_native'></a>
## Fine-tune em PyTorch nativo
<Youtube id="Dh9CL8fyG80"/>
O [`Trainer`] se encarrega do ciclo de treinamento e permite aplicar o fine-tuning a um modelo em uma linha de código apenas.
Para os usuários que preferirem escrever o seu próprio ciclo de treinamento, também é possível aplicar o fine-tuning a um
modelo de 🤗 Transformers em PyTorch nativo.
Neste momento, talvez ocorra a necessidade de reinicar seu notebook ou executar a seguinte linha de código para liberar
memória:
```py
del model
del pytorch_model
del trainer
torch.cuda.empty_cache()
```
Em sequência, faremos um post-processing manual do `tokenized_dataset` e assim prepará-lo para o treinamento.
1. Apague a coluna de `text` porque o modelo não aceita texto cru como entrada:
```py
>>> tokenized_datasets = tokenized_datasets.remove_columns(["text"])
```
2. Troque o nome da coluna `label` para `labels`, pois o modelo espera um argumento de mesmo nome:
```py
>>> tokenized_datasets = tokenized_datasets.rename_column("label", "labels")
```
3. Defina o formato do dataset para retornar tensores do PyTorch no lugar de listas:
```py
>>> tokenized_datasets.set_format("torch")
```
Em sequência, crie um subconjunto menor do dataset, como foi mostrado anteriormente, para acelerá-lo o fine-tuning.
```py
>>> small_train_dataset = tokenized_datasets["train"].shuffle(seed=42).select(range(1000))
>>> small_eval_dataset = tokenized_datasets["test"].shuffle(seed=42).select(range(1000))
```
### DataLoader
Crie um `DataLoader` para os seus datasets de treinamento e de teste para poder iterar sobre batches de dados:
```py
>>> from torch.utils.data import DataLoader
>>> train_dataloader = DataLoader(small_train_dataset, shuffle=True, batch_size=8)
>>> eval_dataloader = DataLoader(small_eval_dataset, batch_size=8)
```
Carregue seu modelo com o número de labels esperados:
```py
>>> from transformers import AutoModelForSequenceClassification
>>> model = AutoModelForSequenceClassification.from_pretrained("google-bert/bert-base-cased", num_labels=5)
```
### Otimização e configuração do Learning Rate
Crie um otimizador e um learning rate para aplicar o fine-tuning ao modelo.
Iremos utilizar o otimizador [`AdamW`](https://pytorch.org/docs/stable/generated/torch.optim.AdamW.html) do PyTorch:
```py
>>> from torch.optim import AdamW
>>> optimizer = AdamW(model.parameters(), lr=5e-5)
```
Defina o learning rate do [`Trainer`]:
```py
>>> from transformers import get_scheduler
>>> num_epochs = 3
>>> num_training_steps = num_epochs * len(train_dataloader)
>>> lr_scheduler = get_scheduler(
... name="linear", optimizer=optimizer, num_warmup_steps=0, num_training_steps=num_training_steps
... )
```
Por último, especifique o `device` do ambiente para utilizar uma GPU se tiver acesso à alguma. Caso contrário, o treinamento
em uma CPU pode acabar levando várias horas em vez de minutos.
```py
>>> import torch
>>> device = torch.device("cuda") if torch.cuda.is_available() else torch.device("cpu")
>>> model.to(device)
```
<Tip>
Se necessário, você pode obter o acesso gratuito a uma GPU na núvem por meio de um notebook no
[Colaboratory](https://colab.research.google.com/) ou [SageMaker StudioLab](https://studiolab.sagemaker.aws/)
se não tiver esse recurso de forma local.
</Tip>
Perfeito, agora estamos prontos para começar o treinamento! 🥳
### Ciclo de treinamento
Para visualizar melhor o processo de treinamento, utilize a biblioteca [tqdm](https://tqdm.github.io/) para adicionar
uma barra de progresso sobre o número de passos percorridos no treinamento atual:
```py
>>> from tqdm.auto import tqdm
>>> progress_bar = tqdm(range(num_training_steps))
>>> model.train()
>>> for epoch in range(num_epochs):
... for batch in train_dataloader:
... batch = {k: v.to(device) for k, v in batch.items()}
... outputs = model(**batch)
... loss = outputs.loss
... loss.backward()
... optimizer.step()
... lr_scheduler.step()
... optimizer.zero_grad()
... progress_bar.update(1)
```
### Métricas
Da mesma forma que é necessário adicionar uma função de avaliação ao [`Trainer`], é necessário fazer o mesmo quando
escrevendo o próprio ciclo de treinamento. Contudo, em vez de calcular e retornar a métrica final de cada época,
você deverá adicionar todos os batches com [`add_batch`](https://huggingface.co/docs/datasets/package_reference/main_classes?highlight=add_batch#datasets.Metric.add_batch)
e calcular a métrica apenas no final.
```py
>>> metric = load_metric("accuracy")
>>> model.eval()
>>> for batch in eval_dataloader:
... batch = {k: v.to(device) for k, v in batch.items()}
... with torch.no_grad():
... outputs = model(**batch)
... logits = outputs.logits
... predictions = torch.argmax(logits, dim=-1)
... metric.add_batch(predictions=predictions, references=batch["labels"])
>>> metric.compute()
```
<a id='additional-resources'></a>
## Recursos adicionais
Para mais exemplos de fine-tuning acesse:
- [🤗 Transformers Examples](https://github.com/huggingface/transformers/tree/main/examples) inclui scripts
para treinas tarefas comuns de NLP em PyTorch e TensorFlow.
- [🤗 Transformers Notebooks](notebooks) contém vários notebooks sobre como aplicar o fine-tuning a um modelo
para tarefas específicas no PyTorch e TensorFlow.
| 0
|
mavonic_private_repos/transformers/docs/source
|
mavonic_private_repos/transformers/docs/source/pt/serialization.md
|
<!--Copyright 2020 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
specific language governing permissions and limitations under the License.
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer.
-->
# Exportando modelos para ONNX
Se você precisar implantar modelos 🤗 Transformers em ambientes de produção, recomendamos
exporta-los para um formato serializado que pode ser carregado e executado em
tempos de execução e hardware. Neste guia, mostraremos como exportar modelos 🤗 Transformers
para [ONNX (Open Neural Network eXchange)](http://onnx.ai).
<Tip>
Uma vez exportado, um modelo pode ser otimizado para inferência por meio de técnicas como
quantização e poda. Se você estiver interessado em otimizar seus modelos para serem executados com
máxima eficiência, confira a biblioteca [🤗 Optimum
](https://github.com/huggingface/optimum).
</Tip>
ONNX é um padrão aberto que define um conjunto comum de operadores e um formato de arquivo comum
para representar modelos de aprendizado profundo em uma ampla variedade de estruturas, incluindo PyTorch e
TensorFlow. Quando um modelo é exportado para o formato ONNX, esses operadores são usados para
construir um grafo computacional (muitas vezes chamado de _representação intermediária_) que
representa o fluxo de dados através da rede neural.
Ao expor um grafo com operadores e tipos de dados padronizados, o ONNX facilita a
alternar entre os frameworks. Por exemplo, um modelo treinado em PyTorch pode ser exportado para
formato ONNX e depois importado no TensorFlow (e vice-versa).
🤗 Transformers fornece um pacote [`transformers.onnx`](main_classes/onnx) que permite
que você converta os checkpoints do modelo em um grafo ONNX aproveitando os objetos de configuração.
Esses objetos de configuração vêm prontos para várias arquiteturas de modelo e são
projetado para ser facilmente extensível a outras arquiteturas.
As configurações prontas incluem as seguintes arquiteturas:
<!--This table is automatically generated by `make fix-copies`, do not fill manually!-->
- ALBERT
- BART
- BEiT
- BERT
- BigBird
- BigBird-Pegasus
- Blenderbot
- BlenderbotSmall
- BLOOM
- CamemBERT
- CLIP
- CodeGen
- Conditional DETR
- ConvBERT
- ConvNeXT
- ConvNeXTV2
- Data2VecText
- Data2VecVision
- DeBERTa
- DeBERTa-v2
- DeiT
- DETR
- DistilBERT
- ELECTRA
- ERNIE
- FlauBERT
- GPT Neo
- GPT-J
- GroupViT
- I-BERT
- LayoutLM
- LayoutLMv3
- LeViT
- Longformer
- LongT5
- M2M100
- Marian
- mBART
- MobileBERT
- MobileViT
- MT5
- OpenAI GPT-2
- OWL-ViT
- Perceiver
- PLBart
- ResNet
- RoBERTa
- RoFormer
- SegFormer
- SqueezeBERT
- Swin Transformer
- T5
- Table Transformer
- Vision Encoder decoder
- ViT
- XLM
- XLM-RoBERTa
- XLM-RoBERTa-XL
- YOLOS
Nas próximas duas seções, mostraremos como:
* Exportar um modelo suportado usando o pacote `transformers.onnx`.
* Exportar um modelo personalizado para uma arquitetura sem suporte.
## Exportando um modelo para ONNX
Para exportar um modelo 🤗 Transformers para o ONNX, primeiro você precisa instalar algumas
dependências extras:
```bash
pip install transformers[onnx]
```
O pacote `transformers.onnx` pode então ser usado como um módulo Python:
```bash
python -m transformers.onnx --help
usage: Hugging Face Transformers ONNX exporter [-h] -m MODEL [--feature {causal-lm, ...}] [--opset OPSET] [--atol ATOL] output
positional arguments:
output Path indicating where to store generated ONNX model.
optional arguments:
-h, --help show this help message and exit
-m MODEL, --model MODEL
Model ID on huggingface.co or path on disk to load model from.
--feature {causal-lm, ...}
The type of features to export the model with.
--opset OPSET ONNX opset version to export the model with.
--atol ATOL Absolute difference tolerance when validating the model.
```
A exportação de um checkpoint usando uma configuração pronta pode ser feita da seguinte forma:
```bash
python -m transformers.onnx --model=distilbert/distilbert-base-uncased onnx/
```
Você deve ver os seguintes logs:
```bash
Validating ONNX model...
-[✓] ONNX model output names match reference model ({'last_hidden_state'})
- Validating ONNX Model output "last_hidden_state":
-[✓] (2, 8, 768) matches (2, 8, 768)
-[✓] all values close (atol: 1e-05)
All good, model saved at: onnx/model.onnx
```
Isso exporta um grafo ONNX do ponto de verificação definido pelo argumento `--model`. Nisso
Por exemplo, é `distilbert/distilbert-base-uncased`, mas pode ser qualquer checkpoint no Hugging
Face Hub ou um armazenado localmente.
O arquivo `model.onnx` resultante pode ser executado em um dos [muitos
aceleradores](https://onnx.ai/supported-tools.html#deployModel) que suportam o ONNX
padrão. Por exemplo, podemos carregar e executar o modelo com [ONNX
Tempo de execução](https://onnxruntime.ai/) da seguinte forma:
```python
>>> from transformers import AutoTokenizer
>>> from onnxruntime import InferenceSession
>>> tokenizer = AutoTokenizer.from_pretrained("distilbert/distilbert-base-uncased")
>>> session = InferenceSession("onnx/model.onnx")
>>> # ONNX Runtime expects NumPy arrays as input
>>> inputs = tokenizer("Using DistilBERT with ONNX Runtime!", return_tensors="np")
>>> outputs = session.run(output_names=["last_hidden_state"], input_feed=dict(inputs))
```
Os nomes de saída necessários (como `["last_hidden_state"]`) podem ser obtidos pegando uma
configuração ONNX de cada modelo. Por exemplo, para DistilBERT temos:
```python
>>> from transformers.models.distilbert import DistilBertConfig, DistilBertOnnxConfig
>>> config = DistilBertConfig()
>>> onnx_config = DistilBertOnnxConfig(config)
>>> print(list(onnx_config.outputs.keys()))
["last_hidden_state"]
```
O processo é idêntico para os checkpoints do TensorFlow no Hub. Por exemplo, podemos
exportar um checkpoint TensorFlow puro do [Keras
](https://huggingface.co/keras-io) da seguinte forma:
```bash
python -m transformers.onnx --model=keras-io/transformers-qa onnx/
```
Para exportar um modelo armazenado localmente, você precisará ter os pesos e
arquivos tokenizer armazenados em um diretório. Por exemplo, podemos carregar e salvar um checkpoint como:
```python
>>> from transformers import AutoTokenizer, AutoModelForSequenceClassification
>>> # Load tokenizer and PyTorch weights form the Hub
>>> tokenizer = AutoTokenizer.from_pretrained("distilbert/distilbert-base-uncased")
>>> pt_model = AutoModelForSequenceClassification.from_pretrained("distilbert/distilbert-base-uncased")
>>> # Save to disk
>>> tokenizer.save_pretrained("local-pt-checkpoint")
>>> pt_model.save_pretrained("local-pt-checkpoint")
```
Uma vez que o checkpoint é salvo, podemos exportá-lo para o ONNX apontando o `--model`
argumento do pacote `transformers.onnx` para o diretório desejado:
```bash
python -m transformers.onnx --model=local-pt-checkpoint onnx/
```
```python
>>> from transformers import AutoTokenizer, TFAutoModelForSequenceClassification
>>> # Load tokenizer and TensorFlow weights from the Hub
>>> tokenizer = AutoTokenizer.from_pretrained("distilbert/distilbert-base-uncased")
>>> tf_model = TFAutoModelForSequenceClassification.from_pretrained("distilbert/distilbert-base-uncased")
>>> # Save to disk
>>> tokenizer.save_pretrained("local-tf-checkpoint")
>>> tf_model.save_pretrained("local-tf-checkpoint")
```
Uma vez que o checkpoint é salvo, podemos exportá-lo para o ONNX apontando o `--model`
argumento do pacote `transformers.onnx` para o diretório desejado:
```bash
python -m transformers.onnx --model=local-tf-checkpoint onnx/
```
## Selecionando features para diferentes tarefas do modelo
Cada configuração pronta vem com um conjunto de _features_ que permitem exportar
modelos para diferentes tipos de tarefas. Conforme mostrado na tabela abaixo, cada recurso é
associado a uma `AutoClass` diferente:
| Feature | Auto Class |
| ------------------------------------ | ------------------------------------ |
| `causal-lm`, `causal-lm-with-past` | `AutoModelForCausalLM` |
| `default`, `default-with-past` | `AutoModel` |
| `masked-lm` | `AutoModelForMaskedLM` |
| `question-answering` | `AutoModelForQuestionAnswering` |
| `seq2seq-lm`, `seq2seq-lm-with-past` | `AutoModelForSeq2SeqLM` |
| `sequence-classification` | `AutoModelForSequenceClassification` |
| `token-classification` | `AutoModelForTokenClassification` |
Para cada configuração, você pode encontrar a lista de recursos suportados por meio do
[`~transformers.onnx.FeaturesManager`]. Por exemplo, para DistilBERT temos:
```python
>>> from transformers.onnx.features import FeaturesManager
>>> distilbert_features = list(FeaturesManager.get_supported_features_for_model_type("distilbert").keys())
>>> print(distilbert_features)
["default", "masked-lm", "causal-lm", "sequence-classification", "token-classification", "question-answering"]
```
Você pode então passar um desses recursos para o argumento `--feature` no
pacote `transformers.onnx`. Por exemplo, para exportar um modelo de classificação de texto, podemos
escolher um modelo ajustado no Hub e executar:
```bash
python -m transformers.onnx --model=distilbert/distilbert-base-uncased-finetuned-sst-2-english \
--feature=sequence-classification onnx/
```
Isso exibe os seguintes logs:
```bash
Validating ONNX model...
-[✓] ONNX model output names match reference model ({'logits'})
- Validating ONNX Model output "logits":
-[✓] (2, 2) matches (2, 2)
-[✓] all values close (atol: 1e-05)
All good, model saved at: onnx/model.onnx
```
Observe que, neste caso, os nomes de saída do modelo ajustado são `logits`
em vez do `last_hidden_state` que vimos com o checkpoint `distilbert/distilbert-base-uncased`
mais cedo. Isso é esperado, pois o modelo ajustado (fine-tuned) possui uma cabeça de classificação de sequência.
<Tip>
Os recursos que têm um sufixo `with-pass` (como `causal-lm-with-pass`) correspondem a
classes de modelo com estados ocultos pré-computados (chave e valores nos blocos de atenção)
que pode ser usado para decodificação autorregressiva rápida.
</Tip>
<Tip>
Para modelos do tipo `VisionEncoderDecoder`, as partes do codificador e do decodificador são
exportados separadamente como dois arquivos ONNX chamados `encoder_model.onnx` e `decoder_model.onnx` respectivamente.
</Tip>
## Exportando um modelo para uma arquitetura sem suporte
Se você deseja exportar um modelo cuja arquitetura não é suportada nativamente pela
biblioteca, há três etapas principais a seguir:
1. Implemente uma configuração ONNX personalizada.
2. Exporte o modelo para o ONNX.
3. Valide as saídas do PyTorch e dos modelos exportados.
Nesta seção, veremos como o DistilBERT foi implementado para mostrar o que está envolvido
em cada passo.
### Implementando uma configuração ONNX personalizada
Vamos começar com o objeto de configuração ONNX. Fornecemos três classes abstratas que
você deve herdar, dependendo do tipo de arquitetura de modelo que deseja exportar:
* Modelos baseados em codificador herdam de [`~onnx.config.OnnxConfig`]
* Modelos baseados em decodificador herdam de [`~onnx.config.OnnxConfigWithPast`]
* Os modelos codificador-decodificador herdam de [`~onnx.config.OnnxSeq2SeqConfigWithPast`]
<Tip>
Uma boa maneira de implementar uma configuração ONNX personalizada é observar as
implementação no arquivo `configuration_<model_name>.py` de uma arquitetura semelhante.
</Tip>
Como o DistilBERT é um modelo baseado em codificador, sua configuração é herdada de
`OnnxConfig`:
```python
>>> from typing import Mapping, OrderedDict
>>> from transformers.onnx import OnnxConfig
>>> class DistilBertOnnxConfig(OnnxConfig):
... @property
... def inputs(self) -> Mapping[str, Mapping[int, str]]:
... return OrderedDict(
... [
... ("input_ids", {0: "batch", 1: "sequence"}),
... ("attention_mask", {0: "batch", 1: "sequence"}),
... ]
... )
```
Todo objeto de configuração deve implementar a propriedade `inputs` e retornar um mapeamento,
onde cada chave corresponde a uma entrada esperada e cada valor indica o eixo
dessa entrada. Para o DistilBERT, podemos ver que duas entradas são necessárias: `input_ids` e
`attention_mask`. Essas entradas têm a mesma forma de `(batch_size, sequence_length)`
é por isso que vemos os mesmos eixos usados na configuração.
<Tip>
Notice that `inputs` property for `DistilBertOnnxConfig` returns an `OrderedDict`. This
ensures that the inputs are matched with their relative position within the
`PreTrainedModel.forward()` method when tracing the graph. We recommend using an
`OrderedDict` for the `inputs` and `outputs` properties when implementing custom ONNX
configurations.
Observe que a propriedade `inputs` para `DistilBertOnnxConfig` retorna um `OrderedDict`. Este
garante que as entradas sejam combinadas com sua posição relativa dentro do
método `PreTrainedModel.forward()` ao traçar o grafo. Recomendamos o uso de um
`OrderedDict` para as propriedades `inputs` e `outputs` ao implementar configurações personalizadas ONNX.
</Tip>
Depois de implementar uma configuração ONNX, você pode instanciá-la fornecendo a
configuração do modelo base da seguinte forma:
```python
>>> from transformers import AutoConfig
>>> config = AutoConfig.from_pretrained("distilbert/distilbert-base-uncased")
>>> onnx_config = DistilBertOnnxConfig(config)
```
O objeto resultante tem várias propriedades úteis. Por exemplo, você pode visualizar o conjunto de operadores ONNX
que será usado durante a exportação:
```python
>>> print(onnx_config.default_onnx_opset)
11
```
Você também pode visualizar as saídas associadas ao modelo da seguinte forma:
```python
>>> print(onnx_config.outputs)
OrderedDict([("last_hidden_state", {0: "batch", 1: "sequence"})])
```
Observe que a propriedade outputs segue a mesma estrutura das entradas; ele retorna um
`OrderedDict` de saídas nomeadas e suas formas. A estrutura de saída está ligada a
escolha do recurso com o qual a configuração é inicializada. Por padrão, a configuração do ONNX
é inicializada com o recurso `default` que corresponde à exportação de um
modelo carregado com a classe `AutoModel`. Se você deseja exportar um modelo para outra tarefa,
apenas forneça um recurso diferente para o argumento `task` quando você inicializar a configuração ONNX
. Por exemplo, se quisermos exportar o DistilBERT com uma sequência
de classificação, poderíamos usar:
```python
>>> from transformers import AutoConfig
>>> config = AutoConfig.from_pretrained("distilbert/distilbert-base-uncased")
>>> onnx_config_for_seq_clf = DistilBertOnnxConfig(config, task="sequence-classification")
>>> print(onnx_config_for_seq_clf.outputs)
OrderedDict([('logits', {0: 'batch'})])
```
<Tip>
Todas as propriedades e métodos básicos associados a [`~onnx.config.OnnxConfig`] e
as outras classes de configuração podem ser substituídas se necessário. Confira [`BartOnnxConfig`]
para um exemplo avançado.
</Tip>
### Exportando um modelo
Depois de ter implementado a configuração do ONNX, o próximo passo é exportar o modelo.
Aqui podemos usar a função `export()` fornecida pelo pacote `transformers.onnx`.
Esta função espera a configuração do ONNX, juntamente com o modelo base e o tokenizer,
e o caminho para salvar o arquivo exportado:
```python
>>> from pathlib import Path
>>> from transformers.onnx import export
>>> from transformers import AutoTokenizer, AutoModel
>>> onnx_path = Path("model.onnx")
>>> model_ckpt = "distilbert/distilbert-base-uncased"
>>> base_model = AutoModel.from_pretrained(model_ckpt)
>>> tokenizer = AutoTokenizer.from_pretrained(model_ckpt)
>>> onnx_inputs, onnx_outputs = export(tokenizer, base_model, onnx_config, onnx_config.default_onnx_opset, onnx_path)
```
Os `onnx_inputs` e `onnx_outputs` retornados pela função `export()` são listas de
chaves definidas nas propriedades `inputs` e `outputs` da configuração. Uma vez que o
modelo é exportado, você pode testar se o modelo está bem formado da seguinte forma:
```python
>>> import onnx
>>> onnx_model = onnx.load("model.onnx")
>>> onnx.checker.check_model(onnx_model)
```
<Tip>
Se o seu modelo for maior que 2GB, você verá que muitos arquivos adicionais são criados
durante a exportação. Isso é _esperado_ porque o ONNX usa [Protocol
Buffers](https://developers.google.com/protocol-buffers/) para armazenar o modelo e estes
têm um limite de tamanho de 2GB. Veja a [ONNX
documentação](https://github.com/onnx/onnx/blob/master/docs/ExternalData.md) para
instruções sobre como carregar modelos com dados externos.
</Tip>
### Validando a saída dos modelos
A etapa final é validar se as saídas do modelo base e exportado concordam
dentro de alguma tolerância absoluta. Aqui podemos usar a função `validate_model_outputs()`
fornecida pelo pacote `transformers.onnx` da seguinte forma:
```python
>>> from transformers.onnx import validate_model_outputs
>>> validate_model_outputs(
... onnx_config, tokenizer, base_model, onnx_path, onnx_outputs, onnx_config.atol_for_validation
... )
```
Esta função usa o método [`~transformers.onnx.OnnxConfig.generate_dummy_inputs`] para
gerar entradas para o modelo base e o exportado, e a tolerância absoluta pode ser
definida na configuração. Geralmente encontramos concordância numérica em 1e-6 a 1e-4
de alcance, embora qualquer coisa menor que 1e-3 provavelmente esteja OK.
## Contribuindo com uma nova configuração para 🤗 Transformers
Estamos procurando expandir o conjunto de configurações prontas e receber contribuições
da comunidade! Se você gostaria de contribuir para a biblioteca, você
precisará:
* Implemente a configuração do ONNX no arquivo `configuration_<model_name>.py` correspondente
Arquivo
* Incluir a arquitetura do modelo e recursos correspondentes em
[`~onnx.features.FeatureManager`]
* Adicione sua arquitetura de modelo aos testes em `test_onnx_v2.py`
Confira como ficou a configuração do [IBERT
](https://github.com/huggingface/transformers/pull/14868/files) para obter uma
idéia do que está envolvido.
| 0
|
mavonic_private_repos/transformers/docs/source
|
mavonic_private_repos/transformers/docs/source/pt/accelerate.md
|
<!--Copyright 2022 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
specific language governing permissions and limitations under the License.
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer.
-->
# Treinamento distribuído com o 🤗 Accelerate
O paralelismo surgiu como uma estratégia para treinar modelos grandes em hardware limitado e aumentar a velocidade
de treinamento em várias órdens de magnitude. Na Hugging Face criamos a biblioteca [🤗 Accelerate](https://huggingface.co/docs/accelerate)
para ajudar os usuários a treinar modelos 🤗 Transformers com qualquer configuração distribuída, seja em uma máquina
com múltiplos GPUs ou em múltiplos GPUs distribuidos entre muitas máquinas. Neste tutorial, você irá aprender como
personalizar seu laço de treinamento de PyTorch para poder treinar em ambientes distribuídos.
## Configuração
De início, instale o 🤗 Accelerate:
```bash
pip install accelerate
```
Logo, devemos importar e criar um objeto [`Accelerator`](https://huggingface.co/docs/accelerate/package_reference/accelerator#accelerate.Accelerator).
O `Accelerator` detectará automáticamente a configuração distribuída disponível e inicializará todos os
componentes necessários para o treinamento. Não há necessidade portanto de especificar o dispositivo onde deve colocar seu modelo.
```py
>>> from accelerate import Accelerator
>>> accelerator = Accelerator()
```
## Preparando a aceleração
Passe todos os objetos relevantes ao treinamento para o método [`prepare`](https://huggingface.co/docs/accelerate/package_reference/accelerator#accelerate.Accelerator.prepare).
Isto inclui os DataLoaders de treino e evaluação, um modelo e um otimizador:
```py
>>> train_dataloader, eval_dataloader, model, optimizer = accelerator.prepare(
... train_dataloader, eval_dataloader, model, optimizer
... )
```
## Backward
Por último, substitua o `loss.backward()` padrão em seu laço de treinamento com o método [`backward`](https://huggingface.co/docs/accelerate/package_reference/accelerator#accelerate.Accelerator.backward) do 🤗 Accelerate:
```py
>>> for epoch in range(num_epochs):
... for batch in train_dataloader:
... outputs = model(**batch)
... loss = outputs.loss
... accelerator.backward(loss)
... optimizer.step()
... lr_scheduler.step()
... optimizer.zero_grad()
... progress_bar.update(1)
```
Como se poder ver no seguinte código, só precisará adicionar quatro linhas de código ao seu laço de treinamento
para habilitar o treinamento distribuído!
```diff
+ from accelerate import Accelerator
from transformers import AdamW, AutoModelForSequenceClassification, get_scheduler
+ accelerator = Accelerator()
model = AutoModelForSequenceClassification.from_pretrained(checkpoint, num_labels=2)
optimizer = AdamW(model.parameters(), lr=3e-5)
- device = torch.device("cuda") if torch.cuda.is_available() else torch.device("cpu")
- model.to(device)
+ train_dataloader, eval_dataloader, model, optimizer = accelerator.prepare(
+ train_dataloader, eval_dataloader, model, optimizer
+ )
num_epochs = 3
num_training_steps = num_epochs * len(train_dataloader)
lr_scheduler = get_scheduler(
"linear",
optimizer=optimizer,
num_warmup_steps=0,
num_training_steps=num_training_steps
)
progress_bar = tqdm(range(num_training_steps))
model.train()
for epoch in range(num_epochs):
for batch in train_dataloader:
- batch = {k: v.to(device) for k, v in batch.items()}
outputs = model(**batch)
loss = outputs.loss
- loss.backward()
+ accelerator.backward(loss)
optimizer.step()
lr_scheduler.step()
optimizer.zero_grad()
progress_bar.update(1)
```
## Treinamento
Quando tiver adicionado as linhas de código relevantes, inicie o treinamento por um script ou notebook como o Colab.
### Treinamento em um Script
Se estiver rodando seu treinamento em um Script, execute o seguinte comando para criar e guardar um arquivo de configuração:
```bash
accelerate config
```
Comece o treinamento com:
```bash
accelerate launch train.py
```
### Treinamento em um Notebook
O 🤗 Accelerate pode rodar em um notebook, por exemplo, se estiver planejando usar as TPUs do Google Colab.
Encapsule o código responsável pelo treinamento de uma função e passe-o ao `notebook_launcher`:
```py
>>> from accelerate import notebook_launcher
>>> notebook_launcher(training_function)
```
Para obter mais informações sobre o 🤗 Accelerate e suas numerosas funções, consulte a [documentación](https://huggingface.co/docs/accelerate/index).
| 0
|
mavonic_private_repos/transformers/docs/source
|
mavonic_private_repos/transformers/docs/source/pt/installation.md
|
<!---
Copyright 2022 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer.
-->
# Guia de Instalação
Neste guia poderá encontrar informações para a instalação do 🤗 Transformers para qualquer biblioteca de
Machine Learning com a qual esteja a trabalhar. Além disso, poderá encontrar informações sobre como gerar cachês e
configurar o 🤗 Transformers para execução em modo offline (opcional).
🤗 Transformers foi testado com Python 3.6+, PyTorch 1.1.0+, TensorFlow 2.0+, e Flax. Para instalar a biblioteca de
deep learning com que deseja trabalhar, siga as instruções correspondentes listadas a seguir:
* [PyTorch](https://pytorch.org/get-started/locally/)
* [TensorFlow 2.0](https://www.tensorflow.org/install/pip)
* [Flax](https://flax.readthedocs.io/en/latest/)
## Instalação pelo Pip
É sugerido instalar o 🤗 Transformers num [ambiente virtual](https://docs.python.org/3/library/venv.html). Se precisar
de mais informações sobre ambientes virtuais em Python, consulte este [guia](https://packaging.python.org/guides/installing-using-pip-and-virtual-environments/).
Um ambiente virtual facilitará a manipulação e organização de projetos e evita problemas de compatibilidade entre dependências.
Comece criando um ambiente virtual no diretório do seu projeto:
```bash
python -m venv .env
```
E para ativar o ambiente virtual:
```bash
source .env/bin/activate
```
Agora É possível instalar o 🤗 Transformers com o comando a seguir:
```bash
pip install transformers
```
Somente para a CPU, é possível instalar o 🤗 Transformers e a biblioteca de deep learning respectiva apenas numa linha.
Por exemplo, para instalar o 🤗 Transformers e o PyTorch, digite:
```bash
pip install transformers[torch]
```
🤗 Transformers e TensorFlow 2.0:
```bash
pip install transformers[tf-cpu]
```
🤗 Transformers e Flax:
```bash
pip install transformers[flax]
```
Por último, verifique se o 🤗 Transformers foi instalado com sucesso usando o seguinte comando para baixar um modelo pré-treinado:
```bash
python -c "from transformers import pipeline; print(pipeline('sentiment-analysis')('we love you'))"
```
Em seguida, imprima um rótulo e sua pontuação:
```bash
[{'label': 'POSITIVE', 'score': 0.9998704791069031}]
```
## Instalação usando a fonte
Para instalar o 🤗 Transformers a partir da fonte use o seguinte comando:
```bash
pip install git+https://github.com/huggingface/transformers
```
O comando acima instalará a versão `master` mais atual em vez da última versão estável. A versão `master` é útil para
utilizar os últimos updates contidos em 🤗 Transformers. Por exemplo, um erro recente pode ter sido corrigido somente
após a última versão estável, antes que houvesse um novo lançamento. No entanto, há a possibilidade que a versão `master` não esteja estável.
A equipa trata de mantér a versão `master` operacional e a maioria dos erros são resolvidos em poucas horas ou dias.
Se encontrar quaisquer problemas, por favor abra um [Issue](https://github.com/huggingface/transformers/issues) para que o
mesmo possa ser corrigido o mais rápido possível.
Verifique que o 🤗 Transformers está instalado corretamente usando o seguinte comando:
```bash
python -c "from transformers import pipeline; print(pipeline('sentiment-analysis')('I love you'))"
```
## Instalação editável
Uma instalação editável será necessária caso desejas um dos seguintes:
* Usar a versão `master` do código fonte.
* Contribuir ao 🤗 Transformers e precisa testar mudanças ao código.
Para tal, clone o repositório e instale o 🤗 Transformers com os seguintes comandos:
```bash
git clone https://github.com/huggingface/transformers.git
cd transformers
pip install -e .
```
Estes comandos vão ligar o diretório para o qual foi clonado o repositório ao caminho de bibliotecas do Python.
O Python agora buscará dentro dos arquivos que foram clonados além dos caminhos normais da biblioteca.
Por exemplo, se os pacotes do Python se encontram instalados no caminho `~/anaconda3/envs/main/lib/python3.7/site-packages/`,
o Python também buscará módulos no diretório onde clonamos o repositório `~/transformers/`.
<Tip warning={true}>
É necessário manter o diretório `transformers` se desejas continuar usando a biblioteca.
</Tip>
Assim, É possível atualizar sua cópia local para com a última versão do 🤗 Transformers com o seguinte comando:
```bash
cd ~/transformers/
git pull
```
O ambiente de Python que foi criado para a instalação do 🤗 Transformers encontrará a versão `master` em execuções seguintes.
## Instalação usando o Conda
É possível instalar o 🤗 Transformers a partir do canal conda `conda-forge` com o seguinte comando:
```bash
conda install conda-forge::transformers
```
## Configuração do Cachê
Os modelos pré-treinados são baixados e armazenados no cachê local, encontrado em `~/.cache/huggingface/transformers/`.
Este é o diretório padrão determinado pela variável `TRANSFORMERS_CACHE` dentro do shell.
No Windows, este diretório pré-definido é dado por `C:\Users\username\.cache\huggingface\transformers`.
É possível mudar as variáveis dentro do shell em ordem de prioridade para especificar um diretório de cachê diferente:
1. Variável de ambiente do shell (por padrão): `TRANSFORMERS_CACHE`.
2. Variável de ambiente do shell:`HF_HOME` + `transformers/`.
3. Variável de ambiente do shell: `XDG_CACHE_HOME` + `/huggingface/transformers`.
<Tip>
O 🤗 Transformers usará as variáveis de ambiente do shell `PYTORCH_TRANSFORMERS_CACHE` ou `PYTORCH_PRETRAINED_BERT_CACHE`
se estiver vindo de uma versão anterior da biblioteca que tenha configurado essas variáveis de ambiente, a menos que
você especifique a variável de ambiente do shell `TRANSFORMERS_CACHE`.
</Tip>
## Modo Offline
O 🤗 Transformers também pode ser executado num ambiente de firewall ou fora da rede (offline) usando arquivos locais.
Para tal, configure a variável de ambiente de modo que `TRANSFORMERS_OFFLINE=1`.
<Tip>
Você pode adicionar o [🤗 Datasets](https://huggingface.co/docs/datasets/) ao pipeline de treinamento offline declarando
a variável de ambiente `HF_DATASETS_OFFLINE=1`.
</Tip>
Segue um exemplo de execução do programa numa rede padrão com firewall para instâncias externas, usando o seguinte comando:
```bash
python examples/pytorch/translation/run_translation.py --model_name_or_path google-t5/t5-small --dataset_name wmt16 --dataset_config ro-en ...
```
Execute esse mesmo programa numa instância offline com o seguinte comando:
```bash
HF_DATASETS_OFFLINE=1 TRANSFORMERS_OFFLINE=1 \
python examples/pytorch/translation/run_translation.py --model_name_or_path google-t5/t5-small --dataset_name wmt16 --dataset_config ro-en ...
```
O script agora deve ser executado sem travar ou expirar, pois procurará apenas por arquivos locais.
### Obtendo modelos e tokenizers para uso offline
Outra opção para usar o 🤗 Transformers offline é baixar os arquivos antes e depois apontar para o caminho local onde estão localizados. Existem três maneiras de fazer isso:
* Baixe um arquivo por meio da interface de usuário do [Model Hub](https://huggingface.co/models) clicando no ícone ↓.
* Use o pipeline do [`PreTrainedModel.from_pretrained`] e [`PreTrainedModel.save_pretrained`]:
1. Baixa os arquivos previamente com [`PreTrainedModel.from_pretrained`]:
```py
>>> from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
>>> tokenizer = AutoTokenizer.from_pretrained("bigscience/T0_3B")
>>> model = AutoModelForSeq2SeqLM.from_pretrained("bigscience/T0_3B")
```
2. Salve os arquivos em um diretório específico com [`PreTrainedModel.save_pretrained`]:
```py
>>> tokenizer.save_pretrained("./your/path/bigscience_t0")
>>> model.save_pretrained("./your/path/bigscience_t0")
```
3. Quando estiver offline, acesse os arquivos com [`PreTrainedModel.from_pretrained`] do diretório especificado:
```py
>>> tokenizer = AutoTokenizer.from_pretrained("./your/path/bigscience_t0")
>>> model = AutoModel.from_pretrained("./your/path/bigscience_t0")
```
* Baixando arquivos programaticamente com a biblioteca [huggingface_hub](https://github.com/huggingface/huggingface_hub/tree/main/src/huggingface_hub):
1. Instale a biblioteca [huggingface_hub](https://github.com/huggingface/huggingface_hub/tree/main/src/huggingface_hub) em seu ambiente virtual:
```bash
python -m pip install huggingface_hub
```
2. Utiliza a função [`hf_hub_download`](https://huggingface.co/docs/hub/adding-a-library#download-files-from-the-hub) para baixar um arquivo para um caminho específico. Por exemplo, o comando a seguir baixará o arquivo `config.json` para o modelo [T0](https://huggingface.co/bigscience/T0_3B) no caminho desejado:
```py
>>> from huggingface_hub import hf_hub_download
>>> hf_hub_download(repo_id="bigscience/T0_3B", filename="config.json", cache_dir="./your/path/bigscience_t0")
```
Depois que o arquivo for baixado e armazenado no cachê local, especifique seu caminho local para carregá-lo e usá-lo:
```py
>>> from transformers import AutoConfig
>>> config = AutoConfig.from_pretrained("./your/path/bigscience_t0/config.json")
```
<Tip>
Para obter mais detalhes sobre como baixar arquivos armazenados no Hub, consulte a seção [How to download files from the Hub](https://huggingface.co/docs/hub/how-to-downstream).
</Tip>
| 0
|
mavonic_private_repos/transformers/docs/source
|
mavonic_private_repos/transformers/docs/source/pt/quicktour.md
|
<!--Copyright 2022 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
specific language governing permissions and limitations under the License.
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer.
-->
# Tour rápido
[[open-in-colab]]
Comece a trabalhar com 🤗 Transformers! Comece usando [`pipeline`] para rápida inferência e facilmente carregue um modelo pré-treinado e um tokenizer com [AutoClass](./model_doc/auto) para resolver tarefas de texto, visão ou áudio.
<Tip>
Todos os exemplos de código apresentados na documentação têm um botão no canto superior direito para escolher se você deseja ocultar ou mostrar o código no Pytorch ou no TensorFlow. Caso contrário, é esperado que funcione para ambos back-ends sem nenhuma alteração.
</Tip>
## Pipeline
[`pipeline`] é a maneira mais fácil de usar um modelo pré-treinado para uma dada tarefa.
<Youtube id="tiZFewofSLM"/>
A [`pipeline`] apoia diversas tarefas fora da caixa:
**Texto**:
* Análise sentimental: classifica a polaridade de um texto.
* Geração de texto (em Inglês): gera texto a partir de uma entrada.
* Reconhecimento de entidade mencionada: legenda cada palavra com uma classe que a representa (pessoa, data, local, etc...)
* Respostas: extrai uma resposta dado algum contexto e uma questão
* Máscara de preenchimento: preenche o espaço, dado um texto com máscaras de palavras.
* Sumarização: gera o resumo de um texto longo ou documento.
* Tradução: traduz texto para outra língua.
* Extração de características: cria um tensor que representa o texto.
**Imagem**:
* Classificação de imagens: classifica uma imagem.
* Segmentação de imagem: classifica cada pixel da imagem.
* Detecção de objetos: detecta objetos em uma imagem.
**Audio**:
* Classficação de áudio: legenda um trecho de áudio fornecido.
* Reconhecimento de fala automático: transcreve audio em texto.
<Tip>
Para mais detalhes sobre a [`pipeline`] e tarefas associadas, siga a documentação [aqui](./main_classes/pipelines).
</Tip>
### Uso da pipeline
No exemplo a seguir, você usará [`pipeline`] para análise sentimental.
Instale as seguintes dependências se você ainda não o fez:
<frameworkcontent>
<pt>
```bash
pip install torch
```
</pt>
<tf>
```bash
pip install tensorflow
```
</tf>
</frameworkcontent>
Importe [`pipeline`] e especifique a tarefa que deseja completar:
```py
>>> from transformers import pipeline
>>> classifier = pipeline("sentiment-analysis")
```
A pipeline baixa and armazena um [modelo pré-treinado](https://huggingface.co/distilbert/distilbert-base-uncased-finetuned-sst-2-english) padrão e tokenizer para análise sentimental. Agora você pode usar `classifier` no texto alvo:
```py
>>> classifier("We are very happy to show you the 🤗 Transformers library.")
[{'label': 'POSITIVE', 'score': 0.9998}]
```
Para mais de uma sentença, passe uma lista para a [`pipeline`], a qual retornará uma lista de dicionários:
```py
>>> results = classifier(["We are very happy to show you the 🤗 Transformers library.", "We hope you don't hate it."])
>>> for result in results:
... print(f"label: {result['label']}, with score: {round(result['score'], 4)}")
label: POSITIVE, with score: 0.9998
label: NEGATIVE, with score: 0.5309
```
A [`pipeline`] também pode iterar sobre um Dataset inteiro. Comece instalando a biblioteca de [🤗 Datasets](https://huggingface.co/docs/datasets/):
```bash
pip install datasets
```
Crie uma [`pipeline`] com a tarefa que deseja resolver e o modelo que deseja usar.
```py
>>> import torch
>>> from transformers import pipeline
>>> speech_recognizer = pipeline("automatic-speech-recognition", model="facebook/wav2vec2-base-960h")
```
A seguir, carregue uma base de dados (confira a 🤗 [Iniciação em Datasets](https://huggingface.co/docs/datasets/quickstart) para mais detalhes) que você gostaria de iterar sobre. Por exemplo, vamos carregar o dataset [MInDS-14](https://huggingface.co/datasets/PolyAI/minds14):
```py
>>> from datasets import load_dataset, Audio
>>> dataset = load_dataset("PolyAI/minds14", name="en-US", split="train") # doctest: +IGNORE_RESULT
```
Precisamos garantir que a taxa de amostragem do conjunto de dados corresponda à taxa de amostragem em que o facebook/wav2vec2-base-960h foi treinado.
```py
>>> dataset = dataset.cast_column("audio", Audio(sampling_rate=speech_recognizer.feature_extractor.sampling_rate))
```
Os arquivos de áudio são carregados e re-amostrados automaticamente ao chamar a coluna `"audio"`.
Vamos extrair as arrays de formas de onda originais das primeiras 4 amostras e passá-las como uma lista para o pipeline:
```py
>>> result = speech_recognizer(dataset[:4]["audio"])
>>> print([d["text"] for d in result])
['I WOULD LIKE TO SET UP A JOINT ACCOUNT WITH MY PARTNER HOW DO I PROCEED WITH DOING THAT', "FONDERING HOW I'D SET UP A JOIN TO HET WITH MY WIFE AND WHERE THE AP MIGHT BE", "I I'D LIKE TOY SET UP A JOINT ACCOUNT WITH MY PARTNER I'M NOT SEEING THE OPTION TO DO IT ON THE APSO I CALLED IN TO GET SOME HELP CAN I JUST DO IT OVER THE PHONE WITH YOU AND GIVE YOU THE INFORMATION OR SHOULD I DO IT IN THE AP AND I'M MISSING SOMETHING UQUETTE HAD PREFERRED TO JUST DO IT OVER THE PHONE OF POSSIBLE THINGS", 'HOW DO I TURN A JOIN A COUNT']
```
Para um conjunto de dados maior onde as entradas são maiores (como em fala ou visão), será necessário passar um gerador em vez de uma lista que carregue todas as entradas na memória. Consulte a [documentação do pipeline](./main_classes/pipelines) para mais informações.
### Use outro modelo e tokenizer na pipeline
A [`pipeline`] pode acomodar qualquer modelo do [Model Hub](https://huggingface.co/models), facilitando sua adaptação para outros casos de uso. Por exemplo, se você quiser um modelo capaz de lidar com texto em francês, use as tags no Model Hub para filtrar um modelo apropriado. O principal resultado filtrado retorna um [modelo BERT](https://huggingface.co/nlptown/bert-base-multilingual-uncased-sentiment) bilíngue ajustado para análise de sentimentos. Ótimo, vamos usar este modelo!
```py
>>> model_name = "nlptown/bert-base-multilingual-uncased-sentiment"
```
<frameworkcontent>
<pt>
Use o [`AutoModelForSequenceClassification`] e [`AutoTokenizer`] para carregar o modelo pré-treinado e seu tokenizer associado (mais em `AutoClass` abaixo):
```py
>>> from transformers import AutoTokenizer, AutoModelForSequenceClassification
>>> model = AutoModelForSequenceClassification.from_pretrained(model_name)
>>> tokenizer = AutoTokenizer.from_pretrained(model_name)
```
</pt>
<tf>
Use o [`TFAutoModelForSequenceClassification`] and [`AutoTokenizer`] para carregar o modelo pré-treinado e o tokenizer associado (mais em `TFAutoClass` abaixo):
```py
>>> from transformers import AutoTokenizer, TFAutoModelForSequenceClassification
>>> model = TFAutoModelForSequenceClassification.from_pretrained(model_name)
>>> tokenizer = AutoTokenizer.from_pretrained(model_name)
```
</tf>
</frameworkcontent>
Então você pode especificar o modelo e o tokenizador na [`pipeline`] e aplicar o `classifier` no seu texto alvo:
```py
>>> classifier = pipeline("sentiment-analysis", model=model, tokenizer=tokenizer)
>>> classifier("Nous sommes très heureux de vous présenter la bibliothèque 🤗 Transformers.")
[{'label': '5 stars', 'score': 0.7273}]
```
Se você não conseguir achar um modelo para o seu caso de uso, precisará usar fine-tune em um modelo pré-treinado nos seus dados. Veja nosso [tutorial de fine-tuning](./training) para descobrir como. Finalmente, depois que você tiver usado esse processo em seu modelo, considere compartilhá-lo conosco (veja o tutorial [aqui](./model_sharing)) na plataforma Model Hub afim de democratizar NLP! 🤗
## AutoClass
<Youtube id="AhChOFRegn4"/>
Por baixo dos panos, as classes [`AutoModelForSequenceClassification`] e [`AutoTokenizer`] trabalham juntas para fortificar o [`pipeline`]. Um [AutoClass](./model_doc/auto) é um atalho que automaticamente recupera a arquitetura de um modelo pré-treinado a partir de seu nome ou caminho. Basta selecionar a `AutoClass` apropriada para sua tarefa e seu tokenizer associado com [`AutoTokenizer`].
Vamos voltar ao nosso exemplo e ver como você pode usar a `AutoClass` para replicar os resultados do [`pipeline`].
### AutoTokenizer
Um tokenizer é responsável por pré-processar o texto em um formato que seja compreensível para o modelo. Primeiro, o tokenizer dividirá o texto em palavras chamadas *tokens*. Existem várias regras que regem o processo de tokenização, incluindo como dividir uma palavra e em que nível (saiba mais sobre tokenização [aqui](./tokenizer_summary)). A coisa mais importante a lembrar, porém, é que você precisa instanciar o tokenizer com o mesmo nome do modelo para garantir que está usando as mesmas regras de tokenização com as quais um modelo foi pré-treinado.
Carregue um tokenizer com [`AutoTokenizer`]:
```py
>>> from transformers import AutoTokenizer
>>> model_name = "nlptown/bert-base-multilingual-uncased-sentiment"
>>> tokenizer = AutoTokenizer.from_pretrained(model_name)
```
Em seguida, o tokenizer converte os tokens em números para construir um tensor como entrada para o modelo. Isso é conhecido como o *vocabulário* do modelo.
Passe o texto para o tokenizer:
```py
>>> encoding = tokenizer("We are very happy to show you the 🤗 Transformers library.")
>>> print(encoding)
{'input_ids': [101, 11312, 10320, 12495, 19308, 10114, 11391, 10855, 10103, 100, 58263, 13299, 119, 102],
'token_type_ids': [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
'attention_mask': [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]}
```
O tokenizer retornará um dicionário contendo:
* [input_ids](./glossary#input-ids): representações numéricas de seus tokens.
* [atttention_mask](.glossary#attention-mask): indica quais tokens devem ser atendidos.
Assim como o [`pipeline`], o tokenizer aceitará uma lista de entradas. Além disso, o tokenizer também pode preencher e truncar o texto para retornar um lote com comprimento uniforme:
<frameworkcontent>
<pt>
```py
>>> pt_batch = tokenizer(
... ["We are very happy to show you the 🤗 transformers library.", "We hope you don't hate it."],
... padding=True,
... truncation=True,
... max_length=512,
... return_tensors="pt",
... )
```
</pt>
<tf>
```py
>>> tf_batch = tokenizer(
... ["We are very happy to show you the 🤗 Transformers library.", "We hope you don't hate it."],
... padding=True,
... truncation=True,
... max_length=512,
... return_tensors="tf",
... )
```
</tf>
</frameworkcontent>
Leia o tutorial de [pré-processamento](./pré-processamento) para obter mais detalhes sobre tokenização.
### AutoModel
<frameworkcontent>
<pt>
🤗 Transformers fornecem uma maneira simples e unificada de carregar instâncias pré-treinadas. Isso significa que você pode carregar um [`AutoModel`] como carregaria um [`AutoTokenizer`]. A única diferença é selecionar o [`AutoModel`] correto para a tarefa. Como você está fazendo classificação de texto ou sequência, carregue [`AutoModelForSequenceClassification`]:
```py
>>> from transformers import AutoModelForSequenceClassification
>>> model_name = "nlptown/bert-base-multilingual-uncased-sentiment"
>>> pt_model = AutoModelForSequenceClassification.from_pretrained(model_name)
```
<Tip>
Veja o [sumário de tarefas](./task_summary) para qual classe de [`AutoModel`] usar para cada tarefa.
</Tip>
Agora você pode passar seu grupo de entradas pré-processadas diretamente para o modelo. Você apenas tem que descompactar o dicionário usando `**`:
```py
>>> pt_outputs = pt_model(**pt_batch)
```
O modelo gera as ativações finais no atributo `logits`. Aplique a função softmax aos `logits` para recuperar as probabilidades:
```py
>>> from torch import nn
>>> pt_predictions = nn.functional.softmax(pt_outputs.logits, dim=-1)
>>> print(pt_predictions)
tensor([[0.0021, 0.0018, 0.0115, 0.2121, 0.7725],
[0.2084, 0.1826, 0.1969, 0.1755, 0.2365]], grad_fn=<SoftmaxBackward0>)
```
</pt>
<tf>
🤗 Transformers fornecem uma maneira simples e unificada de carregar instâncias pré-treinadas. Isso significa que você pode carregar um [`TFAutoModel`] como carregaria um [`AutoTokenizer`]. A única diferença é selecionar o [`TFAutoModel`] correto para a tarefa. Como você está fazendo classificação de texto ou sequência, carregue [`TFAutoModelForSequenceClassification`]:
```py
>>> from transformers import TFAutoModelForSequenceClassification
>>> model_name = "nlptown/bert-base-multilingual-uncased-sentiment"
>>> tf_model = TFAutoModelForSequenceClassification.from_pretrained(model_name)
```
<Tip>
Veja o [sumário de tarefas](./task_summary) para qual classe de [`AutoModel`] usar para cada tarefa.
</Tip>
Agora você pode passar seu grupo de entradas pré-processadas diretamente para o modelo através da passagem de chaves de dicionários ao tensor.
```py
>>> tf_outputs = tf_model(tf_batch)
```
O modelo gera as ativações finais no atributo `logits`. Aplique a função softmax aos `logits` para recuperar as probabilidades:
```py
>>> import tensorflow as tf
>>> tf_predictions = tf.nn.softmax(tf_outputs.logits, axis=-1)
>>> tf_predictions # doctest: +IGNORE_RESULT
```
</tf>
</frameworkcontent>
<Tip>
Todos os modelos de 🤗 Transformers (PyTorch ou TensorFlow) geram tensores *antes* da função de ativação final (como softmax) pois essa função algumas vezes é fundida com a perda.
</Tip>
Os modelos são um standard [`torch.nn.Module`](https://pytorch.org/docs/stable/nn.html#torch.nn.Module) ou um [`tf.keras.Model`](https://www.tensorflow.org/api_docs/python/tf/keras/Model) para que você possa usá-los em seu loop de treinamento habitual. No entanto, para facilitar as coisas, 🤗 Transformers fornece uma classe [`Trainer`] para PyTorch que adiciona funcionalidade para treinamento distribuído, precisão mista e muito mais. Para o TensorFlow, você pode usar o método `fit` de [Keras](https://keras.io/). Consulte o [tutorial de treinamento](./training) para obter mais detalhes.
<Tip>
As saídas do modelo 🤗 Transformers são classes de dados especiais para que seus atributos sejam preenchidos automaticamente em um IDE.
As saídas do modelo também se comportam como uma tupla ou um dicionário (por exemplo, você pode indexar com um inteiro, uma parte ou uma string), caso em que os atributos `None` são ignorados.
</Tip>
### Salvar um modelo
<frameworkcontent>
<pt>
Uma vez que seu modelo estiver afinado, você pode salvá-lo com seu Tokenizer usando [`PreTrainedModel.save_pretrained`]:
```py
>>> pt_save_directory = "./pt_save_pretrained"
>>> tokenizer.save_pretrained(pt_save_directory) # doctest: +IGNORE_RESULT
>>> pt_model.save_pretrained(pt_save_directory)
```
Quando você estiver pronto para usá-lo novamente, recarregue com [`PreTrainedModel.from_pretrained`]:
```py
>>> pt_model = AutoModelForSequenceClassification.from_pretrained("./pt_save_pretrained")
```
</pt>
<tf>
Uma vez que seu modelo estiver afinado, você pode salvá-lo com seu Tokenizer usando [`TFPreTrainedModel.save_pretrained`]:
```py
>>> tf_save_directory = "./tf_save_pretrained"
>>> tokenizer.save_pretrained(tf_save_directory) # doctest: +IGNORE_RESULT
>>> tf_model.save_pretrained(tf_save_directory)
```
Quando você estiver pronto para usá-lo novamente, recarregue com [`TFPreTrainedModel.from_pretrained`]
```py
>>> tf_model = TFAutoModelForSequenceClassification.from_pretrained("./tf_save_pretrained")
```
</tf>
</frameworkcontent>
Um recurso particularmente interessante dos 🤗 Transformers é a capacidade de salvar um modelo e recarregá-lo como um modelo PyTorch ou TensorFlow. Use `from_pt` ou `from_tf` para converter o modelo de um framework para outro:
<frameworkcontent>
<pt>
```py
>>> from transformers import AutoModel
>>> tokenizer = AutoTokenizer.from_pretrained(tf_save_directory)
>>> pt_model = AutoModelForSequenceClassification.from_pretrained(tf_save_directory, from_tf=True)
```
</pt>
<tf>
```py
>>> from transformers import TFAutoModel
>>> tokenizer = AutoTokenizer.from_pretrained(pt_save_directory)
>>> tf_model = TFAutoModelForSequenceClassification.from_pretrained(pt_save_directory, from_pt=True)
```
</tf>
</frameworkcontent>
| 0
|
mavonic_private_repos/transformers/docs/source
|
mavonic_private_repos/transformers/docs/source/pt/converting_tensorflow_models.md
|
<!--Copyright 2020 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
specific language governing permissions and limitations under the License.
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer.
-->
# Convertendo checkpoints do TensorFlow para Pytorch
Uma interface de linha de comando é fornecida para converter os checkpoints originais Bert/GPT/GPT-2/Transformer-XL/XLNet/XLM em modelos
que podem ser carregados usando os métodos `from_pretrained` da biblioteca.
<Tip>
A partir da versão 2.3.0 o script de conversão agora faz parte do transformers CLI (**transformers-cli**) disponível em qualquer instalação
transformers >= 2.3.0.
A documentação abaixo reflete o formato do comando **transformers-cli convert**.
</Tip>
## BERT
Você pode converter qualquer checkpoint do BERT em TensorFlow (em particular [os modelos pré-treinados lançados pelo Google](https://github.com/google-research/bert#pre-trained-models)) em um arquivo PyTorch usando um
[convert_bert_original_tf_checkpoint_to_pytorch.py](https://github.com/huggingface/transformers/tree/main/src/transformers/models/bert/convert_bert_original_tf_checkpoint_to_pytorch.py) script.
Esta Interface de Linha de Comando (CLI) recebe como entrada um checkpoint do TensorFlow (três arquivos começando com `bert_model.ckpt`) e o
arquivo de configuração (`bert_config.json`), e então cria um modelo PyTorch para esta configuração, carrega os pesos
do checkpoint do TensorFlow no modelo PyTorch e salva o modelo resultante em um arquivo PyTorch que pode
ser importado usando `from_pretrained()` (veja o exemplo em [quicktour](quicktour) , [run_glue.py](https://github.com/huggingface/transformers/tree/main/examples/pytorch/text-classification/run_glue.py) ).
Você só precisa executar este script de conversão **uma vez** para obter um modelo PyTorch. Você pode então desconsiderar o checkpoint em
TensorFlow (os três arquivos começando com `bert_model.ckpt`), mas certifique-se de manter o arquivo de configuração (\
`bert_config.json`) e o arquivo de vocabulário (`vocab.txt`), pois eles também são necessários para o modelo PyTorch.
Para executar este script de conversão específico, você precisará ter o TensorFlow e o PyTorch instalados (`pip install tensorflow`). O resto do repositório requer apenas o PyTorch.
Aqui está um exemplo do processo de conversão para um modelo `BERT-Base Uncased` pré-treinado:
```bash
export BERT_BASE_DIR=/path/to/bert/uncased_L-12_H-768_A-12
transformers-cli convert --model_type bert \
--tf_checkpoint $BERT_BASE_DIR/bert_model.ckpt \
--config $BERT_BASE_DIR/bert_config.json \
--pytorch_dump_output $BERT_BASE_DIR/pytorch_model.bin
```
Você pode baixar os modelos pré-treinados do Google para a conversão [aqui](https://github.com/google-research/bert#pre-trained-models).
## ALBERT
Converta os checkpoints do modelo ALBERT em TensorFlow para PyTorch usando o
[convert_albert_original_tf_checkpoint_to_pytorch.py](https://github.com/huggingface/transformers/tree/main/src/transformers/models/albert/convert_albert_original_tf_checkpoint_to_pytorch.py) script.
A Interface de Linha de Comando (CLI) recebe como entrada um checkpoint do TensorFlow (três arquivos começando com `model.ckpt-best`) e o
arquivo de configuração (`albert_config.json`), então cria e salva um modelo PyTorch. Para executar esta conversão, você
precisa ter o TensorFlow e o PyTorch instalados.
Aqui está um exemplo do processo de conversão para o modelo `ALBERT Base` pré-treinado:
```bash
export ALBERT_BASE_DIR=/path/to/albert/albert_base
transformers-cli convert --model_type albert \
--tf_checkpoint $ALBERT_BASE_DIR/model.ckpt-best \
--config $ALBERT_BASE_DIR/albert_config.json \
--pytorch_dump_output $ALBERT_BASE_DIR/pytorch_model.bin
```
Você pode baixar os modelos pré-treinados do Google para a conversão [aqui](https://github.com/google-research/albert#pre-trained-models).
## OpenAI GPT
Aqui está um exemplo do processo de conversão para um modelo OpenAI GPT pré-treinado, supondo que seu checkpoint NumPy
foi salvo com o mesmo formato do modelo pré-treinado OpenAI (veja [aqui](https://github.com/openai/finetune-transformer-lm)\
)
```bash
export OPENAI_GPT_CHECKPOINT_FOLDER_PATH=/path/to/openai/pretrained/numpy/weights
transformers-cli convert --model_type gpt \
--tf_checkpoint $OPENAI_GPT_CHECKPOINT_FOLDER_PATH \
--pytorch_dump_output $PYTORCH_DUMP_OUTPUT \
[--config OPENAI_GPT_CONFIG] \
[--finetuning_task_name OPENAI_GPT_FINETUNED_TASK] \
```
## OpenAI GPT-2
Aqui está um exemplo do processo de conversão para um modelo OpenAI GPT-2 pré-treinado (consulte [aqui](https://github.com/openai/gpt-2))
```bash
export OPENAI_GPT2_CHECKPOINT_PATH=/path/to/openai-community/gpt2/pretrained/weights
transformers-cli convert --model_type gpt2 \
--tf_checkpoint $OPENAI_GPT2_CHECKPOINT_PATH \
--pytorch_dump_output $PYTORCH_DUMP_OUTPUT \
[--config OPENAI_GPT2_CONFIG] \
[--finetuning_task_name OPENAI_GPT2_FINETUNED_TASK]
```
## XLNet
Aqui está um exemplo do processo de conversão para um modelo XLNet pré-treinado:
```bash
export TRANSFO_XL_CHECKPOINT_PATH=/path/to/xlnet/checkpoint
export TRANSFO_XL_CONFIG_PATH=/path/to/xlnet/config
transformers-cli convert --model_type xlnet \
--tf_checkpoint $TRANSFO_XL_CHECKPOINT_PATH \
--config $TRANSFO_XL_CONFIG_PATH \
--pytorch_dump_output $PYTORCH_DUMP_OUTPUT \
[--finetuning_task_name XLNET_FINETUNED_TASK] \
```
## XLM
Aqui está um exemplo do processo de conversão para um modelo XLM pré-treinado:
```bash
export XLM_CHECKPOINT_PATH=/path/to/xlm/checkpoint
transformers-cli convert --model_type xlm \
--tf_checkpoint $XLM_CHECKPOINT_PATH \
--pytorch_dump_output $PYTORCH_DUMP_OUTPUT
[--config XML_CONFIG] \
[--finetuning_task_name XML_FINETUNED_TASK]
```
## T5
Aqui está um exemplo do processo de conversão para um modelo T5 pré-treinado:
```bash
export T5=/path/to/t5/uncased_L-12_H-768_A-12
transformers-cli convert --model_type t5 \
--tf_checkpoint $T5/t5_model.ckpt \
--config $T5/t5_config.json \
--pytorch_dump_output $T5/pytorch_model.bin
```
| 0
|
mavonic_private_repos/transformers/docs/source/pt
|
mavonic_private_repos/transformers/docs/source/pt/tasks/token_classification.md
|
<!--Copyright 2022 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
specific language governing permissions and limitations under the License.
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer.
-->
# Classificação de tokens
<Youtube id="wVHdVlPScxA"/>
A classificação de tokens atribui um rótulo a tokens individuais em uma frase. Uma das tarefas de classificação de tokens mais comuns é o Reconhecimento de Entidade Nomeada, também chamada de NER (sigla em inglês para Named Entity Recognition). O NER tenta encontrar um rótulo para cada entidade em uma frase, como uma pessoa, local ou organização.
Este guia mostrará como realizar o fine-tuning do [DistilBERT](https://huggingface.co/distilbert/distilbert-base-uncased) no conjunto de dados [WNUT 17](https://huggingface.co/datasets/wnut_17) para detectar novas entidades.
<Tip>
Consulte a [página de tarefas de classificação de tokens](https://huggingface.co/tasks/token-classification) para obter mais informações sobre outras formas de classificação de tokens e seus modelos, conjuntos de dados e métricas associadas.
</Tip>
## Carregando o conjunto de dados WNUT 17
Carregue o conjunto de dados WNUT 17 da biblioteca 🤗 Datasets:
```py
>>> from datasets import load_dataset
>>> wnut = load_dataset("wnut_17")
```
E dê uma olhada em um exemplo:
```py
>>> wnut["train"][0]
{'id': '0',
'ner_tags': [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 7, 8, 8, 0, 7, 0, 0, 0, 0, 0, 0, 0, 0],
'tokens': ['@paulwalk', 'It', "'s", 'the', 'view', 'from', 'where', 'I', "'m", 'living', 'for', 'two', 'weeks', '.', 'Empire', 'State', 'Building', '=', 'ESB', '.', 'Pretty', 'bad', 'storm', 'here', 'last', 'evening', '.']
}
```
Cada número em `ner_tags` representa uma entidade. Converta o número em um rótulo para obter mais informações:
```py
>>> label_list = wnut["train"].features[f"ner_tags"].feature.names
>>> label_list
[
"O",
"B-corporation",
"I-corporation",
"B-creative-work",
"I-creative-work",
"B-group",
"I-group",
"B-location",
"I-location",
"B-person",
"I-person",
"B-product",
"I-product",
]
```
O `ner_tag` descreve uma entidade, como uma organização, local ou pessoa. A letra que prefixa cada `ner_tag` indica a posição do token da entidade:
- `B-` indica o início de uma entidade.
- `I-` indica que um token está contido dentro da mesma entidade (por exemplo, o token `State` pode fazer parte de uma entidade como `Empire State Building`).
- `0` indica que o token não corresponde a nenhuma entidade.
## Pré-processamento
<Youtube id="iY2AZYdZAr0"/>
Carregue o tokenizer do DistilBERT para processar os `tokens`:
```py
>>> from transformers import AutoTokenizer
>>> tokenizer = AutoTokenizer.from_pretrained("distilbert/distilbert-base-uncased")
```
Como a entrada já foi dividida em palavras, defina `is_split_into_words=True` para tokenizar as palavras em subpalavras:
```py
>>> tokenized_input = tokenizer(example["tokens"], is_split_into_words=True)
>>> tokens = tokenizer.convert_ids_to_tokens(tokenized_input["input_ids"])
>>> tokens
['[CLS]', '@', 'paul', '##walk', 'it', "'", 's', 'the', 'view', 'from', 'where', 'i', "'", 'm', 'living', 'for', 'two', 'weeks', '.', 'empire', 'state', 'building', '=', 'es', '##b', '.', 'pretty', 'bad', 'storm', 'here', 'last', 'evening', '.', '[SEP]']
```
Ao adicionar os tokens especiais `[CLS]` e `[SEP]` e a tokenização de subpalavras uma incompatibilidade é gerada entre a entrada e os rótulos. Uma única palavra correspondente a um único rótulo pode ser dividida em duas subpalavras. Você precisará realinhar os tokens e os rótulos da seguinte forma:
1. Mapeie todos os tokens para a palavra correspondente com o método [`word_ids`](https://huggingface.co/docs/tokenizers/python/latest/api/reference.html#tokenizers.Encoding.word_ids).
2. Atribuindo o rótulo `-100` aos tokens especiais `[CLS]` e `[SEP]` para que a função de loss do PyTorch ignore eles.
3. Rotular apenas o primeiro token de uma determinada palavra. Atribuindo `-100` a outros subtokens da mesma palavra.
Aqui está como você pode criar uma função para realinhar os tokens e rótulos e truncar sequências para não serem maiores que o comprimento máximo de entrada do DistilBERT:
```py
>>> def tokenize_and_align_labels(examples):
... tokenized_inputs = tokenizer(examples["tokens"], truncation=True, is_split_into_words=True)
... labels = []
... for i, label in enumerate(examples[f"ner_tags"]):
... word_ids = tokenized_inputs.word_ids(batch_index=i) # Map tokens to their respective word.
... previous_word_idx = None
... label_ids = []
... for word_idx in word_ids: # Set the special tokens to -100.
... if word_idx is None:
... label_ids.append(-100)
... elif word_idx != previous_word_idx: # Only label the first token of a given word.
... label_ids.append(label[word_idx])
... else:
... label_ids.append(-100)
... previous_word_idx = word_idx
... labels.append(label_ids)
... tokenized_inputs["labels"] = labels
... return tokenized_inputs
```
Use a função [`map`](https://huggingface.co/docs/datasets/process#map) do 🤗 Datasets para tokenizar e alinhar os rótulos em todo o conjunto de dados. Você pode acelerar a função `map` configurando `batched=True` para processar vários elementos do conjunto de dados de uma só vez:
```py
>>> tokenized_wnut = wnut.map(tokenize_and_align_labels, batched=True)
```
Use o [`DataCollatorForTokenClassification`] para criar um batch de exemplos. Ele também *preencherá dinamicamente* seu texto e rótulos para o comprimento do elemento mais longo em seu batch, para que tenham um comprimento uniforme. Embora seja possível preencher seu texto na função `tokenizer` configurando `padding=True`, o preenchimento dinâmico é mais eficiente.
<frameworkcontent>
<pt>
```py
>>> from transformers import DataCollatorForTokenClassification
>>> data_collator = DataCollatorForTokenClassification(tokenizer=tokenizer)
```
</pt>
<tf>
```py
>>> from transformers import DataCollatorForTokenClassification
>>> data_collator = DataCollatorForTokenClassification(tokenizer=tokenizer, return_tensors="tf")
```
</tf>
</frameworkcontent>
## Treinamento
<frameworkcontent>
<pt>
Carregue o DistilBERT com o [`AutoModelForTokenClassification`] junto com o número de rótulos esperados:
```py
>>> from transformers import AutoModelForTokenClassification, TrainingArguments, Trainer
>>> model = AutoModelForTokenClassification.from_pretrained("distilbert/distilbert-base-uncased", num_labels=14)
```
<Tip>
Se você não estiver familiarizado com o fine-tuning de um modelo com o [`Trainer`], dê uma olhada no tutorial básico [aqui](../training#finetune-with-trainer)!
</Tip>
Nesse ponto, restam apenas três passos:
1. Definir seus hiperparâmetros de treinamento em [`TrainingArguments`].
2. Passar os argumentos de treinamento para o [`Trainer`] junto com o modelo, conjunto de dados, tokenizador e o data collator.
3. Chamar a função [`~Trainer.train`] para executar o fine-tuning do seu modelo.
```py
>>> training_args = TrainingArguments(
... output_dir="./results",
... eval_strategy="epoch",
... learning_rate=2e-5,
... per_device_train_batch_size=16,
... per_device_eval_batch_size=16,
... num_train_epochs=3,
... weight_decay=0.01,
... )
>>> trainer = Trainer(
... model=model,
... args=training_args,
... train_dataset=tokenized_wnut["train"],
... eval_dataset=tokenized_wnut["test"],
... tokenizer=tokenizer,
... data_collator=data_collator,
... )
>>> trainer.train()
```
</pt>
<tf>
Para executar o fine-tuning de um modelo no TensorFlow, comece convertendo seu conjunto de dados para o formato `tf.data.Dataset` com [`to_tf_dataset`](https://huggingface.co/docs/datasets/package_reference/main_classes#datasets.Dataset.to_tf_dataset). Nessa execução você deverá especificar as entradas e rótulos (no parâmetro `columns`), se deseja embaralhar o conjunto de dados, o tamanho do batch e o data collator:
```py
>>> tf_train_set = tokenized_wnut["train"].to_tf_dataset(
... columns=["attention_mask", "input_ids", "labels"],
... shuffle=True,
... batch_size=16,
... collate_fn=data_collator,
... )
>>> tf_validation_set = tokenized_wnut["validation"].to_tf_dataset(
... columns=["attention_mask", "input_ids", "labels"],
... shuffle=False,
... batch_size=16,
... collate_fn=data_collator,
... )
```
<Tip>
Se você não estiver familiarizado com o fine-tuning de um modelo com o Keras, dê uma olhada no tutorial básico [aqui](training#finetune-with-keras)!
</Tip>
Configure o otimizador e alguns hiperparâmetros de treinamento:
```py
>>> from transformers import create_optimizer
>>> batch_size = 16
>>> num_train_epochs = 3
>>> num_train_steps = (len(tokenized_wnut["train"]) // batch_size) * num_train_epochs
>>> optimizer, lr_schedule = create_optimizer(
... init_lr=2e-5,
... num_train_steps=num_train_steps,
... weight_decay_rate=0.01,
... num_warmup_steps=0,
... )
```
Carregue o DistilBERT com o [`TFAutoModelForTokenClassification`] junto com o número de rótulos esperados:
```py
>>> from transformers import TFAutoModelForTokenClassification
>>> model = TFAutoModelForTokenClassification.from_pretrained("distilbert/distilbert-base-uncased", num_labels=2)
```
Configure o modelo para treinamento com o método [`compile`](https://keras.io/api/models/model_training_apis/#compile-method):
```py
>>> import tensorflow as tf
>>> model.compile(optimizer=optimizer)
```
Chame o método [`fit`](https://keras.io/api/models/model_training_apis/#fit-method) para executar o fine-tuning do modelo:
```py
>>> model.fit(x=tf_train_set, validation_data=tf_validation_set, epochs=3)
```
</tf>
</frameworkcontent>
<Tip>
Para obter um exemplo mais aprofundado de como executar o fine-tuning de um modelo para classificação de tokens, dê uma olhada nesse [notebook utilizando PyTorch](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/token_classification.ipynb) ou nesse [notebook utilizando TensorFlow](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/token_classification-tf.ipynb).
</Tip>
| 0
|
mavonic_private_repos/transformers/docs/source/pt
|
mavonic_private_repos/transformers/docs/source/pt/tasks/sequence_classification.md
|
<!--Copyright 2022 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
specific language governing permissions and limitations under the License.
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer.
-->
# Classificação de texto
<Youtube id="leNG9fN9FQU"/>
A classificação de texto é uma tarefa comum de NLP que atribui um rótulo ou classe a um texto. Existem muitas aplicações práticas de classificação de texto amplamente utilizadas em produção por algumas das maiores empresas da atualidade. Uma das formas mais populares de classificação de texto é a análise de sentimento, que atribui um rótulo como positivo, negativo ou neutro a um texto.
Este guia mostrará como realizar o fine-tuning do [DistilBERT](https://huggingface.co/distilbert/distilbert-base-uncased) no conjunto de dados [IMDb](https://huggingface.co/datasets/imdb) para determinar se a crítica de filme é positiva ou negativa.
<Tip>
Consulte a [página de tarefas de classificação de texto](https://huggingface.co/tasks/text-classification) para obter mais informações sobre outras formas de classificação de texto e seus modelos, conjuntos de dados e métricas associados.
</Tip>
## Carregue o conjunto de dados IMDb
Carregue o conjunto de dados IMDb utilizando a biblioteca 🤗 Datasets:
```py
>>> from datasets import load_dataset
>>> imdb = load_dataset("imdb")
```
Em seguida, dê uma olhada em um exemplo:
```py
>>> imdb["test"][0]
{
"label": 0,
"text": "I love sci-fi and am willing to put up with a lot. Sci-fi movies/TV are usually underfunded, under-appreciated and misunderstood. I tried to like this, I really did, but it is to good TV sci-fi as Babylon 5 is to Star Trek (the original). Silly prosthetics, cheap cardboard sets, stilted dialogues, CG that doesn't match the background, and painfully one-dimensional characters cannot be overcome with a 'sci-fi' setting. (I'm sure there are those of you out there who think Babylon 5 is good sci-fi TV. It's not. It's clichéd and uninspiring.) While US viewers might like emotion and character development, sci-fi is a genre that does not take itself seriously (cf. Star Trek). It may treat important issues, yet not as a serious philosophy. It's really difficult to care about the characters here as they are not simply foolish, just missing a spark of life. Their actions and reactions are wooden and predictable, often painful to watch. The makers of Earth KNOW it's rubbish as they have to always say \"Gene Roddenberry's Earth...\" otherwise people would not continue watching. Roddenberry's ashes must be turning in their orbit as this dull, cheap, poorly edited (watching it without advert breaks really brings this home) trudging Trabant of a show lumbers into space. Spoiler. So, kill off a main character. And then bring him back as another actor. Jeeez! Dallas all over again.",
}
```
Existem dois campos neste dataset:
- `text`: uma string contendo o texto da crítica do filme.
- `label`: um valor que pode ser `0` para uma crítica negativa ou `1` para uma crítica positiva.
## Pré-processamento dos dados
Carregue o tokenizador do DistilBERT para processar o campo `text`:
```py
>>> from transformers import AutoTokenizer
>>> tokenizer = AutoTokenizer.from_pretrained("distilbert/distilbert-base-uncased")
```
Crie uma função de pré-processamento para tokenizar o campo `text` e truncar as sequências para que não sejam maiores que o comprimento máximo de entrada do DistilBERT:
```py
>>> def preprocess_function(examples):
... return tokenizer(examples["text"], truncation=True)
```
Use a função [`map`](https://huggingface.co/docs/datasets/process#map) do 🤗 Datasets para aplicar a função de pré-processamento em todo o conjunto de dados. Você pode acelerar a função `map` definindo `batched=True` para processar vários elementos do conjunto de dados de uma só vez:
```py
tokenized_imdb = imdb.map(preprocess_function, batched=True)
```
Use o [`DataCollatorWithPadding`] para criar um batch de exemplos. Ele também *preencherá dinamicamente* seu texto até o comprimento do elemento mais longo em seu batch, para que os exemplos do batch tenham um comprimento uniforme. Embora seja possível preencher seu texto com a função `tokenizer` definindo `padding=True`, o preenchimento dinâmico utilizando um data collator é mais eficiente.
<frameworkcontent>
<pt>
```py
>>> from transformers import DataCollatorWithPadding
>>> data_collator = DataCollatorWithPadding(tokenizer=tokenizer)
```
</pt>
<tf>
```py
>>> from transformers import DataCollatorWithPadding
>>> data_collator = DataCollatorWithPadding(tokenizer=tokenizer, return_tensors="tf")
```
</tf>
</frameworkcontent>
## Train
<frameworkcontent>
<pt>
Carregue o DistilBERT com [`AutoModelForSequenceClassification`] junto com o número de rótulos esperados:
```py
>>> from transformers import AutoModelForSequenceClassification, TrainingArguments, Trainer
>>> model = AutoModelForSequenceClassification.from_pretrained("distilbert/distilbert-base-uncased", num_labels=2)
```
<Tip>
Se você não estiver familiarizado com o fine-tuning de um modelo com o [`Trainer`], dê uma olhada no tutorial básico [aqui](../training#finetune-with-trainer)!
</Tip>
Nesse ponto, restam apenas três passos:
1. Definir seus hiperparâmetros de treinamento em [`TrainingArguments`].
2. Passar os argumentos de treinamento para o [`Trainer`] junto com o modelo, conjunto de dados, tokenizador e o data collator.
3. Chamar a função [`~Trainer.train`] para executar o fine-tuning do seu modelo.
```py
>>> training_args = TrainingArguments(
... output_dir="./results",
... learning_rate=2e-5,
... per_device_train_batch_size=16,
... per_device_eval_batch_size=16,
... num_train_epochs=5,
... weight_decay=0.01,
... )
>>> trainer = Trainer(
... model=model,
... args=training_args,
... train_dataset=tokenized_imdb["train"],
... eval_dataset=tokenized_imdb["test"],
... tokenizer=tokenizer,
... data_collator=data_collator,
... )
>>> trainer.train()
```
<Tip>
O [`Trainer`] aplicará o preenchimento dinâmico por padrão quando você definir o argumento `tokenizer` dele. Nesse caso, você não precisa especificar um data collator explicitamente.
</Tip>
</pt>
<tf>
Para executar o fine-tuning de um modelo no TensorFlow, comece convertendo seu conjunto de dados para o formato `tf.data.Dataset` com [`to_tf_dataset`](https://huggingface.co/docs/datasets/package_reference/main_classes#datasets.Dataset.to_tf_dataset). Nessa execução você deverá especificar as entradas e rótulos (no parâmetro `columns`), se deseja embaralhar o conjunto de dados, o tamanho do batch e o data collator:
```py
>>> tf_train_set = tokenized_imdb["train"].to_tf_dataset(
... columns=["attention_mask", "input_ids", "label"],
... shuffle=True,
... batch_size=16,
... collate_fn=data_collator,
... )
>>> tf_validation_set = tokenized_imdb["test"].to_tf_dataset(
... columns=["attention_mask", "input_ids", "label"],
... shuffle=False,
... batch_size=16,
... collate_fn=data_collator,
... )
```
<Tip>
Se você não estiver familiarizado com o fine-tuning de um modelo com o Keras, dê uma olhada no tutorial básico [aqui](training#finetune-with-keras)!
</Tip>
Configure o otimizador e alguns hiperparâmetros de treinamento:
```py
>>> from transformers import create_optimizer
>>> import tensorflow as tf
>>> batch_size = 16
>>> num_epochs = 5
>>> batches_per_epoch = len(tokenized_imdb["train"]) // batch_size
>>> total_train_steps = int(batches_per_epoch * num_epochs)
>>> optimizer, schedule = create_optimizer(init_lr=2e-5, num_warmup_steps=0, num_train_steps=total_train_steps)
```
Carregue o DistilBERT com [`TFAutoModelForSequenceClassification`] junto com o número de rótulos esperados:
```py
>>> from transformers import TFAutoModelForSequenceClassification
>>> model = TFAutoModelForSequenceClassification.from_pretrained("distilbert/distilbert-base-uncased", num_labels=2)
```
Configure o modelo para treinamento com o método [`compile`](https://keras.io/api/models/model_training_apis/#compile-method):
```py
>>> import tensorflow as tf
>>> model.compile(optimizer=optimizer)
```
Chame o método [`fit`](https://keras.io/api/models/model_training_apis/#fit-method) para executar o fine-tuning do modelo:
```py
>>> model.fit(x=tf_train_set, validation_data=tf_validation_set, epochs=3)
```
</tf>
</frameworkcontent>
<Tip>
Para obter um exemplo mais aprofundado de como executar o fine-tuning de um modelo para classificação de texto, dê uma olhada nesse [notebook utilizando PyTorch](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/text_classification.ipynb) ou nesse [notebook utilizando TensorFlow](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/text_classification-tf.ipynb).
</Tip>
| 0
|
mavonic_private_repos/transformers/docs/source
|
mavonic_private_repos/transformers/docs/source/zh/fast_tokenizers.md
|
<!--Copyright 2020 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
specific language governing permissions and limitations under the License.
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer.
-->
# 使用 🤗 Tokenizers 中的分词器
[`PreTrainedTokenizerFast`] 依赖于 [🤗 Tokenizers](https://huggingface.co/docs/tokenizers) 库。从 🤗 Tokenizers 库获得的分词器可以被轻松地加载到 🤗 Transformers 中。
在了解具体内容之前，让我们先用几行代码创建一个虚拟的分词器：
```python
>>> from tokenizers import Tokenizer
>>> from tokenizers.models import BPE
>>> from tokenizers.trainers import BpeTrainer
>>> from tokenizers.pre_tokenizers import Whitespace
>>> tokenizer = Tokenizer(BPE(unk_token="[UNK]"))
>>> trainer = BpeTrainer(special_tokens=["[UNK]", "[CLS]", "[SEP]", "[PAD]", "[MASK]"])
>>> tokenizer.pre_tokenizer = Whitespace()
>>> files = [...]
>>> tokenizer.train(files, trainer)
```
现在，我们拥有了一个针对我们定义的文件进行训练的分词器。我们可以在当前运行时中继续使用它，或者将其保存到一个 JSON 文件以供将来重复使用。
## 直接从分词器对象加载
让我们看看如何利用 🤗 Transformers 库中的这个分词器对象。[`PreTrainedTokenizerFast`] 类允许通过接受已实例化的 *tokenizer* 对象作为参数，进行轻松实例化：
```python
>>> from transformers import PreTrainedTokenizerFast
>>> fast_tokenizer = PreTrainedTokenizerFast(tokenizer_object=tokenizer)
```
现在可以使用这个对象，使用 🤗 Transformers 分词器共享的所有方法！前往[分词器页面](main_classes/tokenizer)了解更多信息。
## 从 JSON 文件加载
为了从 JSON 文件中加载分词器，让我们先保存我们的分词器：
```python
>>> tokenizer.save("tokenizer.json")
```
我们保存此文件的路径可以通过 `tokenizer_file` 参数传递给 [`PreTrainedTokenizerFast`] 初始化方法：
```python
>>> from transformers import PreTrainedTokenizerFast
>>> fast_tokenizer = PreTrainedTokenizerFast(tokenizer_file="tokenizer.json")
```
现在可以使用这个对象，使用 🤗 Transformers 分词器共享的所有方法！前往[分词器页面](main_classes/tokenizer)了解更多信息。
| 0
|
mavonic_private_repos/transformers/docs/source
|
mavonic_private_repos/transformers/docs/source/zh/run_scripts.md
|
<!--Copyright 2022 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
specific language governing permissions and limitations under the License.
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer.
-->
# 使用脚本进行训练
除了 🤗 Transformers [notebooks](./notebooks)，还有示例脚本演示了如何使用[PyTorch](https://github.com/huggingface/transformers/tree/main/examples/pytorch)、[TensorFlow](https://github.com/huggingface/transformers/tree/main/examples/tensorflow)或[JAX/Flax](https://github.com/huggingface/transformers/tree/main/examples/flax)训练模型以解决特定任务。
您还可以在这些示例中找到我们在[研究项目](https://github.com/huggingface/transformers/tree/main/examples/research_projects)和[遗留示例](https://github.com/huggingface/transformers/tree/main/examples/legacy)中使用过的脚本，这些脚本主要是由社区贡献的。这些脚本已不再被积极维护，需要使用特定版本的🤗 Transformers， 可能与库的最新版本不兼容。
示例脚本可能无法在初始配置下直接解决每个问题，您可能需要根据要解决的问题调整脚本。为了帮助您，大多数脚本都完全暴露了数据预处理的方式，允许您根据需要对其进行编辑。
如果您想在示例脚本中实现任何功能，请在[论坛](https://discuss.huggingface.co/)或[issue](https://github.com/huggingface/transformers/issues)上讨论，然后再提交Pull Request。虽然我们欢迎修复错误，但不太可能合并添加更多功能的Pull Request，因为这会降低可读性。
本指南将向您展示如何在[PyTorch](https://github.com/huggingface/transformers/tree/main/examples/pytorch/summarization)和[TensorFlow](https://github.com/huggingface/transformers/tree/main/examples/tensorflow/summarization)中运行示例摘要训练脚本。除非另有说明，否则所有示例都可以在两个框架中工作。
## 设置
要成功运行示例脚本的最新版本，您必须在新虚拟环境中**从源代码安装 🤗 Transformers**：
```bash
git clone https://github.com/huggingface/transformers
cd transformers
pip install .
```
对于旧版本的示例脚本，请点击下面的切换按钮：
<details>
<summary>老版本🤗 Transformers示例 </summary>
<ul>
<li><a href="https://github.com/huggingface/transformers/tree/v4.5.1/examples">v4.5.1</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v4.4.2/examples">v4.4.2</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v4.3.3/examples">v4.3.3</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v4.2.2/examples">v4.2.2</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v4.1.1/examples">v4.1.1</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v4.0.1/examples">v4.0.1</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v3.5.1/examples">v3.5.1</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v3.4.0/examples">v3.4.0</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v3.3.1/examples">v3.3.1</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v3.2.0/examples">v3.2.0</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v3.1.0/examples">v3.1.0</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v3.0.2/examples">v3.0.2</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v2.11.0/examples">v2.11.0</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v2.10.0/examples">v2.10.0</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v2.9.1/examples">v2.9.1</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v2.8.0/examples">v2.8.0</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v2.7.0/examples">v2.7.0</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v2.6.0/examples">v2.6.0</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v2.5.1/examples">v2.5.1</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v2.4.0/examples">v2.4.0</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v2.3.0/examples">v2.3.0</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v2.2.0/examples">v2.2.0</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v2.1.0/examples">v2.1.1</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v2.0.0/examples">v2.0.0</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v1.2.0/examples">v1.2.0</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v1.1.0/examples">v1.1.0</a></li>
<li><a href="https://github.com/huggingface/transformers/tree/v1.0.0/examples">v1.0.0</a></li>
</ul>
</details>
然后切换您clone的 🤗 Transformers 仓到特定的版本，例如v3.5.1：
```bash
git checkout tags/v3.5.1
```
在安装了正确的库版本后，进入您选择的版本的`example`文件夹并安装例子要求的环境：
```bash
pip install -r requirements.txt
```
## 运行脚本
<frameworkcontent>
<pt>
示例脚本从🤗 [Datasets](https://huggingface.co/docs/datasets/)库下载并预处理数据集。然后，脚本通过[Trainer](https://huggingface.co/docs/transformers/main_classes/trainer)使用支持摘要任务的架构对数据集进行微调。以下示例展示了如何在[CNN/DailyMail](https://huggingface.co/datasets/cnn_dailymail)数据集上微调[T5-small](https://huggingface.co/google-t5/t5-small)。由于T5模型的训练方式，它需要一个额外的`source_prefix`参数。这个提示让T5知道这是一个摘要任务。
```bash
python examples/pytorch/summarization/run_summarization.py \
--model_name_or_path google-t5/t5-small \
--do_train \
--do_eval \
--dataset_name cnn_dailymail \
--dataset_config "3.0.0" \
--source_prefix "summarize: " \
--output_dir /tmp/tst-summarization \
--per_device_train_batch_size=4 \
--per_device_eval_batch_size=4 \
--overwrite_output_dir \
--predict_with_generate
```
</pt>
<tf>
示例脚本从 🤗 [Datasets](https://huggingface.co/docs/datasets/) 库下载并预处理数据集。然后，脚本使用 Keras 在支持摘要的架构上微调数据集。以下示例展示了如何在 [CNN/DailyMail](https://huggingface.co/datasets/cnn_dailymail) 数据集上微调 [T5-small](https://huggingface.co/google-t5/t5-small)。T5 模型由于训练方式需要额外的 `source_prefix` 参数。这个提示让 T5 知道这是一个摘要任务。
```bash
python examples/tensorflow/summarization/run_summarization.py \
--model_name_or_path google-t5/t5-small \
--dataset_name cnn_dailymail \
--dataset_config "3.0.0" \
--output_dir /tmp/tst-summarization \
--per_device_train_batch_size 8 \
--per_device_eval_batch_size 16 \
--num_train_epochs 3 \
--do_train \
--do_eval
```
</tf>
</frameworkcontent>
## 分布式训练和混合精度
[Trainer](https://huggingface.co/docs/transformers/main_classes/trainer) 支持分布式训练和混合精度，这意味着你也可以在脚本中使用它。要启用这两个功能，可以做如下设置：
- 添加 `fp16` 参数以启用混合精度。
- 使用 `nproc_per_node` 参数设置使用的GPU数量。
```bash
torchrun \
--nproc_per_node 8 pytorch/summarization/run_summarization.py \
--fp16 \
--model_name_or_path google-t5/t5-small \
--do_train \
--do_eval \
--dataset_name cnn_dailymail \
--dataset_config "3.0.0" \
--source_prefix "summarize: " \
--output_dir /tmp/tst-summarization \
--per_device_train_batch_size=4 \
--per_device_eval_batch_size=4 \
--overwrite_output_dir \
--predict_with_generate
```
TensorFlow脚本使用[`MirroredStrategy`](https://www.tensorflow.org/guide/distributed_training#mirroredstrategy)进行分布式训练，您无需在训练脚本中添加任何其他参数。如果可用，TensorFlow脚本将默认使用多个GPU。
## 在TPU上运行脚本
<frameworkcontent>
<pt>
张量处理单元（TPUs）是专门设计用于加速性能的。PyTorch使用[XLA](https://www.tensorflow.org/xla)深度学习编译器支持TPU（更多细节请参见[这里](https://github.com/pytorch/xla/blob/master/README.md)）。要使用TPU，请启动`xla_spawn.py`脚本并使用`num_cores`参数设置要使用的TPU核心数量。
```bash
python xla_spawn.py --num_cores 8 \
summarization/run_summarization.py \
--model_name_or_path google-t5/t5-small \
--do_train \
--do_eval \
--dataset_name cnn_dailymail \
--dataset_config "3.0.0" \
--source_prefix "summarize: " \
--output_dir /tmp/tst-summarization \
--per_device_train_batch_size=4 \
--per_device_eval_batch_size=4 \
--overwrite_output_dir \
--predict_with_generate
```
</pt>
<tf>
张量处理单元（TPUs）是专门设计用于加速性能的。TensorFlow脚本使用[`TPUStrategy`](https://www.tensorflow.org/guide/distributed_training#tpustrategy)在TPU上进行训练。要使用TPU，请将TPU资源的名称传递给`tpu`参数。
```bash
python run_summarization.py \
--tpu name_of_tpu_resource \
--model_name_or_path google-t5/t5-small \
--dataset_name cnn_dailymail \
--dataset_config "3.0.0" \
--output_dir /tmp/tst-summarization \
--per_device_train_batch_size 8 \
--per_device_eval_batch_size 16 \
--num_train_epochs 3 \
--do_train \
--do_eval
```
</tf>
</frameworkcontent>
## 基于🤗 Accelerate运行脚本
🤗 [Accelerate](https://huggingface.co/docs/accelerate) 是一个仅支持 PyTorch 的库，它提供了一种统一的方法来在不同类型的设置（仅 CPU、多个 GPU、多个TPU）上训练模型，同时保持对 PyTorch 训练循环的完全可见性。如果你还没有安装 🤗 Accelerate，请确保你已经安装了它：
> 注意：由于 Accelerate 正在快速发展，因此必须安装 git 版本的 accelerate 来运行脚本。
```bash
pip install git+https://github.com/huggingface/accelerate
```
你需要使用`run_summarization_no_trainer.py`脚本，而不是`run_summarization.py`脚本。🤗 Accelerate支持的脚本需要在文件夹中有一个`task_no_trainer.py`文件。首先运行以下命令以创建并保存配置文件：
```bash
accelerate config
```
检测您的设置以确保配置正确：
```bash
accelerate test
```
现在您可以开始训练模型了：
```bash
accelerate launch run_summarization_no_trainer.py \
--model_name_or_path google-t5/t5-small \
--dataset_name cnn_dailymail \
--dataset_config "3.0.0" \
--source_prefix "summarize: " \
--output_dir ~/tmp/tst-summarization
```
## 使用自定义数据集
摘要脚本支持自定义数据集，只要它们是CSV或JSON Line文件。当你使用自己的数据集时，需要指定一些额外的参数：
- `train_file` 和 `validation_file` 分别指定您的训练和验证文件的路径。
- `text_column` 是输入要进行摘要的文本。
- `summary_column` 是目标输出的文本。
使用自定义数据集的摘要脚本看起来是这样的：
```bash
python examples/pytorch/summarization/run_summarization.py \
--model_name_or_path google-t5/t5-small \
--do_train \
--do_eval \
--train_file path_to_csv_or_jsonlines_file \
--validation_file path_to_csv_or_jsonlines_file \
--text_column text_column_name \
--summary_column summary_column_name \
--source_prefix "summarize: " \
--output_dir /tmp/tst-summarization \
--overwrite_output_dir \
--per_device_train_batch_size=4 \
--per_device_eval_batch_size=4 \
--predict_with_generate
```
## 测试脚本
通常，在提交整个数据集之前，最好先在较少的数据集示例上运行脚本，以确保一切按预期工作,因为完整数据集的处理可能需要花费几个小时的时间。使用以下参数将数据集截断为最大样本数：
- `max_train_samples`
- `max_eval_samples`
- `max_predict_samples`
```bash
python examples/pytorch/summarization/run_summarization.py \
--model_name_or_path google-t5/t5-small \
--max_train_samples 50 \
--max_eval_samples 50 \
--max_predict_samples 50 \
--do_train \
--do_eval \
--dataset_name cnn_dailymail \
--dataset_config "3.0.0" \
--source_prefix "summarize: " \
--output_dir /tmp/tst-summarization \
--per_device_train_batch_size=4 \
--per_device_eval_batch_size=4 \
--overwrite_output_dir \
--predict_with_generate
```
并非所有示例脚本都支持`max_predict_samples`参数。如果您不确定您的脚本是否支持此参数，请添加`-h`参数进行检查：
```bash
examples/pytorch/summarization/run_summarization.py -h
```
## 从checkpoint恢复训练
另一个有用的选项是从之前的checkpoint恢复训练。这将确保在训练中断时，您可以从之前停止的地方继续进行，而无需重新开始。有两种方法可以从checkpoint恢复训练。
第一种方法使用`output_dir previous_output_dir`参数从存储在`output_dir`中的最新的checkpoint恢复训练。在这种情况下，您应该删除`overwrite_output_dir`：
```bash
python examples/pytorch/summarization/run_summarization.py
--model_name_or_path google-t5/t5-small \
--do_train \
--do_eval \
--dataset_name cnn_dailymail \
--dataset_config "3.0.0" \
--source_prefix "summarize: " \
--output_dir /tmp/tst-summarization \
--per_device_train_batch_size=4 \
--per_device_eval_batch_size=4 \
--output_dir previous_output_dir \
--predict_with_generate
```
第二种方法使用`resume_from_checkpoint path_to_specific_checkpoint`参数从特定的checkpoint文件夹恢复训练。
```bash
python examples/pytorch/summarization/run_summarization.py
--model_name_or_path google-t5/t5-small \
--do_train \
--do_eval \
--dataset_name cnn_dailymail \
--dataset_config "3.0.0" \
--source_prefix "summarize: " \
--output_dir /tmp/tst-summarization \
--per_device_train_batch_size=4 \
--per_device_eval_batch_size=4 \
--overwrite_output_dir \
--resume_from_checkpoint path_to_specific_checkpoint \
--predict_with_generate
```
## 分享模型
所有脚本都可以将您的最终模型上传到[Model Hub](https://huggingface.co/models)。在开始之前，请确保您已登录Hugging Face：
```bash
huggingface-cli login
```
然后，在脚本中添加`push_to_hub`参数。这个参数会创建一个带有您Hugging Face用户名和`output_dir`中指定的文件夹名称的仓库。
为了给您的仓库指定一个特定的名称，使用`push_to_hub_model_id`参数来添加它。该仓库将自动列出在您的命名空间下。
以下示例展示了如何上传具有特定仓库名称的模型：
```bash
python examples/pytorch/summarization/run_summarization.py
--model_name_or_path google-t5/t5-small \
--do_train \
--do_eval \
--dataset_name cnn_dailymail \
--dataset_config "3.0.0" \
--source_prefix "summarize: " \
--push_to_hub \
--push_to_hub_model_id finetuned-t5-cnn_dailymail \
--output_dir /tmp/tst-summarization \
--per_device_train_batch_size=4 \
--per_device_eval_batch_size=4 \
--overwrite_output_dir \
--predict_with_generate
```
| 0
|
mavonic_private_repos/transformers/docs/source
|
mavonic_private_repos/transformers/docs/source/zh/create_a_model.md
|
<!--Copyright 2022 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
specific language governing permissions and limitations under the License.
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer.
-->
# 创建自定义架构
[`AutoClass`](model_doc/auto) 自动推断模型架构并下载预训练的配置和权重。一般来说，我们建议使用 `AutoClass` 生成与检查点（checkpoint）无关的代码。希望对特定模型参数有更多控制的用户，可以仅从几个基类创建自定义的 🤗 Transformers 模型。这对于任何有兴趣学习、训练或试验 🤗 Transformers 模型的人可能特别有用。通过本指南，深入了解如何不通过 `AutoClass` 创建自定义模型。了解如何：
- 加载并自定义模型配置。
- 创建模型架构。
- 为文本创建慢速和快速分词器。
- 为视觉任务创建图像处理器。
- 为音频任务创建特征提取器。
- 为多模态任务创建处理器。
## 配置
[配置](main_classes/configuration) 涉及到模型的具体属性。每个模型配置都有不同的属性；例如，所有 NLP 模型都共享 `hidden_size`、`num_attention_heads`、 `num_hidden_layers` 和 `vocab_size` 属性。这些属性用于指定构建模型时的注意力头数量或隐藏层层数。
访问 [`DistilBertConfig`] 以更近一步了解 [DistilBERT](model_doc/distilbert)，检查它的属性：
```py
>>> from transformers import DistilBertConfig
>>> config = DistilBertConfig()
>>> print(config)
DistilBertConfig {
"activation": "gelu",
"attention_dropout": 0.1,
"dim": 768,
"dropout": 0.1,
"hidden_dim": 3072,
"initializer_range": 0.02,
"max_position_embeddings": 512,
"model_type": "distilbert",
"n_heads": 12,
"n_layers": 6,
"pad_token_id": 0,
"qa_dropout": 0.1,
"seq_classif_dropout": 0.2,
"sinusoidal_pos_embds": false,
"transformers_version": "4.16.2",
"vocab_size": 30522
}
```
[`DistilBertConfig`] 显示了构建基础 [`DistilBertModel`] 所使用的所有默认属性。所有属性都可以进行自定义，为实验创造了空间。例如，您可以将默认模型自定义为：
- 使用 `activation` 参数尝试不同的激活函数。
- 使用 `attention_dropout` 参数为 attention probabilities 使用更高的 dropout ratio。
```py
>>> my_config = DistilBertConfig(activation="relu", attention_dropout=0.4)
>>> print(my_config)
DistilBertConfig {
"activation": "relu",
"attention_dropout": 0.4,
"dim": 768,
"dropout": 0.1,
"hidden_dim": 3072,
"initializer_range": 0.02,
"max_position_embeddings": 512,
"model_type": "distilbert",
"n_heads": 12,
"n_layers": 6,
"pad_token_id": 0,
"qa_dropout": 0.1,
"seq_classif_dropout": 0.2,
"sinusoidal_pos_embds": false,
"transformers_version": "4.16.2",
"vocab_size": 30522
}
```
预训练模型的属性可以在 [`~PretrainedConfig.from_pretrained`] 函数中进行修改：
```py
>>> my_config = DistilBertConfig.from_pretrained("distilbert/distilbert-base-uncased", activation="relu", attention_dropout=0.4)
```
当你对模型配置满意时，可以使用 [`~PretrainedConfig.save_pretrained`] 来保存配置。你的配置文件将以 JSON 文件的形式存储在指定的保存目录中：
```py
>>> my_config.save_pretrained(save_directory="./your_model_save_path")
```
要重用配置文件，请使用 [`~PretrainedConfig.from_pretrained`] 进行加载：
```py
>>> my_config = DistilBertConfig.from_pretrained("./your_model_save_path/config.json")
```
<Tip>
你还可以将配置文件保存为字典，甚至只保存自定义配置属性与默认配置属性之间的差异！有关更多详细信息，请参阅 [配置](main_classes/configuration) 文档。
</Tip>
## 模型
接下来，创建一个[模型](main_classes/models)。模型，也可泛指架构，定义了每一层网络的行为以及进行的操作。配置中的 `num_hidden_layers` 等属性用于定义架构。每个模型都共享基类 [`PreTrainedModel`] 和一些常用方法，例如调整输入嵌入的大小和修剪自注意力头。此外，所有模型都是 [`torch.nn.Module`](https://pytorch.org/docs/stable/generated/torch.nn.Module.html)、[`tf.keras.Model`](https://www.tensorflow.org/api_docs/python/tf/keras/Model) 或 [`flax.linen.Module`](https://flax.readthedocs.io/en/latest/api_reference/flax.linen/module.html) 的子类。这意味着模型与各自框架的用法兼容。
<frameworkcontent>
<pt>
将自定义配置属性加载到模型中：
```py
>>> from transformers import DistilBertModel
>>> my_config = DistilBertConfig.from_pretrained("./your_model_save_path/config.json")
>>> model = DistilBertModel(my_config)
```
这段代码创建了一个具有随机参数而不是预训练权重的模型。在训练该模型之前，您还无法将该模型用于任何用途。训练是一项昂贵且耗时的过程。通常来说，最好使用预训练模型来更快地获得更好的结果，同时仅使用训练所需资源的一小部分。
使用 [`~PreTrainedModel.from_pretrained`] 创建预训练模型：
```py
>>> model = DistilBertModel.from_pretrained("distilbert/distilbert-base-uncased")
```
当加载预训练权重时，如果模型是由 🤗 Transformers 提供的，将自动加载默认模型配置。然而，如果你愿意，仍然可以将默认模型配置的某些或者所有属性替换成你自己的配置：
```py
>>> model = DistilBertModel.from_pretrained("distilbert/distilbert-base-uncased", config=my_config)
```
</pt>
<tf>
将自定义配置属性加载到模型中：
```py
>>> from transformers import TFDistilBertModel
>>> my_config = DistilBertConfig.from_pretrained("./your_model_save_path/my_config.json")
>>> tf_model = TFDistilBertModel(my_config)
```
这段代码创建了一个具有随机参数而不是预训练权重的模型。在训练该模型之前，您还无法将该模型用于任何用途。训练是一项昂贵且耗时的过程。通常来说，最好使用预训练模型来更快地获得更好的结果，同时仅使用训练所需资源的一小部分。
使用 [`~TFPreTrainedModel.from_pretrained`] 创建预训练模型：
```py
>>> tf_model = TFDistilBertModel.from_pretrained("distilbert/distilbert-base-uncased")
```
当加载预训练权重时，如果模型是由 🤗 Transformers 提供的，将自动加载默认模型配置。然而，如果你愿意，仍然可以将默认模型配置的某些或者所有属性替换成自己的配置：
```py
>>> tf_model = TFDistilBertModel.from_pretrained("distilbert/distilbert-base-uncased", config=my_config)
```
</tf>
</frameworkcontent>
### 模型头（Model heads）
此时，你已经有了一个输出*隐藏状态*的基础 DistilBERT 模型。隐藏状态作为输入传递到模型头以生成最终输出。🤗 Transformers 为每个任务提供不同的模型头，只要模型支持该任务（即，您不能使用 DistilBERT 来执行像翻译这样的序列到序列任务）。
<frameworkcontent>
<pt>
例如，[`DistilBertForSequenceClassification`] 是一个带有序列分类头（sequence classification head）的基础 DistilBERT 模型。序列分类头是池化输出之上的线性层。
```py
>>> from transformers import DistilBertForSequenceClassification
>>> model = DistilBertForSequenceClassification.from_pretrained("distilbert/distilbert-base-uncased")
```
通过切换到不同的模型头，可以轻松地将此检查点重复用于其他任务。对于问答任务，你可以使用 [`DistilBertForQuestionAnswering`] 模型头。问答头（question answering head）与序列分类头类似，不同点在于它是隐藏状态输出之上的线性层。
```py
>>> from transformers import DistilBertForQuestionAnswering
>>> model = DistilBertForQuestionAnswering.from_pretrained("distilbert/distilbert-base-uncased")
```
</pt>
<tf>
例如，[`TFDistilBertForSequenceClassification`] 是一个带有序列分类头（sequence classification head）的基础 DistilBERT 模型。序列分类头是池化输出之上的线性层。
```py
>>> from transformers import TFDistilBertForSequenceClassification
>>> tf_model = TFDistilBertForSequenceClassification.from_pretrained("distilbert/distilbert-base-uncased")
```
通过切换到不同的模型头,可以轻松地将此检查点重复用于其他任务。对于问答任务，你可以使用 [`TFDistilBertForQuestionAnswering`] 模型头。问答头（question answering head）与序列分类头类似，不同点在于它是隐藏状态输出之上的线性层。
```py
>>> from transformers import TFDistilBertForQuestionAnswering
>>> tf_model = TFDistilBertForQuestionAnswering.from_pretrained("distilbert/distilbert-base-uncased")
```
</tf>
</frameworkcontent>
## 分词器
在将模型用于文本数据之前，你需要的最后一个基类是 [tokenizer](main_classes/tokenizer)，它用于将原始文本转换为张量。🤗 Transformers 支持两种类型的分词器：
- [`PreTrainedTokenizer`]：分词器的Python实现
- [`PreTrainedTokenizerFast`]：来自我们基于 Rust 的 [🤗 Tokenizer](https://huggingface.co/docs/tokenizers/python/latest/) 库的分词器。因为其使用了 Rust 实现，这种分词器类型的速度要快得多，尤其是在批量分词（batch tokenization）的时候。快速分词器还提供其他的方法，例如*偏移映射（offset mapping）*，它将标记（token）映射到其原始单词或字符。
这两种分词器都支持常用的方法，如编码和解码、添加新标记以及管理特殊标记。
<Tip warning={true}>
并非每个模型都支持快速分词器。参照这张 [表格](index#supported-frameworks) 查看模型是否支持快速分词器。
</Tip>
如果您训练了自己的分词器，则可以从*词表*文件创建一个分词器：
```py
>>> from transformers import DistilBertTokenizer
>>> my_tokenizer = DistilBertTokenizer(vocab_file="my_vocab_file.txt", do_lower_case=False, padding_side="left")
```
请务必记住，自定义分词器生成的词表与预训练模型分词器生成的词表是不同的。如果使用预训练模型，则需要使用预训练模型的词表，否则输入将没有意义。 使用 [`DistilBertTokenizer`] 类创建具有预训练模型词表的分词器：
```py
>>> from transformers import DistilBertTokenizer
>>> slow_tokenizer = DistilBertTokenizer.from_pretrained("distilbert/distilbert-base-uncased")
```
使用 [`DistilBertTokenizerFast`] 类创建快速分词器：
```py
>>> from transformers import DistilBertTokenizerFast
>>> fast_tokenizer = DistilBertTokenizerFast.from_pretrained("distilbert/distilbert-base-uncased")
```
<Tip>
默认情况下，[`AutoTokenizer`] 将尝试加载快速标记生成器。你可以通过在 `from_pretrained` 中设置 `use_fast=False` 以禁用此行为。
</Tip>
## 图像处理器
图像处理器用于处理视觉输入。它继承自 [`~image_processing_utils.ImageProcessingMixin`] 基类。
要使用它，需要创建一个与你使用的模型关联的图像处理器。例如，如果你使用 [ViT](model_doc/vit) 进行图像分类，可以创建一个默认的 [`ViTImageProcessor`]：
```py
>>> from transformers import ViTImageProcessor
>>> vit_extractor = ViTImageProcessor()
>>> print(vit_extractor)
ViTImageProcessor {
"do_normalize": true,
"do_resize": true,
"image_processor_type": "ViTImageProcessor",
"image_mean": [
0.5,
0.5,
0.5
],
"image_std": [
0.5,
0.5,
0.5
],
"resample": 2,
"size": 224
}
```
<Tip>
如果您不需要进行任何自定义，只需使用 `from_pretrained` 方法加载模型的默认图像处理器参数。
</Tip>
修改任何 [`ViTImageProcessor`] 参数以创建自定义图像处理器：
```py
>>> from transformers import ViTImageProcessor
>>> my_vit_extractor = ViTImageProcessor(resample="PIL.Image.BOX", do_normalize=False, image_mean=[0.3, 0.3, 0.3])
>>> print(my_vit_extractor)
ViTImageProcessor {
"do_normalize": false,
"do_resize": true,
"image_processor_type": "ViTImageProcessor",
"image_mean": [
0.3,
0.3,
0.3
],
"image_std": [
0.5,
0.5,
0.5
],
"resample": "PIL.Image.BOX",
"size": 224
}
```
## 特征提取器
特征提取器用于处理音频输入。它继承自 [`~feature_extraction_utils.FeatureExtractionMixin`] 基类，亦可继承 [`SequenceFeatureExtractor`] 类来处理音频输入。
要使用它，创建一个与你使用的模型关联的特征提取器。例如，如果你使用 [Wav2Vec2](model_doc/wav2vec2) 进行音频分类，可以创建一个默认的 [`Wav2Vec2FeatureExtractor`]：
```py
>>> from transformers import Wav2Vec2FeatureExtractor
>>> w2v2_extractor = Wav2Vec2FeatureExtractor()
>>> print(w2v2_extractor)
Wav2Vec2FeatureExtractor {
"do_normalize": true,
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
"feature_size": 1,
"padding_side": "right",
"padding_value": 0.0,
"return_attention_mask": false,
"sampling_rate": 16000
}
```
<Tip>
如果您不需要进行任何自定义，只需使用 `from_pretrained` 方法加载模型的默认特征提取器参数。
</Tip>
修改任何 [`Wav2Vec2FeatureExtractor`] 参数以创建自定义特征提取器：
```py
>>> from transformers import Wav2Vec2FeatureExtractor
>>> w2v2_extractor = Wav2Vec2FeatureExtractor(sampling_rate=8000, do_normalize=False)
>>> print(w2v2_extractor)
Wav2Vec2FeatureExtractor {
"do_normalize": false,
"feature_extractor_type": "Wav2Vec2FeatureExtractor",
"feature_size": 1,
"padding_side": "right",
"padding_value": 0.0,
"return_attention_mask": false,
"sampling_rate": 8000
}
```
## 处理器
对于支持多模式任务的模型，🤗 Transformers 提供了一个处理器类，可以方便地将特征提取器和分词器等处理类包装到单个对象中。例如，让我们使用 [`Wav2Vec2Processor`] 来执行自动语音识别任务 (ASR)。 ASR 将音频转录为文本，因此您将需要一个特征提取器和一个分词器。
创建一个特征提取器来处理音频输入：
```py
>>> from transformers import Wav2Vec2FeatureExtractor
>>> feature_extractor = Wav2Vec2FeatureExtractor(padding_value=1.0, do_normalize=True)
```
创建一个分词器来处理文本输入：
```py
>>> from transformers import Wav2Vec2CTCTokenizer
>>> tokenizer = Wav2Vec2CTCTokenizer(vocab_file="my_vocab_file.txt")
```
将特征提取器和分词器合并到 [`Wav2Vec2Processor`] 中：
```py
>>> from transformers import Wav2Vec2Processor
>>> processor = Wav2Vec2Processor(feature_extractor=feature_extractor, tokenizer=tokenizer)
```
通过两个基类 - 配置类和模型类 - 以及一个附加的预处理类（分词器、图像处理器、特征提取器或处理器），你可以创建 🤗 Transformers 支持的任何模型。 每个基类都是可配置的，允许你使用所需的特定属性。 你可以轻松设置模型进行训练或修改现有的预训练模型进行微调。
| 0
|
mavonic_private_repos/transformers/docs/source
|
mavonic_private_repos/transformers/docs/source/zh/tf_xla.md
|
<!--Copyright 2023 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
specific language governing permissions and limitations under the License.
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer.
-->
# 用于 TensorFlow 模型的 XLA 集成
[[open-in-colab]]
加速线性代数，也称为XLA，是一个用于加速TensorFlow模型运行时间的编译器。从[官方文档](https://www.tensorflow.org/xla)中可以看到：
XLA（加速线性代数）是一种针对线性代数的特定领域编译器，可以在可能不需要更改源代码的情况下加速TensorFlow模型。
在TensorFlow中使用XLA非常简单——它包含在`tensorflow`库中，并且可以使用任何图创建函数中的`jit_compile`参数来触发，例如[`tf.function`](https://www.tensorflow.org/guide/intro_to_graphs)。在使用Keras方法如`fit()`和`predict()`时，只需将`jit_compile`参数传递给`model.compile()`即可启用XLA。然而，XLA不仅限于这些方法 - 它还可以用于加速任何任意的`tf.function`。
在🤗 Transformers中，几个TensorFlow方法已经被重写为与XLA兼容，包括[GPT2](https://huggingface.co/docs/transformers/model_doc/gpt2)、[T5](https://huggingface.co/docs/transformers/model_doc/t5)和[OPT](https://huggingface.co/docs/transformers/model_doc/opt)等文本生成模型，以及[Whisper](https://huggingface.co/docs/transformers/model_doc/whisper)等语音处理模型。
虽然确切的加速倍数很大程度上取决于模型，但对于🤗 Transformers中的TensorFlow文本生成模型，我们注意到速度提高了约100倍。本文档将解释如何在这些模型上使用XLA获得最大的性能。如果您有兴趣了解更多关于基准测试和我们在XLA集成背后的设计哲学的信息，我们还将提供额外的资源链接。
## 使用 XLA 运行 TensorFlow 函数
让我们考虑以下TensorFlow 中的模型：
```py
import tensorflow as tf
model = tf.keras.Sequential(
[tf.keras.layers.Dense(10, input_shape=(10,), activation="relu"), tf.keras.layers.Dense(5, activation="softmax")]
)
```
上述模型接受维度为 `(10,)` 的输入。我们可以像下面这样使用模型进行前向传播：
```py
# Generate random inputs for the model.
batch_size = 16
input_vector_dim = 10
random_inputs = tf.random.normal((batch_size, input_vector_dim))
# Run a forward pass.
_ = model(random_inputs)
```
为了使用 XLA 编译的函数运行前向传播，我们需要执行以下操作：
```py
xla_fn = tf.function(model, jit_compile=True)
_ = xla_fn(random_inputs)
```
`model`的默认`call()`函数用于编译XLA图。但如果你想将其他模型函数编译成XLA，也是可以的，如下所示：
```py
my_xla_fn = tf.function(model.my_xla_fn, jit_compile=True)
```
## 在🤗 Transformers库中使用XLA运行TensorFlow文本生成模型
要在🤗 Transformers中启用XLA加速生成，您需要安装最新版本的`transformers`。您可以通过运行以下命令来安装它：
```bash
pip install transformers --upgrade
```
然后您可以运行以下代码：
```py
import tensorflow as tf
from transformers import AutoTokenizer, TFAutoModelForCausalLM
# Will error if the minimal version of Transformers is not installed.
from transformers.utils import check_min_version
check_min_version("4.21.0")
tokenizer = AutoTokenizer.from_pretrained("openai-community/gpt2", padding_side="left", pad_token="</s>")
model = TFAutoModelForCausalLM.from_pretrained("openai-community/gpt2")
input_string = ["TensorFlow is"]
# One line to create an XLA generation function
xla_generate = tf.function(model.generate, jit_compile=True)
tokenized_input = tokenizer(input_string, return_tensors="tf")
generated_tokens = xla_generate(**tokenized_input, num_beams=2)
decoded_text = tokenizer.decode(generated_tokens[0], skip_special_tokens=True)
print(f"Generated -- {decoded_text}")
# Generated -- TensorFlow is an open-source, open-source, distributed-source application # framework for the
```
正如您所注意到的，在`generate()`上启用XLA只需要一行代码。其余部分代码保持不变。然而，上面的代码片段中有一些与XLA相关的注意事项。您需要了解这些注意事项，以充分利用XLA可能带来的性能提升。我们将在下面的部分讨论这些内容。
## 需要关注的注意事项
当您首次执行启用XLA的函数（如上面的`xla_generate()`）时，它将在内部尝试推断计算图，这是一个耗时的过程。这个过程被称为[“tracing”](https://www.tensorflow.org/guide/intro_to_graphs#when_is_a_function_tracing)。
您可能会注意到生成时间并不快。连续调用`xla_generate()`（或任何其他启用了XLA的函数）不需要再次推断计算图，只要函数的输入与最初构建计算图时的形状相匹配。对于具有固定输入形状的模态（例如图像），这不是问题，但如果您正在处理具有可变输入形状的模态（例如文本），则必须注意。
为了确保`xla_generate()`始终使用相同的输入形状，您可以在调用`tokenizer`时指定`padding`参数。
```py
import tensorflow as tf
from transformers import AutoTokenizer, TFAutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("openai-community/gpt2", padding_side="left", pad_token="</s>")
model = TFAutoModelForCausalLM.from_pretrained("openai-community/gpt2")
input_string = ["TensorFlow is"]
xla_generate = tf.function(model.generate, jit_compile=True)
# Here, we call the tokenizer with padding options.
tokenized_input = tokenizer(input_string, pad_to_multiple_of=8, padding=True, return_tensors="tf")
generated_tokens = xla_generate(**tokenized_input, num_beams=2)
decoded_text = tokenizer.decode(generated_tokens[0], skip_special_tokens=True)
print(f"Generated -- {decoded_text}")
```
通过这种方式，您可以确保`xla_generate()`的输入始终具有它跟踪的形状，从而加速生成时间。您可以使用以下代码来验证这一点：
```py
import time
import tensorflow as tf
from transformers import AutoTokenizer, TFAutoModelForCausalLM
tokenizer = AutoTokenizer.from_pretrained("openai-community/gpt2", padding_side="left", pad_token="</s>")
model = TFAutoModelForCausalLM.from_pretrained("openai-community/gpt2")
xla_generate = tf.function(model.generate, jit_compile=True)
for input_string in ["TensorFlow is", "TensorFlow is a", "TFLite is a"]:
tokenized_input = tokenizer(input_string, pad_to_multiple_of=8, padding=True, return_tensors="tf")
start = time.time_ns()
generated_tokens = xla_generate(**tokenized_input, num_beams=2)
end = time.time_ns()
print(f"Execution time -- {(end - start) / 1e6:.1f} ms\n")
```
在Tesla T4 GPU上，您可以期望如下的输出：
```bash
Execution time -- 30819.6 ms
Execution time -- 79.0 ms
Execution time -- 78.9 ms
```
第一次调用`xla_generate()`会因为`tracing`而耗时，但后续的调用会快得多。请注意，任何时候对生成选项的更改都会触发重新`tracing`，从而导致生成时间减慢。
在本文档中，我们没有涵盖🤗 Transformers提供的所有文本生成选项。我们鼓励您阅读文档以了解高级用例。
## 附加资源
以下是一些附加资源，如果您想深入了解在🤗 Transformers和其他库下使用XLA：
* [这个Colab Notebook](https://colab.research.google.com/github/huggingface/blog/blob/main/notebooks/91_tf_xla_generate.ipynb) 提供了一个互动演示，让您可以尝试使用XLA兼容的编码器-解码器（例如[T5](https://huggingface.co/docs/transformers/model_doc/t5)）和仅解码器（例如[GPT2](https://huggingface.co/docs/transformers/model_doc/gpt2)）文本生成模型。
* [这篇博客文章](https://huggingface.co/blog/tf-xla-generate) 提供了XLA兼容模型的比较基准概述，以及关于在TensorFlow中使用XLA的友好介绍。
* [这篇博客文章](https://blog.tensorflow.org/2022/11/how-hugging-face-improved-text-generation-performance-with-xla.html) 讨论了我们在🤗 Transformers中为TensorFlow模型添加XLA支持的设计理念。
* 推荐用于更多学习XLA和TensorFlow图的资源：
* [XLA：面向机器学习的优化编译器](https://www.tensorflow.org/xla)
* [图和tf.function简介](https://www.tensorflow.org/guide/intro_to_graphs)
* [使用tf.function获得更好的性能](https://www.tensorflow.org/guide/function)
| 0
|
mavonic_private_repos/transformers/docs/source
|
mavonic_private_repos/transformers/docs/source/zh/transformers_agents.md
|
<!--Copyright 2023 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
specific language governing permissions and limitations under the License.
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer.
-->
# Transformers Agents
<Tip warning={true}>
`Transformers Agents`是一个实验性的随时可能发生变化的API。由于API或底层模型可能发生变化，`agents`返回的结果也会有所不同。
</Tip>
Transformers版本`v4.29.0`基于`tools`和`agents`概念构建。您可以在[此Colab链接](https://colab.research.google.com/drive/1c7MHD-T1forUPGcC_jlwsIptOzpG3hSj)中进行测试。
简而言之，它在`Transformers`之上提供了一个自然语言API：我们定义了一组经过筛选的`tools`，并设计了一个`agents`来解读自然语言并使用这些工具。它具有很强的可扩展性；我们筛选了一些相关的`tools`，但我们将向您展示如何通过社区开发的`tool`轻松地扩展系统。
让我们从一些可以通过这个新API实现的示例开始。在处理多模态任务时它尤其强大，因此让我们快速试着生成图像并大声朗读文本。
```py
agent.run("Caption the following image", image=image)
```
| **输入** | **输出** |
|-----------------------------------------------------------------------------------------------------------------------------|-----------------------------------|
| <img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/beaver.png" width=200> | A beaver is swimming in the water |
---
```py
agent.run("Read the following text out loud", text=text)
```
| **输入** | **输出** |
|-----------------------------------|----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------|
| A beaver is swimming in the water | <audio controls><source src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/tts_example.wav" type="audio/wav"> your browser does not support the audio element. </audio>
---
```py
agent.run(
"In the following `document`, where will the TRRF Scientific Advisory Council Meeting take place?",
document=document,
)
```
| **输入** | **输出** |
|-----------------------------------------------------------------------------------------------------------------------------|----------------|
| <img src="https://datasets-server.huggingface.co/assets/hf-internal-testing/example-documents/--/hf-internal-testing--example-documents/test/0/image/image.jpg" width=200> | ballroom foyer |
## 快速入门
要使用 `agent.run`，您需要实例化一个`agent`，它是一个大型语言模型（LLM）。我们支持OpenAI模型以及来自BigCode和OpenAssistant的开源替代方案。OpenAI模型性能更好（但需要您拥有OpenAI API密钥，因此无法免费使用），Hugging Face为BigCode和OpenAssistant模型提供了免费访问端点。
一开始请安装`agents`附加模块，以安装所有默认依赖项。
```bash
pip install transformers[agents]
```
要使用OpenAI模型，您可以在安装`openai`依赖项后实例化一个`OpenAiAgent`：
```bash
pip install openai
```
```py
from transformers import OpenAiAgent
agent = OpenAiAgent(model="text-davinci-003", api_key="<your_api_key>")
```
要使用BigCode或OpenAssistant，请首先登录以访问Inference API：
```py
from huggingface_hub import login
login("<YOUR_TOKEN>")
```
然后，实例化`agent`：
```py
from transformers import HfAgent
# Starcoder
agent = HfAgent("https://api-inference.huggingface.co/models/bigcode/starcoder")
# StarcoderBase
# agent = HfAgent("https://api-inference.huggingface.co/models/bigcode/starcoderbase")
# OpenAssistant
# agent = HfAgent(url_endpoint="https://api-inference.huggingface.co/models/OpenAssistant/oasst-sft-4-pythia-12b-epoch-3.5")
```
此示例使用了目前Hugging Face免费提供的推理API。如果你有自己的推理端点用于此模型（或其他模型），你可以用你的URL替换上面的URL。
<Tip>
StarCoder和OpenAssistant可以免费使用，并且在简单任务上表现出色。然而，当处理更复杂的提示时就不再有效。如果你遇到这样的问题，我们建议尝试使用OpenAI模型，尽管遗憾的是它不是开源的，但它在目前情况下表现更好。
</Tip>
现在，您已经可以开始使用了！让我们深入了解您现在可以使用的两个API。
### 单次执行(run)
单次执行方法是使用`agent`的 `~Agent.run`：
```py
agent.run("Draw me a picture of rivers and lakes.")
```
<img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/rivers_and_lakes.png" width=200>
它会自动选择适合您要执行的任务的`tool`（或`tools`），并以适当的方式运行它们。它可以在同一指令中执行一个或多个任务（尽管您的指令越复杂，`agent`失败的可能性就越大）。
```py
agent.run("Draw me a picture of the sea then transform the picture to add an island")
```
<img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/sea_and_island.png" width=200>
<br/>
每个 [`~Agent.run`] 操作都是独立的，因此您可以多次连续运行 [`~Agent.run`]并执行不同的任务。
请注意，您的 `agent` 只是一个大型语言模型，因此您略有变化的提示可能会产生完全不同的结果。重要的是尽可能清晰地解释您要执行的任务。我们在[这里](../en/custom_tools#writing-good-user-inputs)更深入地讨论了如何编写良好的提示。
如果您想在多次执行之间保持同一状态或向`agent`传递非文本对象，可以通过指定`agent`要使用的变量来实现。例如，您可以生成有关河流和湖泊的第一幅图像，并要求模型通过执行以下操作向该图片添加一个岛屿：
```python
picture = agent.run("Generate a picture of rivers and lakes.")
updated_picture = agent.run("Transform the image in `picture` to add an island to it.", picture=picture)
```
<Tip>
当模型无法理解您的请求和库中的工具时，这可能会有所帮助。例如：
```py
agent.run("Draw me the picture of a capybara swimming in the sea")
```
在这种情况下，模型可以以两种方式理解您的请求：
- 使用`text-to-image` 生成在大海中游泳的大水獭
- 或者，使用`text-to-image`生成大水獭，然后使用`image-transformation`工具使其在大海中游泳
如果您想强制使用第一种情景，可以通过将提示作为参数传递给它来实现：
```py
agent.run("Draw me a picture of the `prompt`", prompt="a capybara swimming in the sea")
```
</Tip>
### 基于交流的执行 (chat)
基于交流的执行（chat）方式是使用 [`~Agent.chat`]：
```py
agent.chat("Generate a picture of rivers and lakes")
```
<img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/rivers_and_lakes.png" width=200>
```py
agent.chat("Transform the picture so that there is a rock in there")
```
<img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/rivers_and_lakes_and_beaver.png" width=200>
<br/>
当您希望在不同指令之间保持同一状态时，这会是一个有趣的方法。它更适合用于单个指令，而不是复杂的多步指令（`~Agent.run` 方法更适合处理这种情况）。
这种方法也可以接受参数，以便您可以传递非文本类型或特定提示。
### ⚠️ 远程执行
出于演示目的以便适用于所有设置，我们为发布版本的少数默认工具创建了远程执行器。这些工具是使用推理终端（inference endpoints）创建的。
目前我们已将其关闭，但为了了解如何自行设置远程执行器工具，我们建议阅读[自定义工具指南](./custom_tools)。
### 这里发生了什么？什么是`tools`，什么是`agents`？
<img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/diagram.png">
#### Agents
这里的`Agents`是一个大型语言模型，我们通过提示它以访问特定的工具集。
大型语言模型在生成小代码示例方面表现出色，因此这个API利用这一特点，通过提示LLM生成一个使用`tools`集合的小代码示例。然后，根据您给`Agents`的任务和`tools`的描述来完成此提示。这种方式让它能够访问工具的文档，特别是它们的期望输入和输出，以生成相关的代码。
#### Tools
`Tools`非常简单：它们是有名称和描述的单个函数。然后，我们使用这些`tools`的描述来提示代理。通过提示，我们向`agent`展示如何使用`tool`来执行查询语言中请求的操作。
这是使用全新`tools`而不是`pipelines`，因为`agent`编写的代码更好，具有非常原子化的`tools`。`pipelines`经常被重构，并且通常将多个任务合并为一个。`tools`旨在专注于一个非常简单的任务。
#### 代码执行？
然后，这段代码基于`tools`的输入被我们的小型Python解释器执行。我们听到你在后面大声呼喊“任意代码执行！”，但让我们解释为什么情况并非如此。
只能您提供的`tools`和打印函数可以被执行，因此您已经受到了执行的限制。如果仅限于 Hugging Face 工具，那么您应该是安全的。
然后，我们不允许任何属性查找或导入（无论如何都不需要将输入/输出传递给一小组函数），因此所有最明显的攻击（并且您需要提示LLM无论如何输出它们）不应该是一个问题。如果你想超级安全，你可以使用附加参数 return_code=True 执行 run() 方法，在这种情况下，`agent`将只返回要执行的代码，你可以决定是否执行。
如果`agent`生成的代码存在任何尝试执行非法操作的行为，或者代码中出现了常规Python错误，执行将停止。
### 一组经过精心筛选的`tools`
我们确定了一组可以赋予这些`agent`强大能力的`tools`。以下是我们在`transformers`中集成的`tools`的更新列表：
- **文档问答**：给定一个图像格式的文档（例如PDF），回答该文档上的问题（[Donut](../en/model_doc/donut)）
- **文本问答**：给定一段长文本和一个问题，回答文本中的问题（[Flan-T5](../en/model_doc/flan-t5)）
- **无条件图像字幕**：为图像添加字幕！（[BLIP](../en/model_doc/blip)）
- **图像问答**：给定一张图像，回答该图像上的问题（[VILT](../en/model_doc/vilt)）
- **图像分割**：给定一张图像和一个提示，输出该提示的分割掩模（[CLIPSeg](../en/model_doc/clipseg)）
- **语音转文本**：给定一个人说话的音频录音，将演讲内容转录为文本（[Whisper](../en/model_doc/whisper)）
- **文本转语音**：将文本转换为语音（[SpeechT5](../en/model_doc/speecht5)）
- **Zero-Shot文本分类**：给定一个文本和一个标签列表，确定文本最符合哪个标签（[BART](../en/model_doc/bart)）
- **文本摘要**：总结长文本为一两句话（[BART](../en/model_doc/bart)）
- **翻译**：将文本翻译为指定语言（[NLLB](../en/model_doc/nllb)）
这些`tools`已在transformers中集成，并且也可以手动使用，例如：
```py
from transformers import load_tool
tool = load_tool("text-to-speech")
audio = tool("This is a text to speech tool")
```
### 自定义工具
尽管我们确定了一组经过筛选的`tools`，但我们坚信，此实现提供的主要价值在于能够快速创建和共享自定义`tool`。
通过将工具的代码上传到Hugging Face空间或模型repository，您可以直接通过`agent`使用`tools`。我们已经添加了一些**与transformers无关**的`tools`到[`huggingface-tools`组织](https://huggingface.co/huggingface-tools)中：
- **文本下载器**：从Web URL下载文本
- **文本到图像**：根据提示生成图像，利用`stable diffusion`
- **图像转换**：根据初始图像和提示修改图像，利用`instruct pix2pix stable diffusion`
- **文本到视频**：根据提示生成小视频，利用`damo-vilab`
从一开始就一直在使用的文本到图像`tool`是一个远程`tool `，位于[*huggingface-tools/text-to-image*](https://huggingface.co/spaces/huggingface-tools/text-to-image)！我们将继续在此组织和其他组织上发布此类`tool`，以进一步增强此实现。
`agents`默认可以访问存储在[`huggingface-tools`](https://huggingface.co/huggingface-tools)上的`tools`。我们将在后续指南中解释如何编写和共享自定义`tools`，以及如何利用Hub上存在的任何自定义`tools`。
### 代码生成
到目前为止，我们已经展示了如何使用`agents`来为您执行操作。但是，`agents`仅使用非常受限Python解释器执行的代码。如果您希望在不同的环境中使用生成的代码，可以提示`agents`返回代码，以及`tools`的定义和准确的导入信息。
例如，以下指令
```python
agent.run("Draw me a picture of rivers and lakes", return_code=True)
```
返回以下代码
```python
from transformers import load_tool
image_generator = load_tool("huggingface-tools/text-to-image")
image = image_generator(prompt="rivers and lakes")
```
然后你就可以调整并执行代码
| 0
|
mavonic_private_repos/transformers/docs/source
|
mavonic_private_repos/transformers/docs/source/zh/perf_torch_compile.md
|
<!--Copyright 2023 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer.
-->
# 使用 torch.compile() 优化推理
本指南旨在为使用[`torch.compile()`](https://pytorch.org/tutorials/intermediate/torch_compile_tutorial.html)在[🤗 Transformers中的计算机视觉模型](https://huggingface.co/models?pipeline_tag=image-classification&library=transformers&sort=trending)中引入的推理速度提升提供一个基准。
## torch.compile 的优势
根据模型和GPU的不同，`torch.compile()`在推理过程中可以提高多达30%的速度。要使用`torch.compile()`，只需安装2.0及以上版本的`torch`即可。
编译模型需要时间，因此如果您只需要编译一次模型而不是每次推理都编译，那么它非常有用。
要编译您选择的任何计算机视觉模型，请按照以下方式调用`torch.compile()`：
```diff
from transformers import AutoModelForImageClassification
model = AutoModelForImageClassification.from_pretrained(MODEL_ID).to("cuda")
+ model = torch.compile(model)
```
`compile()` 提供了多种编译模式，它们在编译时间和推理开销上有所不同。`max-autotune` 比 `reduce-overhead` 需要更长的时间，但会得到更快的推理速度。默认模式在编译时最快，但在推理时间上与 `reduce-overhead` 相比效率较低。在本指南中，我们使用了默认模式。您可以在[这里](https://pytorch.org/get-started/pytorch-2.0/#user-experience)了解更多信息。
我们在 PyTorch 2.0.1 版本上使用不同的计算机视觉模型、任务、硬件类型和数据批量大小对 `torch.compile` 进行了基准测试。
## 基准测试代码
以下是每个任务的基准测试代码。我们在推理之前”预热“GPU，并取300次推理的平均值，每次使用相同的图像。
### 使用 ViT 进行图像分类
```python
import torch
from PIL import Image
import requests
import numpy as np
from transformers import AutoImageProcessor, AutoModelForImageClassification
url = 'http://images.cocodataset.org/val2017/000000039769.jpg'
image = Image.open(requests.get(url, stream=True).raw)
processor = AutoImageProcessor.from_pretrained("google/vit-base-patch16-224")
model = AutoModelForImageClassification.from_pretrained("google/vit-base-patch16-224").to("cuda")
model = torch.compile(model)
processed_input = processor(image, return_tensors='pt').to(device="cuda")
with torch.no_grad():
_ = model(**processed_input)
```
#### 使用 DETR 进行目标检测
```python
from transformers import AutoImageProcessor, AutoModelForObjectDetection
processor = AutoImageProcessor.from_pretrained("facebook/detr-resnet-50")
model = AutoModelForObjectDetection.from_pretrained("facebook/detr-resnet-50").to("cuda")
model = torch.compile(model)
texts = ["a photo of a cat", "a photo of a dog"]
inputs = processor(text=texts, images=image, return_tensors="pt").to("cuda")
with torch.no_grad():
_ = model(**inputs)
```
#### 使用 Segformer 进行图像分割
```python
from transformers import SegformerImageProcessor, SegformerForSemanticSegmentation
processor = SegformerImageProcessor.from_pretrained("nvidia/segformer-b0-finetuned-ade-512-512")
model = SegformerForSemanticSegmentation.from_pretrained("nvidia/segformer-b0-finetuned-ade-512-512").to("cuda")
model = torch.compile(model)
seg_inputs = processor(images=image, return_tensors="pt").to("cuda")
with torch.no_grad():
_ = model(**seg_inputs)
```
以下是我们进行基准测试的模型列表。
**图像分类**
- [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224)
- [microsoft/beit-base-patch16-224-pt22k-ft22k](https://huggingface.co/microsoft/beit-base-patch16-224-pt22k-ft22k)
- [facebook/convnext-large-224](https://huggingface.co/facebook/convnext-large-224)
- [microsoft/resnet-50](https://huggingface.co/)
**图像分割**
- [nvidia/segformer-b0-finetuned-ade-512-512](https://huggingface.co/nvidia/segformer-b0-finetuned-ade-512-512)
- [facebook/mask2former-swin-tiny-coco-panoptic](https://huggingface.co/facebook/mask2former-swin-tiny-coco-panoptic)
- [facebook/maskformer-swin-base-ade](https://huggingface.co/facebook/maskformer-swin-base-ade)
- [google/deeplabv3_mobilenet_v2_1.0_513](https://huggingface.co/google/deeplabv3_mobilenet_v2_1.0_513)
**目标检测**
- [google/owlvit-base-patch32](https://huggingface.co/google/owlvit-base-patch32)
- [facebook/detr-resnet-101](https://huggingface.co/facebook/detr-resnet-101)
- [microsoft/conditional-detr-resnet-50](https://huggingface.co/microsoft/conditional-detr-resnet-50)
下面是使用和不使用`torch.compile()`的推理持续时间可视化，以及每个模型在不同硬件和数据批量大小下的改进百分比。
<div class="flex">
<div>
<img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/torch_compile/a100_batch_comp.png" />
</div>
<div>
<img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/torch_compile/v100_batch_comp.png" />
</div>
<div>
<img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/torch_compile/t4_batch_comp.png" />
</div>
</div>
<div class="flex">
<div>
<img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/torch_compile/A100_1_duration.png" />
</div>
<div>
<img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/torch_compile/A100_1_percentage.png" />
</div>
</div>
下面可以找到每个模型使用和不使用`compile()`的推理时间（毫秒）。请注意，OwlViT在大批量大小下会导致内存溢出。
### A100 (batch size: 1)
| **Task/Model** | **torch 2.0 - <br>no compile** | **torch 2.0 - <br>compile** |
|:---:|:---:|:---:|
| Image Classification/ViT | 9.325 | 7.584 |
| Image Segmentation/Segformer | 11.759 | 10.500 |
| Object Detection/OwlViT | 24.978 | 18.420 |
| Image Classification/BeiT | 11.282 | 8.448 |
| Object Detection/DETR | 34.619 | 19.040 |
| Image Classification/ConvNeXT | 10.410 | 10.208 |
| Image Classification/ResNet | 6.531 | 4.124 |
| Image Segmentation/Mask2former | 60.188 | 49.117 |
| Image Segmentation/Maskformer | 75.764 | 59.487 |
| Image Segmentation/MobileNet | 8.583 | 3.974 |
| Object Detection/Resnet-101 | 36.276 | 18.197 |
| Object Detection/Conditional-DETR | 31.219 | 17.993 |
### A100 (batch size: 4)
| **Task/Model** | **torch 2.0 - <br>no compile** | **torch 2.0 - <br>compile** |
|:---:|:---:|:---:|
| Image Classification/ViT | 14.832 | 14.499 |
| Image Segmentation/Segformer | 18.838 | 16.476 |
| Image Classification/BeiT | 13.205 | 13.048 |
| Object Detection/DETR | 48.657 | 32.418|
| Image Classification/ConvNeXT | 22.940 | 21.631 |
| Image Classification/ResNet | 6.657 | 4.268 |
| Image Segmentation/Mask2former | 74.277 | 61.781 |
| Image Segmentation/Maskformer | 180.700 | 159.116 |
| Image Segmentation/MobileNet | 14.174 | 8.515 |
| Object Detection/Resnet-101 | 68.101 | 44.998 |
| Object Detection/Conditional-DETR | 56.470 | 35.552 |
### A100 (batch size: 16)
| **Task/Model** | **torch 2.0 - <br>no compile** | **torch 2.0 - <br>compile** |
|:---:|:---:|:---:|
| Image Classification/ViT | 40.944 | 40.010 |
| Image Segmentation/Segformer | 37.005 | 31.144 |
| Image Classification/BeiT | 41.854 | 41.048 |
| Object Detection/DETR | 164.382 | 161.902 |
| Image Classification/ConvNeXT | 82.258 | 75.561 |
| Image Classification/ResNet | 7.018 | 5.024 |
| Image Segmentation/Mask2former | 178.945 | 154.814 |
| Image Segmentation/Maskformer | 638.570 | 579.826 |
| Image Segmentation/MobileNet | 51.693 | 30.310 |
| Object Detection/Resnet-101 | 232.887 | 155.021 |
| Object Detection/Conditional-DETR | 180.491 | 124.032 |
### V100 (batch size: 1)
| **Task/Model** | **torch 2.0 - <br>no compile** | **torch 2.0 - <br>compile** |
|:---:|:---:|:---:|
| Image Classification/ViT | 10.495 | 6.00 |
| Image Segmentation/Segformer | 13.321 | 5.862 |
| Object Detection/OwlViT | 25.769 | 22.395 |
| Image Classification/BeiT | 11.347 | 7.234 |
| Object Detection/DETR | 33.951 | 19.388 |
| Image Classification/ConvNeXT | 11.623 | 10.412 |
| Image Classification/ResNet | 6.484 | 3.820 |
| Image Segmentation/Mask2former | 64.640 | 49.873 |
| Image Segmentation/Maskformer | 95.532 | 72.207 |
| Image Segmentation/MobileNet | 9.217 | 4.753 |
| Object Detection/Resnet-101 | 52.818 | 28.367 |
| Object Detection/Conditional-DETR | 39.512 | 20.816 |
### V100 (batch size: 4)
| **Task/Model** | **torch 2.0 - <br>no compile** | **torch 2.0 - <br>compile** |
|:---:|:---:|:---:|
| Image Classification/ViT | 15.181 | 14.501 |
| Image Segmentation/Segformer | 16.787 | 16.188 |
| Image Classification/BeiT | 15.171 | 14.753 |
| Object Detection/DETR | 88.529 | 64.195 |
| Image Classification/ConvNeXT | 29.574 | 27.085 |
| Image Classification/ResNet | 6.109 | 4.731 |
| Image Segmentation/Mask2former | 90.402 | 76.926 |
| Image Segmentation/Maskformer | 234.261 | 205.456 |
| Image Segmentation/MobileNet | 24.623 | 14.816 |
| Object Detection/Resnet-101 | 134.672 | 101.304 |
| Object Detection/Conditional-DETR | 97.464 | 69.739 |
### V100 (batch size: 16)
| **Task/Model** | **torch 2.0 - <br>no compile** | **torch 2.0 - <br>compile** |
|:---:|:---:|:---:|
| Image Classification/ViT | 52.209 | 51.633 |
| Image Segmentation/Segformer | 61.013 | 55.499 |
| Image Classification/BeiT | 53.938 | 53.581 |
| Object Detection/DETR | OOM | OOM |
| Image Classification/ConvNeXT | 109.682 | 100.771 |
| Image Classification/ResNet | 14.857 | 12.089 |
| Image Segmentation/Mask2former | 249.605 | 222.801 |
| Image Segmentation/Maskformer | 831.142 | 743.645 |
| Image Segmentation/MobileNet | 93.129 | 55.365 |
| Object Detection/Resnet-101 | 482.425 | 361.843 |
| Object Detection/Conditional-DETR | 344.661 | 255.298 |
### T4 (batch size: 1)
| **Task/Model** | **torch 2.0 - <br>no compile** | **torch 2.0 - <br>compile** |
|:---:|:---:|:---:|
| Image Classification/ViT | 16.520 | 15.786 |
| Image Segmentation/Segformer | 16.116 | 14.205 |
| Object Detection/OwlViT | 53.634 | 51.105 |
| Image Classification/BeiT | 16.464 | 15.710 |
| Object Detection/DETR | 73.100 | 53.99 |
| Image Classification/ConvNeXT | 32.932 | 30.845 |
| Image Classification/ResNet | 6.031 | 4.321 |
| Image Segmentation/Mask2former | 79.192 | 66.815 |
| Image Segmentation/Maskformer | 200.026 | 188.268 |
| Image Segmentation/MobileNet | 18.908 | 11.997 |
| Object Detection/Resnet-101 | 106.622 | 82.566 |
| Object Detection/Conditional-DETR | 77.594 | 56.984 |
### T4 (batch size: 4)
| **Task/Model** | **torch 2.0 - <br>no compile** | **torch 2.0 - <br>compile** |
|:---:|:---:|:---:|
| Image Classification/ViT | 43.653 | 43.626 |
| Image Segmentation/Segformer | 45.327 | 42.445 |
| Image Classification/BeiT | 52.007 | 51.354 |
| Object Detection/DETR | 277.850 | 268.003 |
| Image Classification/ConvNeXT | 119.259 | 105.580 |
| Image Classification/ResNet | 13.039 | 11.388 |
| Image Segmentation/Mask2former | 201.540 | 184.670 |
| Image Segmentation/Maskformer | 764.052 | 711.280 |
| Image Segmentation/MobileNet | 74.289 | 48.677 |
| Object Detection/Resnet-101 | 421.859 | 357.614 |
| Object Detection/Conditional-DETR | 289.002 | 226.945 |
### T4 (batch size: 16)
| **Task/Model** | **torch 2.0 - <br>no compile** | **torch 2.0 - <br>compile** |
|:---:|:---:|:---:|
| Image Classification/ViT | 163.914 | 160.907 |
| Image Segmentation/Segformer | 192.412 | 163.620 |
| Image Classification/BeiT | 188.978 | 187.976 |
| Object Detection/DETR | OOM | OOM |
| Image Classification/ConvNeXT | 422.886 | 388.078 |
| Image Classification/ResNet | 44.114 | 37.604 |
| Image Segmentation/Mask2former | 756.337 | 695.291 |
| Image Segmentation/Maskformer | 2842.940 | 2656.88 |
| Image Segmentation/MobileNet | 299.003 | 201.942 |
| Object Detection/Resnet-101 | 1619.505 | 1262.758 |
| Object Detection/Conditional-DETR | 1137.513 | 897.390|
## PyTorch Nightly
我们还在 PyTorch Nightly 版本（2.1.0dev）上进行了基准测试，可以在[这里](https://download.pytorch.org/whl/nightly/cu118)找到 Nightly 版本的安装包，并观察到了未编译和编译模型的延迟性能改善。
### A100
| **Task/Model** | **Batch Size** | **torch 2.0 - no compile** | **torch 2.0 -<br> compile** |
|:---:|:---:|:---:|:---:|
| Image Classification/BeiT | Unbatched | 12.462 | 6.954 |
| Image Classification/BeiT | 4 | 14.109 | 12.851 |
| Image Classification/BeiT | 16 | 42.179 | 42.147 |
| Object Detection/DETR | Unbatched | 30.484 | 15.221 |
| Object Detection/DETR | 4 | 46.816 | 30.942 |
| Object Detection/DETR | 16 | 163.749 | 163.706 |
### T4
| **Task/Model** | **Batch Size** | **torch 2.0 - <br>no compile** | **torch 2.0 - <br>compile** |
|:---:|:---:|:---:|:---:|
| Image Classification/BeiT | Unbatched | 14.408 | 14.052 |
| Image Classification/BeiT | 4 | 47.381 | 46.604 |
| Image Classification/BeiT | 16 | 42.179 | 42.147 |
| Object Detection/DETR | Unbatched | 68.382 | 53.481 |
| Object Detection/DETR | 4 | 269.615 | 204.785 |
| Object Detection/DETR | 16 | OOM | OOM |
### V100
| **Task/Model** | **Batch Size** | **torch 2.0 - <br>no compile** | **torch 2.0 - <br>compile** |
|:---:|:---:|:---:|:---:|
| Image Classification/BeiT | Unbatched | 13.477 | 7.926 |
| Image Classification/BeiT | 4 | 15.103 | 14.378 |
| Image Classification/BeiT | 16 | 52.517 | 51.691 |
| Object Detection/DETR | Unbatched | 28.706 | 19.077 |
| Object Detection/DETR | 4 | 88.402 | 62.949|
| Object Detection/DETR | 16 | OOM | OOM |
## 降低开销
我们在 PyTorch Nightly 版本中为 A100 和 T4 进行了 `reduce-overhead` 编译模式的性能基准测试。
### A100
| **Task/Model** | **Batch Size** | **torch 2.0 - <br>no compile** | **torch 2.0 - <br>compile** |
|:---:|:---:|:---:|:---:|
| Image Classification/ConvNeXT | Unbatched | 11.758 | 7.335 |
| Image Classification/ConvNeXT | 4 | 23.171 | 21.490 |
| Image Classification/ResNet | Unbatched | 7.435 | 3.801 |
| Image Classification/ResNet | 4 | 7.261 | 2.187 |
| Object Detection/Conditional-DETR | Unbatched | 32.823 | 11.627 |
| Object Detection/Conditional-DETR | 4 | 50.622 | 33.831 |
| Image Segmentation/MobileNet | Unbatched | 9.869 | 4.244 |
| Image Segmentation/MobileNet | 4 | 14.385 | 7.946 |
### T4
| **Task/Model** | **Batch Size** | **torch 2.0 - <br>no compile** | **torch 2.0 - <br>compile** |
|:---:|:---:|:---:|:---:|
| Image Classification/ConvNeXT | Unbatched | 32.137 | 31.84 |
| Image Classification/ConvNeXT | 4 | 120.944 | 110.209 |
| Image Classification/ResNet | Unbatched | 9.761 | 7.698 |
| Image Classification/ResNet | 4 | 15.215 | 13.871 |
| Object Detection/Conditional-DETR | Unbatched | 72.150 | 57.660 |
| Object Detection/Conditional-DETR | 4 | 301.494 | 247.543 |
| Image Segmentation/MobileNet | Unbatched | 22.266 | 19.339 |
| Image Segmentation/MobileNet | 4 | 78.311 | 50.983 |
| 0
|
mavonic_private_repos/transformers/docs/source
|
mavonic_private_repos/transformers/docs/source/zh/torchscript.md
|
<!--
Copyright 2022 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
specific language governing permissions and limitations under the License.
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer.
-->
# 导出为 TorchScript
<Tip>
这是开始使用 TorchScript 进行实验的起点，我们仍在探索其在变量输入大小模型中的能力。
这是我们关注的焦点，我们将在即将发布的版本中深入分析，提供更多的代码示例、更灵活的实现以及比较
Python 代码与编译 TorchScript 的性能基准。
</Tip>
根据 [TorchScript 文档](https://pytorch.org/docs/stable/jit.html)：
> TorchScript 是从 PyTorch 代码创建可序列化和可优化的模型的一种方式。
有两个 PyTorch 模块：[JIT 和 TRACE](https://pytorch.org/docs/stable/jit.html)。
这两个模块允许开发人员将其模型导出到其他程序中重用，比如面向效率的 C++ 程序。
我们提供了一个接口，允许您将 🤗 Transformers 模型导出为 TorchScript，
以便在与基于 PyTorch 的 Python 程序不同的环境中重用。
本文解释如何使用 TorchScript 导出并使用我们的模型。
导出模型需要两个步骤：
- 使用 `torchscript` 参数实例化模型
- 使用虚拟输入进行前向传递
这些必要条件意味着开发人员应该注意以下详细信息。
## TorchScript 参数和绑定权重
`torchscript` 参数是必需的，因为大多数 🤗 Transformers 语言模型的 `Embedding` 层和
`Decoding` 层之间有绑定权重。TorchScript 不允许导出具有绑定权重的模型，因此必须事先解绑和克隆权重。
使用 `torchscript` 参数实例化的模型将其 `Embedding` 层和 `Decoding` 层分开，
这意味着它们不应该在后续进行训练。训练将导致这两层不同步，产生意外结果。
对于没有语言模型头部的模型，情况不同，因为这些模型没有绑定权重。
这些模型可以安全地导出而无需 `torchscript` 参数。
## 虚拟输入和标准长度
虚拟输入用于模型的前向传递。当输入的值传播到各层时，PyTorch 会跟踪在每个张量上执行的不同操作。
然后使用记录的操作来创建模型的 *trace* 。
跟踪是相对于输入的维度创建的。因此，它受到虚拟输入的维度限制，对于任何其他序列长度或批量大小都不起作用。
当尝试使用不同大小时，会引发以下错误：
```text
`The expanded size of the tensor (3) must match the existing size (7) at non-singleton dimension 2`
```
我们建议使用至少与推断期间将馈送到模型的最大输入一样大的虚拟输入大小进行跟踪。
填充可以帮助填补缺失的值。然而，由于模型是使用更大的输入大小进行跟踪的，矩阵的维度也会很大，导致更多的计算。
在每个输入上执行的操作总数要仔细考虑，并在导出不同序列长度模型时密切关注性能。
## 在 Python 中使用 TorchScript
本节演示了如何保存和加载模型以及如何使用 trace 进行推断。
### 保存模型
要使用 TorchScript 导出 `BertModel`，请从 `BertConfig` 类实例化 `BertModel`，
然后将其保存到名为 `traced_bert.pt` 的磁盘文件中：
```python
from transformers import BertModel, BertTokenizer, BertConfig
import torch
enc = BertTokenizer.from_pretrained("google-bert/bert-base-uncased")
# 对输入文本分词
text = "[CLS] Who was Jim Henson ? [SEP] Jim Henson was a puppeteer [SEP]"
tokenized_text = enc.tokenize(text)
# 屏蔽一个输入 token
masked_index = 8
tokenized_text[masked_index] = "[MASK]"
indexed_tokens = enc.convert_tokens_to_ids(tokenized_text)
segments_ids = [0, 0, 0, 0, 0, 0, 0, 1, 1, 1, 1, 1, 1, 1]
# 创建虚拟输入
tokens_tensor = torch.tensor([indexed_tokens])
segments_tensors = torch.tensor([segments_ids])
dummy_input = [tokens_tensor, segments_tensors]
# 使用 torchscript 参数初始化模型
# 即使此模型没有 LM Head，也将参数设置为 True。
config = BertConfig(
vocab_size_or_config_json_file=32000,
hidden_size=768,
num_hidden_layers=12,
num_attention_heads=12,
intermediate_size=3072,
torchscript=True,
)
# 实例化模型
model = BertModel(config)
# 模型需要处于评估模式
model.eval()
# 如果您使用 *from_pretrained* 实例化模型，还可以轻松设置 TorchScript 参数
model = BertModel.from_pretrained("google-bert/bert-base-uncased", torchscript=True)
# 创建 trace
traced_model = torch.jit.trace(model, [tokens_tensor, segments_tensors])
torch.jit.save(traced_model, "traced_bert.pt")
```
### 加载模型
现在，您可以从磁盘加载先前保存的 `BertModel`、`traced_bert.pt`，并在先前初始化的 `dummy_input` 上使用：
```python
loaded_model = torch.jit.load("traced_bert.pt")
loaded_model.eval()
all_encoder_layers, pooled_output = loaded_model(*dummy_input)
```
### 使用 trace 模型进行推断
通过使用其 `__call__` dunder 方法使用 trace 模型进行推断：
```python
traced_model(tokens_tensor, segments_tensors)
```
## 使用 Neuron SDK 将 Hugging Face TorchScript 模型部署到 AWS
AWS 引入了用于云端低成本、高性能机器学习推理的
[Amazon EC2 Inf1](https://aws.amazon.com/ec2/instance-types/inf1/) 实例系列。
Inf1 实例由 AWS Inferentia 芯片提供支持，这是一款专为深度学习推理工作负载而构建的定制硬件加速器。
[AWS Neuron](https://awsdocs-neuron.readthedocs-hosted.com/en/latest/#) 是
Inferentia 的 SDK，支持对 transformers 模型进行跟踪和优化，以便在 Inf1 上部署。Neuron SDK 提供：
1. 简单易用的 API，只需更改一行代码即可为云端推理跟踪和优化 TorchScript 模型。
2. 针对[改进的性能成本](https://awsdocs-neuron.readthedocs-hosted.com/en/latest/neuron-guide/benchmark/)的即插即用性能优化。
3. 支持使用 [PyTorch](https://awsdocs-neuron.readthedocs-hosted.com/en/latest/src/examples/pytorch/bert_tutorial/tutorial_pretrained_bert.html)
或 [TensorFlow](https://awsdocs-neuron.readthedocs-hosted.com/en/latest/src/examples/tensorflow/huggingface_bert/huggingface_bert.html)
构建的 Hugging Face transformers 模型。
### 影响
基于 [BERT（来自 Transformers 的双向编码器表示）](https://huggingface.co/docs/transformers/main/model_doc/bert)架构的
transformers 模型，或其变体，如 [distilBERT](https://huggingface.co/docs/transformers/main/model_doc/distilbert)
和 [roBERTa](https://huggingface.co/docs/transformers/main/model_doc/roberta) 在 Inf1 上运行最佳，
可用于生成抽取式问答、序列分类和标记分类等任务。然而，文本生成任务仍可以适应在 Inf1 上运行，
如这篇 [AWS Neuron MarianMT 教程](https://awsdocs-neuron.readthedocs-hosted.com/en/latest/src/examples/pytorch/transformers-marianmt.html)所述。
有关可以直接在 Inferentia 上转换的模型的更多信息，请参阅 Neuron 文档的[模型架构适配](https://awsdocs-neuron.readthedocs-hosted.com/en/latest/neuron-guide/models/models-inferentia.html#models-inferentia)章节。
### 依赖关系
使用 AWS Neuron 将模型转换为模型需要一个
[Neuron SDK 环境](https://awsdocs-neuron.readthedocs-hosted.com/en/latest/neuron-guide/neuron-frameworks/pytorch-neuron/index.html#installation-guide)，
它已经预先配置在 [AWS 深度学习 AMI](https://docs.aws.amazon.com/dlami/latest/devguide/tutorial-inferentia-launching.html)上。
### 将模型转换为 AWS Neuron
使用与 [Python 中使用 TorchScript](torchscript#using-torchscript-in-python) 相同的代码来跟踪
`BertModel` 以将模型转换为 AWS NEURON。导入 `torch.neuron` 框架扩展以通过 Python API 访问 Neuron SDK 的组件：
```python
from transformers import BertModel, BertTokenizer, BertConfig
import torch
import torch.neuron
```
您只需要修改下面这一行：
```diff
- torch.jit.trace(model, [tokens_tensor, segments_tensors])
+ torch.neuron.trace(model, [token_tensor, segments_tensors])
```
这样就能使 Neuron SDK 跟踪模型并对其进行优化，以在 Inf1 实例上运行。
要了解有关 AWS Neuron SDK 功能、工具、示例教程和最新更新的更多信息，
请参阅 [AWS NeuronSDK 文档](https://awsdocs-neuron.readthedocs-hosted.com/en/latest/index.html)。
| 0
|
mavonic_private_repos/transformers/docs/source
|
mavonic_private_repos/transformers/docs/source/zh/performance.md
|
<!---
Copyright 2021 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer.
-->
# 性能与可扩展性
训练大型transformer模型并将其部署到生产环境会面临各种挑战。
在训练过程中，模型可能需要比可用的GPU内存更多的资源，或者表现出较慢的训练速度。在部署阶段，模型可能在生产环境中难以处理所需的吞吐量。
本文档旨在帮助您克服这些挑战，并找到适合您使用场景的最佳设置。教程分为训练和推理部分，因为每个部分都有不同的挑战和解决方案。在每个部分中，您将找到针对不同硬件配置的单独指南，例如单GPU与多GPU用于训练或CPU与GPU用于推理。
将此文档作为您的起点，进一步导航到与您的情况匹配的方法。
## 训练
高效训练大型transformer模型需要使用加速器硬件，如GPU或TPU。最常见的情况是您只有一个GPU。您应用于单个GPU上提高训练效率的方法可以扩展到其他设置，如多个GPU。然而，也有一些特定于多GPU或CPU训练的技术。我们在单独的部分中介绍它们。
* [在单个GPU上进行高效训练的方法和工具](perf_train_gpu_one)：从这里开始学习常见的方法，可以帮助优化GPU内存利用率、加快训练速度或两者兼备。
* [多GPU训练部分](perf_train_gpu_many)：探索此部分以了解适用于多GPU设置的进一步优化方法，例如数据并行、张量并行和流水线并行。
* [CPU训练部分](perf_train_cpu)：了解在CPU上的混合精度训练。
* [在多个CPU上进行高效训练](perf_train_cpu_many)：了解分布式CPU训练。
* [使用TensorFlow在TPU上进行训练](perf_train_tpu_tf)：如果您对TPU还不熟悉，请参考此部分，了解有关在TPU上进行训练和使用XLA的建议性介绍。
* [自定义硬件进行训练](perf_hardware)：在构建自己的深度学习机器时查找技巧和窍门。
* [使用Trainer API进行超参数搜索](hpo_train)
## 推理
在生产环境中对大型模型进行高效推理可能与训练它们一样具有挑战性。在接下来的部分中，我们将详细介绍如何在CPU和单/多GPU设置上进行推理的步骤。
* [在单个CPU上进行推理](perf_infer_cpu)
* [在单个GPU上进行推理](perf_infer_gpu_one)
* [多GPU推理](perf_infer_gpu_one)
* [TensorFlow模型的XLA集成](tf_xla)
## 训练和推理
在这里，您将找到适用于训练模型或使用它进行推理的技巧、窍门和技巧。
* [实例化大型模型](big_models)
* [解决性能问题](debugging)
## 贡献
这份文档还远远没有完成，还有很多需要添加的内容，所以如果你有补充或更正的内容，请毫不犹豫地提交一个PR（Pull Request），或者如果你不确定，可以创建一个Issue，我们可以在那里讨论细节。
在做出贡献时，如果A比B更好，请尽量包含可重复的基准测试和(或)该信息来源的链接（除非它直接来自您）。
| 0
|
mavonic_private_repos/transformers/docs/source
|
mavonic_private_repos/transformers/docs/source/zh/tflite.md
|
<!--Copyright 2023 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
specific language governing permissions and limitations under the License.
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer.
-->
# 导出为 TFLite
[TensorFlow Lite](https://www.tensorflow.org/lite/guide) 是一个轻量级框架，用于资源受限的设备上，如手机、嵌入式系统和物联网（IoT）设备，部署机器学习模型。TFLite 旨在在计算能力、内存和功耗有限的设备上优化和高效运行模型。模型以一种特殊的高效可移植格式表示，其文件扩展名为 `.tflite`。
🤗 Optimum 通过 `exporters.tflite` 模块提供将 🤗 Transformers 模型导出至 TFLite 格式的功能。请参考 [🤗 Optimum 文档](https://huggingface.co/docs/optimum/exporters/tflite/overview) 以获取支持的模型架构列表。
要将模型导出为 TFLite 格式，请安装所需的依赖项：
```bash
pip install optimum[exporters-tf]
```
请参阅 [🤗 Optimum 文档](https://huggingface.co/docs/optimum/main/en/exporters/tflite/usage_guides/export_a_model) 以查看所有可用参数，或者在命令行中查看帮助：
```bash
optimum-cli export tflite --help
```
运行以下命令，以从 🤗 Hub 导出模型的检查点（checkpoint），以 `google-bert/bert-base-uncased` 为例：
```bash
optimum-cli export tflite --model google-bert/bert-base-uncased --sequence_length 128 bert_tflite/
```
你应该能在日志中看到导出进度以及生成的 `model.tflite` 文件的保存位置，如下所示：
```bash
Validating TFLite model...
-[✓] TFLite model output names match reference model (logits)
- Validating TFLite Model output "logits":
-[✓] (1, 128, 30522) matches (1, 128, 30522)
-[x] values not close enough, max diff: 5.817413330078125e-05 (atol: 1e-05)
The TensorFlow Lite export succeeded with the warning: The maximum absolute difference between the output of the reference model and the TFLite exported model is not within the set tolerance 1e-05:
- logits: max diff = 5.817413330078125e-05.
The exported model was saved at: bert_tflite
```
上面的示例说明了从 🤗 Hub 导出检查点的过程。导出本地模型时，首先需要确保将模型的权重和分词器文件保存在同一目录（`local_path`）中。在使用 CLI（命令行）时，将 `local_path` 传递给 `model` 参数，而不是 🤗 Hub 上的检查点名称。
| 0
|
mavonic_private_repos/transformers/docs/source
|
mavonic_private_repos/transformers/docs/source/zh/autoclass_tutorial.md
|
<!--Copyright 2022 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
specific language governing permissions and limitations under the License.
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer.
-->
# 使用AutoClass加载预训练实例
由于存在许多不同的Transformer架构，因此为您的checkpoint创建一个可用架构可能会具有挑战性。通过`AutoClass`可以自动推断并从给定的checkpoint加载正确的架构, 这也是🤗 Transformers易于使用、简单且灵活核心规则的重要一部分。`from_pretrained()`方法允许您快速加载任何架构的预训练模型，因此您不必花费时间和精力从头开始训练模型。生成这种与checkpoint无关的代码意味着，如果您的代码适用于一个checkpoint，它将适用于另一个checkpoint - 只要它们是为了类似的任务进行训练的 - 即使架构不同。
<Tip>
请记住，架构指的是模型的结构，而checkpoints是给定架构的权重。例如，[BERT](https://huggingface.co/google-bert/bert-base-uncased)是一种架构，而`google-bert/bert-base-uncased`是一个checkpoint。模型是一个通用术语，可以指代架构或checkpoint。
</Tip>
在这个教程中，学习如何：
* 加载预训练的分词器（`tokenizer`）
* 加载预训练的图像处理器(`image processor`)
* 加载预训练的特征提取器(`feature extractor`)
* 加载预训练的处理器(`processor`)
* 加载预训练的模型。
## AutoTokenizer
几乎所有的NLP任务都以`tokenizer`开始。`tokenizer`将您的输入转换为模型可以处理的格式。
使用[`AutoTokenizer.from_pretrained`]加载`tokenizer`：
```py
>>> from transformers import AutoTokenizer
>>> tokenizer = AutoTokenizer.from_pretrained("google-bert/bert-base-uncased")
```
然后按照如下方式对输入进行分词：
```py
>>> sequence = "In a hole in the ground there lived a hobbit."
>>> print(tokenizer(sequence))
{'input_ids': [101, 1999, 1037, 4920, 1999, 1996, 2598, 2045, 2973, 1037, 7570, 10322, 4183, 1012, 102],
'token_type_ids': [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0],
'attention_mask': [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1]}
```
## AutoImageProcessor
对于视觉任务，`image processor`将图像处理成正确的输入格式。
```py
>>> from transformers import AutoImageProcessor
>>> image_processor = AutoImageProcessor.from_pretrained("google/vit-base-patch16-224")
```
## AutoFeatureExtractor
对于音频任务,`feature extractor`将音频信号处理成正确的输入格式。
使用[`AutoFeatureExtractor.from_pretrained`]加载`feature extractor`：
```py
>>> from transformers import AutoFeatureExtractor
>>> feature_extractor = AutoFeatureExtractor.from_pretrained(
... "ehcalabres/wav2vec2-lg-xlsr-en-speech-emotion-recognition"
... )
```
## AutoProcessor
多模态任务需要一种`processor`，将两种类型的预处理工具结合起来。例如，[LayoutLMV2](model_doc/layoutlmv2)模型需要一个`image processor`来处理图像和一个`tokenizer`来处理文本；`processor`将两者结合起来。
使用[`AutoProcessor.from_pretrained`]加载`processor`：
```py
>>> from transformers import AutoProcessor
>>> processor = AutoProcessor.from_pretrained("microsoft/layoutlmv2-base-uncased")
```
## AutoModel
<frameworkcontent>
<pt>
最后，`AutoModelFor`类让你可以加载给定任务的预训练模型（参见[这里](model_doc/auto)获取可用任务的完整列表）。例如，使用[`AutoModelForSequenceClassification.from_pretrained`]加载用于序列分类的模型：
```py
>>> from transformers import AutoModelForSequenceClassification
>>> model = AutoModelForSequenceClassification.from_pretrained("distilbert/distilbert-base-uncased")
```
轻松地重复使用相同的checkpoint来为不同任务加载模型架构：
```py
>>> from transformers import AutoModelForTokenClassification
>>> model = AutoModelForTokenClassification.from_pretrained("distilbert/distilbert-base-uncased")
```
<Tip warning={true}>
对于PyTorch模型，`from_pretrained()`方法使用`torch.load()`，它内部使用已知是不安全的`pickle`。一般来说，永远不要加载来自不可信来源或可能被篡改的模型。对于托管在Hugging Face Hub上的公共模型，这种安全风险在一定程度上得到了缓解，因为每次提交都会进行[恶意软件扫描](https://huggingface.co/docs/hub/security-malware)。请参阅[Hub文档](https://huggingface.co/docs/hub/security)以了解最佳实践，例如使用GPG进行[签名提交验证](https://huggingface.co/docs/hub/security-gpg#signing-commits-with-gpg)。
TensorFlow和Flax的checkpoints不受影响，并且可以在PyTorch架构中使用`from_tf`和`from_flax`关键字参数,通过`from_pretrained`方法进行加载,来绕过此问题。
</Tip>
一般来说，我们建议使用`AutoTokenizer`类和`AutoModelFor`类来加载预训练的模型实例。这样可以确保每次加载正确的架构。在下一个[教程](preprocessing)中，学习如何使用新加载的`tokenizer`, `image processor`, `feature extractor`和`processor`对数据集进行预处理以进行微调。
</pt>
<tf>
最后，`TFAutoModelFor`类允许您加载给定任务的预训练模型（请参阅[这里](model_doc/auto)获取可用任务的完整列表）。例如，使用[`TFAutoModelForSequenceClassification.from_pretrained`]加载用于序列分类的模型：
```py
>>> from transformers import TFAutoModelForSequenceClassification
>>> model = TFAutoModelForSequenceClassification.from_pretrained("distilbert/distilbert-base-uncased")
```
轻松地重复使用相同的checkpoint来为不同任务加载模型架构：
```py
>>> from transformers import TFAutoModelForTokenClassification
>>> model = TFAutoModelForTokenClassification.from_pretrained("distilbert/distilbert-base-uncased")
```
一般来说，我们推荐使用`AutoTokenizer`类和`TFAutoModelFor`类来加载模型的预训练实例。这样可以确保每次加载正确的架构。在下一个[教程](preprocessing)中，学习如何使用新加载的`tokenizer`, `image processor`, `feature extractor`和`processor`对数据集进行预处理以进行微调。
</tf>
</frameworkcontent>
| 0
|
mavonic_private_repos/transformers/docs/source
|
mavonic_private_repos/transformers/docs/source/zh/multilingual.md
|
<!--Copyright 2022 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
specific language governing permissions and limitations under the License.
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer.
-->
# 用于推理的多语言模型
[[open-in-colab]]
🤗 Transformers 中有多种多语言模型，它们的推理用法与单语言模型不同。但是，并非*所有*的多语言模型用法都不同。一些模型，例如 [google-bert/bert-base-multilingual-uncased](https://huggingface.co/google-bert/bert-base-multilingual-uncased) 就可以像单语言模型一样使用。本指南将向您展示如何使用不同用途的多语言模型进行推理。
## XLM
XLM 有十个不同的检查点，其中只有一个是单语言的。剩下的九个检查点可以归为两类：使用语言嵌入的检查点和不使用语言嵌入的检查点。
### 带有语言嵌入的 XLM
以下 XLM 模型使用语言嵌入来指定推理中使用的语言：
- `FacebookAI/xlm-mlm-ende-1024` （掩码语言建模，英语-德语）
- `FacebookAI/xlm-mlm-enfr-1024` （掩码语言建模，英语-法语）
- `FacebookAI/xlm-mlm-enro-1024` （掩码语言建模，英语-罗马尼亚语）
- `FacebookAI/xlm-mlm-xnli15-1024` （掩码语言建模，XNLI 数据集语言）
- `FacebookAI/xlm-mlm-tlm-xnli15-1024` （掩码语言建模+翻译，XNLI 数据集语言）
- `FacebookAI/xlm-clm-enfr-1024` （因果语言建模，英语-法语）
- `FacebookAI/xlm-clm-ende-1024` （因果语言建模，英语-德语）
语言嵌入被表示一个张量，其形状与传递给模型的 `input_ids` 相同。这些张量中的值取决于所使用的语言，并由分词器的 `lang2id` 和 `id2lang` 属性识别。
在此示例中，加载 `FacebookAI/xlm-clm-enfr-1024` 检查点（因果语言建模，英语-法语）：
```py
>>> import torch
>>> from transformers import XLMTokenizer, XLMWithLMHeadModel
>>> tokenizer = XLMTokenizer.from_pretrained("FacebookAI/xlm-clm-enfr-1024")
>>> model = XLMWithLMHeadModel.from_pretrained("FacebookAI/xlm-clm-enfr-1024")
```
分词器的 `lang2id` 属性显示了该模型的语言及其对应的id：
```py
>>> print(tokenizer.lang2id)
{'en': 0, 'fr': 1}
```
接下来，创建一个示例输入：
```py
>>> input_ids = torch.tensor([tokenizer.encode("Wikipedia was used to")]) # batch size 为 1
```
将语言 id 设置为 `"en"` 并用其定义语言嵌入。语言嵌入是一个用 `0` 填充的张量，这个张量应该与 `input_ids` 大小相同。
```py
>>> language_id = tokenizer.lang2id["en"] # 0
>>> langs = torch.tensor([language_id] * input_ids.shape[1]) # torch.tensor([0, 0, 0, ..., 0])
>>> # 我们将其 reshape 为 (batch_size, sequence_length) 大小
>>> langs = langs.view(1, -1) # 现在的形状是 [1, sequence_length] (我们的 batch size 为 1)
```
现在，你可以将 `input_ids` 和语言嵌入传递给模型：
```py
>>> outputs = model(input_ids, langs=langs)
```
[run_generation.py](https://github.com/huggingface/transformers/tree/main/examples/pytorch/text-generation/run_generation.py) 脚本可以使用 `xlm-clm` 检查点生成带有语言嵌入的文本。
### 不带语言嵌入的 XLM
以下 XLM 模型在推理时不需要语言嵌入：
- `FacebookAI/xlm-mlm-17-1280` （掩码语言建模，支持 17 种语言）
- `FacebookAI/xlm-mlm-100-1280` （掩码语言建模，支持 100 种语言）
与之前的 XLM 检查点不同，这些模型用于通用句子表示。
## BERT
以下 BERT 模型可用于多语言任务：
- `google-bert/bert-base-multilingual-uncased` （掩码语言建模 + 下一句预测，支持 102 种语言）
- `google-bert/bert-base-multilingual-cased` （掩码语言建模 + 下一句预测，支持 104 种语言）
这些模型在推理时不需要语言嵌入。它们应该能够从上下文中识别语言并进行相应的推理。
## XLM-RoBERTa
以下 XLM-RoBERTa 模型可用于多语言任务：
- `FacebookAI/xlm-roberta-base` （掩码语言建模，支持 100 种语言）
- `FacebookAI/xlm-roberta-large` （掩码语言建模，支持 100 种语言）
XLM-RoBERTa 使用 100 种语言的 2.5TB 新创建和清理的 CommonCrawl 数据进行了训练。与之前发布的 mBERT 或 XLM 等多语言模型相比，它在分类、序列标记和问答等下游任务上提供了更强大的优势。
## M2M100
以下 M2M100 模型可用于多语言翻译：
- `facebook/m2m100_418M` （翻译）
- `facebook/m2m100_1.2B` （翻译）
在此示例中，加载 `facebook/m2m100_418M` 检查点以将中文翻译为英文。你可以在分词器中设置源语言：
```py
>>> from transformers import M2M100ForConditionalGeneration, M2M100Tokenizer
>>> en_text = "Do not meddle in the affairs of wizards, for they are subtle and quick to anger."
>>> chinese_text = "不要插手巫師的事務, 因為他們是微妙的, 很快就會發怒."
>>> tokenizer = M2M100Tokenizer.from_pretrained("facebook/m2m100_418M", src_lang="zh")
>>> model = M2M100ForConditionalGeneration.from_pretrained("facebook/m2m100_418M")
```
对文本进行分词：
```py
>>> encoded_zh = tokenizer(chinese_text, return_tensors="pt")
```
M2M100 强制将目标语言 id 作为第一个生成的标记，以进行到目标语言的翻译。在 `generate` 方法中将 `forced_bos_token_id` 设置为 `en` 以翻译成英语：
```py
>>> generated_tokens = model.generate(**encoded_zh, forced_bos_token_id=tokenizer.get_lang_id("en"))
>>> tokenizer.batch_decode(generated_tokens, skip_special_tokens=True)
'Do not interfere with the matters of the witches, because they are delicate and will soon be angry.'
```
## MBart
以下 MBart 模型可用于多语言翻译：
- `facebook/mbart-large-50-one-to-many-mmt` （一对多多语言机器翻译，支持 50 种语言）
- `facebook/mbart-large-50-many-to-many-mmt` （多对多多语言机器翻译，支持 50 种语言）
- `facebook/mbart-large-50-many-to-one-mmt` （多对一多语言机器翻译，支持 50 种语言）
- `facebook/mbart-large-50` （多语言翻译，支持 50 种语言）
- `facebook/mbart-large-cc25`
在此示例中，加载 `facebook/mbart-large-50-many-to-many-mmt` 检查点以将芬兰语翻译为英语。 你可以在分词器中设置源语言：
```py
>>> from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
>>> en_text = "Do not meddle in the affairs of wizards, for they are subtle and quick to anger."
>>> fi_text = "Älä sekaannu velhojen asioihin, sillä ne ovat hienovaraisia ja nopeasti vihaisia."
>>> tokenizer = AutoTokenizer.from_pretrained("facebook/mbart-large-50-many-to-many-mmt", src_lang="fi_FI")
>>> model = AutoModelForSeq2SeqLM.from_pretrained("facebook/mbart-large-50-many-to-many-mmt")
```
对文本进行分词：
```py
>>> encoded_en = tokenizer(en_text, return_tensors="pt")
```
MBart 强制将目标语言 id 作为第一个生成的标记，以进行到目标语言的翻译。在 `generate` 方法中将 `forced_bos_token_id` 设置为 `en` 以翻译成英语：
```py
>>> generated_tokens = model.generate(**encoded_en, forced_bos_token_id=tokenizer.lang_code_to_id["en_XX"])
>>> tokenizer.batch_decode(generated_tokens, skip_special_tokens=True)
"Don't interfere with the wizard's affairs, because they are subtle, will soon get angry."
```
如果你使用的是 `facebook/mbart-large-50-many-to-one-mmt` 检查点，则无需强制目标语言 id 作为第一个生成的令牌，否则用法是相同的。
| 0
|
mavonic_private_repos/transformers/docs/source
|
mavonic_private_repos/transformers/docs/source/zh/model_sharing.md
|
<!--Copyright 2022 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
specific language governing permissions and limitations under the License.
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer.
-->
# 分享模型
最后两个教程展示了如何使用PyTorch、Keras和 🤗 Accelerate进行分布式设置来微调模型。下一步是将您的模型与社区分享！在Hugging Face，我们相信公开分享知识和资源，能实现人工智能的普及化，让每个人都能受益。我们鼓励您将您的模型与社区分享，以帮助他人节省时间和精力。
在本教程中，您将学习两种在[Model Hub](https://huggingface.co/models)上共享训练好的或微调的模型的方法：
- 通过编程将文件推送到Hub。
- 使用Web界面将文件拖放到Hub。
<iframe width="560" height="315" src="https://www.youtube.com/embed/XvSGPZFEjDY" title="YouTube video player"
frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope;
picture-in-picture" allowfullscreen></iframe>
<Tip>
要与社区共享模型，您需要在[huggingface.co](https://huggingface.co/join)上拥有一个帐户。您还可以加入现有的组织或创建一个新的组织。
</Tip>
## 仓库功能
Model Hub上的每个仓库都像是一个典型的GitHub仓库。我们的仓库提供版本控制、提交历史记录以及可视化差异的能力。
Model Hub的内置版本控制基于git和[git-lfs](https://git-lfs.github.com/)。换句话说，您可以将一个模型视为一个仓库，从而实现更好的访问控制和可扩展性。版本控制允许使用*修订*方法来固定特定版本的模型，可以使用提交哈希值、标签或分支来标记。
因此，您可以通过`revision`参数加载特定的模型版本：
```py
>>> model = AutoModel.from_pretrained(
... "julien-c/EsperBERTo-small", revision="v2.0.1" # tag name, or branch name, or commit hash
... )
```
文件也可以轻松地在仓库中编辑，您可以查看提交历史记录以及差异：
## 设置
在将模型共享到Hub之前，您需要拥有Hugging Face的凭证。如果您有访问终端的权限，请在安装🤗 Transformers的虚拟环境中运行以下命令。这将在您的Hugging Face缓存文件夹（默认为`~/.cache/`）中存储您的`access token`：
```bash
huggingface-cli login
```
如果您正在使用像Jupyter或Colaboratory这样的`notebook`，请确保您已安装了[`huggingface_hub`](https://huggingface.co/docs/hub/adding-a-library)库。该库允许您以编程方式与Hub进行交互。
```bash
pip install huggingface_hub
```
然后使用`notebook_login`登录到Hub，并按照[这里](https://huggingface.co/settings/token)的链接生成一个token进行登录：
```py
>>> from huggingface_hub import notebook_login
>>> notebook_login()
```
## 转换模型适用于所有框架
为确保您的模型可以被使用不同框架的人使用，我们建议您将PyTorch和TensorFlow `checkpoints`都转换并上传。如果您跳过此步骤，用户仍然可以从其他框架加载您的模型，但速度会变慢，因为🤗 Transformers需要实时转换`checkpoints`。
为另一个框架转换`checkpoints`很容易。确保您已安装PyTorch和TensorFlow（请参阅[此处](installation)的安装说明），然后在其他框架中找到适合您任务的特定模型。
<frameworkcontent>
<pt>
指定`from_tf=True`将checkpoint从TensorFlow转换为PyTorch。
```py
>>> pt_model = DistilBertForSequenceClassification.from_pretrained("path/to/awesome-name-you-picked", from_tf=True)
>>> pt_model.save_pretrained("path/to/awesome-name-you-picked")
```
</pt>
<tf>
指定`from_pt=True`将checkpoint从PyTorch转换为TensorFlow。
```py
>>> tf_model = TFDistilBertForSequenceClassification.from_pretrained("path/to/awesome-name-you-picked", from_pt=True)
```
然后，您可以使用新的checkpoint保存您的新TensorFlow模型：
```py
>>> tf_model.save_pretrained("path/to/awesome-name-you-picked")
```
</tf>
<jax>
如果模型在Flax中可用，您还可以将PyTorch checkpoint转换为Flax：
```py
>>> flax_model = FlaxDistilBertForSequenceClassification.from_pretrained(
... "path/to/awesome-name-you-picked", from_pt=True
... )
```
</jax>
</frameworkcontent>
## 在训练过程中推送模型
<frameworkcontent>
<pt>
<Youtube id="Z1-XMy-GNLQ"/>
将模型分享到Hub就像添加一个额外的参数或回调函数一样简单。请记住，在[微调教程](training)中，`TrainingArguments`类是您指定超参数和附加训练选项的地方。其中一项训练选项包括直接将模型推送到Hub的能力。在您的`TrainingArguments`中设置`push_to_hub=True`：
```py
>>> training_args = TrainingArguments(output_dir="my-awesome-model", push_to_hub=True)
```
像往常一样将您的训练参数传递给[`Trainer`]：
```py
>>> trainer = Trainer(
... model=model,
... args=training_args,
... train_dataset=small_train_dataset,
... eval_dataset=small_eval_dataset,
... compute_metrics=compute_metrics,
... )
```
在您微调完模型后，在[`Trainer`]上调用[`~transformers.Trainer.push_to_hub`]将训练好的模型推送到Hub。🤗 Transformers甚至会自动将训练超参数、训练结果和框架版本添加到你的模型卡片中！
```py
>>> trainer.push_to_hub()
```
</pt>
<tf>
使用[`PushToHubCallback`]将模型分享到Hub。在[`PushToHubCallback`]函数中，添加以下内容：
- 一个用于存储模型的输出目录。
- 一个tokenizer。
- `hub_model_id`，即您的Hub用户名和模型名称。
```py
>>> from transformers import PushToHubCallback
>>> push_to_hub_callback = PushToHubCallback(
... output_dir="./your_model_save_path", tokenizer=tokenizer, hub_model_id="your-username/my-awesome-model"
... )
```
将回调函数添加到 [`fit`](https://keras.io/api/models/model_training_apis/)中，然后🤗 Transformers 会将训练好的模型推送到 Hub：
```py
>>> model.fit(tf_train_dataset, validation_data=tf_validation_dataset, epochs=3, callbacks=push_to_hub_callback)
```
</tf>
</frameworkcontent>
## 使用`push_to_hub`功能
您可以直接在您的模型上调用`push_to_hub`来将其上传到Hub。
在`push_to_hub`中指定你的模型名称：
```py
>>> pt_model.push_to_hub("my-awesome-model")
```
这会在您的用户名下创建一个名为`my-awesome-model`的仓库。用户现在可以使用`from_pretrained`函数加载您的模型：
```py
>>> from transformers import AutoModel
>>> model = AutoModel.from_pretrained("your_username/my-awesome-model")
```
如果您属于一个组织，并希望将您的模型推送到组织名称下，只需将其添加到`repo_id`中：
```py
>>> pt_model.push_to_hub("my-awesome-org/my-awesome-model")
```
`push_to_hub`函数还可以用于向模型仓库添加其他文件。例如，向模型仓库中添加一个`tokenizer`：
```py
>>> tokenizer.push_to_hub("my-awesome-model")
```
或者，您可能希望将您的微调后的PyTorch模型的TensorFlow版本添加进去：
```py
>>> tf_model.push_to_hub("my-awesome-model")
```
现在，当您导航到您的Hugging Face个人资料时，您应该看到您新创建的模型仓库。点击**文件**选项卡将显示您已上传到仓库的所有文件。
有关如何创建和上传文件到仓库的更多详细信息，请参考Hub文档[这里](https://huggingface.co/docs/hub/how-to-upstream)。
## 使用Web界面上传
喜欢无代码方法的用户可以通过Hugging Face的Web界面上传模型。访问[huggingface.co/new](https://huggingface.co/new)创建一个新的仓库：
从这里开始，添加一些关于您的模型的信息：
- 选择仓库的**所有者**。这可以是您本人或者您所属的任何组织。
- 为您的项目选择一个名称，该名称也将成为仓库的名称。
- 选择您的模型是公开还是私有。
- 指定您的模型的许可证使用情况。
现在点击**文件**选项卡，然后点击**添加文件**按钮将一个新文件上传到你的仓库。接着拖放一个文件进行上传，并添加提交信息。
## 添加模型卡片
为了确保用户了解您的模型的能力、限制、潜在偏差和伦理考虑，请在仓库中添加一个模型卡片。模型卡片在`README.md`文件中定义。你可以通过以下方式添加模型卡片：
* 手动创建并上传一个`README.md`文件。
* 在你的模型仓库中点击**编辑模型卡片**按钮。
可以参考DistilBert的[模型卡片](https://huggingface.co/distilbert/distilbert-base-uncased)来了解模型卡片应该包含的信息类型。有关您可以在`README.md`文件中控制的更多选项的细节，例如模型的碳足迹或小部件示例，请参考文档[这里](https://huggingface.co/docs/hub/models-cards)。
| 0
|
mavonic_private_repos/transformers/docs/source
|
mavonic_private_repos/transformers/docs/source/zh/custom_models.md
|
<!--Copyright 2020 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
specific language governing permissions and limitations under the License.
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer.
-->
# 共享自定义模型
🤗 Transformers 库设计得易于扩展。每个模型的代码都在仓库给定的子文件夹中，没有进行抽象，因此你可以轻松复制模型代码文件并根据需要进行调整。
如果你要编写全新的模型，从头开始可能更容易。在本教程中，我们将向你展示如何编写自定义模型及其配置，以便可以在 Transformers 中使用它；以及如何与社区共享它（及其依赖的代码），以便任何人都可以使用，即使它不在 🤗 Transformers 库中。
我们将以 ResNet 模型为例，通过将 [timm 库](https://github.com/rwightman/pytorch-image-models) 的 ResNet 类封装到 [`PreTrainedModel`] 中来进行说明。
## 编写自定义配置
在深入研究模型之前，让我们首先编写其配置。模型的配置是一个对象，其中包含构建模型所需的所有信息。我们将在下一节中看到，模型只能接受一个 `config` 来进行初始化，因此我们很需要使该对象尽可能完整。
我们将采用一些我们可能想要调整的 ResNet 类的参数举例。不同的配置将为我们提供不同类型可能的 ResNet 模型。在确认其中一些参数的有效性后，我们只需存储这些参数。
```python
from transformers import PretrainedConfig
from typing import List
class ResnetConfig(PretrainedConfig):
model_type = "resnet"
def __init__(
self,
block_type="bottleneck",
layers: List[int] = [3, 4, 6, 3],
num_classes: int = 1000,
input_channels: int = 3,
cardinality: int = 1,
base_width: int = 64,
stem_width: int = 64,
stem_type: str = "",
avg_down: bool = False,
**kwargs,
):
if block_type not in ["basic", "bottleneck"]:
raise ValueError(f"`block_type` must be 'basic' or bottleneck', got {block_type}.")
if stem_type not in ["", "deep", "deep-tiered"]:
raise ValueError(f"`stem_type` must be '', 'deep' or 'deep-tiered', got {stem_type}.")
self.block_type = block_type
self.layers = layers
self.num_classes = num_classes
self.input_channels = input_channels
self.cardinality = cardinality
self.base_width = base_width
self.stem_width = stem_width
self.stem_type = stem_type
self.avg_down = avg_down
super().__init__(**kwargs)
```
编写自定义配置时需要记住的三个重要事项如下：
- 必须继承自 `PretrainedConfig`，
- `PretrainedConfig` 的 `__init__` 方法必须接受任何 kwargs，
- 这些 `kwargs` 需要传递给超类的 `__init__` 方法。
继承是为了确保你获得来自 🤗 Transformers 库的所有功能，而另外两个约束源于 `PretrainedConfig` 的字段比你设置的字段多。在使用 `from_pretrained` 方法重新加载配置时，这些字段需要被你的配置接受，然后传递给超类。
为你的配置定义 `model_type`（此处为 `model_type="resnet"`）不是必须的，除非你想使用自动类注册你的模型（请参阅最后一节）。
做完这些以后，就可以像使用库里任何其他模型配置一样，轻松地创建和保存配置。以下代码展示了如何创建并保存 resnet50d 配置：
```py
resnet50d_config = ResnetConfig(block_type="bottleneck", stem_width=32, stem_type="deep", avg_down=True)
resnet50d_config.save_pretrained("custom-resnet")
```
这行代码将在 `custom-resnet` 文件夹内保存一个名为 `config.json` 的文件。然后，你可以使用 `from_pretrained` 方法重新加载配置：
```py
resnet50d_config = ResnetConfig.from_pretrained("custom-resnet")
```
你还可以使用 [`PretrainedConfig`] 类的任何其他方法，例如 [`~PretrainedConfig.push_to_hub`]，直接将配置上传到 Hub。
## 编写自定义模型
有了 ResNet 配置后，就可以继续编写模型了。实际上，我们将编写两个模型：一个模型用于从一批图像中提取隐藏特征（类似于 [`BertModel`]），另一个模型适用于图像分类（类似于 [`BertForSequenceClassification`]）。
正如之前提到的，我们只会编写一个松散的模型包装，以使示例保持简洁。在编写此类之前，只需要建立起块类型（block types）与实际块类（block classes）之间的映射。然后，通过将所有内容传递给ResNet类，从配置中定义模型：
```py
from transformers import PreTrainedModel
from timm.models.resnet import BasicBlock, Bottleneck, ResNet
from .configuration_resnet import ResnetConfig
BLOCK_MAPPING = {"basic": BasicBlock, "bottleneck": Bottleneck}
class ResnetModel(PreTrainedModel):
config_class = ResnetConfig
def __init__(self, config):
super().__init__(config)
block_layer = BLOCK_MAPPING[config.block_type]
self.model = ResNet(
block_layer,
config.layers,
num_classes=config.num_classes,
in_chans=config.input_channels,
cardinality=config.cardinality,
base_width=config.base_width,
stem_width=config.stem_width,
stem_type=config.stem_type,
avg_down=config.avg_down,
)
def forward(self, tensor):
return self.model.forward_features(tensor)
```
对用于进行图像分类的模型，我们只需更改前向方法：
```py
import torch
class ResnetModelForImageClassification(PreTrainedModel):
config_class = ResnetConfig
def __init__(self, config):
super().__init__(config)
block_layer = BLOCK_MAPPING[config.block_type]
self.model = ResNet(
block_layer,
config.layers,
num_classes=config.num_classes,
in_chans=config.input_channels,
cardinality=config.cardinality,
base_width=config.base_width,
stem_width=config.stem_width,
stem_type=config.stem_type,
avg_down=config.avg_down,
)
def forward(self, tensor, labels=None):
logits = self.model(tensor)
if labels is not None:
loss = torch.nn.cross_entropy(logits, labels)
return {"loss": loss, "logits": logits}
return {"logits": logits}
```
在这两种情况下，请注意我们如何继承 `PreTrainedModel` 并使用 `config` 调用了超类的初始化（有点像编写常规的torch.nn.Module）。设置 `config_class` 的那行代码不是必须的，除非你想使用自动类注册你的模型（请参阅最后一节）。
<Tip>
如果你的模型与库中的某个模型非常相似，你可以重用与该模型相同的配置。
</Tip>
你可以让模型返回任何你想要的内容，但是像我们为 `ResnetModelForImageClassification` 做的那样返回一个字典，并在传递标签时包含loss，可以使你的模型能够在 [`Trainer`] 类中直接使用。只要你计划使用自己的训练循环或其他库进行训练，也可以使用其他输出格式。
现在我们已经有了模型类，让我们创建一个：
```py
resnet50d = ResnetModelForImageClassification(resnet50d_config)
```
同样的，你可以使用 [`PreTrainedModel`] 的任何方法，比如 [`~PreTrainedModel.save_pretrained`] 或者 [`~PreTrainedModel.push_to_hub`]。我们将在下一节中使用第二种方法，并了解如何如何使用我们的模型的代码推送模型权重。但首先，让我们在模型内加载一些预训练权重。
在你自己的用例中，你可能会在自己的数据上训练自定义模型。为了快速完成本教程，我们将使用 resnet50d 的预训练版本。由于我们的模型只是它的包装，转移这些权重将会很容易：
```py
import timm
pretrained_model = timm.create_model("resnet50d", pretrained=True)
resnet50d.model.load_state_dict(pretrained_model.state_dict())
```
现在让我们看看，如何确保在执行 [`~PreTrainedModel.save_pretrained`] 或 [`~PreTrainedModel.push_to_hub`] 时，模型的代码被保存。
## 将代码发送到 Hub
<Tip warning={true}>
此 API 是实验性的，未来的发布中可能会有一些轻微的不兼容更改。
</Tip>
首先，确保你的模型在一个 `.py` 文件中完全定义。只要所有文件都位于同一目录中，它就可以依赖于某些其他文件的相对导入（目前我们还不为子模块支持此功能）。对于我们的示例，我们将在当前工作目录中名为 `resnet_model` 的文件夹中定义一个 `modeling_resnet.py` 文件和一个 `configuration_resnet.py` 文件。 配置文件包含 `ResnetConfig` 的代码，模型文件包含 `ResnetModel` 和 `ResnetModelForImageClassification` 的代码。
```
.
└── resnet_model
├── __init__.py
├── configuration_resnet.py
└── modeling_resnet.py
```
`__init__.py` 可以为空，它的存在只是为了让 Python 检测到 `resnet_model` 可以用作模块。
<Tip warning={true}>
如果从库中复制模型文件，你需要将文件顶部的所有相对导入替换为从 `transformers` 包中的导入。
</Tip>
请注意，你可以重用（或子类化）现有的配置/模型。
要与社区共享您的模型，请参照以下步骤：首先从新创建的文件中导入ResNet模型和配置：
```py
from resnet_model.configuration_resnet import ResnetConfig
from resnet_model.modeling_resnet import ResnetModel, ResnetModelForImageClassification
```
接下来，你需要告诉库，当使用 `save_pretrained` 方法时，你希望复制这些对象的代码文件，并将它们正确注册到给定的 Auto 类（特别是对于模型），只需要运行以下代码：
```py
ResnetConfig.register_for_auto_class()
ResnetModel.register_for_auto_class("AutoModel")
ResnetModelForImageClassification.register_for_auto_class("AutoModelForImageClassification")
```
请注意，对于配置（只有一个自动类 [`AutoConfig`]），不需要指定自动类，但对于模型来说情况不同。 你的自定义模型可能适用于许多不同的任务，因此你必须指定哪一个自动类适合你的模型。
接下来，让我们像之前一样创建配置和模型：
```py
resnet50d_config = ResnetConfig(block_type="bottleneck", stem_width=32, stem_type="deep", avg_down=True)
resnet50d = ResnetModelForImageClassification(resnet50d_config)
pretrained_model = timm.create_model("resnet50d", pretrained=True)
resnet50d.model.load_state_dict(pretrained_model.state_dict())
```
现在要将模型推送到集线器，请确保你已登录。你看可以在终端中运行以下命令：
```bash
huggingface-cli login
```
或者在笔记本中运行以下代码：
```py
from huggingface_hub import notebook_login
notebook_login()
```
然后，可以这样将模型推送到自己的命名空间（或你所属的组织）：
```py
resnet50d.push_to_hub("custom-resnet50d")
```
除了模型权重和 JSON 格式的配置外，这行代码也会复制 `custom-resnet50d` 文件夹内的模型以及配置的 `.py` 文件并将结果上传至 Hub。你可以在此[模型仓库](https://huggingface.co/sgugger/custom-resnet50d)中查看结果。
有关推推送至 Hub 方法的更多信息，请参阅[共享教程](model_sharing)。
## 使用带有自定义代码的模型
可以使用自动类（auto-classes）和 `from_pretrained` 方法，使用模型仓库里带有自定义代码的配置、模型或分词器文件。所有上传到 Hub 的文件和代码都会进行恶意软件扫描（有关更多信息，请参阅 [Hub 安全](https://huggingface.co/docs/hub/security#malware-scanning) 文档）, 但你仍应查看模型代码和作者，以避免在你的计算机上执行恶意代码。 设置 `trust_remote_code=True` 以使用带有自定义代码的模型：
```py
from transformers import AutoModelForImageClassification
model = AutoModelForImageClassification.from_pretrained("sgugger/custom-resnet50d", trust_remote_code=True)
```
我们强烈建议为 `revision` 参数传递提交哈希（commit hash），以确保模型的作者没有使用一些恶意的代码行更新了代码（除非您完全信任模型的作者）。
```py
commit_hash = "ed94a7c6247d8aedce4647f00f20de6875b5b292"
model = AutoModelForImageClassification.from_pretrained(
"sgugger/custom-resnet50d", trust_remote_code=True, revision=commit_hash
)
```
在 Hub 上浏览模型仓库的提交历史时，有一个按钮可以轻松复制任何提交的提交哈希。
## 将自定义代码的模型注册到自动类
如果你在编写一个扩展 🤗 Transformers 的库，你可能想要扩展自动类以包含您自己的模型。这与将代码推送到 Hub 不同，因为用户需要导入你的库才能获取自定义模型（与从 Hub 自动下载模型代码相反）。
只要你的配置 `model_type` 属性与现有模型类型不同，并且你的模型类有正确的 `config_class` 属性，你可以像这样将它们添加到自动类中：
```py
from transformers import AutoConfig, AutoModel, AutoModelForImageClassification
AutoConfig.register("resnet", ResnetConfig)
AutoModel.register(ResnetConfig, ResnetModel)
AutoModelForImageClassification.register(ResnetConfig, ResnetModelForImageClassification)
```
请注意，将自定义配置注册到 [`AutoConfig`] 时，使用的第一个参数需要与自定义配置的 `model_type` 匹配；而将自定义模型注册到任何自动模型类时，使用的第一个参数需要与 `config_class` 匹配。
| 0
|
mavonic_private_repos/transformers/docs/source
|
mavonic_private_repos/transformers/docs/source/zh/fsdp.md
|
<!--
Copyright 2023 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
specific language governing permissions and limitations under the License.
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer.
-->
# 完全分片数据并行
[完全分片数据并行（FSDP）](https://pytorch.org/blog/introducing-pytorch-fully-sharded-data-parallel-api/)是一种数据并行方法，
它将模型的参数、梯度和优化器状态在可用 GPU（也称为 Worker 或 *rank*）的数量上进行分片。
与[分布式数据并行（DDP）](https://pytorch.org/docs/stable/generated/torch.nn.parallel.DistributedDataParallel.html)不同，
FSDP 减少了内存使用量，因为模型在每个 GPU 上都被复制了一次。这就提高了 GPU 内存效率，
使您能够用较少的 GPU 训练更大的模型。FSDP 已经集成到 Accelerate 中，
这是一个用于在分布式环境中轻松管理训练的库，这意味着可以从 [`Trainer`] 类中调用这个库。
在开始之前，请确保已安装 Accelerate，并且至少使用 PyTorch 2.1.0 或更高版本。
```bash
pip install accelerate
```
## FSDP 配置
首先，运行 [`accelerate config`](https://huggingface.co/docs/accelerate/package_reference/cli#accelerate-config)
命令为您的训练环境创建一个配置文件。Accelerate 使用此配置文件根据您在 `accelerate config`
中选择的训练选项来自动搭建正确的训练环境。
```bash
accelerate config
```
运行 `accelerate config` 时，您将被提示一系列选项来配置训练环境。
本节涵盖了一些最重要的 FSDP 选项。要了解有关其他可用的 FSDP 选项的更多信息，
请查阅 [fsdp_config](https://huggingface.co/docs/transformers/main_classes/trainer#transformers.TrainingArguments.fsdp_config) 参数。
### 分片策略
FSDP 提供了多种可选择的分片策略：
- `FULL_SHARD` - 将模型参数、梯度和优化器状态跨 Worker 进行分片；为此选项选择 `1`
- `SHARD_GRAD_OP`- 将梯度和优化器状态跨 Worker 进行分片；为此选项选择 `2`
- `NO_SHARD` - 不分片任何内容（这等同于 DDP）；为此选项选择 `3`
- `HYBRID_SHARD` - 在每个 Worker 中分片模型参数、梯度和优化器状态，其中每个 Worker 也有完整副本；为此选项选择 `4`
- `HYBRID_SHARD_ZERO2` - 在每个 Worker 中分片梯度和优化器状态，其中每个 Worker 也有完整副本；为此选项选择 `5`
这由 `fsdp_sharding_strategy` 标志启用。
### CPU 卸载
当参数和梯度在不使用时可以卸载到 CPU 上，以节省更多 GPU 内存并帮助您适应即使 FSDP 也不足以容纳大型模型的情况。
在运行 `accelerate config` 时，通过设置 `fsdp_offload_params: true` 来启用此功能。
### 包装策略
FSDP 是通过包装网络中的每个层来应用的。通常，包装是以嵌套方式应用的，其中完整的权重在每次前向传递后被丢弃，
以便在下一层使用内存。**自动包装**策略是实现这一点的最简单方法，您不需要更改任何代码。
您应该选择 `fsdp_auto_wrap_policy: TRANSFORMER_BASED_WRAP` 来包装一个 Transformer 层，
并且 `fsdp_transformer_layer_cls_to_wrap` 来指定要包装的层（例如 `BertLayer`）。
否则，您可以选择基于大小的包装策略，其中如果一层的参数超过一定数量，则应用 FSDP。通过设置
`fsdp_wrap_policy: SIZE_BASED_WRAP` 和 `min_num_param` 来启用此功能，将参数设置为所需的大小阈值。
### 检查点
应该使用 `fsdp_state_dict_type: SHARDED_STATE_DICT` 来保存中间检查点，
因为在排名 0 上保存完整状态字典需要很长时间，通常会导致 `NCCL Timeout` 错误，因为在广播过程中会无限期挂起。
您可以使用 [`~accelerate.Accelerator.load_state`]` 方法加载分片状态字典以恢复训练。
```py
# 包含检查点的目录
accelerator.load_state("ckpt")
```
然而，当训练结束时，您希望保存完整状态字典，因为分片状态字典仅与 FSDP 兼容。
```py
if trainer.is_fsdp_enabled:
trainer.accelerator.state.fsdp_plugin.set_state_dict_type("FULL_STATE_DICT")
trainer.save_model(script_args.output_dir)
```
### TPU
[PyTorch XLA](https://pytorch.org/xla/release/2.1/index.html) 支持用于 TPUs 的 FSDP 训练，
可以通过修改由 `accelerate config` 生成的 FSDP 配置文件来启用。除了上面指定的分片策略和包装选项外，
您还可以将以下参数添加到文件中。
```yaml
xla: True # 必须设置为 True 以启用 PyTorch/XLA
xla_fsdp_settings: # XLA 特定的 FSDP 参数
xla_fsdp_grad_ckpt: True # 使用梯度检查点
```
[`xla_fsdp_settings`](https://github.com/pytorch/xla/blob/2e6e183e0724818f137c8135b34ef273dea33318/torch_xla/distributed/fsdp/xla_fully_sharded_data_parallel.py#L128)
允许您配置用于 FSDP 的额外 XLA 特定参数。
## 启动训练
FSDP 配置文件示例如下所示：
```yaml
compute_environment: LOCAL_MACHINE
debug: false
distributed_type: FSDP
downcast_bf16: "no"
fsdp_config:
fsdp_auto_wrap_policy: TRANSFORMER_BASED_WRAP
fsdp_backward_prefetch_policy: BACKWARD_PRE
fsdp_cpu_ram_efficient_loading: true
fsdp_forward_prefetch: false
fsdp_offload_params: true
fsdp_sharding_strategy: 1
fsdp_state_dict_type: SHARDED_STATE_DICT
fsdp_sync_module_states: true
fsdp_transformer_layer_cls_to_wrap: BertLayer
fsdp_use_orig_params: true
machine_rank: 0
main_training_function: main
mixed_precision: bf16
num_machines: 1
num_processes: 2
rdzv_backend: static
same_network: true
tpu_env: []
tpu_use_cluster: false
tpu_use_sudo: false
use_cpu: false
```
要启动训练，请运行 [`accelerate launch`](https://huggingface.co/docs/accelerate/package_reference/cli#accelerate-launch)
命令，它将自动使用您之前使用 `accelerate config` 创建的配置文件。
```bash
accelerate launch my-trainer-script.py
```
```bash
accelerate launch --fsdp="full shard" --fsdp_config="path/to/fsdp_config/ my-trainer-script.py
```
## 下一步
FSDP 在大规模模型训练方面是一个强大的工具，您可以使用多个 GPU 或 TPU。
通过分片模型参数、优化器和梯度状态，甚至在它们不活动时将其卸载到 CPU 上，
FSDP 可以减少大规模训练的高成本。如果您希望了解更多信息，下面的内容可能会有所帮助：
- 深入参考 Accelerate 指南，了解有关
[FSDP](https://huggingface.co/docs/accelerate/usage_guides/fsdp)的更多信息。
- 阅读[介绍 PyTorch 完全分片数据并行（FSDP）API](https://pytorch.org/blog/introducing-pytorch-fully-sharded-data-parallel-api/) 博文。
- 阅读[使用 FSDP 在云 TPU 上扩展 PyTorch 模型](https://pytorch.org/blog/scaling-pytorch-models-on-cloud-tpus-with-fsdp/)博文。
| 0
|
mavonic_private_repos/transformers/docs/source
|
mavonic_private_repos/transformers/docs/source/zh/pipeline_tutorial.md
|
<!--Copyright 2022 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
specific language governing permissions and limitations under the License.
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer.
-->
# 推理pipeline
[`pipeline`] 让使用[Hub](https://huggingface.co/models)上的任何模型进行任何语言、计算机视觉、语音以及多模态任务的推理变得非常简单。即使您对特定的模态没有经验，或者不熟悉模型的源码，您仍然可以使用[`pipeline`]进行推理！本教程将教您：
- 如何使用[`pipeline`] 进行推理。
- 如何使用特定的`tokenizer`(分词器)或模型。
- 如何使用[`pipeline`] 进行音频、视觉和多模态任务的推理。
<Tip>
请查看[`pipeline`]文档以获取已支持的任务和可用参数的完整列表。
</Tip>
## Pipeline使用
虽然每个任务都有一个关联的[`pipeline`]，但使用通用的抽象的[`pipeline`]更加简单，其中包含所有特定任务的`pipelines`。[`pipeline`]会自动加载一个默认模型和一个能够进行任务推理的预处理类。让我们以使用[`pipeline`]进行自动语音识别（ASR）或语音转文本为例。
1. 首先，创建一个[`pipeline`]并指定推理任务：
```py
>>> from transformers import pipeline
>>> transcriber = pipeline(task="automatic-speech-recognition")
```
2. 将您的输入传递给[`pipeline`]。对于语音识别，这通常是一个音频输入文件：
```py
>>> transcriber("https://huggingface.co/datasets/Narsil/asr_dummy/resolve/main/mlk.flac")
{'text': 'I HAVE A DREAM BUT ONE DAY THIS NATION WILL RISE UP LIVE UP THE TRUE MEANING OF ITS TREES'}
```
您没有得到您期望的结果？可以在Hub上查看一些[最受欢迎的自动语音识别模型](https://huggingface.co/models?pipeline_tag=automatic-speech-recognition&sort=trending)
，看看是否可以获得更好的转录。
让我们尝试来自 OpenAI 的[Whisper large-v2](https://huggingface.co/openai/whisper-large) 模型。Whisperb比Wav2Vec2晚2年发布，使用接近10倍的数据进行了训练。因此，它在大多数下游基准测试上击败了Wav2Vec2。
它还具有预测标点和大小写的附加优势，而Wav2Vec2则无法实现这些功能。
让我们在这里尝试一下，看看它的表现如何：
```py
>>> transcriber = pipeline(model="openai/whisper-large-v2")
>>> transcriber("https://huggingface.co/datasets/Narsil/asr_dummy/resolve/main/mlk.flac")
{'text': ' I have a dream that one day this nation will rise up and live out the true meaning of its creed.'}
```
现在这个结果看起来更准确了！要进行深入的Wav2Vec2与Whisper比较，请参阅[音频变换器课程](https://huggingface.co/learn/audio-course/chapter5/asr_models)。
我们鼓励您在 Hub 上查看不同语言的模型，以及专业领域的模型等。您可以在Hub上直接查看并比较模型的结果，以确定是否适合或处理边缘情况是否比其他模型更好。如果您没有找到适用于您的用例的模型，您始终可以[训练](training)自己的模型！
如果您有多个输入，您可以将输入作为列表传递：
```py
transcriber(
[
"https://huggingface.co/datasets/Narsil/asr_dummy/resolve/main/mlk.flac",
"https://huggingface.co/datasets/Narsil/asr_dummy/resolve/main/1.flac",
]
)
```
`Pipelines`非常适合用于测试，因为从一个模型切换到另一个模型非常琐碎；但是，还有一些方法可以将它们优化后用于大型工作负载而不仅仅是测试。请查看以下指南，深入探讨如何迭代整个数据集或在Web服务器中使用`Pipelines`：
* [在数据集上使用流水线](#using-pipelines-on-a-dataset)
* [在Web服务器中使用流水线](./pipeline_webserver)
## 参数
[`pipeline`] 支持许多参数；有些是适用于特定任务的，而有些适用于所有`pipeline`。通常情况下，您可以在任何地方指定对应参数：
```py
transcriber = pipeline(model="openai/whisper-large-v2", my_parameter=1)
out = transcriber(...) # This will use `my_parameter=1`.
out = transcriber(..., my_parameter=2) # This will override and use `my_parameter=2`.
out = transcriber(...) # This will go back to using `my_parameter=1`.
```
让我们查看其中的三个重要参数：
### 设备
如果您使用 `device=n`，`pipeline`会自动将模型放在指定的设备上。无论您使用PyTorch还是Tensorflow，这都可以工作。
```py
transcriber = pipeline(model="openai/whisper-large-v2", device=0)
```
如果模型对于单个GPU来说过于庞大，并且您正在使用PyTorch，您可以设置 `device_map="auto"` 以自动确定如何加载和存储模型权重。使用 `device_map` 参数需要安装🤗 [Accelerate](https://huggingface.co/docs/accelerate) 软件包：
```bash
pip install --upgrade accelerate
```
以下代码会自动在各个设备上加载和存储模型权重：
```py
transcriber = pipeline(model="openai/whisper-large-v2", device_map="auto")
```
请注意，如果传递了 `device_map="auto"`，在实例化您的 `pipeline` 时不需要添加 `device=device` 参数，否则可能会遇到一些意外的状况！
### 批量大小
默认情况下，`pipelines`不会进行批量推理，原因在[这里](https://huggingface.co/docs/transformers/main_classes/pipelines#pipeline-batching)详细解释。因为批处理不一定更快，实际上在某些情况下可能会更慢。
但如果在您的用例中起作用，您可以使用：
```py
transcriber = pipeline(model="openai/whisper-large-v2", device=0, batch_size=2)
audio_filenames = [f"https://huggingface.co/datasets/Narsil/asr_dummy/resolve/main/{i}.flac" for i in range(1, 5)]
texts = transcriber(audio_filenames)
```
以上代码会在提供的4个音频文件上运行`pipeline`，它会将它们以2个一组的批次传递给模型（模型在GPU上，此时批处理更有可能有所帮助），而您无需编写额外的代码。输出应始终与没有批处理时收到的结果相一致。它只是一种帮助您更快地使用`pipeline`的方式。
`pipeline`也可以减轻一些批处理的复杂性，因为对于某些`pipeline`，需要将单个项目（如长音频文件）分成多个部分以供模型处理。`pipeline`为您执行这种[*chunk batching*](./main_classes/pipelines#pipeline-chunk-batching)。
### 任务特定参数
所有任务都提供了特定于任务的参数，这些参数提供额外的灵活性和选择，以帮助您完成工作。
例如，[`transformers.AutomaticSpeechRecognitionPipeline.__call__`] 方法具有一个 `return_timestamps` 参数，对于字幕视频似乎很有帮助：
```py
>>> transcriber = pipeline(model="openai/whisper-large-v2", return_timestamps=True)
>>> transcriber("https://huggingface.co/datasets/Narsil/asr_dummy/resolve/main/mlk.flac")
{'text': ' I have a dream that one day this nation will rise up and live out the true meaning of its creed.', 'chunks': [{'timestamp': (0.0, 11.88), 'text': ' I have a dream that one day this nation will rise up and live out the true meaning of its'}, {'timestamp': (11.88, 12.38), 'text': ' creed.'}]}
```
正如您所看到的，模型推断出了文本，还输出了各个句子发音的**时间**。
每个任务都有许多可用的参数，因此请查看每个任务的API参考，以了解您可以进行哪些调整！例如，[`~transformers.AutomaticSpeechRecognitionPipeline`] 具有 `chunk_length_s` 参数，对于处理非常长的音频文件（例如，为整部电影或长达一小时的视频配字幕）非常有帮助，这通常是模型无法单独处理的：
```python
>>> transcriber = pipeline(model="openai/whisper-large-v2", chunk_length_s=30, return_timestamps=True)
>>> transcriber("https://huggingface.co/datasets/sanchit-gandhi/librispeech_long/resolve/main/audio.wav")
{'text': " Chapter 16. I might have told you of the beginning of this liaison in a few lines, but I wanted you to see every step by which we came. I, too, agree to whatever Marguerite wished, Marguerite to be unable to live apart from me. It was the day after the evening...
```
如果您找不到一个真正有帮助的参数，欢迎[提出请求](https://github.com/huggingface/transformers/issues/new?assignees=&labels=feature&template=feature-request.yml)！
## 在数据集上使用pipelines
`pipelines`也可以对大型数据集进行推理。我们建议使用迭代器来完成这一任务，这是最简单的方法：
```py
def data():
for i in range(1000):
yield f"My example {i}"
pipe = pipeline(model="openai-community/gpt2", device=0)
generated_characters = 0
for out in pipe(data()):
generated_characters += len(out[0]["generated_text"])
```
迭代器 `data()` 会产生每个结果，`pipelines`会自动识别输入为可迭代对象，并在GPU上处理数据的同时开始获取数据（在底层使用[DataLoader](https://pytorch.org/docs/stable/data.html#torch.utils.data.DataLoader)）。这一点非常重要，因为您不必为整个数据集分配内存，可以尽可能快地将数据传送到GPU。
由于批处理可以加速处理，因此在这里尝试调整 `batch_size` 参数可能会很有用。
迭代数据集的最简单方法就是从🤗 [Datasets](https://github.com/huggingface/datasets/) 中加载数据集：
```py
# KeyDataset is a util that will just output the item we're interested in.
from transformers.pipelines.pt_utils import KeyDataset
from datasets import load_dataset
pipe = pipeline(model="hf-internal-testing/tiny-random-wav2vec2", device=0)
dataset = load_dataset("hf-internal-testing/librispeech_asr_dummy", "clean", split="validation[:10]")
for out in pipe(KeyDataset(dataset, "audio")):
print(out)
```
## 在Web服务器上使用pipelines
<Tip>
创建推理引擎是一个复杂的主题，值得有自己的页面。
</Tip>
[链接](./pipeline_webserver)
## 视觉流水线
对于视觉任务，使用[`pipeline`] 几乎是相同的。
指定您的任务并将图像传递给分类器。图像可以是链接、本地路径或base64编码的图像。例如，下面显示的是哪种品种的猫？
```py
>>> from transformers import pipeline
>>> vision_classifier = pipeline(model="google/vit-base-patch16-224")
>>> preds = vision_classifier(
... images="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/pipeline-cat-chonk.jpeg"
... )
>>> preds = [{"score": round(pred["score"], 4), "label": pred["label"]} for pred in preds]
>>> preds
[{'score': 0.4335, 'label': 'lynx, catamount'}, {'score': 0.0348, 'label': 'cougar, puma, catamount, mountain lion, painter, panther, Felis concolor'}, {'score': 0.0324, 'label': 'snow leopard, ounce, Panthera uncia'}, {'score': 0.0239, 'label': 'Egyptian cat'}, {'score': 0.0229, 'label': 'tiger cat'}]
```
## 文本流水线
对于NLP任务，使用[`pipeline`] 几乎是相同的。
```py
>>> from transformers import pipeline
>>> # This model is a `zero-shot-classification` model.
>>> # It will classify text, except you are free to choose any label you might imagine
>>> classifier = pipeline(model="facebook/bart-large-mnli")
>>> classifier(
... "I have a problem with my iphone that needs to be resolved asap!!",
... candidate_labels=["urgent", "not urgent", "phone", "tablet", "computer"],
... )
{'sequence': 'I have a problem with my iphone that needs to be resolved asap!!', 'labels': ['urgent', 'phone', 'computer', 'not urgent', 'tablet'], 'scores': [0.504, 0.479, 0.013, 0.003, 0.002]}
```
## 多模态流水线
[`pipeline`] 支持多个模态。例如，视觉问题回答（VQA）任务结合了文本和图像。请随意使用您喜欢的任何图像链接和您想要问关于该图像的问题。图像可以是URL或图像的本地路径。
例如，如果您使用这个[invoice image](https://huggingface.co/spaces/impira/docquery/resolve/2359223c1837a7587402bda0f2643382a6eefeab/invoice.png)：
```py
>>> from transformers import pipeline
>>> vqa = pipeline(model="impira/layoutlm-document-qa")
>>> output = vqa(
... image="https://huggingface.co/spaces/impira/docquery/resolve/2359223c1837a7587402bda0f2643382a6eefeab/invoice.png",
... question="What is the invoice number?",
... )
>>> output[0]["score"] = round(output[0]["score"], 3)
>>> output
[{'score': 0.425, 'answer': 'us-001', 'start': 16, 'end': 16}]
```
<Tip>
要运行上面的示例，除了🤗 Transformers之外，您需要安装[`pytesseract`](https://pypi.org/project/pytesseract/)。
```bash
sudo apt install -y tesseract-ocr
pip install pytesseract
```
</Tip>
## 在大模型上使用🤗 `accelerate`和`pipeline`：
您可以轻松地使用🤗 `accelerate`在大模型上运行 `pipeline`！首先确保您已经使用 `pip install accelerate` 安装了 `accelerate`。
首先使用 `device_map="auto"` 加载您的模型！我们将在示例中使用 `facebook/opt-1.3b`。
```py
# pip install accelerate
import torch
from transformers import pipeline
pipe = pipeline(model="facebook/opt-1.3b", torch_dtype=torch.bfloat16, device_map="auto")
output = pipe("This is a cool example!", do_sample=True, top_p=0.95)
```
如果安装 `bitsandbytes` 并添加参数 `load_in_8bit=True`，您还可以传递8位加载的模型。
```py
# pip install accelerate bitsandbytes
import torch
from transformers import pipeline
pipe = pipeline(model="facebook/opt-1.3b", device_map="auto", model_kwargs={"load_in_8bit": True})
output = pipe("This is a cool example!", do_sample=True, top_p=0.95)
```
请注意，您可以将`checkpoint `替换为任何支持大模型加载的Hugging Face模型，比如BLOOM！
| 0
|
mavonic_private_repos/transformers/docs/source
|
mavonic_private_repos/transformers/docs/source/zh/add_new_pipeline.md
|
<!--
Copyright 2020 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer.
-->
# 如何创建自定义流水线？
在本指南中，我们将演示如何创建一个自定义流水线并分享到 [Hub](https://hf.co/models)，或将其添加到 🤗 Transformers 库中。
首先，你需要决定流水线将能够接受的原始条目。它可以是字符串、原始字节、字典或任何看起来最可能是期望的输入。
尽量保持输入为纯 Python 语言，因为这样可以更容易地实现兼容性（甚至通过 JSON 在其他语言之间）。
这些将是流水线 (`preprocess`) 的 `inputs`。
然后定义 `outputs`。与 `inputs` 相同的策略。越简单越好。这些将是 `postprocess` 方法的输出。
首先继承基类 `Pipeline`，其中包含实现 `preprocess`、`_forward`、`postprocess` 和 `_sanitize_parameters` 所需的 4 个方法。
```python
from transformers import Pipeline
class MyPipeline(Pipeline):
def _sanitize_parameters(self, **kwargs):
preprocess_kwargs = {}
if "maybe_arg" in kwargs:
preprocess_kwargs["maybe_arg"] = kwargs["maybe_arg"]
return preprocess_kwargs, {}, {}
def preprocess(self, inputs, maybe_arg=2):
model_input = Tensor(inputs["input_ids"])
return {"model_input": model_input}
def _forward(self, model_inputs):
# model_inputs == {"model_input": model_input}
outputs = self.model(**model_inputs)
# Maybe {"logits": Tensor(...)}
return outputs
def postprocess(self, model_outputs):
best_class = model_outputs["logits"].softmax(-1)
return best_class
```
这种分解的结构旨在为 CPU/GPU 提供相对无缝的支持，同时支持在不同线程上对 CPU 进行预处理/后处理。
`preprocess` 将接受最初定义的输入，并将其转换为可供模型输入的内容。它可能包含更多信息，通常是一个 `Dict`。
`_forward` 是实现细节，不应直接调用。`forward` 是首选的调用方法，因为它包含保障措施，以确保一切都在预期的设备上运作。
如果任何内容与实际模型相关，它应该属于 `_forward` 方法，其他内容应该在 preprocess/postprocess 中。
`postprocess` 方法将接受 `_forward` 的输出，并将其转换为之前确定的最终输出。
`_sanitize_parameters` 存在是为了允许用户在任何时候传递任何参数，无论是在初始化时 `pipeline(...., maybe_arg=4)`
还是在调用时 `pipe = pipeline(...); output = pipe(...., maybe_arg=4)`。
`_sanitize_parameters` 的返回值是将直接传递给 `preprocess`、`_forward` 和 `postprocess` 的 3 个关键字参数字典。
如果调用方没有使用任何额外参数调用，则不要填写任何内容。这样可以保留函数定义中的默认参数，这总是更"自然"的。
在分类任务中，一个经典的例子是在后处理中使用 `top_k` 参数。
```python
>>> pipe = pipeline("my-new-task")
>>> pipe("This is a test")
[{"label": "1-star", "score": 0.8}, {"label": "2-star", "score": 0.1}, {"label": "3-star", "score": 0.05}
{"label": "4-star", "score": 0.025}, {"label": "5-star", "score": 0.025}]
>>> pipe("This is a test", top_k=2)
[{"label": "1-star", "score": 0.8}, {"label": "2-star", "score": 0.1}]
```
为了实现这一点，我们将更新我们的 `postprocess` 方法，将默认参数设置为 `5`，
并编辑 `_sanitize_parameters` 方法，以允许这个新参数。
```python
def postprocess(self, model_outputs, top_k=5):
best_class = model_outputs["logits"].softmax(-1)
# Add logic to handle top_k
return best_class
def _sanitize_parameters(self, **kwargs):
preprocess_kwargs = {}
if "maybe_arg" in kwargs:
preprocess_kwargs["maybe_arg"] = kwargs["maybe_arg"]
postprocess_kwargs = {}
if "top_k" in kwargs:
postprocess_kwargs["top_k"] = kwargs["top_k"]
return preprocess_kwargs, {}, postprocess_kwargs
```
尽量保持简单输入/输出，最好是可 JSON 序列化的，因为这样可以使流水线的使用非常简单，而不需要用户了解新的对象类型。
通常也相对常见地支持许多不同类型的参数以便使用（例如音频文件，可以是文件名、URL 或纯字节）。
## 将其添加到支持的任务列表中
要将你的 `new-task` 注册到支持的任务列表中，你需要将其添加到 `PIPELINE_REGISTRY` 中：
```python
from transformers.pipelines import PIPELINE_REGISTRY
PIPELINE_REGISTRY.register_pipeline(
"new-task",
pipeline_class=MyPipeline,
pt_model=AutoModelForSequenceClassification,
)
```
如果需要，你可以指定一个默认模型，此时它应该带有一个特定的修订版本（可以是分支名称或提交哈希，这里我们使用了 `"abcdef"`），以及类型：
```python
PIPELINE_REGISTRY.register_pipeline(
"new-task",
pipeline_class=MyPipeline,
pt_model=AutoModelForSequenceClassification,
default={"pt": ("user/awesome_model", "abcdef")},
type="text", # current support type: text, audio, image, multimodal
)
```
## 在 Hub 上分享你的流水线
要在 Hub 上分享你的自定义流水线，你只需要将 `Pipeline` 子类的自定义代码保存在一个 Python 文件中。
例如，假设我们想使用一个自定义流水线进行句对分类，如下所示：
```py
import numpy as np
from transformers import Pipeline
def softmax(outputs):
maxes = np.max(outputs, axis=-1, keepdims=True)
shifted_exp = np.exp(outputs - maxes)
return shifted_exp / shifted_exp.sum(axis=-1, keepdims=True)
class PairClassificationPipeline(Pipeline):
def _sanitize_parameters(self, **kwargs):
preprocess_kwargs = {}
if "second_text" in kwargs:
preprocess_kwargs["second_text"] = kwargs["second_text"]
return preprocess_kwargs, {}, {}
def preprocess(self, text, second_text=None):
return self.tokenizer(text, text_pair=second_text, return_tensors=self.framework)
def _forward(self, model_inputs):
return self.model(**model_inputs)
def postprocess(self, model_outputs):
logits = model_outputs.logits[0].numpy()
probabilities = softmax(logits)
best_class = np.argmax(probabilities)
label = self.model.config.id2label[best_class]
score = probabilities[best_class].item()
logits = logits.tolist()
return {"label": label, "score": score, "logits": logits}
```
这个实现与框架无关，适用于 PyTorch 和 TensorFlow 模型。如果我们将其保存在一个名为
`pair_classification.py` 的文件中，然后我们可以像这样导入并注册它：
```py
from pair_classification import PairClassificationPipeline
from transformers.pipelines import PIPELINE_REGISTRY
from transformers import AutoModelForSequenceClassification, TFAutoModelForSequenceClassification
PIPELINE_REGISTRY.register_pipeline(
"pair-classification",
pipeline_class=PairClassificationPipeline,
pt_model=AutoModelForSequenceClassification,
tf_model=TFAutoModelForSequenceClassification,
)
```
完成这些步骤后，我们可以将其与预训练模型一起使用。例如，`sgugger/finetuned-bert-mrpc`
已经在 MRPC 数据集上进行了微调，用于将句子对分类为是释义或不是释义。
```py
from transformers import pipeline
classifier = pipeline("pair-classification", model="sgugger/finetuned-bert-mrpc")
```
然后，我们可以通过在 `Repository` 中使用 `save_pretrained` 方法将其分享到 Hub 上：
```py
from huggingface_hub import Repository
repo = Repository("test-dynamic-pipeline", clone_from="{your_username}/test-dynamic-pipeline")
classifier.save_pretrained("test-dynamic-pipeline")
repo.push_to_hub()
```
这将会复制包含你定义的 `PairClassificationPipeline` 的文件到文件夹 `"test-dynamic-pipeline"` 中，
同时保存流水线的模型和分词器，然后将所有内容推送到仓库 `{your_username}/test-dynamic-pipeline` 中。
之后，只要提供选项 `trust_remote_code=True`，任何人都可以使用它：
```py
from transformers import pipeline
classifier = pipeline(model="{your_username}/test-dynamic-pipeline", trust_remote_code=True)
```
## 将流水线添加到 🤗 Transformers
如果你想将你的流水线贡献给 🤗 Transformers，你需要在 `pipelines` 子模块中添加一个新模块，
其中包含你的流水线的代码，然后将其添加到 `pipelines/__init__.py` 中定义的任务列表中。
然后，你需要添加测试。创建一个新文件 `tests/test_pipelines_MY_PIPELINE.py`，其中包含其他测试的示例。
`run_pipeline_test` 函数将非常通用，并在每种可能的架构上运行小型随机模型，如 `model_mapping` 和 `tf_model_mapping` 所定义。
这对于测试未来的兼容性非常重要，这意味着如果有人为 `XXXForQuestionAnswering` 添加了一个新模型，
流水线测试将尝试在其上运行。由于模型是随机的，所以不可能检查实际值，这就是为什么有一个帮助函数 `ANY`，它只是尝试匹配流水线的输出类型。
你还 **需要** 实现 2（最好是 4）个测试。
- `test_small_model_pt`：为这个流水线定义一个小型模型（结果是否合理并不重要），并测试流水线的输出。
结果应该与 `test_small_model_tf` 的结果相同。
- `test_small_model_tf`：为这个流水线定义一个小型模型（结果是否合理并不重要），并测试流水线的输出。
结果应该与 `test_small_model_pt` 的结果相同。
- `test_large_model_pt`（可选）：在一个真实的流水线上测试流水线，结果应该是有意义的。
这些测试速度较慢，应该被如此标记。这里的目标是展示流水线，并确保在未来的发布中没有漂移。
- `test_large_model_tf`（可选）：在一个真实的流水线上测试流水线，结果应该是有意义的。
这些测试速度较慢，应该被如此标记。这里的目标是展示流水线，并确保在未来的发布中没有漂移。
| 0
|
mavonic_private_repos/transformers/docs/source
|
mavonic_private_repos/transformers/docs/source/zh/_toctree.yml
|
- sections:
- local: index
title: 🤗 Transformers 简介
- local: quicktour
title: 快速上手
- local: installation
title: 安装
title: 开始使用
- sections:
- local: pipeline_tutorial
title: 使用pipelines进行推理
- local: autoclass_tutorial
title: 使用AutoClass编写可移植的代码
- local: preprocessing
title: 预处理数据
- local: training
title: 微调预训练模型
- local: run_scripts
title: 通过脚本训练模型
- local: accelerate
title: 使用🤗Accelerate进行分布式训练
- local: peft
title: 使用🤗 PEFT加载和训练adapters
- local: model_sharing
title: 分享您的模型
- local: transformers_agents
title: agents教程
- local: llm_tutorial
title: 使用LLMs进行生成
title: 教程
- sections:
- isExpanded: false
sections:
- local: tasks/asr
title: 自动语音识别
- sections:
- local: fast_tokenizers
title: 使用 🤗 Tokenizers 中的分词器
- local: multilingual
title: 使用多语言模型进行推理
- local: create_a_model
title: 使用特定于模型的 API
- local: custom_models
title: 共享自定义模型
- local: chat_templating
title: 聊天模型的模板
- local: serialization
title: 导出为 ONNX
- local: tflite
title: 导出为 TFLite
- local: torchscript
title: 导出为 TorchScript
title: 开发者指南
- sections:
- local: performance
title: 综述
- sections:
- local: fsdp
title: 完全分片数据并行
- local: perf_hardware
title: 用于训练的定制硬件
- local: hpo_train
title: 使用Trainer API 进行超参数搜索
title: 高效训练技术
- local: big_models
title: 实例化大模型
- local: debugging
title: 问题定位及解决
- local: tf_xla
title: TensorFlow模型的XLA集成
- local: perf_torch_compile
title: 使用 `torch.compile()` 优化推理
title: 性能和可扩展性
- sections:
- local: contributing
title: 如何为 🤗 Transformers 做贡献？
- local: add_new_pipeline
title: 如何将流水线添加到 🤗 Transformers？
title: 贡献
- sections:
- local: task_summary
title: 🤗Transformers能做什么
- local: tokenizer_summary
title: 分词器的摘要
title: 概念指南
- sections:
- sections:
- local: main_classes/agent
title: Agents和工具
- local: main_classes/callback
title: Callbacks
- local: main_classes/configuration
title: Configuration
- local: main_classes/data_collator
title: Data Collator
- local: main_classes/keras_callbacks
title: Keras callbacks
- local: main_classes/logging
title: Logging
- local: main_classes/model
title: 模型
- local: main_classes/text_generation
title: 文本生成
- local: main_classes/onnx
title: ONNX
- local: main_classes/optimizer_schedules
title: Optimization
- local: main_classes/output
title: 模型输出
- local: main_classes/pipelines
title: Pipelines
- local: main_classes/processors
title: Processors
- local: main_classes/quantization
title: Quantization
- local: main_classes/tokenizer
title: Tokenizer
- local: main_classes/trainer
title: Trainer
- local: main_classes/deepspeed
title: DeepSpeed集成
- local: main_classes/feature_extractor
title: Feature Extractor
- local: main_classes/image_processor
title: Image Processor
title: 主要类
- sections:
- local: internal/modeling_utils
title: 自定义层和工具
- local: internal/pipelines_utils
title: pipelines工具
- local: internal/tokenization_utils
title: Tokenizers工具
- local: internal/trainer_utils
title: 训练器工具
- local: internal/generation_utils
title: 生成工具
- local: internal/image_processing_utils
title: 图像处理工具
- local: internal/audio_utils
title: 音频处理工具
- local: internal/file_utils
title: 通用工具
- local: internal/time_series_utils
title: 时序数据工具
title: 内部辅助工具
title: 应用程序接口 (API)
| 0
|
mavonic_private_repos/transformers/docs/source
|
mavonic_private_repos/transformers/docs/source/zh/debugging.md
|
<!--Copyright 2021 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
specific language governing permissions and limitations under the License.
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer.
-->
# 调试
## 多GPU网络问题调试
当使用`DistributedDataParallel`和多个GPU进行训练或推理时，如果遇到进程和（或）节点之间的互联问题，您可以使用以下脚本来诊断网络问题。
```bash
wget https://raw.githubusercontent.com/huggingface/transformers/main/scripts/distributed/torch-distributed-gpu-test.py
```
例如，要测试两个GPU之间的互联，请执行以下操作：
```bash
python -m torch.distributed.run --nproc_per_node 2 --nnodes 1 torch-distributed-gpu-test.py
```
如果两个进程能够相互通信并分配GPU内存，它们各自将打印出 "OK" 状态。
对于更多的GPU或节点，可以根据脚本中的参数进行调整。
在诊断脚本内部，您将找到更多详细信息，甚至有关如何在SLURM环境中运行它的说明。
另一种级别的调试是添加 `NCCL_DEBUG=INFO` 环境变量，如下所示：
```bash
NCCL_DEBUG=INFO python -m torch.distributed.run --nproc_per_node 2 --nnodes 1 torch-distributed-gpu-test.py
```
这将产生大量与NCCL相关的调试信息，如果发现有问题报告，您可以在线搜索以获取相关信息。或者，如果您不确定如何解释输出，可以在`issue`中分享日志文件。
## 下溢和上溢检测
<Tip>
目前，此功能仅适用于PyTorch。
</Tip>
<Tip>
对于多GPU训练，它需要使用DDP（`torch.distributed.launch`）。
</Tip>
<Tip>
此功能可以与任何基于`nn.Module`的模型一起使用。
</Tip>
如果您开始发现`loss=NaN`或模型因激活值或权重中的`inf`或`nan`而出现一些异常行为，就需要发现第一个下溢或上溢发生的地方以及导致它的原因。幸运的是，您可以通过激活一个特殊模块来自动进行检测。
如果您正在使用[`Trainer`]，只需把以下内容：
```bash
--debug underflow_overflow
```
添加到常规命令行参数中，或在创建[`TrainingArguments`]对象时传递 `debug="underflow_overflow"`。
如果您正在使用自己的训练循环或其他Trainer，您可以通过以下方式实现相同的功能：
```python
from transformers.debug_utils import DebugUnderflowOverflow
debug_overflow = DebugUnderflowOverflow(model)
```
[`debug_utils.DebugUnderflowOverflow`] 将`hooks`插入模型，紧跟在每次前向调用之后，进而测试输入和输出变量，以及相应模块的权重。一旦在激活值或权重的至少一个元素中检测到`inf`或`nan`，程序将执行`assert`并打印报告，就像这样（这是在`google/mt5-small`下使用fp16混合精度捕获的）：
```
Detected inf/nan during batch_number=0
Last 21 forward frames:
abs min abs max metadata
encoder.block.1.layer.1.DenseReluDense.dropout Dropout
0.00e+00 2.57e+02 input[0]
0.00e+00 2.85e+02 output
[...]
encoder.block.2.layer.0 T5LayerSelfAttention
6.78e-04 3.15e+03 input[0]
2.65e-04 3.42e+03 output[0]
None output[1]
2.25e-01 1.00e+04 output[2]
encoder.block.2.layer.1.layer_norm T5LayerNorm
8.69e-02 4.18e-01 weight
2.65e-04 3.42e+03 input[0]
1.79e-06 4.65e+00 output
encoder.block.2.layer.1.DenseReluDense.wi_0 Linear
2.17e-07 4.50e+00 weight
1.79e-06 4.65e+00 input[0]
2.68e-06 3.70e+01 output
encoder.block.2.layer.1.DenseReluDense.wi_1 Linear
8.08e-07 2.66e+01 weight
1.79e-06 4.65e+00 input[0]
1.27e-04 2.37e+02 output
encoder.block.2.layer.1.DenseReluDense.dropout Dropout
0.00e+00 8.76e+03 input[0]
0.00e+00 9.74e+03 output
encoder.block.2.layer.1.DenseReluDense.wo Linear
1.01e-06 6.44e+00 weight
0.00e+00 9.74e+03 input[0]
3.18e-04 6.27e+04 output
encoder.block.2.layer.1.DenseReluDense T5DenseGatedGeluDense
1.79e-06 4.65e+00 input[0]
3.18e-04 6.27e+04 output
encoder.block.2.layer.1.dropout Dropout
3.18e-04 6.27e+04 input[0]
0.00e+00 inf output
```
由于篇幅原因，示例输出中间的部分已经被缩减。
第二列显示了绝对最大元素的值，因此，如果您仔细查看最后`frame`，输入和输出都在`1e4`的范围内。因此，在使用fp16混合精度进行训练时，最后一步发生了溢出（因为在`fp16`下，在`inf`之前的最大数字是`64e3`）。为了避免在`fp16`下发生溢出，激活值必须保持低于`1e4`，因为`1e4 * 1e4 = 1e8`，因此任何具有大激活值的矩阵乘法都会导致数值溢出。
在跟踪的开始处，您可以发现问题发生在哪个批次（这里的`Detected inf/nan during batch_number=0`表示问题发生在第一个批次）。
每个报告的`frame`都以声明相应模块的层信息为开头，说明这一`frame`是为哪个模块报告的。如果只看这个`frame`：
```
encoder.block.2.layer.1.layer_norm T5LayerNorm
8.69e-02 4.18e-01 weight
2.65e-04 3.42e+03 input[0]
1.79e-06 4.65e+00 output
```
在这里，`encoder.block.2.layer.1.layer_norm` 表示它是编码器的第二个块中第一层的`layer norm`。而 `forward` 的具体调用是 `T5LayerNorm`。
让我们看看该报告的最后几个`frame`：
```
Detected inf/nan during batch_number=0
Last 21 forward frames:
abs min abs max metadata
[...]
encoder.block.2.layer.1.DenseReluDense.wi_0 Linear
2.17e-07 4.50e+00 weight
1.79e-06 4.65e+00 input[0]
2.68e-06 3.70e+01 output
encoder.block.2.layer.1.DenseReluDense.wi_1 Linear
8.08e-07 2.66e+01 weight
1.79e-06 4.65e+00 input[0]
1.27e-04 2.37e+02 output
encoder.block.2.layer.1.DenseReluDense.wo Linear
1.01e-06 6.44e+00 weight
0.00e+00 9.74e+03 input[0]
3.18e-04 6.27e+04 output
encoder.block.2.layer.1.DenseReluDense T5DenseGatedGeluDense
1.79e-06 4.65e+00 input[0]
3.18e-04 6.27e+04 output
encoder.block.2.layer.1.dropout Dropout
3.18e-04 6.27e+04 input[0]
0.00e+00 inf output
```
最后一个`frame`报告了`Dropout.forward`函数，第一个条目是唯一的输入，第二个条目是唯一的输出。您可以看到，它是从`DenseReluDense`类内的属性`dropout`中调用的。我们可以看到它发生在第2个块的第1层，也就是在第一个批次期间。最后，绝对最大的输入元素值为`6.27e+04`，输出也是`inf`。
您可以在这里看到，`T5DenseGatedGeluDense.forward`产生了输出激活值，其绝对最大值约为62.7K，非常接近fp16的上限64K。在下一个`frame`中，我们有`Dropout`对权重进行重新归一化，之后将某些元素归零，将绝对最大值推到了64K以上，导致溢出（`inf`）。
正如你所看到的，我们需要查看前面的`frame`, 从那里fp16数字开始变得非常大。
让我们将报告与`models/t5/modeling_t5.py`中的代码匹配：
```python
class T5DenseGatedGeluDense(nn.Module):
def __init__(self, config):
super().__init__()
self.wi_0 = nn.Linear(config.d_model, config.d_ff, bias=False)
self.wi_1 = nn.Linear(config.d_model, config.d_ff, bias=False)
self.wo = nn.Linear(config.d_ff, config.d_model, bias=False)
self.dropout = nn.Dropout(config.dropout_rate)
self.gelu_act = ACT2FN["gelu_new"]
def forward(self, hidden_states):
hidden_gelu = self.gelu_act(self.wi_0(hidden_states))
hidden_linear = self.wi_1(hidden_states)
hidden_states = hidden_gelu * hidden_linear
hidden_states = self.dropout(hidden_states)
hidden_states = self.wo(hidden_states)
return hidden_states
```
现在很容易看到`dropout`调用，以及所有之前的调用。
由于检测是在前向`hook`中进行的，这些报告将立即在每个`forward`返回后打印出来。
回到完整的报告，要采取措施并解决问题，我们需要往回看几个`frame`，在那里数字开始上升，并且最有可能切换到fp32模式以便在乘法或求和时数字不会溢出。当然，可能还有其他解决方案。例如，如果启用了`amp`，我们可以在将原始`forward`移到`helper wrapper`中后，暂时关闭它，如下所示：
```python
def _forward(self, hidden_states):
hidden_gelu = self.gelu_act(self.wi_0(hidden_states))
hidden_linear = self.wi_1(hidden_states)
hidden_states = hidden_gelu * hidden_linear
hidden_states = self.dropout(hidden_states)
hidden_states = self.wo(hidden_states)
return hidden_states
import torch
def forward(self, hidden_states):
if torch.is_autocast_enabled():
with torch.cuda.amp.autocast(enabled=False):
return self._forward(hidden_states)
else:
return self._forward(hidden_states)
```
由于自动检测器仅报告完整`frame`的输入和输出，一旦知道在哪里查找，您可能还希望分析特定`forward`函数的中间阶段。在这种情况下，您可以使用`detect_overflow`辅助函数将检测器放到希望的位置，例如：
```python
from debug_utils import detect_overflow
class T5LayerFF(nn.Module):
[...]
def forward(self, hidden_states):
forwarded_states = self.layer_norm(hidden_states)
detect_overflow(forwarded_states, "after layer_norm")
forwarded_states = self.DenseReluDense(forwarded_states)
detect_overflow(forwarded_states, "after DenseReluDense")
return hidden_states + self.dropout(forwarded_states)
```
可以看到，我们添加了2个检测器，现在我们可以跟踪是否在`forwarded_states`中间的某个地方检测到了`inf`或`nan`。
实际上，检测器已经报告了这些，因为上面示例中的每个调用都是一个`nn.Module`，但假设如果您有一些本地的直接计算，这就是您将如何执行的方式。
此外，如果您在自己的代码中实例化调试器，您可以调整从其默认打印的`frame`数，例如：
```python
from transformers.debug_utils import DebugUnderflowOverflow
debug_overflow = DebugUnderflowOverflow(model, max_frames_to_save=100)
```
### 特定批次的绝对最小值和最大值跟踪
当关闭下溢/上溢检测功能, 同样的调试类可以用于批处理跟踪。
假设您想要监视给定批次的每个`forward`调用的所有成分的绝对最小值和最大值，并且仅对批次1和3执行此操作，您可以这样实例化这个类：
```python
debug_overflow = DebugUnderflowOverflow(model, trace_batch_nums=[1, 3])
```
现在，完整的批次1和3将以与下溢/上溢检测器相同的格式进行跟踪。
批次从0开始计数。
如果您知道程序在某个批次编号之后开始出现问题，那么您可以直接快进到该区域。以下是一个截取的配置示例输出：
```
*** Starting batch number=1 ***
abs min abs max metadata
shared Embedding
1.01e-06 7.92e+02 weight
0.00e+00 2.47e+04 input[0]
5.36e-05 7.92e+02 output
[...]
decoder.dropout Dropout
1.60e-07 2.27e+01 input[0]
0.00e+00 2.52e+01 output
decoder T5Stack
not a tensor output
lm_head Linear
1.01e-06 7.92e+02 weight
0.00e+00 1.11e+00 input[0]
6.06e-02 8.39e+01 output
T5ForConditionalGeneration
not a tensor output
*** Starting batch number=3 ***
abs min abs max metadata
shared Embedding
1.01e-06 7.92e+02 weight
0.00e+00 2.78e+04 input[0]
5.36e-05 7.92e+02 output
[...]
```
在这里，您将获得大量的`frame`被`dump` - 与您的模型中的前向调用一样多，它有可能符合也可能不符合您的要求，但有时对于调试目的来说，它可能比正常的调试器更容易使用。例如，如果问题开始发生在批次号150上，您可以`dump`批次149和150的跟踪，并比较数字开始发散的地方。
你还可以使用以下命令指定停止训练的批次号：
```python
debug_overflow = DebugUnderflowOverflow(model, trace_batch_nums=[1, 3], abort_after_batch_num=3)
```
| 0
|
mavonic_private_repos/transformers/docs/source
|
mavonic_private_repos/transformers/docs/source/zh/index.md
|
<!--Copyright 2020 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
specific language governing permissions and limitations under the License.
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer.
-->
# 🤗 Transformers简介
为 [PyTorch](https://pytorch.org/)、[TensorFlow](https://www.tensorflow.org/) 和 [JAX](https://jax.readthedocs.io/en/latest/) 打造的先进的机器学习工具.
🤗 Transformers 提供了可以轻松地下载并且训练先进的预训练模型的 API 和工具。使用预训练模型可以减少计算消耗和碳排放，并且节省从头训练所需要的时间和资源。这些模型支持不同模态中的常见任务，比如：
📝 **自然语言处理**：文本分类、命名实体识别、问答、语言建模、摘要、翻译、多项选择和文本生成。<br>
🖼️ **机器视觉**：图像分类、目标检测和语义分割。<br>
🗣️ **音频**：自动语音识别和音频分类。<br>
🐙 **多模态**：表格问答、光学字符识别、从扫描文档提取信息、视频分类和视觉问答。
🤗 Transformers 支持在 PyTorch、TensorFlow 和 JAX 上的互操作性. 这给在模型的每个阶段使用不同的框架带来了灵活性；在一个框架中使用几行代码训练一个模型，然后在另一个框架中加载它并进行推理。模型也可以被导出为 ONNX 和 TorchScript 格式，用于在生产环境中部署。
马上加入在 [Hub](https://huggingface.co/models)、[论坛](https://discuss.huggingface.co/) 或者 [Discord](https://discord.com/invite/JfAtkvEtRb) 上正在快速发展的社区吧！
## 如果你需要来自 Hugging Face 团队的个性化支持
<a target="_blank" href="https://huggingface.co/support">
<img alt="HuggingFace Expert Acceleration Program" src="https://cdn-media.huggingface.co/marketing/transformers/new-support-improved.png" style="width: 100%; max-width: 600px; border: 1px solid #eee; border-radius: 4px; box-shadow: 0 1px 2px 0 rgba(0, 0, 0, 0.05);">
</a>
## 目录
这篇文档由以下 5 个章节组成：
- **开始使用** 包含了库的快速上手和安装说明，便于配置和运行。
- **教程** 是一个初学者开始的好地方。本章节将帮助你获得你会用到的使用这个库的基本技能。
- **操作指南** 向你展示如何实现一个特定目标，比如为语言建模微调一个预训练模型或者如何创造并分享个性化模型。
- **概念指南** 对 🤗 Transformers 的模型，任务和设计理念背后的基本概念和思想做了更多的讨论和解释。
- **API 介绍** 描述了所有的类和函数：
- **主要类别** 详述了配置（configuration）、模型（model）、分词器（tokenizer）和流水线（pipeline）这几个最重要的类。
- **模型** 详述了在这个库中和每个模型实现有关的类和函数。
- **内部帮助** 详述了内部使用的工具类和函数。
### 支持的模型和框架
下表展示了库中对每个模型的支持情况，如是否具有 Python 分词器（表中的“Tokenizer slow”）、是否具有由 🤗 Tokenizers 库支持的快速分词器（表中的“Tokenizer fast”）、是否支持 Jax（通过 Flax）、PyTorch 与 TensorFlow。
<!--This table is updated automatically from the auto modules with _make fix-copies_. Do not update manually!-->
| 模型 | PyTorch 支持 | TensorFlow 支持 | Flax 支持 |
|:------------------------------------------------------------------------:|:---------------:|:------------------:|:------------:|
| [ALBERT](../en/model_doc/albert.md) | ✅ | ✅ | ✅ |
| [ALIGN](../en/model_doc/align.md) | ✅ | ❌ | ❌ |
| [AltCLIP](../en/model_doc/altclip) | ✅ | ❌ | ❌ |
| [Audio Spectrogram Transformer](../en/model_doc/audio-spectrogram-transformer) | ✅ | ❌ | ❌ |
| [Autoformer](../en/model_doc/autoformer) | ✅ | ❌ | ❌ |
| [Bark](../en/model_doc/bark) | ✅ | ❌ | ❌ |
| [BART](../en/model_doc/bart) | ✅ | ✅ | ✅ |
| [BARThez](../en/model_doc/barthez) | ✅ | ✅ | ✅ |
| [BARTpho](../en/model_doc/bartpho) | ✅ | ✅ | ✅ |
| [BEiT](../en/model_doc/beit) | ✅ | ❌ | ✅ |
| [BERT](../en/model_doc/bert) | ✅ | ✅ | ✅ |
| [Bert Generation](../en/model_doc/bert-generation) | ✅ | ❌ | ❌ |
| [BertJapanese](../en/model_doc/bert-japanese) | ✅ | ✅ | ✅ |
| [BERTweet](../en/model_doc/bertweet) | ✅ | ✅ | ✅ |
| [BigBird](../en/model_doc/big_bird) | ✅ | ❌ | ✅ |
| [BigBird-Pegasus](../en/model_doc/bigbird_pegasus) | ✅ | ❌ | ❌ |
| [BioGpt](../en/model_doc/biogpt) | ✅ | ❌ | ❌ |
| [BiT](../en/model_doc/bit) | ✅ | ❌ | ❌ |
| [Blenderbot](../en/model_doc/blenderbot) | ✅ | ✅ | ✅ |
| [BlenderbotSmall](../en/model_doc/blenderbot-small) | ✅ | ✅ | ✅ |
| [BLIP](../en/model_doc/blip) | ✅ | ✅ | ❌ |
| [BLIP-2](../en/model_doc/blip-2) | ✅ | ❌ | ❌ |
| [BLOOM](../en/model_doc/bloom) | ✅ | ❌ | ✅ |
| [BORT](../en/model_doc/bort) | ✅ | ✅ | ✅ |
| [BridgeTower](../en/model_doc/bridgetower) | ✅ | ❌ | ❌ |
| [BROS](../en/model_doc/bros) | ✅ | ❌ | ❌ |
| [ByT5](../en/model_doc/byt5) | ✅ | ✅ | ✅ |
| [CamemBERT](../en/model_doc/camembert) | ✅ | ✅ | ❌ |
| [CANINE](../en/model_doc/canine) | ✅ | ❌ | ❌ |
| [Chinese-CLIP](../en/model_doc/chinese_clip) | ✅ | ❌ | ❌ |
| [CLAP](../en/model_doc/clap) | ✅ | ❌ | ❌ |
| [CLIP](../en/model_doc/clip) | ✅ | ✅ | ✅ |
| [CLIPSeg](../en/model_doc/clipseg) | ✅ | ❌ | ❌ |
| [CLVP](../en/model_doc/clvp) | ✅ | ❌ | ❌ |
| [CodeGen](../en/model_doc/codegen) | ✅ | ❌ | ❌ |
| [CodeLlama](../en/model_doc/code_llama) | ✅ | ❌ | ✅ |
| [Conditional DETR](../en/model_doc/conditional_detr) | ✅ | ❌ | ❌ |
| [ConvBERT](../en/model_doc/convbert) | ✅ | ✅ | ❌ |
| [ConvNeXT](../en/model_doc/convnext) | ✅ | ✅ | ❌ |
| [ConvNeXTV2](../en/model_doc/convnextv2) | ✅ | ✅ | ❌ |
| [CPM](../en/model_doc/cpm) | ✅ | ✅ | ✅ |
| [CPM-Ant](../en/model_doc/cpmant) | ✅ | ❌ | ❌ |
| [CTRL](../en/model_doc/ctrl) | ✅ | ✅ | ❌ |
| [CvT](../en/model_doc/cvt) | ✅ | ✅ | ❌ |
| [Data2VecAudio](../en/model_doc/data2vec) | ✅ | ❌ | ❌ |
| [Data2VecText](../en/model_doc/data2vec) | ✅ | ❌ | ❌ |
| [Data2VecVision](../en/model_doc/data2vec) | ✅ | ✅ | ❌ |
| [DeBERTa](../en/model_doc/deberta) | ✅ | ✅ | ❌ |
| [DeBERTa-v2](../en/model_doc/deberta-v2) | ✅ | ✅ | ❌ |
| [Decision Transformer](../en/model_doc/decision_transformer) | ✅ | ❌ | ❌ |
| [Deformable DETR](../en/model_doc/deformable_detr) | ✅ | ❌ | ❌ |
| [DeiT](../en/model_doc/deit) | ✅ | ✅ | ❌ |
| [DePlot](../en/model_doc/deplot) | ✅ | ❌ | ❌ |
| [Depth Anything](../en/model_doc/depth_anything) | ✅ | ❌ | ❌ |
| [DETA](../en/model_doc/deta) | ✅ | ❌ | ❌ |
| [DETR](../en/model_doc/detr) | ✅ | ❌ | ❌ |
| [DialoGPT](../en/model_doc/dialogpt) | ✅ | ✅ | ✅ |
| [DiNAT](../en/model_doc/dinat) | ✅ | ❌ | ❌ |
| [DINOv2](../en/model_doc/dinov2) | ✅ | ❌ | ❌ |
| [DistilBERT](../en/model_doc/distilbert) | ✅ | ✅ | ✅ |
| [DiT](../en/model_doc/dit) | ✅ | ❌ | ✅ |
| [DonutSwin](../en/model_doc/donut) | ✅ | ❌ | ❌ |
| [DPR](../en/model_doc/dpr) | ✅ | ✅ | ❌ |
| [DPT](../en/model_doc/dpt) | ✅ | ❌ | ❌ |
| [EfficientFormer](../en/model_doc/efficientformer) | ✅ | ✅ | ❌ |
| [EfficientNet](../en/model_doc/efficientnet) | ✅ | ❌ | ❌ |
| [ELECTRA](../en/model_doc/electra) | ✅ | ✅ | ✅ |
| [EnCodec](../en/model_doc/encodec) | ✅ | ❌ | ❌ |
| [Encoder decoder](../en/model_doc/encoder-decoder) | ✅ | ✅ | ✅ |
| [ERNIE](../en/model_doc/ernie) | ✅ | ❌ | ❌ |
| [ErnieM](../en/model_doc/ernie_m) | ✅ | ❌ | ❌ |
| [ESM](../en/model_doc/esm) | ✅ | ✅ | ❌ |
| [FairSeq Machine-Translation](../en/model_doc/fsmt) | ✅ | ❌ | ❌ |
| [Falcon](../en/model_doc/falcon) | ✅ | ❌ | ❌ |
| [FastSpeech2Conformer](../en/model_doc/fastspeech2_conformer) | ✅ | ❌ | ❌ |
| [FLAN-T5](../en/model_doc/flan-t5) | ✅ | ✅ | ✅ |
| [FLAN-UL2](../en/model_doc/flan-ul2) | ✅ | ✅ | ✅ |
| [FlauBERT](../en/model_doc/flaubert) | ✅ | ✅ | ❌ |
| [FLAVA](../en/model_doc/flava) | ✅ | ❌ | ❌ |
| [FNet](../en/model_doc/fnet) | ✅ | ❌ | ❌ |
| [FocalNet](../en/model_doc/focalnet) | ✅ | ❌ | ❌ |
| [Funnel Transformer](../en/model_doc/funnel) | ✅ | ✅ | ❌ |
| [Fuyu](../en/model_doc/fuyu) | ✅ | ❌ | ❌ |
| [Gemma](../en/model_doc/gemma) | ✅ | ❌ | ✅ |
| [GIT](../en/model_doc/git) | ✅ | ❌ | ❌ |
| [GLPN](../en/model_doc/glpn) | ✅ | ❌ | ❌ |
| [GPT Neo](../en/model_doc/gpt_neo) | ✅ | ❌ | ✅ |
| [GPT NeoX](../en/model_doc/gpt_neox) | ✅ | ❌ | ❌ |
| [GPT NeoX Japanese](../en/model_doc/gpt_neox_japanese) | ✅ | ❌ | ❌ |
| [GPT-J](../en/model_doc/gptj) | ✅ | ✅ | ✅ |
| [GPT-Sw3](../en/model_doc/gpt-sw3) | ✅ | ✅ | ✅ |
| [GPTBigCode](../en/model_doc/gpt_bigcode) | ✅ | ❌ | ❌ |
| [GPTSAN-japanese](../en/model_doc/gptsan-japanese) | ✅ | ❌ | ❌ |
| [Graphormer](../en/model_doc/graphormer) | ✅ | ❌ | ❌ |
| [GroupViT](../en/model_doc/groupvit) | ✅ | ✅ | ❌ |
| [HerBERT](../en/model_doc/herbert) | ✅ | ✅ | ✅ |
| [Hubert](../en/model_doc/hubert) | ✅ | ✅ | ❌ |
| [I-BERT](../en/model_doc/ibert) | ✅ | ❌ | ❌ |
| [IDEFICS](../en/model_doc/idefics) | ✅ | ❌ | ❌ |
| [ImageGPT](../en/model_doc/imagegpt) | ✅ | ❌ | ❌ |
| [Informer](../en/model_doc/informer) | ✅ | ❌ | ❌ |
| [InstructBLIP](../en/model_doc/instructblip) | ✅ | ❌ | ❌ |
| [Jukebox](../en/model_doc/jukebox) | ✅ | ❌ | ❌ |
| [KOSMOS-2](../en/model_doc/kosmos-2) | ✅ | ❌ | ❌ |
| [LayoutLM](../en/model_doc/layoutlm) | ✅ | ✅ | ❌ |
| [LayoutLMv2](../en/model_doc/layoutlmv2) | ✅ | ❌ | ❌ |
| [LayoutLMv3](../en/model_doc/layoutlmv3) | ✅ | ✅ | ❌ |
| [LayoutXLM](../en/model_doc/layoutxlm) | ✅ | ❌ | ❌ |
| [LED](../en/model_doc/led) | ✅ | ✅ | ❌ |
| [LeViT](../en/model_doc/levit) | ✅ | ❌ | ❌ |
| [LiLT](../en/model_doc/lilt) | ✅ | ❌ | ❌ |
| [LLaMA](../en/model_doc/llama) | ✅ | ❌ | ✅ |
| [Llama2](../en/model_doc/llama2) | ✅ | ❌ | ✅ |
| [LLaVa](../en/model_doc/llava) | ✅ | ❌ | ❌ |
| [Longformer](../en/model_doc/longformer) | ✅ | ✅ | ❌ |
| [LongT5](../en/model_doc/longt5) | ✅ | ❌ | ✅ |
| [LUKE](../en/model_doc/luke) | ✅ | ❌ | ❌ |
| [LXMERT](../en/model_doc/lxmert) | ✅ | ✅ | ❌ |
| [M-CTC-T](../en/model_doc/mctct) | ✅ | ❌ | ❌ |
| [M2M100](../en/model_doc/m2m_100) | ✅ | ❌ | ❌ |
| [MADLAD-400](../en/model_doc/madlad-400) | ✅ | ✅ | ✅ |
| [Marian](../en/model_doc/marian) | ✅ | ✅ | ✅ |
| [MarkupLM](../en/model_doc/markuplm) | ✅ | ❌ | ❌ |
| [Mask2Former](../en/model_doc/mask2former) | ✅ | ❌ | ❌ |
| [MaskFormer](../en/model_doc/maskformer) | ✅ | ❌ | ❌ |
| [MatCha](../en/model_doc/matcha) | ✅ | ❌ | ❌ |
| [mBART](../en/model_doc/mbart) | ✅ | ✅ | ✅ |
| [mBART-50](../en/model_doc/mbart50) | ✅ | ✅ | ✅ |
| [MEGA](../en/model_doc/mega) | ✅ | ❌ | ❌ |
| [Megatron-BERT](../en/model_doc/megatron-bert) | ✅ | ❌ | ❌ |
| [Megatron-GPT2](../en/model_doc/megatron_gpt2) | ✅ | ✅ | ✅ |
| [MGP-STR](../en/model_doc/mgp-str) | ✅ | ❌ | ❌ |
| [Mistral](../en/model_doc/mistral) | ✅ | ❌ | ✅ |
| [Mixtral](../en/model_doc/mixtral) | ✅ | ❌ | ❌ |
| [mLUKE](../en/model_doc/mluke) | ✅ | ❌ | ❌ |
| [MMS](../en/model_doc/mms) | ✅ | ✅ | ✅ |
| [MobileBERT](../en/model_doc/mobilebert) | ✅ | ✅ | ❌ |
| [MobileNetV1](../en/model_doc/mobilenet_v1) | ✅ | ❌ | ❌ |
| [MobileNetV2](../en/model_doc/mobilenet_v2) | ✅ | ❌ | ❌ |
| [MobileViT](../en/model_doc/mobilevit) | ✅ | ✅ | ❌ |
| [MobileViTV2](../en/model_doc/mobilevitv2) | ✅ | ❌ | ❌ |
| [MPNet](../en/model_doc/mpnet) | ✅ | ✅ | ❌ |
| [MPT](../en/model_doc/mpt) | ✅ | ❌ | ❌ |
| [MRA](../en/model_doc/mra) | ✅ | ❌ | ❌ |
| [MT5](../en/model_doc/mt5) | ✅ | ✅ | ✅ |
| [MusicGen](../en/model_doc/musicgen) | ✅ | ❌ | ❌ |
| [MVP](../en/model_doc/mvp) | ✅ | ❌ | ❌ |
| [NAT](../en/model_doc/nat) | ✅ | ❌ | ❌ |
| [Nezha](../en/model_doc/nezha) | ✅ | ❌ | ❌ |
| [NLLB](../en/model_doc/nllb) | ✅ | ❌ | ❌ |
| [NLLB-MOE](../en/model_doc/nllb-moe) | ✅ | ❌ | ❌ |
| [Nougat](../en/model_doc/nougat) | ✅ | ✅ | ✅ |
| [Nyströmformer](../en/model_doc/nystromformer) | ✅ | ❌ | ❌ |
| [OneFormer](../en/model_doc/oneformer) | ✅ | ❌ | ❌ |
| [OpenAI GPT](../en/model_doc/openai-gpt) | ✅ | ✅ | ❌ |
| [OpenAI GPT-2](../en/model_doc/gpt2) | ✅ | ✅ | ✅ |
| [OpenLlama](../en/model_doc/open-llama) | ✅ | ❌ | ❌ |
| [OPT](../en/model_doc/opt) | ✅ | ✅ | ✅ |
| [OWL-ViT](../en/model_doc/owlvit) | ✅ | ❌ | ❌ |
| [OWLv2](../en/model_doc/owlv2) | ✅ | ❌ | ❌ |
| [PatchTSMixer](../en/model_doc/patchtsmixer) | ✅ | ❌ | ❌ |
| [PatchTST](../en/model_doc/patchtst) | ✅ | ❌ | ❌ |
| [Pegasus](../en/model_doc/pegasus) | ✅ | ✅ | ✅ |
| [PEGASUS-X](../en/model_doc/pegasus_x) | ✅ | ❌ | ❌ |
| [Perceiver](../en/model_doc/perceiver) | ✅ | ❌ | ❌ |
| [Persimmon](../en/model_doc/persimmon) | ✅ | ❌ | ❌ |
| [Phi](../en/model_doc/phi) | ✅ | ❌ | ❌ |
| [PhoBERT](../en/model_doc/phobert) | ✅ | ✅ | ✅ |
| [Pix2Struct](../en/model_doc/pix2struct) | ✅ | ❌ | ❌ |
| [PLBart](../en/model_doc/plbart) | ✅ | ❌ | ❌ |
| [PoolFormer](../en/model_doc/poolformer) | ✅ | ❌ | ❌ |
| [Pop2Piano](../en/model_doc/pop2piano) | ✅ | ❌ | ❌ |
| [ProphetNet](../en/model_doc/prophetnet) | ✅ | ❌ | ❌ |
| [PVT](../en/model_doc/pvt) | ✅ | ❌ | ❌ |
| [QDQBert](../en/model_doc/qdqbert) | ✅ | ❌ | ❌ |
| [Qwen2](../en/model_doc/qwen2) | ✅ | ❌ | ❌ |
| [RAG](../en/model_doc/rag) | ✅ | ✅ | ❌ |
| [REALM](../en/model_doc/realm) | ✅ | ❌ | ❌ |
| [Reformer](../en/model_doc/reformer) | ✅ | ❌ | ❌ |
| [RegNet](../en/model_doc/regnet) | ✅ | ✅ | ✅ |
| [RemBERT](../en/model_doc/rembert) | ✅ | ✅ | ❌ |
| [ResNet](../en/model_doc/resnet) | ✅ | ✅ | ✅ |
| [RetriBERT](../en/model_doc/retribert) | ✅ | ❌ | ❌ |
| [RoBERTa](../en/model_doc/roberta) | ✅ | ✅ | ✅ |
| [RoBERTa-PreLayerNorm](../en/model_doc/roberta-prelayernorm) | ✅ | ✅ | ✅ |
| [RoCBert](../en/model_doc/roc_bert) | ✅ | ❌ | ❌ |
| [RoFormer](../en/model_doc/roformer) | ✅ | ✅ | ✅ |
| [RWKV](../en/model_doc/rwkv) | ✅ | ❌ | ❌ |
| [SAM](../en/model_doc/sam) | ✅ | ✅ | ❌ |
| [SeamlessM4T](../en/model_doc/seamless_m4t) | ✅ | ❌ | ❌ |
| [SeamlessM4Tv2](../en/model_doc/seamless_m4t_v2) | ✅ | ❌ | ❌ |
| [SegFormer](../en/model_doc/segformer) | ✅ | ✅ | ❌ |
| [SegGPT](../en/model_doc/seggpt) | ✅ | ❌ | ❌ |
| [SEW](../en/model_doc/sew) | ✅ | ❌ | ❌ |
| [SEW-D](../en/model_doc/sew-d) | ✅ | ❌ | ❌ |
| [SigLIP](../en/model_doc/siglip) | ✅ | ❌ | ❌ |
| [Speech Encoder decoder](../en/model_doc/speech-encoder-decoder) | ✅ | ❌ | ✅ |
| [Speech2Text](../en/model_doc/speech_to_text) | ✅ | ✅ | ❌ |
| [SpeechT5](../en/model_doc/speecht5) | ✅ | ❌ | ❌ |
| [Splinter](../en/model_doc/splinter) | ✅ | ❌ | ❌ |
| [SqueezeBERT](../en/model_doc/squeezebert) | ✅ | ❌ | ❌ |
| [StableLm](../en/model_doc/stablelm) | ✅ | ❌ | ❌ |
| [Starcoder2](../en/model_doc/starcoder2) | ✅ | ❌ | ❌ |
| [SwiftFormer](../en/model_doc/swiftformer) | ✅ | ❌ | ❌ |
| [Swin Transformer](../en/model_doc/swin) | ✅ | ✅ | ❌ |
| [Swin Transformer V2](../en/model_doc/swinv2) | ✅ | ❌ | ❌ |
| [Swin2SR](../en/model_doc/swin2sr) | ✅ | ❌ | ❌ |
| [SwitchTransformers](../en/model_doc/switch_transformers) | ✅ | ❌ | ❌ |
| [T5](../en/model_doc/t5) | ✅ | ✅ | ✅ |
| [T5v1.1](../en/model_doc/t5v1.1) | ✅ | ✅ | ✅ |
| [Table Transformer](../en/model_doc/table-transformer) | ✅ | ❌ | ❌ |
| [TAPAS](../en/model_doc/tapas) | ✅ | ✅ | ❌ |
| [TAPEX](../en/model_doc/tapex) | ✅ | ✅ | ✅ |
| [Time Series Transformer](../en/model_doc/time_series_transformer) | ✅ | ❌ | ❌ |
| [TimeSformer](../en/model_doc/timesformer) | ✅ | ❌ | ❌ |
| [Trajectory Transformer](../en/model_doc/trajectory_transformer) | ✅ | ❌ | ❌ |
| [Transformer-XL](../en/model_doc/transfo-xl) | ✅ | ✅ | ❌ |
| [TrOCR](../en/model_doc/trocr) | ✅ | ❌ | ❌ |
| [TVLT](../en/model_doc/tvlt) | ✅ | ❌ | ❌ |
| [TVP](../en/model_doc/tvp) | ✅ | ❌ | ❌ |
| [UL2](../en/model_doc/ul2) | ✅ | ✅ | ✅ |
| [UMT5](../en/model_doc/umt5) | ✅ | ❌ | ❌ |
| [UniSpeech](../en/model_doc/unispeech) | ✅ | ❌ | ❌ |
| [UniSpeechSat](../en/model_doc/unispeech-sat) | ✅ | ❌ | ❌ |
| [UnivNet](../en/model_doc/univnet) | ✅ | ❌ | ❌ |
| [UPerNet](../en/model_doc/upernet) | ✅ | ❌ | ❌ |
| [VAN](../en/model_doc/van) | ✅ | ❌ | ❌ |
| [VideoMAE](../en/model_doc/videomae) | ✅ | ❌ | ❌ |
| [ViLT](../en/model_doc/vilt) | ✅ | ❌ | ❌ |
| [VipLlava](../en/model_doc/vipllava) | ✅ | ❌ | ❌ |
| [Vision Encoder decoder](../en/model_doc/vision-encoder-decoder) | ✅ | ✅ | ✅ |
| [VisionTextDualEncoder](../en/model_doc/vision-text-dual-encoder) | ✅ | ✅ | ✅ |
| [VisualBERT](../en/model_doc/visual_bert) | ✅ | ❌ | ❌ |
| [ViT](../en/model_doc/vit) | ✅ | ✅ | ✅ |
| [ViT Hybrid](../en/model_doc/vit_hybrid) | ✅ | ❌ | ❌ |
| [VitDet](../en/model_doc/vitdet) | ✅ | ❌ | ❌ |
| [ViTMAE](../en/model_doc/vit_mae) | ✅ | ✅ | ❌ |
| [ViTMatte](../en/model_doc/vitmatte) | ✅ | ❌ | ❌ |
| [ViTMSN](../en/model_doc/vit_msn) | ✅ | ❌ | ❌ |
| [VITS](../en/model_doc/vits) | ✅ | ❌ | ❌ |
| [ViViT](../en/model_doc/vivit) | ✅ | ❌ | ❌ |
| [Wav2Vec2](../en/model_doc/wav2vec2) | ✅ | ✅ | ✅ |
| [Wav2Vec2-BERT](../en/model_doc/wav2vec2-bert) | ✅ | ❌ | ❌ |
| [Wav2Vec2-Conformer](../en/model_doc/wav2vec2-conformer) | ✅ | ❌ | ❌ |
| [Wav2Vec2Phoneme](../en/model_doc/wav2vec2_phoneme) | ✅ | ✅ | ✅ |
| [WavLM](../en/model_doc/wavlm) | ✅ | ❌ | ❌ |
| [Whisper](../en/model_doc/whisper) | ✅ | ✅ | ✅ |
| [X-CLIP](../en/model_doc/xclip) | ✅ | ❌ | ❌ |
| [X-MOD](../en/model_doc/xmod) | ✅ | ❌ | ❌ |
| [XGLM](../en/model_doc/xglm) | ✅ | ✅ | ✅ |
| [XLM](../en/model_doc/xlm) | ✅ | ✅ | ❌ |
| [XLM-ProphetNet](../en/model_doc/xlm-prophetnet) | ✅ | ❌ | ❌ |
| [XLM-RoBERTa](../en/model_doc/xlm-roberta) | ✅ | ✅ | ✅ |
| [XLM-RoBERTa-XL](../en/model_doc/xlm-roberta-xl) | ✅ | ❌ | ❌ |
| [XLM-V](../en/model_doc/xlm-v) | ✅ | ✅ | ✅ |
| [XLNet](../en/model_doc/xlnet) | ✅ | ✅ | ❌ |
| [XLS-R](../en/model_doc/xls_r) | ✅ | ✅ | ✅ |
| [XLSR-Wav2Vec2](../en/model_doc/xlsr_wav2vec2) | ✅ | ✅ | ✅ |
| [YOLOS](../en/model_doc/yolos) | ✅ | ❌ | ❌ |
| [YOSO](../en/model_doc/yoso) | ✅ | ❌ | ❌ |
<!-- End table-->
| 0
|
mavonic_private_repos/transformers/docs/source
|
mavonic_private_repos/transformers/docs/source/zh/big_models.md
|
<!--Copyright 2022 The HuggingFace Team. All rights reserved.
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
specific language governing permissions and limitations under the License.
⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
rendered properly in your Markdown viewer.
-->
# 实例化大型模型
当你想使用一个非常大的预训练模型时，一个挑战是尽量减少对内存的使用。通常从PyTorch开始的工作流程如下：
1. 用随机权重创建你的模型。
2. 加载你的预训练权重。
3. 将这些预训练权重放入你的随机模型中。
步骤1和2都需要完整版本的模型在内存中，这在大多数情况下不是问题，但如果你的模型开始达到几个GB的大小，这两个副本可能会让你超出内存的限制。更糟糕的是，如果你使用`torch.distributed`来启动分布式训练，每个进程都会加载预训练模型并将这两个副本存储在内存中。
<Tip>
请注意，随机创建的模型使用“空”张量进行初始化，这些张量占用内存空间但不填充它（因此随机值是给定时间内该内存块中的任何内容）。在第3步之后，对未初始化的权重执行适合模型/参数种类的随机初始化（例如正态分布），以尽可能提高速度！
</Tip>
在本指南中，我们将探讨 Transformers 提供的解决方案来处理这个问题。请注意，这是一个积极开发的领域，因此这里解释的API在将来可能会略有变化。
## 分片checkpoints
自4.18.0版本起，占用空间超过10GB的模型检查点将自动分成较小的片段。在使用`model.save_pretrained(save_dir)`时，您最终会得到几个部分`checkpoints`（每个的大小都小于10GB）以及一个索引，该索引将参数名称映射到存储它们的文件。
您可以使用`max_shard_size`参数来控制分片之前的最大大小。为了示例的目的，我们将使用具有较小分片大小的普通大小的模型：让我们以传统的BERT模型为例。
```py
from transformers import AutoModel
model = AutoModel.from_pretrained("google-bert/bert-base-cased")
```
如果您使用 [`PreTrainedModel.save_pretrained`](模型预训练保存) 进行保存，您将得到一个新的文件夹，其中包含两个文件：模型的配置和权重：
```py
>>> import os
>>> import tempfile
>>> with tempfile.TemporaryDirectory() as tmp_dir:
... model.save_pretrained(tmp_dir)
... print(sorted(os.listdir(tmp_dir)))
['config.json', 'pytorch_model.bin']
```
现在让我们使用最大分片大小为200MB：
```py
>>> with tempfile.TemporaryDirectory() as tmp_dir:
... model.save_pretrained(tmp_dir, max_shard_size="200MB")
... print(sorted(os.listdir(tmp_dir)))
['config.json', 'pytorch_model-00001-of-00003.bin', 'pytorch_model-00002-of-00003.bin', 'pytorch_model-00003-of-00003.bin', 'pytorch_model.bin.index.json']
```
在模型配置文件最上方，我们可以看到三个不同的权重文件，以及一个`index.json`索引文件。这样的`checkpoint`可以使用[`~PreTrainedModel.from_pretrained`]方法完全重新加载：
```py
>>> with tempfile.TemporaryDirectory() as tmp_dir:
... model.save_pretrained(tmp_dir, max_shard_size="200MB")
... new_model = AutoModel.from_pretrained(tmp_dir)
```
对于大型模型来说，这样做的主要优点是在上述工作流程的步骤2中，每个`checkpoint`的分片在前一个分片之后加载，从而将内存中的内存使用限制在模型大小加上最大分片的大小。
在后台，索引文件用于确定`checkpoint`中包含哪些键以及相应的权重存储在哪里。我们可以像加载任何json一样加载该索引，并获得一个字典：
```py
>>> import json
>>> with tempfile.TemporaryDirectory() as tmp_dir:
... model.save_pretrained(tmp_dir, max_shard_size="200MB")
... with open(os.path.join(tmp_dir, "pytorch_model.bin.index.json"), "r") as f:
... index = json.load(f)
>>> print(index.keys())
dict_keys(['metadata', 'weight_map'])
```
目前元数据仅包括模型的总大小。我们计划在将来添加其他信息：
```py
>>> index["metadata"]
{'total_size': 433245184}
```
权重映射是该索引的主要部分，它将每个参数的名称（通常在PyTorch模型的`state_dict`中找到）映射到存储该参数的文件：
```py
>>> index["weight_map"]
{'embeddings.LayerNorm.bias': 'pytorch_model-00001-of-00003.bin',
'embeddings.LayerNorm.weight': 'pytorch_model-00001-of-00003.bin',
...
```
如果您想直接在模型内部加载这样的分片`checkpoint`，而不使用 [`PreTrainedModel.from_pretrained`](就像您会为完整`checkpoint`执行 `model.load_state_dict()` 一样)，您应该使用 [`modeling_utils.load_sharded_checkpoint`]：
```py
>>> from transformers.modeling_utils import load_sharded_checkpoint
>>> with tempfile.TemporaryDirectory() as tmp_dir:
... model.save_pretrained(tmp_dir, max_shard_size="200MB")
... load_sharded_checkpoint(model, tmp_dir)
```
## 低内存加载
分片`checkpoints`在上述工作流的第2步中降低了内存使用，但为了在低内存环境中使用该模型，我们建议使用基于 Accelerate 库的工具。
请阅读以下指南以获取更多信息：[使用 Accelerate 进行大模型加载](./main_classes/model#large-model-loading)
| 0
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.