entry_point
stringlengths 1
65
| original_triton_python_code
stringlengths 208
619k
| optimised_triton_code
stringlengths 1.15k
275k
| repo_name
stringlengths 7
115
| module_name
stringlengths 1
65
| synthetic
bool 1
class | uuid
int64 0
18.5k
| licenses
listlengths 1
6
| stars
int64 0
19.8k
| sha
stringlengths 40
40
| repo_link
stringlengths 72
180
|
|---|---|---|---|---|---|---|---|---|---|---|
Normalize
|
import torch
import torch.nn as nn
from itertools import product as product
import torch.onnx
class Normalize(nn.Module):
def __init__(self, n_channels, scale=1.0):
super(Normalize, self).__init__()
self.n_channels = n_channels
self.scale = scale
self.eps = 1e-10
self.weight = nn.Parameter(torch.Tensor(self.n_channels))
self.weight.data *= 0.0
self.weight.data += self.scale
self.register_parameter('bias', None)
def __repr__(self):
return 'Normalize(channels=%d, scale=%f)' % (self.n_channels, self.
scale)
def forward(self, x):
norm = x.pow(2).sum(dim=1, keepdim=True).sqrt() + self.eps
x = x / norm * self.weight.view(1, -1, 1, 1)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'n_channels': 4}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice
import torch.nn as nn
from itertools import product as product
import torch.onnx
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_add_div_mul_pow_sqrt_sum_0(in_ptr0, in_ptr1, out_ptr0,
xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x0 = xindex % 16
x2 = xindex // 64
x1 = xindex // 16 % 4
tmp0 = tl.load(in_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + (x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp3 = tl.load(in_ptr0 + (16 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp6 = tl.load(in_ptr0 + (32 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp9 = tl.load(in_ptr0 + (48 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp16 = tl.load(in_ptr1 + x1, xmask, eviction_policy='evict_last')
tmp2 = tmp1 * tmp1
tmp4 = tmp3 * tmp3
tmp5 = tmp2 + tmp4
tmp7 = tmp6 * tmp6
tmp8 = tmp5 + tmp7
tmp10 = tmp9 * tmp9
tmp11 = tmp8 + tmp10
tmp12 = libdevice.sqrt(tmp11)
tmp13 = 1e-10
tmp14 = tmp12 + tmp13
tmp15 = tmp0 / tmp14
tmp17 = tmp15 * tmp16
tl.store(out_ptr0 + x3, tmp17, xmask)
def call(args):
primals_1, primals_2 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_add_div_mul_pow_sqrt_sum_0[grid(256)](primals_1,
primals_2, buf0, 256, XBLOCK=256, num_warps=4, num_stages=1)
del primals_2
return buf0, primals_1
class NormalizeNew(nn.Module):
def __init__(self, n_channels, scale=1.0):
super(NormalizeNew, self).__init__()
self.n_channels = n_channels
self.scale = scale
self.eps = 1e-10
self.weight = nn.Parameter(torch.Tensor(self.n_channels))
self.weight.data *= 0.0
self.weight.data += self.scale
self.register_parameter('bias', None)
def __repr__(self):
return 'Normalize(channels=%d, scale=%f)' % (self.n_channels, self.
scale)
def forward(self, input_0):
primals_2 = self.weight
primals_1 = input_0
output = call([primals_1, primals_2])
return output[0]
|
Janus1984/Msnhnet
|
Normalize
| false
| 13,872
|
[
"MIT"
] | 546
|
4e09f2501ba8db789f0a20441a357de3ba468f10
|
https://github.com/Janus1984/Msnhnet/tree/4e09f2501ba8db789f0a20441a357de3ba468f10
|
GeLU
|
from torch.nn import Module
import functools
import math
import torch
import torch.utils.data
import torch.nn as nn
from torchvision.models import *
import torch.nn.init
class GeLU(Module):
def forward(self, x):
return 0.5 * x * (1 + torch.tanh(math.sqrt(2 / math.pi) * (x +
0.044715 * torch.pow(x, 3))))
class PrePostInitMeta(type):
"""A metaclass that calls optional `__pre_init__` and `__post_init__` methods"""
def __new__(cls, name, bases, dct):
x = super().__new__(cls, name, bases, dct)
old_init = x.__init__
def _pass(self):
pass
@functools.wraps(old_init)
def _init(self, *args, **kwargs):
self.__pre_init__()
old_init(self, *args, **kwargs)
self.__post_init__()
x.__init__ = _init
if not hasattr(x, '__pre_init__'):
x.__pre_init__ = _pass
if not hasattr(x, '__post_init__'):
x.__post_init__ = _pass
return x
class Module(nn.Module, metaclass=PrePostInitMeta):
"""Same as `nn.Module`, but no need for subclasses to call `super().__init__`"""
def __pre_init__(self):
super().__init__()
def __init__(self):
pass
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice
from torch.nn import Module
import functools
import torch.utils.data
import torch.nn as nn
from torchvision.models import *
import torch.nn.init
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_add_mul_pow_tanh_0(in_ptr0, out_ptr0, xnumel, XBLOCK:
tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = 0.5
tmp2 = tmp0 * tmp1
tmp3 = tmp0 * tmp0
tmp4 = tmp3 * tmp0
tmp5 = 0.044715
tmp6 = tmp4 * tmp5
tmp7 = tmp0 + tmp6
tmp8 = 0.7978845608028654
tmp9 = tmp7 * tmp8
tmp10 = libdevice.tanh(tmp9)
tmp11 = 1.0
tmp12 = tmp10 + tmp11
tmp13 = tmp2 * tmp12
tl.store(out_ptr0 + x0, tmp13, xmask)
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_add_mul_pow_tanh_0[grid(256)](arg0_1, buf0, 256,
XBLOCK=256, num_warps=4, num_stages=1)
del arg0_1
return buf0,
class GeLUNew(Module):
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
class PrePostInitMeta(type):
"""A metaclass that calls optional `__pre_init__` and `__post_init__` methods"""
def __new__(cls, name, bases, dct):
x = super().__new__(cls, name, bases, dct)
old_init = x.__init__
def _pass(self):
pass
@functools.wraps(old_init)
def _init(self, *args, **kwargs):
self.__pre_init__()
old_init(self, *args, **kwargs)
self.__post_init__()
x.__init__ = _init
if not hasattr(x, '__pre_init__'):
x.__pre_init__ = _pass
if not hasattr(x, '__post_init__'):
x.__post_init__ = _pass
return x
class Module(nn.Module, metaclass=PrePostInitMeta):
"""Same as `nn.Module`, but no need for subclasses to call `super().__init__`"""
def __pre_init__(self):
super().__init__()
def __init__(self):
pass
|
JiahuaWU/fastai
|
GeLU
| false
| 13,873
|
[
"Apache-2.0"
] | 59
|
13a2df812d875abf0558004283392ab40d9bdea1
|
https://github.com/JiahuaWU/fastai/tree/13a2df812d875abf0558004283392ab40d9bdea1
|
Scale
|
import torch
import torch.nn as nn
from torch.nn.parameter import Parameter
from itertools import product as product
import torch.onnx
class Scale(nn.Module):
def __init__(self, channels):
super(Scale, self).__init__()
self.weight = Parameter(torch.Tensor(channels))
self.bias = Parameter(torch.Tensor(channels))
self.channels = channels
def __repr__(self):
return 'Scale(channels = %d)' % self.channels
def forward(self, x):
if x.dim() == 2:
nB = x.size(0)
nC = x.size(1)
x = x * self.weight.view(1, nC).expand(nB, nC) + self.bias.view(
1, nC).expand(nB, nC)
else:
nB = x.size(0)
nC = x.size(1)
nH = x.size(2)
nW = x.size(3)
x = x * self.weight.view(1, nC, 1, 1).expand(nB, nC, nH, nW
) + self.bias.view(1, nC, 1, 1).expand(nB, nC, nH, nW)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'channels': 4}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
import torch.nn as nn
from torch.nn.parameter import Parameter
from itertools import product as product
import torch.onnx
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_add_mul_0(in_ptr0, in_ptr1, in_ptr2, out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = xindex // 16 % 4
tmp0 = tl.load(in_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr1 + x1, xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr2 + x1, xmask, eviction_policy='evict_last')
tmp2 = tmp0 * tmp1
tmp4 = tmp2 + tmp3
tl.store(out_ptr0 + x3, tmp4, xmask)
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_add_mul_0[grid(256)](primals_1, primals_2,
primals_3, buf0, 256, XBLOCK=128, num_warps=4, num_stages=1)
del primals_2
del primals_3
return buf0, primals_1
class ScaleNew(nn.Module):
def __init__(self, channels):
super(ScaleNew, self).__init__()
self.weight = Parameter(torch.Tensor(channels))
self.bias = Parameter(torch.Tensor(channels))
self.channels = channels
def __repr__(self):
return 'Scale(channels = %d)' % self.channels
def forward(self, input_0):
primals_2 = self.weight
primals_3 = self.bias
primals_1 = input_0
output = call([primals_1, primals_2, primals_3])
return output[0]
|
Janus1984/Msnhnet
|
Scale
| false
| 13,874
|
[
"MIT"
] | 546
|
4e09f2501ba8db789f0a20441a357de3ba468f10
|
https://github.com/Janus1984/Msnhnet/tree/4e09f2501ba8db789f0a20441a357de3ba468f10
|
Ecgclient
|
import torch
import torch.nn as nn
class Ecgclient(nn.Module):
def __init__(self):
super(Ecgclient, self).__init__()
self.conv1 = nn.Conv1d(1, 16, 7, padding=3)
self.relu1 = nn.LeakyReLU()
self.pool1 = nn.MaxPool1d(2)
self.conv2 = nn.Conv1d(16, 16, 5, padding=2)
self.relu2 = nn.LeakyReLU()
def forward(self, x):
x = self.conv1(x)
x = self.relu1(x)
x = self.pool1(x)
x = self.conv2(x)
x = self.relu2(x)
return x
def get_inputs():
return [torch.rand([4, 1, 64])]
def get_init_inputs():
return [[], {}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_convolution_leaky_relu_0(in_ptr0, in_ptr1, out_ptr0,
out_ptr1, xnumel, XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x3 = xindex
x1 = xindex // 64 % 16
tmp0 = tl.load(in_ptr0 + x3, None)
tmp1 = tl.load(in_ptr1 + x1, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = 0.0
tmp4 = tmp2 > tmp3
tmp5 = 0.01
tmp6 = tmp2 * tmp5
tmp7 = tl.where(tmp4, tmp2, tmp6)
tl.store(out_ptr0 + x3, tmp4, None)
tl.store(out_ptr1 + x3, tmp7, None)
@triton.jit
def triton_poi_fused_max_pool2d_with_indices_1(in_ptr0, out_ptr0, out_ptr1,
xnumel, XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x0 = xindex
tmp0 = tl.load(in_ptr0 + 2 * x0, None, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr0 + (1 + 2 * x0), None, eviction_policy='evict_last')
tmp2 = tmp1 > tmp0
tmp3 = tl.full([1], 1, tl.int8)
tmp4 = tl.full([1], 0, tl.int8)
tmp5 = tl.where(tmp2, tmp3, tmp4)
tmp6 = triton_helpers.maximum(tmp1, tmp0)
tl.store(out_ptr0 + x0, tmp5, None)
tl.store(out_ptr1 + x0, tmp6, None)
@triton.jit
def triton_poi_fused_convolution_leaky_relu_2(in_ptr0, in_ptr1, out_ptr0,
out_ptr1, xnumel, XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x3 = xindex
x1 = xindex // 32 % 16
tmp0 = tl.load(in_ptr0 + x3, None)
tmp1 = tl.load(in_ptr1 + x1, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = 0.0
tmp4 = tmp2 > tmp3
tmp5 = 0.01
tmp6 = tmp2 * tmp5
tmp7 = tl.where(tmp4, tmp2, tmp6)
tl.store(out_ptr0 + x3, tmp4, None)
tl.store(out_ptr1 + x3, tmp7, None)
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5 = args
args.clear()
assert_size_stride(primals_1, (16, 1, 7), (7, 7, 1))
assert_size_stride(primals_2, (16,), (1,))
assert_size_stride(primals_3, (4, 1, 64), (64, 64, 1))
assert_size_stride(primals_4, (16, 16, 5), (80, 5, 1))
assert_size_stride(primals_5, (16,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(primals_3, primals_1, stride=(1,),
padding=(3,), dilation=(1,), transposed=False, output_padding=(
0,), groups=1, bias=None)
assert_size_stride(buf0, (4, 16, 64), (1024, 64, 1))
buf1 = empty_strided_cuda((4, 16, 64), (1024, 64, 1), torch.bool)
buf2 = empty_strided_cuda((4, 16, 64), (1024, 64, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_convolution_leaky_relu_0[grid(4096)](buf0,
primals_2, buf1, buf2, 4096, XBLOCK=128, num_warps=4, num_stages=1)
del buf0
del primals_2
buf3 = empty_strided_cuda((4, 16, 1, 32), (512, 32, 32, 1), torch.int8)
buf4 = empty_strided_cuda((4, 16, 1, 32), (512, 32, 32, 1), torch.
float32)
triton_poi_fused_max_pool2d_with_indices_1[grid(2048)](buf2, buf3,
buf4, 2048, XBLOCK=256, num_warps=4, num_stages=1)
buf5 = extern_kernels.convolution(reinterpret_tensor(buf4, (4, 16,
32), (512, 32, 1), 0), primals_4, stride=(1,), padding=(2,),
dilation=(1,), transposed=False, output_padding=(0,), groups=1,
bias=None)
assert_size_stride(buf5, (4, 16, 32), (512, 32, 1))
buf6 = empty_strided_cuda((4, 16, 32), (512, 32, 1), torch.bool)
buf7 = empty_strided_cuda((4, 16, 32), (512, 32, 1), torch.float32)
triton_poi_fused_convolution_leaky_relu_2[grid(2048)](buf5,
primals_5, buf6, buf7, 2048, XBLOCK=256, num_warps=4, num_stages=1)
del buf5
del primals_5
return buf7, primals_1, primals_3, primals_4, buf1, reinterpret_tensor(buf2
, (4, 16, 1, 64), (1024, 64, 64, 1), 0), buf3, reinterpret_tensor(buf4,
(4, 16, 32), (512, 32, 1), 0), buf6
class EcgclientNew(nn.Module):
def __init__(self):
super(EcgclientNew, self).__init__()
self.conv1 = nn.Conv1d(1, 16, 7, padding=3)
self.relu1 = nn.LeakyReLU()
self.pool1 = nn.MaxPool1d(2)
self.conv2 = nn.Conv1d(16, 16, 5, padding=2)
self.relu2 = nn.LeakyReLU()
def forward(self, input_0):
primals_1 = self.conv1.weight
primals_2 = self.conv1.bias
primals_4 = self.conv2.weight
primals_5 = self.conv2.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4, primals_5])
return output[0]
|
JayDigvijay/Federated-Learning-and-Split-Learning-with-raspberry-pi
|
Ecgclient
| false
| 13,875
|
[
"MIT"
] | 48
|
314a9618fc6be2ba1b9b7bdf93b126d49a2519ee
|
https://github.com/JayDigvijay/Federated-Learning-and-Split-Learning-with-raspberry-pi/tree/314a9618fc6be2ba1b9b7bdf93b126d49a2519ee
|
CELoss
|
import torch
import torch.nn.functional as F
from torch import nn
def reduce_loss(loss, reduction):
"""Reduce loss as specified.
Args:
loss (Tensor): Elementwise loss tensor.
reduction (str): Options are "none", "mean" and "sum".
Return:
Tensor: Reduced loss tensor.
"""
reduction_enum = F._Reduction.get_enum(reduction)
if reduction_enum == 0:
return loss
elif reduction_enum == 1:
return loss.mean()
elif reduction_enum == 2:
return loss.sum()
def weight_reduce_loss(loss, weight=None, reduction='mean', avg_factor=None):
"""Apply element-wise weight and reduce loss.
Args:
loss (Tensor): Element-wise loss.
weight (Tensor): Element-wise weights.
reduction (str): Same as built-in losses of PyTorch.
avg_factor (float): Avarage factor when computing the mean of losses.
Returns:
Tensor: Processed loss values.
"""
if weight is not None:
loss = loss * weight
if avg_factor is None:
loss = reduce_loss(loss, reduction)
elif reduction == 'mean':
loss = loss.sum() / avg_factor
elif reduction != 'none':
raise ValueError('avg_factor can not be used with reduction="sum"')
return loss
def cross_entropy(pred, label, weight=None, reduction='mean', avg_factor=None):
"""Calculate the CrossEntropy loss.
Args:
pred (torch.Tensor): The prediction with shape (N, C), C is the number
of classes.
label (torch.Tensor): The gt label of the prediction.
weight (torch.Tensor, optional): Sample-wise loss weight.
reduction (str): The method used to reduce the loss.
avg_factor (int, optional): Average factor that is used to average
the loss. Defaults to None.
Returns:
torch.Tensor: The calculated loss
"""
loss = F.cross_entropy(pred, label, reduction='none')
if weight is not None:
weight = weight.float()
loss = weight_reduce_loss(loss, weight=weight, reduction=reduction,
avg_factor=avg_factor)
return loss
class CELoss(nn.Module):
def __init__(self, reduction='mean', loss_weight=1.0):
super(CELoss, self).__init__()
self.reduction = reduction
self.loss_weight = loss_weight
self.cls_criterion = cross_entropy
def forward(self, cls_score, label, weight=None, avg_factor=None,
reduction_override=None, **kwargs):
assert reduction_override in (None, 'none', 'mean', 'sum')
reduction = (reduction_override if reduction_override else self.
reduction)
n_pred_ch, n_target_ch = cls_score.shape[1], label.shape[1]
if n_pred_ch == n_target_ch:
label = torch.argmax(label, dim=1)
else:
label = torch.squeeze(label, dim=1)
label = label.long()
loss_cls = self.loss_weight * self.cls_criterion(cls_score, label,
weight, reduction=reduction, avg_factor=avg_factor, **kwargs)
return loss_cls
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import math as tl_math
import torch.nn.functional as F
from torch import nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused__log_softmax_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.
constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x0 = xindex % 16
x2 = xindex // 64
tmp0 = tl.load(in_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + (x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp2 = tl.load(in_ptr0 + (16 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp4 = tl.load(in_ptr0 + (32 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp6 = tl.load(in_ptr0 + (48 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp3 = triton_helpers.maximum(tmp1, tmp2)
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp0 - tmp7
tl.store(out_ptr0 + x3, tmp8, xmask)
@triton.jit
def triton_per_fused_argmax_mean_mul_nll_loss2d_forward_1(in_out_ptr0,
in_ptr0, in_ptr1, xnumel, rnumel, XBLOCK: tl.constexpr):
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex % 16
r1 = rindex // 16
tmp0 = tl.load(in_ptr0 + (r0 + 64 * r1), None)
tmp1 = tl.load(in_ptr0 + (16 + r0 + 64 * r1), None)
tmp17 = tl.load(in_ptr0 + (32 + r0 + 64 * r1), None)
tmp32 = tl.load(in_ptr0 + (48 + r0 + 64 * r1), None)
tmp56 = tl.load(in_ptr1 + (r0 + 64 * r1), None)
tmp58 = tl.load(in_ptr1 + (16 + r0 + 64 * r1), None)
tmp61 = tl.load(in_ptr1 + (32 + r0 + 64 * r1), None)
tmp64 = tl.load(in_ptr1 + (48 + r0 + 64 * r1), None)
tmp2 = tmp0 > tmp1
tmp3 = tmp0 == tmp1
tmp4 = tmp0 != tmp0
tmp5 = tmp1 != tmp1
tmp6 = tmp4 > tmp5
tmp7 = tmp2 | tmp6
tmp8 = tmp4 & tmp5
tmp9 = tmp3 | tmp8
tmp10 = tl.full([1, 1], 0, tl.int64)
tmp11 = tl.full([1, 1], 1, tl.int64)
tmp12 = tmp10 < tmp11
tmp13 = tmp9 & tmp12
tmp14 = tmp7 | tmp13
tmp15 = tl.where(tmp14, tmp0, tmp1)
tmp16 = tl.where(tmp14, tmp10, tmp11)
tmp18 = tmp15 > tmp17
tmp19 = tmp15 == tmp17
tmp20 = tmp15 != tmp15
tmp21 = tmp17 != tmp17
tmp22 = tmp20 > tmp21
tmp23 = tmp18 | tmp22
tmp24 = tmp20 & tmp21
tmp25 = tmp19 | tmp24
tmp26 = tl.full([1, 1], 2, tl.int64)
tmp27 = tmp16 < tmp26
tmp28 = tmp25 & tmp27
tmp29 = tmp23 | tmp28
tmp30 = tl.where(tmp29, tmp15, tmp17)
tmp31 = tl.where(tmp29, tmp16, tmp26)
tmp33 = tmp30 > tmp32
tmp34 = tmp30 == tmp32
tmp35 = tmp30 != tmp30
tmp36 = tmp32 != tmp32
tmp37 = tmp35 > tmp36
tmp38 = tmp33 | tmp37
tmp39 = tmp35 & tmp36
tmp40 = tmp34 | tmp39
tmp41 = tl.full([1, 1], 3, tl.int64)
tmp42 = tmp31 < tmp41
tmp43 = tmp40 & tmp42
tmp44 = tmp38 | tmp43
tl.where(tmp44, tmp30, tmp32)
tmp46 = tl.where(tmp44, tmp31, tmp41)
tmp47 = tl.full([1, 1], -100, tl.int64)
tmp48 = tmp46 != tmp47
tmp49 = tl.where(tmp48, tmp46, tmp10)
tmp50 = tl.full([XBLOCK, RBLOCK], 4, tl.int32)
tmp51 = tmp49 + tmp50
tmp52 = tmp49 < 0
tmp53 = tl.where(tmp52, tmp51, tmp49)
tl.device_assert((0 <= tmp53) & (tmp53 < 4),
'index out of bounds: 0 <= tmp53 < 4')
tmp55 = tl.load(in_ptr1 + (r0 + 16 * tmp53 + 64 * r1), None)
tmp57 = tl_math.exp(tmp56)
tmp59 = tl_math.exp(tmp58)
tmp60 = tmp57 + tmp59
tmp62 = tl_math.exp(tmp61)
tmp63 = tmp60 + tmp62
tmp65 = tl_math.exp(tmp64)
tmp66 = tmp63 + tmp65
tmp67 = tl_math.log(tmp66)
tmp68 = tmp55 - tmp67
tmp69 = -tmp68
tmp70 = 0.0
tmp71 = tl.where(tmp48, tmp69, tmp70)
tmp72 = tl.broadcast_to(tmp71, [XBLOCK, RBLOCK])
tmp74 = tl.sum(tmp72, 1)[:, None]
tmp75 = 64.0
tmp76 = tmp74 / tmp75
tmp77 = 1.0
tmp78 = tmp76 * tmp77
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([XBLOCK, 1], 0, tl.int32), tmp78, None)
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf1 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused__log_softmax_0[grid(256)](arg0_1, buf1, 256,
XBLOCK=128, num_warps=4, num_stages=1)
del arg0_1
buf2 = empty_strided_cuda((), (), torch.float32)
buf3 = buf2
del buf2
triton_per_fused_argmax_mean_mul_nll_loss2d_forward_1[grid(1)](buf3,
arg1_1, buf1, 1, 64, XBLOCK=1, num_warps=2, num_stages=1)
del arg1_1
del buf1
return buf3,
def reduce_loss(loss, reduction):
"""Reduce loss as specified.
Args:
loss (Tensor): Elementwise loss tensor.
reduction (str): Options are "none", "mean" and "sum".
Return:
Tensor: Reduced loss tensor.
"""
reduction_enum = F._Reduction.get_enum(reduction)
if reduction_enum == 0:
return loss
elif reduction_enum == 1:
return loss.mean()
elif reduction_enum == 2:
return loss.sum()
def weight_reduce_loss(loss, weight=None, reduction='mean', avg_factor=None):
"""Apply element-wise weight and reduce loss.
Args:
loss (Tensor): Element-wise loss.
weight (Tensor): Element-wise weights.
reduction (str): Same as built-in losses of PyTorch.
avg_factor (float): Avarage factor when computing the mean of losses.
Returns:
Tensor: Processed loss values.
"""
if weight is not None:
loss = loss * weight
if avg_factor is None:
loss = reduce_loss(loss, reduction)
elif reduction == 'mean':
loss = loss.sum() / avg_factor
elif reduction != 'none':
raise ValueError('avg_factor can not be used with reduction="sum"')
return loss
def cross_entropy(pred, label, weight=None, reduction='mean', avg_factor=None):
"""Calculate the CrossEntropy loss.
Args:
pred (torch.Tensor): The prediction with shape (N, C), C is the number
of classes.
label (torch.Tensor): The gt label of the prediction.
weight (torch.Tensor, optional): Sample-wise loss weight.
reduction (str): The method used to reduce the loss.
avg_factor (int, optional): Average factor that is used to average
the loss. Defaults to None.
Returns:
torch.Tensor: The calculated loss
"""
loss = F.cross_entropy(pred, label, reduction='none')
if weight is not None:
weight = weight.float()
loss = weight_reduce_loss(loss, weight=weight, reduction=reduction,
avg_factor=avg_factor)
return loss
class CELossNew(nn.Module):
def __init__(self, reduction='mean', loss_weight=1.0):
super(CELossNew, self).__init__()
self.reduction = reduction
self.loss_weight = loss_weight
self.cls_criterion = cross_entropy
def forward(self, input_0, input_1):
arg0_1 = input_0
arg1_1 = input_1
output = call([arg0_1, arg1_1])
return output[0]
|
JiYuanFeng/MCTrans
|
CELoss
| false
| 13,876
|
[
"Apache-2.0"
] | 84
|
9b8b5677eef584b423d5e1630680a4b667cbe823
|
https://github.com/JiYuanFeng/MCTrans/tree/9b8b5677eef584b423d5e1630680a4b667cbe823
|
EdgeFeaturesLayer
|
import torch
import torch.nn as nn
class EdgeFeaturesLayer(nn.Module):
def __init__(self, d_model, d_edge, h, dropout):
super(EdgeFeaturesLayer, self).__init__()
assert d_model % h == 0
d_model // h
self.linear = nn.Linear(d_edge, 1, bias=False)
with torch.no_grad():
self.linear.weight.fill_(0.25)
def forward(self, x):
p_edge = x.permute(0, 2, 3, 1)
p_edge = self.linear(p_edge).permute(0, 3, 1, 2)
return torch.relu(p_edge)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'d_model': 4, 'd_edge': 4, 'h': 4, 'dropout': 0.5}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_clone_0(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK: tl.
constexpr, XBLOCK: tl.constexpr):
ynumel = 64
xnumel = 4
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 16
y1 = yindex // 16
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + 16 * x2 + 64 * y1), xmask & ymask,
eviction_policy='evict_last')
tl.store(out_ptr0 + (x2 + 4 * y3), tmp0, xmask & ymask)
@triton.jit
def triton_poi_fused_relu_threshold_backward_1(in_out_ptr0, out_ptr0,
xnumel, XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_out_ptr0 + x0, xmask)
tmp1 = tl.full([1], 0, tl.int32)
tmp2 = triton_helpers.maximum(tmp1, tmp0)
tmp3 = 0.0
tmp4 = tmp2 <= tmp3
tl.store(in_out_ptr0 + x0, tmp2, xmask)
tl.store(out_ptr0 + x0, tmp4, xmask)
def call(args):
primals_1, primals_2 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (1, 4), (4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_clone_0[grid(64, 4)](primals_1, buf0, 64, 4,
XBLOCK=4, YBLOCK=32, num_warps=4, num_stages=1)
del primals_1
buf1 = empty_strided_cuda((64, 1), (1, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf0, (64, 4), (4, 1), 0),
reinterpret_tensor(primals_2, (4, 1), (1, 4), 0), out=buf1)
del primals_2
buf2 = reinterpret_tensor(buf1, (4, 1, 4, 4), (16, 1, 4, 1), 0)
del buf1
buf3 = empty_strided_cuda((4, 1, 4, 4), (16, 1, 4, 1), torch.bool)
triton_poi_fused_relu_threshold_backward_1[grid(64)](buf2, buf3, 64,
XBLOCK=64, num_warps=1, num_stages=1)
return buf2, reinterpret_tensor(buf0, (64, 4), (4, 1), 0), buf3
class EdgeFeaturesLayerNew(nn.Module):
def __init__(self, d_model, d_edge, h, dropout):
super(EdgeFeaturesLayerNew, self).__init__()
assert d_model % h == 0
d_model // h
self.linear = nn.Linear(d_edge, 1, bias=False)
with torch.no_grad():
self.linear.weight.fill_(0.25)
def forward(self, input_0):
primals_2 = self.linear.weight
primals_1 = input_0
output = call([primals_1, primals_2])
return output[0]
|
Jh-SYSU/MolRep
|
EdgeFeaturesLayer
| false
| 13,877
|
[
"MIT"
] | 57
|
b2c802d18d41d7db26c19c6dd644098f945e48a1
|
https://github.com/Jh-SYSU/MolRep/tree/b2c802d18d41d7db26c19c6dd644098f945e48a1
|
PositionGenerator
|
import torch
import torch.nn as nn
class LayerNorm(nn.Module):
def __init__(self, hidden_size, variance_epsilon=1e-12):
super(LayerNorm, self).__init__()
self.gamma = nn.Parameter(torch.ones(hidden_size))
self.beta = nn.Parameter(torch.zeros(hidden_size))
self.variance_epsilon = variance_epsilon
def forward(self, x):
u = x.mean(-1, keepdim=True)
s = (x - u).pow(2).mean(-1, keepdim=True)
x = (x - u) / torch.sqrt(s + self.variance_epsilon)
return self.gamma * x + self.beta
class PositionGenerator(nn.Module):
"""Define standard linear + softmax generation step."""
def __init__(self, d_model):
super(PositionGenerator, self).__init__()
self.norm = LayerNorm(d_model)
self.proj = nn.Linear(d_model, 3)
def forward(self, x, mask):
mask = mask.unsqueeze(-1).float()
out_masked = self.norm(x) * mask
projected = self.proj(out_masked)
return projected
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'d_model': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_mean_sub_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + 4 * x1, xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + 4 * x1), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (2 + 4 * x1), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (3 + 4 * x1), xmask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp5 = tmp3 + tmp4
tmp7 = tmp5 + tmp6
tmp8 = 4.0
tmp9 = tmp7 / tmp8
tmp10 = tmp0 - tmp9
tl.store(out_ptr0 + x2, tmp10, xmask)
@triton.jit
def triton_poi_fused_add_div_mean_mul_pow_sqrt_1(in_ptr0, in_ptr1, in_ptr2,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 4
x2 = xindex
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + x2, xmask)
tmp2 = tl.load(in_ptr1 + 4 * x1, xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr1 + (1 + 4 * x1), xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr1 + (2 + 4 * x1), xmask, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr1 + (3 + 4 * x1), xmask, eviction_policy='evict_last'
)
tmp20 = tl.load(in_ptr2 + x0, xmask, eviction_policy='evict_last')
tmp3 = tmp2 * tmp2
tmp5 = tmp4 * tmp4
tmp6 = tmp3 + tmp5
tmp8 = tmp7 * tmp7
tmp9 = tmp6 + tmp8
tmp11 = tmp10 * tmp10
tmp12 = tmp9 + tmp11
tmp13 = 4.0
tmp14 = tmp12 / tmp13
tmp15 = 1e-12
tmp16 = tmp14 + tmp15
tmp17 = libdevice.sqrt(tmp16)
tmp18 = tmp1 / tmp17
tmp19 = tmp0 * tmp18
tmp21 = tmp19 + tmp20
tl.store(out_ptr0 + x2, tmp21, xmask)
@triton.jit
def triton_poi_fused_add_div_mean_mul_pow_sqrt_2(in_ptr0, in_ptr1, out_ptr0,
xnumel, XBLOCK: tl.constexpr):
xnumel = 1024
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex % 256
x4 = xindex // 4
x5 = xindex
tmp0 = tl.load(in_ptr0 + x3, xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + x4, xmask, eviction_policy='evict_last')
tmp2 = tmp0 * tmp1
tl.store(out_ptr0 + x5, tmp2, xmask)
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_3, (4,), (1,))
assert_size_stride(primals_4, (4,), (1,))
assert_size_stride(primals_5, (3, 4), (4, 1))
assert_size_stride(primals_6, (3,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_mean_sub_0[grid(256)](primals_2, buf0, 256, XBLOCK
=128, num_warps=4, num_stages=1)
buf1 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused_add_div_mean_mul_pow_sqrt_1[grid(256)](primals_3,
buf0, primals_4, buf1, 256, XBLOCK=128, num_warps=4, num_stages=1)
del buf0
del primals_3
del primals_4
buf2 = empty_strided_cuda((4, 4, 4, 4, 4), (256, 64, 16, 4, 1),
torch.float32)
triton_poi_fused_add_div_mean_mul_pow_sqrt_2[grid(1024)](buf1,
primals_1, buf2, 1024, XBLOCK=256, num_warps=4, num_stages=1)
del buf1
buf3 = empty_strided_cuda((256, 3), (3, 1), torch.float32)
extern_kernels.addmm(primals_6, reinterpret_tensor(buf2, (256, 4),
(4, 1), 0), reinterpret_tensor(primals_5, (4, 3), (1, 4), 0),
alpha=1, beta=1, out=buf3)
del primals_6
return reinterpret_tensor(buf3, (4, 4, 4, 4, 3), (192, 48, 12, 3, 1), 0
), primals_1, primals_2, reinterpret_tensor(buf2, (256, 4), (4, 1), 0
), primals_5
class LayerNorm(nn.Module):
def __init__(self, hidden_size, variance_epsilon=1e-12):
super(LayerNorm, self).__init__()
self.gamma = nn.Parameter(torch.ones(hidden_size))
self.beta = nn.Parameter(torch.zeros(hidden_size))
self.variance_epsilon = variance_epsilon
def forward(self, x):
u = x.mean(-1, keepdim=True)
s = (x - u).pow(2).mean(-1, keepdim=True)
x = (x - u) / torch.sqrt(s + self.variance_epsilon)
return self.gamma * x + self.beta
class PositionGeneratorNew(nn.Module):
"""Define standard linear + softmax generation step."""
def __init__(self, d_model):
super(PositionGeneratorNew, self).__init__()
self.norm = LayerNorm(d_model)
self.proj = nn.Linear(d_model, 3)
def forward(self, input_0, input_1):
primals_3 = self.norm.gamma
primals_4 = self.norm.beta
primals_5 = self.proj.weight
primals_6 = self.proj.bias
primals_1 = input_0
primals_2 = input_1
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6])
return output[0]
|
Jh-SYSU/MolRep
|
PositionGenerator
| false
| 13,878
|
[
"MIT"
] | 57
|
b2c802d18d41d7db26c19c6dd644098f945e48a1
|
https://github.com/Jh-SYSU/MolRep/tree/b2c802d18d41d7db26c19c6dd644098f945e48a1
|
LNN
|
import math
import torch
import torch.utils.data
import torch.nn.functional as F
class LNN(torch.nn.Module):
"""
A pytorch implementation of LNN layer
Input shape
- A 3D tensor with shape: ``(batch_size,field_size,embedding_size)``.
Output shape
- 2D tensor with shape:``(batch_size,LNN_dim*embedding_size)``.
Arguments
- **in_features** : Embedding of feature.
- **num_fields**: int.The field size of feature.
- **LNN_dim**: int.The number of Logarithmic neuron.
- **bias**: bool.Whether or not use bias in LNN.
"""
def __init__(self, num_fields, embed_dim, LNN_dim, bias=False):
super(LNN, self).__init__()
self.num_fields = num_fields
self.embed_dim = embed_dim
self.LNN_dim = LNN_dim
self.lnn_output_dim = LNN_dim * embed_dim
self.weight = torch.nn.Parameter(torch.Tensor(LNN_dim, num_fields))
if bias:
self.bias = torch.nn.Parameter(torch.Tensor(LNN_dim, embed_dim))
else:
self.register_parameter('bias', None)
self.reset_parameters()
def reset_parameters(self):
stdv = 1.0 / math.sqrt(self.weight.size(1))
self.weight.data.uniform_(-stdv, stdv)
if self.bias is not None:
self.bias.data.uniform_(-stdv, stdv)
def forward(self, x):
"""
:param x: Long tensor of size ``(batch_size, num_fields, embedding_size)``
"""
embed_x_abs = torch.abs(x)
embed_x_afn = torch.add(embed_x_abs, 1e-07)
embed_x_log = torch.log1p(embed_x_afn)
lnn_out = torch.matmul(self.weight, embed_x_log)
if self.bias is not None:
lnn_out += self.bias
lnn_exp = torch.expm1(lnn_out)
output = F.relu(lnn_exp).contiguous().view(-1, self.lnn_output_dim)
return output
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'num_fields': 4, 'embed_dim': 4, 'LNN_dim': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
import math
import torch.utils.data
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_clone_0(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK: tl.
constexpr, XBLOCK: tl.constexpr):
ynumel = 64
xnumel = 4
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 4
y1 = yindex // 4
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + 4 * x2 + 16 * y1), xmask & ymask,
eviction_policy='evict_last')
tmp1 = tl_math.abs(tmp0)
tmp2 = 1e-07
tmp3 = tmp1 + tmp2
tmp4 = libdevice.log1p(tmp3)
tl.store(out_ptr0 + (x2 + 4 * y3), tmp4, xmask & ymask)
@triton.jit
def triton_poi_fused_clone_expm1_relu_threshold_backward_1(in_ptr0,
out_ptr0, out_ptr1, ynumel, xnumel, YBLOCK: tl.constexpr, XBLOCK: tl.
constexpr):
ynumel = 64
xnumel = 4
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 4
y1 = yindex // 4
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + 4 * x2 + 16 * y1), xmask & ymask,
eviction_policy='evict_last')
tmp1 = libdevice.expm1(tmp0)
tmp2 = tl.full([1, 1], 0, tl.int32)
tmp3 = triton_helpers.maximum(tmp2, tmp1)
tmp4 = 0.0
tmp5 = tmp3 <= tmp4
tl.store(out_ptr0 + (x2 + 4 * y3), tmp3, xmask & ymask)
tl.store(out_ptr1 + (x2 + 4 * y3), tmp5, xmask & ymask)
def call(args):
primals_1, primals_2 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4, 4), (4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_clone_0[grid(64, 4)](primals_1, buf0, 64, 4,
XBLOCK=4, YBLOCK=32, num_warps=4, num_stages=1)
del primals_1
buf1 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf0, (64, 4), (4, 1), 0),
reinterpret_tensor(primals_2, (4, 4), (1, 4), 0), out=buf1)
del primals_2
buf2 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
buf3 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
triton_poi_fused_clone_expm1_relu_threshold_backward_1[grid(64, 4)](
buf1, buf2, buf3, 64, 4, XBLOCK=4, YBLOCK=32, num_warps=4,
num_stages=1)
return reinterpret_tensor(buf2, (16, 16), (16, 1), 0), reinterpret_tensor(
buf0, (64, 4), (4, 1), 0), buf1, buf3
class LNNNew(torch.nn.Module):
"""
A pytorch implementation of LNN layer
Input shape
- A 3D tensor with shape: ``(batch_size,field_size,embedding_size)``.
Output shape
- 2D tensor with shape:``(batch_size,LNN_dim*embedding_size)``.
Arguments
- **in_features** : Embedding of feature.
- **num_fields**: int.The field size of feature.
- **LNN_dim**: int.The number of Logarithmic neuron.
- **bias**: bool.Whether or not use bias in LNN.
"""
def __init__(self, num_fields, embed_dim, LNN_dim, bias=False):
super(LNNNew, self).__init__()
self.num_fields = num_fields
self.embed_dim = embed_dim
self.LNN_dim = LNN_dim
self.lnn_output_dim = LNN_dim * embed_dim
self.weight = torch.nn.Parameter(torch.Tensor(LNN_dim, num_fields))
if bias:
self.bias = torch.nn.Parameter(torch.Tensor(LNN_dim, embed_dim))
else:
self.register_parameter('bias', None)
self.reset_parameters()
def reset_parameters(self):
stdv = 1.0 / math.sqrt(self.weight.size(1))
self.weight.data.uniform_(-stdv, stdv)
if self.bias is not None:
self.bias.data.uniform_(-stdv, stdv)
def forward(self, input_0):
primals_2 = self.weight
primals_1 = input_0
output = call([primals_1, primals_2])
return output[0]
|
JazonJiao/pytorch-fm
|
LNN
| false
| 13,879
|
[
"MIT"
] | 734
|
7192e7861fa54341d5b2df995f92858f583ea09e
|
https://github.com/JazonJiao/pytorch-fm/tree/7192e7861fa54341d5b2df995f92858f583ea09e
|
FactorizationMachine
|
import torch
import torch.utils.data
class FactorizationMachine(torch.nn.Module):
def __init__(self, reduce_sum=True):
super().__init__()
self.reduce_sum = reduce_sum
def forward(self, x):
"""
:param x: Float tensor of size ``(batch_size, num_fields, embed_dim)``
"""
square_of_sum = torch.sum(x, dim=1) ** 2
sum_of_square = torch.sum(x ** 2, dim=1)
ix = square_of_sum - sum_of_square
if self.reduce_sum:
ix = torch.sum(ix, dim=1, keepdim=True)
return 0.5 * ix
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
import torch.utils.data
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_mul_pow_sub_sum_0(in_out_ptr0, in_ptr0, xnumel, XBLOCK:
tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 4
x1 = xindex // 4
x2 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + 64 * x1), xmask)
tmp1 = tl.load(in_ptr0 + (16 + x0 + 64 * x1), xmask)
tmp3 = tl.load(in_ptr0 + (32 + x0 + 64 * x1), xmask)
tmp5 = tl.load(in_ptr0 + (48 + x0 + 64 * x1), xmask)
tmp16 = tl.load(in_ptr0 + (4 + x0 + 64 * x1), xmask)
tmp17 = tl.load(in_ptr0 + (20 + x0 + 64 * x1), xmask)
tmp19 = tl.load(in_ptr0 + (36 + x0 + 64 * x1), xmask)
tmp21 = tl.load(in_ptr0 + (52 + x0 + 64 * x1), xmask)
tmp33 = tl.load(in_ptr0 + (8 + x0 + 64 * x1), xmask)
tmp34 = tl.load(in_ptr0 + (24 + x0 + 64 * x1), xmask)
tmp36 = tl.load(in_ptr0 + (40 + x0 + 64 * x1), xmask)
tmp38 = tl.load(in_ptr0 + (56 + x0 + 64 * x1), xmask)
tmp50 = tl.load(in_ptr0 + (12 + x0 + 64 * x1), xmask)
tmp51 = tl.load(in_ptr0 + (28 + x0 + 64 * x1), xmask)
tmp53 = tl.load(in_ptr0 + (44 + x0 + 64 * x1), xmask)
tmp55 = tl.load(in_ptr0 + (60 + x0 + 64 * x1), xmask)
tmp2 = tmp0 + tmp1
tmp4 = tmp2 + tmp3
tmp6 = tmp4 + tmp5
tmp7 = tmp6 * tmp6
tmp8 = tmp0 * tmp0
tmp9 = tmp1 * tmp1
tmp10 = tmp8 + tmp9
tmp11 = tmp3 * tmp3
tmp12 = tmp10 + tmp11
tmp13 = tmp5 * tmp5
tmp14 = tmp12 + tmp13
tmp15 = tmp7 - tmp14
tmp18 = tmp16 + tmp17
tmp20 = tmp18 + tmp19
tmp22 = tmp20 + tmp21
tmp23 = tmp22 * tmp22
tmp24 = tmp16 * tmp16
tmp25 = tmp17 * tmp17
tmp26 = tmp24 + tmp25
tmp27 = tmp19 * tmp19
tmp28 = tmp26 + tmp27
tmp29 = tmp21 * tmp21
tmp30 = tmp28 + tmp29
tmp31 = tmp23 - tmp30
tmp32 = tmp15 + tmp31
tmp35 = tmp33 + tmp34
tmp37 = tmp35 + tmp36
tmp39 = tmp37 + tmp38
tmp40 = tmp39 * tmp39
tmp41 = tmp33 * tmp33
tmp42 = tmp34 * tmp34
tmp43 = tmp41 + tmp42
tmp44 = tmp36 * tmp36
tmp45 = tmp43 + tmp44
tmp46 = tmp38 * tmp38
tmp47 = tmp45 + tmp46
tmp48 = tmp40 - tmp47
tmp49 = tmp32 + tmp48
tmp52 = tmp50 + tmp51
tmp54 = tmp52 + tmp53
tmp56 = tmp54 + tmp55
tmp57 = tmp56 * tmp56
tmp58 = tmp50 * tmp50
tmp59 = tmp51 * tmp51
tmp60 = tmp58 + tmp59
tmp61 = tmp53 * tmp53
tmp62 = tmp60 + tmp61
tmp63 = tmp55 * tmp55
tmp64 = tmp62 + tmp63
tmp65 = tmp57 - tmp64
tmp66 = tmp49 + tmp65
tmp67 = 0.5
tmp68 = tmp66 * tmp67
tl.store(in_out_ptr0 + x2, tmp68, xmask)
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 1, 4), (4, 16, 1), torch.float32)
buf1 = reinterpret_tensor(buf0, (4, 1, 4), (4, 4, 1), 0)
del buf0
get_raw_stream(0)
triton_poi_fused_mul_pow_sub_sum_0[grid(16)](buf1, arg0_1, 16,
XBLOCK=16, num_warps=1, num_stages=1)
del arg0_1
return buf1,
class FactorizationMachineNew(torch.nn.Module):
def __init__(self, reduce_sum=True):
super().__init__()
self.reduce_sum = reduce_sum
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
|
JazonJiao/pytorch-fm
|
FactorizationMachine
| false
| 13,880
|
[
"MIT"
] | 734
|
7192e7861fa54341d5b2df995f92858f583ea09e
|
https://github.com/JazonJiao/pytorch-fm/tree/7192e7861fa54341d5b2df995f92858f583ea09e
|
Linear_2L_KFRA
|
import torch
import torch.nn as nn
import torch.utils.data
def sample_K_laplace_MN(MAP, upper_Qinv, lower_HHinv):
Z = MAP.data.new(MAP.size()).normal_(mean=0, std=1)
all_mtx_sample = MAP + torch.matmul(torch.matmul(lower_HHinv, Z),
upper_Qinv)
weight_mtx_sample = all_mtx_sample[:, :-1]
bias_mtx_sample = all_mtx_sample[:, -1]
return weight_mtx_sample, bias_mtx_sample
class Linear_2L_KFRA(nn.Module):
def __init__(self, input_dim, output_dim, n_hid):
super(Linear_2L_KFRA, self).__init__()
self.n_hid = n_hid
self.input_dim = input_dim
self.output_dim = output_dim
self.fc1 = nn.Linear(input_dim, self.n_hid)
self.fc2 = nn.Linear(self.n_hid, self.n_hid)
self.fc3 = nn.Linear(self.n_hid, output_dim)
self.act = nn.ReLU(inplace=True)
self.one = None
self.a2 = None
self.h2 = None
self.a1 = None
self.h1 = None
self.a0 = None
def forward(self, x):
self.one = x.new(x.shape[0], 1).fill_(1)
a0 = x.view(-1, self.input_dim)
self.a0 = torch.cat((a0.data, self.one), dim=1)
h1 = self.fc1(a0)
self.h1 = h1.data
a1 = self.act(h1)
self.a1 = torch.cat((a1.data, self.one), dim=1)
h2 = self.fc2(a1)
self.h2 = h2.data
a2 = self.act(h2)
self.a2 = torch.cat((a2.data, self.one), dim=1)
h3 = self.fc3(a2)
return h3
def sample_predict(self, x, Nsamples, Qinv1, HHinv1, MAP1, Qinv2,
HHinv2, MAP2, Qinv3, HHinv3, MAP3):
predictions = x.data.new(Nsamples, x.shape[0], self.output_dim)
x = x.view(-1, self.input_dim)
for i in range(Nsamples):
w1, b1 = sample_K_laplace_MN(MAP1, Qinv1, HHinv1)
a = torch.matmul(x, torch.t(w1)) + b1.unsqueeze(0)
a = self.act(a)
w2, b2 = sample_K_laplace_MN(MAP2, Qinv2, HHinv2)
a = torch.matmul(a, torch.t(w2)) + b2.unsqueeze(0)
a = self.act(a)
w3, b3 = sample_K_laplace_MN(MAP3, Qinv3, HHinv3)
y = torch.matmul(a, torch.t(w3)) + b3.unsqueeze(0)
predictions[i] = y
return predictions
def get_inputs():
return [torch.rand([4, 4])]
def get_init_inputs():
return [[], {'input_dim': 4, 'output_dim': 4, 'n_hid': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
import torch.nn as nn
import torch.utils.data
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_fill_0(out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = 1.0
tl.store(out_ptr0 + x0, tmp0, xmask)
@triton.jit
def triton_poi_fused_cat_1(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 20
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 5
x1 = xindex // 5
x2 = xindex
tmp0 = x0
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (4 * x1 + x0), tmp4 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp6 = tmp0 >= tmp3
tl.full([1], 5, tl.int64)
tmp9 = 1.0
tmp10 = tl.full(tmp9.shape, 0.0, tmp9.dtype)
tmp11 = tl.where(tmp6, tmp9, tmp10)
tmp12 = tl.where(tmp4, tmp5, tmp11)
tl.store(out_ptr0 + x2, tmp12, xmask)
@triton.jit
def triton_poi_fused_relu_2(in_out_ptr0, in_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_out_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + x2, tmp4, xmask)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7) = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4, 4), (4, 1))
assert_size_stride(primals_3, (4,), (1,))
assert_size_stride(primals_4, (4, 4), (4, 1))
assert_size_stride(primals_5, (4,), (1,))
assert_size_stride(primals_6, (4, 4), (4, 1))
assert_size_stride(primals_7, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 1), (1, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_fill_0[grid(4)](buf0, 4, XBLOCK=4, num_warps=1,
num_stages=1)
buf1 = empty_strided_cuda((4, 5), (5, 1), torch.float32)
triton_poi_fused_cat_1[grid(20)](primals_1, buf1, 20, XBLOCK=32,
num_warps=1, num_stages=1)
buf2 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.mm(primals_1, reinterpret_tensor(primals_2, (4, 4),
(1, 4), 0), out=buf2)
del primals_2
buf3 = buf2
del buf2
triton_poi_fused_relu_2[grid(16)](buf3, primals_3, 16, XBLOCK=16,
num_warps=1, num_stages=1)
del primals_3
buf4 = empty_strided_cuda((4, 5), (5, 1), torch.float32)
triton_poi_fused_cat_1[grid(20)](buf3, buf4, 20, XBLOCK=32,
num_warps=1, num_stages=1)
buf5 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.mm(buf3, reinterpret_tensor(primals_4, (4, 4), (1, 4
), 0), out=buf5)
buf6 = buf5
del buf5
triton_poi_fused_relu_2[grid(16)](buf6, primals_5, 16, XBLOCK=16,
num_warps=1, num_stages=1)
del primals_5
buf7 = empty_strided_cuda((4, 5), (5, 1), torch.float32)
triton_poi_fused_cat_1[grid(20)](buf6, buf7, 20, XBLOCK=32,
num_warps=1, num_stages=1)
buf8 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_7, buf6, reinterpret_tensor(primals_6,
(4, 4), (1, 4), 0), alpha=1, beta=1, out=buf8)
del primals_7
return (buf8, buf7, buf6, buf4, buf3, buf1, buf0, primals_1, buf3, buf6,
primals_6, primals_4)
def sample_K_laplace_MN(MAP, upper_Qinv, lower_HHinv):
Z = MAP.data.new(MAP.size()).normal_(mean=0, std=1)
all_mtx_sample = MAP + torch.matmul(torch.matmul(lower_HHinv, Z),
upper_Qinv)
weight_mtx_sample = all_mtx_sample[:, :-1]
bias_mtx_sample = all_mtx_sample[:, -1]
return weight_mtx_sample, bias_mtx_sample
class Linear_2L_KFRANew(nn.Module):
def __init__(self, input_dim, output_dim, n_hid):
super(Linear_2L_KFRANew, self).__init__()
self.n_hid = n_hid
self.input_dim = input_dim
self.output_dim = output_dim
self.fc1 = nn.Linear(input_dim, self.n_hid)
self.fc2 = nn.Linear(self.n_hid, self.n_hid)
self.fc3 = nn.Linear(self.n_hid, output_dim)
self.act = nn.ReLU(inplace=True)
self.one = None
self.a2 = None
self.h2 = None
self.a1 = None
self.h1 = None
self.a0 = None
def sample_predict(self, x, Nsamples, Qinv1, HHinv1, MAP1, Qinv2,
HHinv2, MAP2, Qinv3, HHinv3, MAP3):
predictions = x.data.new(Nsamples, x.shape[0], self.output_dim)
x = x.view(-1, self.input_dim)
for i in range(Nsamples):
w1, b1 = sample_K_laplace_MN(MAP1, Qinv1, HHinv1)
a = torch.matmul(x, torch.t(w1)) + b1.unsqueeze(0)
a = self.act(a)
w2, b2 = sample_K_laplace_MN(MAP2, Qinv2, HHinv2)
a = torch.matmul(a, torch.t(w2)) + b2.unsqueeze(0)
a = self.act(a)
w3, b3 = sample_K_laplace_MN(MAP3, Qinv3, HHinv3)
y = torch.matmul(a, torch.t(w3)) + b3.unsqueeze(0)
predictions[i] = y
return predictions
def forward(self, input_0):
primals_1 = self.fc1.weight
primals_3 = self.fc1.bias
primals_2 = self.fc2.weight
primals_5 = self.fc2.bias
primals_4 = self.fc3.weight
primals_7 = self.fc3.bias
primals_6 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7])
return output[0]
|
JavierAntoran/Bayesain-Neural-Networks
|
Linear_2L_KFRA
| false
| 13,881
|
[
"MIT"
] | 1,299
|
1f867a5bcbd1abfecede99807eb0b5f97ed8be7c
|
https://github.com/JavierAntoran/Bayesain-Neural-Networks/tree/1f867a5bcbd1abfecede99807eb0b5f97ed8be7c
|
ScaleNorm
|
import math
import torch
import torch.nn as nn
class ScaleNorm(nn.Module):
"""ScaleNorm"""
"""All g’s in SCALE NORM are initialized to sqrt(d)"""
def __init__(self, scale, eps=1e-05):
super(ScaleNorm, self).__init__()
self.scale = nn.Parameter(torch.tensor(math.sqrt(scale)))
self.eps = eps
def forward(self, x):
norm = self.scale / torch.norm(x, dim=-1, keepdim=True).clamp(min=
self.eps)
return x * norm
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'scale': 1.0}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import libdevice
import math
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_clamp_div_linalg_vector_norm_0(in_ptr0, in_ptr1,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + 0)
tmp1 = tl.broadcast_to(tmp0, [XBLOCK])
tmp2 = tl.load(in_ptr1 + 4 * x0, xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr1 + (1 + 4 * x0), xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr1 + (2 + 4 * x0), xmask, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr1 + (3 + 4 * x0), xmask, eviction_policy='evict_last'
)
tmp3 = tmp2 * tmp2
tmp5 = tmp4 * tmp4
tmp6 = tmp3 + tmp5
tmp8 = tmp7 * tmp7
tmp9 = tmp6 + tmp8
tmp11 = tmp10 * tmp10
tmp12 = tmp9 + tmp11
tmp13 = libdevice.sqrt(tmp12)
tmp14 = 1e-05
tmp15 = triton_helpers.maximum(tmp13, tmp14)
tmp16 = tmp1 / tmp15
tl.store(out_ptr0 + x0, tmp16, xmask)
@triton.jit
def triton_poi_fused_clamp_div_linalg_vector_norm_mul_1(in_ptr0, in_ptr1,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr1 + x1, xmask, eviction_policy='evict_last')
tmp2 = tmp0 * tmp1
tl.store(out_ptr0 + x2, tmp2, xmask)
def call(args):
primals_1, primals_2 = args
args.clear()
assert_size_stride(primals_1, (), ())
assert_size_stride(primals_2, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 64), torch.float32)
get_raw_stream(0)
triton_poi_fused_clamp_div_linalg_vector_norm_0[grid(64)](primals_1,
primals_2, buf0, 64, XBLOCK=64, num_warps=1, num_stages=1)
del primals_1
buf1 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused_clamp_div_linalg_vector_norm_mul_1[grid(256)](
primals_2, buf0, buf1, 256, XBLOCK=128, num_warps=4, num_stages=1)
del buf0
return buf1, primals_2
class ScaleNormNew(nn.Module):
"""ScaleNorm"""
"""All g’s in SCALE NORM are initialized to sqrt(d)"""
def __init__(self, scale, eps=1e-05):
super(ScaleNormNew, self).__init__()
self.scale = nn.Parameter(torch.tensor(math.sqrt(scale)))
self.eps = eps
def forward(self, input_0):
primals_1 = self.scale
primals_2 = input_0
output = call([primals_1, primals_2])
return output[0]
|
Jh-SYSU/MolRep
|
ScaleNorm
| false
| 13,882
|
[
"MIT"
] | 57
|
b2c802d18d41d7db26c19c6dd644098f945e48a1
|
https://github.com/Jh-SYSU/MolRep/tree/b2c802d18d41d7db26c19c6dd644098f945e48a1
|
AsymmetricLoss
|
import torch
import torch.nn as nn
import torch.nn.functional as F
def reduce_loss(loss, reduction):
"""Reduce loss as specified.
Args:
loss (Tensor): Elementwise loss tensor.
reduction (str): Options are "none", "mean" and "sum".
Return:
Tensor: Reduced loss tensor.
"""
reduction_enum = F._Reduction.get_enum(reduction)
if reduction_enum == 0:
return loss
elif reduction_enum == 1:
return loss.mean()
elif reduction_enum == 2:
return loss.sum()
def weight_reduce_loss(loss, weight=None, reduction='mean', avg_factor=None):
"""Apply element-wise weight and reduce loss.
Args:
loss (Tensor): Element-wise loss.
weight (Tensor): Element-wise weights.
reduction (str): Same as built-in losses of PyTorch.
avg_factor (float): Average factor when computing the mean of losses.
Returns:
Tensor: Processed loss values.
"""
if weight is not None:
loss = loss * weight
if avg_factor is None:
loss = reduce_loss(loss, reduction)
elif reduction == 'mean':
loss = loss.sum() / avg_factor
elif reduction != 'none':
raise ValueError('avg_factor can not be used with reduction="sum"')
return loss
def asymmetric_loss(pred, target, weight=None, gamma_pos=1.0, gamma_neg=4.0,
clip=0.05, reduction='mean', avg_factor=None, use_sigmoid=True, eps=1e-08):
"""asymmetric loss.
Please refer to the `paper <https://arxiv.org/abs/2009.14119>`__ for
details.
Args:
pred (torch.Tensor): The prediction with shape (N, \\*).
target (torch.Tensor): The ground truth label of the prediction with
shape (N, \\*).
weight (torch.Tensor, optional): Sample-wise loss weight with shape
(N, ). Defaults to None.
gamma_pos (float): positive focusing parameter. Defaults to 0.0.
gamma_neg (float): Negative focusing parameter. We usually set
gamma_neg > gamma_pos. Defaults to 4.0.
clip (float, optional): Probability margin. Defaults to 0.05.
reduction (str): The method used to reduce the loss.
Options are "none", "mean" and "sum". If reduction is 'none' , loss
is same shape as pred and label. Defaults to 'mean'.
avg_factor (int, optional): Average factor that is used to average
the loss. Defaults to None.
use_sigmoid (bool): Whether the prediction uses sigmoid instead
of softmax. Defaults to True.
eps (float): The minimum value of the argument of logarithm. Defaults
to 1e-8.
Returns:
torch.Tensor: Loss.
"""
assert pred.shape == target.shape, 'pred and target should be in the same shape.'
if use_sigmoid:
pred_sigmoid = pred.sigmoid()
else:
pred_sigmoid = nn.functional.softmax(pred, dim=-1)
target = target.type_as(pred)
if clip and clip > 0:
pt = (1 - pred_sigmoid + clip).clamp(max=1) * (1 - target
) + pred_sigmoid * target
else:
pt = (1 - pred_sigmoid) * (1 - target) + pred_sigmoid * target
asymmetric_weight = (1 - pt).pow(gamma_pos * target + gamma_neg * (1 -
target))
loss = -torch.log(pt.clamp(min=eps)) * asymmetric_weight
if weight is not None:
assert weight.dim() == 1
weight = weight.float()
if pred.dim() > 1:
weight = weight.reshape(-1, 1)
loss = weight_reduce_loss(loss, weight, reduction, avg_factor)
return loss
def convert_to_one_hot(targets: 'torch.Tensor', classes) ->torch.Tensor:
"""This function converts target class indices to one-hot vectors, given
the number of classes.
Args:
targets (Tensor): The ground truth label of the prediction
with shape (N, 1)
classes (int): the number of classes.
Returns:
Tensor: Processed loss values.
"""
assert torch.max(targets).item(
) < classes, 'Class Index must be less than number of classes'
one_hot_targets = torch.zeros((targets.shape[0], classes), dtype=torch.
long, device=targets.device)
one_hot_targets.scatter_(1, targets.long(), 1)
return one_hot_targets
class AsymmetricLoss(nn.Module):
"""asymmetric loss.
Args:
gamma_pos (float): positive focusing parameter.
Defaults to 0.0.
gamma_neg (float): Negative focusing parameter. We
usually set gamma_neg > gamma_pos. Defaults to 4.0.
clip (float, optional): Probability margin. Defaults to 0.05.
reduction (str): The method used to reduce the loss into
a scalar.
loss_weight (float): Weight of loss. Defaults to 1.0.
use_sigmoid (bool): Whether the prediction uses sigmoid instead
of softmax. Defaults to True.
eps (float): The minimum value of the argument of logarithm. Defaults
to 1e-8.
"""
def __init__(self, gamma_pos=0.0, gamma_neg=4.0, clip=0.05, reduction=
'mean', loss_weight=1.0, use_sigmoid=True, eps=1e-08):
super(AsymmetricLoss, self).__init__()
self.gamma_pos = gamma_pos
self.gamma_neg = gamma_neg
self.clip = clip
self.reduction = reduction
self.loss_weight = loss_weight
self.use_sigmoid = use_sigmoid
self.eps = eps
def forward(self, pred, target, weight=None, avg_factor=None,
reduction_override=None):
"""asymmetric loss.
Args:
pred (torch.Tensor): The prediction with shape (N, \\*).
target (torch.Tensor): The ground truth label of the prediction
with shape (N, \\*), N or (N,1).
weight (torch.Tensor, optional): Sample-wise loss weight with shape
(N, \\*). Defaults to None.
avg_factor (int, optional): Average factor that is used to average
the loss. Defaults to None.
reduction_override (str, optional): The method used to reduce the
loss into a scalar. Options are "none", "mean" and "sum".
Defaults to None.
Returns:
torch.Tensor: Loss.
"""
assert reduction_override in (None, 'none', 'mean', 'sum')
reduction = (reduction_override if reduction_override else self.
reduction)
if target.dim() == 1 or target.dim() == 2 and target.shape[1] == 1:
target = convert_to_one_hot(target.view(-1, 1), pred.shape[-1])
loss_cls = self.loss_weight * asymmetric_loss(pred, target, weight,
gamma_pos=self.gamma_pos, gamma_neg=self.gamma_neg, clip=self.
clip, reduction=reduction, avg_factor=avg_factor, use_sigmoid=
self.use_sigmoid, eps=self.eps)
return loss_cls
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
import torch.nn as nn
import torch.nn.functional as F
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_per_fused_add_clamp_log_mean_mul_neg_pow_rsub_sigmoid_0(in_out_ptr0,
in_ptr0, in_ptr1, xnumel, rnumel):
XBLOCK: tl.constexpr = 1
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
tl.full([1], xoffset, tl.int32)
tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
tl.full([RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp7 = tl.load(in_ptr1 + r0, None)
tmp1 = tl.sigmoid(tmp0)
tmp2 = 1.0
tmp3 = tmp2 - tmp1
tmp4 = 0.05
tmp5 = tmp3 + tmp4
tmp6 = triton_helpers.minimum(tmp5, tmp2)
tmp8 = tmp2 - tmp7
tmp9 = tmp6 * tmp8
tmp10 = tmp1 * tmp7
tmp11 = tmp9 + tmp10
tmp12 = 1e-08
tmp13 = triton_helpers.maximum(tmp11, tmp12)
tmp14 = tl_math.log(tmp13)
tmp15 = -tmp14
tmp16 = tmp2 - tmp11
tmp17 = 0.0
tmp18 = tmp7 * tmp17
tmp19 = 4.0
tmp20 = tmp8 * tmp19
tmp21 = tmp18 + tmp20
tmp22 = libdevice.pow(tmp16, tmp21)
tmp23 = tmp15 * tmp22
tmp24 = tl.broadcast_to(tmp23, [RBLOCK])
tmp26 = triton_helpers.promote_to_tensor(tl.sum(tmp24, 0))
tmp27 = 256.0
tmp28 = tmp26 / tmp27
tmp29 = tmp28 * tmp2
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([1], 0, tl.int32), tmp29, None)
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((), (), torch.float32)
buf1 = buf0
del buf0
get_raw_stream(0)
triton_per_fused_add_clamp_log_mean_mul_neg_pow_rsub_sigmoid_0[grid(1)
](buf1, arg1_1, arg0_1, 1, 256, num_warps=2, num_stages=1)
del arg0_1
del arg1_1
return buf1,
def reduce_loss(loss, reduction):
"""Reduce loss as specified.
Args:
loss (Tensor): Elementwise loss tensor.
reduction (str): Options are "none", "mean" and "sum".
Return:
Tensor: Reduced loss tensor.
"""
reduction_enum = F._Reduction.get_enum(reduction)
if reduction_enum == 0:
return loss
elif reduction_enum == 1:
return loss.mean()
elif reduction_enum == 2:
return loss.sum()
def weight_reduce_loss(loss, weight=None, reduction='mean', avg_factor=None):
"""Apply element-wise weight and reduce loss.
Args:
loss (Tensor): Element-wise loss.
weight (Tensor): Element-wise weights.
reduction (str): Same as built-in losses of PyTorch.
avg_factor (float): Average factor when computing the mean of losses.
Returns:
Tensor: Processed loss values.
"""
if weight is not None:
loss = loss * weight
if avg_factor is None:
loss = reduce_loss(loss, reduction)
elif reduction == 'mean':
loss = loss.sum() / avg_factor
elif reduction != 'none':
raise ValueError('avg_factor can not be used with reduction="sum"')
return loss
def asymmetric_loss(pred, target, weight=None, gamma_pos=1.0, gamma_neg=4.0,
clip=0.05, reduction='mean', avg_factor=None, use_sigmoid=True, eps=1e-08):
"""asymmetric loss.
Please refer to the `paper <https://arxiv.org/abs/2009.14119>`__ for
details.
Args:
pred (torch.Tensor): The prediction with shape (N, \\*).
target (torch.Tensor): The ground truth label of the prediction with
shape (N, \\*).
weight (torch.Tensor, optional): Sample-wise loss weight with shape
(N, ). Defaults to None.
gamma_pos (float): positive focusing parameter. Defaults to 0.0.
gamma_neg (float): Negative focusing parameter. We usually set
gamma_neg > gamma_pos. Defaults to 4.0.
clip (float, optional): Probability margin. Defaults to 0.05.
reduction (str): The method used to reduce the loss.
Options are "none", "mean" and "sum". If reduction is 'none' , loss
is same shape as pred and label. Defaults to 'mean'.
avg_factor (int, optional): Average factor that is used to average
the loss. Defaults to None.
use_sigmoid (bool): Whether the prediction uses sigmoid instead
of softmax. Defaults to True.
eps (float): The minimum value of the argument of logarithm. Defaults
to 1e-8.
Returns:
torch.Tensor: Loss.
"""
assert pred.shape == target.shape, 'pred and target should be in the same shape.'
if use_sigmoid:
pred_sigmoid = pred.sigmoid()
else:
pred_sigmoid = nn.functional.softmax(pred, dim=-1)
target = target.type_as(pred)
if clip and clip > 0:
pt = (1 - pred_sigmoid + clip).clamp(max=1) * (1 - target
) + pred_sigmoid * target
else:
pt = (1 - pred_sigmoid) * (1 - target) + pred_sigmoid * target
asymmetric_weight = (1 - pt).pow(gamma_pos * target + gamma_neg * (1 -
target))
loss = -torch.log(pt.clamp(min=eps)) * asymmetric_weight
if weight is not None:
assert weight.dim() == 1
weight = weight.float()
if pred.dim() > 1:
weight = weight.reshape(-1, 1)
loss = weight_reduce_loss(loss, weight, reduction, avg_factor)
return loss
def convert_to_one_hot(targets: 'torch.Tensor', classes) ->torch.Tensor:
"""This function converts target class indices to one-hot vectors, given
the number of classes.
Args:
targets (Tensor): The ground truth label of the prediction
with shape (N, 1)
classes (int): the number of classes.
Returns:
Tensor: Processed loss values.
"""
assert torch.max(targets).item(
) < classes, 'Class Index must be less than number of classes'
one_hot_targets = torch.zeros((targets.shape[0], classes), dtype=torch.
long, device=targets.device)
one_hot_targets.scatter_(1, targets.long(), 1)
return one_hot_targets
class AsymmetricLossNew(nn.Module):
"""asymmetric loss.
Args:
gamma_pos (float): positive focusing parameter.
Defaults to 0.0.
gamma_neg (float): Negative focusing parameter. We
usually set gamma_neg > gamma_pos. Defaults to 4.0.
clip (float, optional): Probability margin. Defaults to 0.05.
reduction (str): The method used to reduce the loss into
a scalar.
loss_weight (float): Weight of loss. Defaults to 1.0.
use_sigmoid (bool): Whether the prediction uses sigmoid instead
of softmax. Defaults to True.
eps (float): The minimum value of the argument of logarithm. Defaults
to 1e-8.
"""
def __init__(self, gamma_pos=0.0, gamma_neg=4.0, clip=0.05, reduction=
'mean', loss_weight=1.0, use_sigmoid=True, eps=1e-08):
super(AsymmetricLossNew, self).__init__()
self.gamma_pos = gamma_pos
self.gamma_neg = gamma_neg
self.clip = clip
self.reduction = reduction
self.loss_weight = loss_weight
self.use_sigmoid = use_sigmoid
self.eps = eps
def forward(self, input_0, input_1):
arg0_1 = input_0
arg1_1 = input_1
output = call([arg0_1, arg1_1])
return output[0]
|
JiYuanFeng/mmclassification
|
AsymmetricLoss
| false
| 13,883
|
[
"Apache-2.0"
] | 1,190
|
b337ef1f11b85148cca4b6fb0c4da3f8cc2eede6
|
https://github.com/JiYuanFeng/mmclassification/tree/b337ef1f11b85148cca4b6fb0c4da3f8cc2eede6
|
Generator
|
import math
import torch
import torch.nn as nn
class LayerNorm(nn.Module):
def __init__(self, hidden_size, variance_epsilon=1e-12):
super(LayerNorm, self).__init__()
self.gamma = nn.Parameter(torch.ones(hidden_size))
self.beta = nn.Parameter(torch.zeros(hidden_size))
self.variance_epsilon = variance_epsilon
def forward(self, x):
u = x.mean(-1, keepdim=True)
s = (x - u).pow(2).mean(-1, keepdim=True)
x = (x - u) / torch.sqrt(s + self.variance_epsilon)
return self.gamma * x + self.beta
class ScaleNorm(nn.Module):
"""ScaleNorm"""
"""All g’s in SCALE NORM are initialized to sqrt(d)"""
def __init__(self, scale, eps=1e-05):
super(ScaleNorm, self).__init__()
self.scale = nn.Parameter(torch.tensor(math.sqrt(scale)))
self.eps = eps
def forward(self, x):
norm = self.scale / torch.norm(x, dim=-1, keepdim=True).clamp(min=
self.eps)
return x * norm
class Generator(nn.Module):
"""Define standard linear + softmax generation step."""
def __init__(self, d_model, aggregation_type='mean', n_output=1,
n_layers=1, leaky_relu_slope=0.01, dropout=0.0, scale_norm=False):
super(Generator, self).__init__()
if n_layers == 1:
self.proj = nn.Linear(d_model, n_output)
else:
self.proj = []
for i in range(n_layers - 1):
self.proj.append(nn.Linear(d_model, d_model))
self.proj.append(nn.LeakyReLU(leaky_relu_slope))
self.proj.append(ScaleNorm(d_model) if scale_norm else
LayerNorm(d_model))
self.proj.append(nn.Dropout(dropout))
self.proj.append(nn.Linear(d_model, n_output))
self.proj = torch.nn.Sequential(*self.proj)
self.aggregation_type = aggregation_type
def forward(self, x, mask):
mask = mask.unsqueeze(-1).float()
out_masked = x * mask
if self.aggregation_type == 'mean':
out_sum = out_masked.sum(dim=1)
mask_sum = mask.sum(dim=1)
out_avg_pooling = out_sum / mask_sum
elif self.aggregation_type == 'sum':
out_sum = out_masked.sum(dim=1)
out_avg_pooling = out_sum
elif self.aggregation_type == 'dummy_node':
out_avg_pooling = out_masked[:, 0]
projected = self.proj(out_avg_pooling)
return projected
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'d_model': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
import math
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_div_mul_sum_0(in_ptr0, in_ptr1, out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex % 64
x1 = xindex // 4 % 16
x2 = xindex // 64
x4 = xindex
tmp0 = tl.load(in_ptr0 + x3, xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (x1 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp3 = tl.load(in_ptr0 + (64 + x3), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr1 + (16 + x1 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp7 = tl.load(in_ptr0 + (128 + x3), xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr1 + (32 + x1 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp11 = tl.load(in_ptr0 + (192 + x3), xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr1 + (48 + x1 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp2 = tmp0 * tmp1
tmp5 = tmp3 * tmp4
tmp6 = tmp2 + tmp5
tmp9 = tmp7 * tmp8
tmp10 = tmp6 + tmp9
tmp13 = tmp11 * tmp12
tmp14 = tmp10 + tmp13
tmp15 = tmp1 + tmp4
tmp16 = tmp15 + tmp8
tmp17 = tmp16 + tmp12
tmp18 = tmp14 / tmp17
tl.store(out_ptr0 + x4, tmp18, xmask)
def call(args):
primals_1, primals_2, primals_3, primals_4 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_3, (1, 4), (4, 1))
assert_size_stride(primals_4, (1,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_div_mul_sum_0[grid(256)](primals_2, primals_1,
buf0, 256, XBLOCK=256, num_warps=4, num_stages=1)
del primals_1
del primals_2
buf2 = empty_strided_cuda((64, 1), (1, 1), torch.float32)
extern_kernels.addmm(primals_4, reinterpret_tensor(buf0, (64, 4), (
4, 1), 0), reinterpret_tensor(primals_3, (4, 1), (1, 4), 0),
alpha=1, beta=1, out=buf2)
del primals_3
del primals_4
return reinterpret_tensor(buf2, (4, 4, 4, 1), (16, 4, 1, 1), 0
), reinterpret_tensor(buf0, (64, 4), (4, 1), 0)
class LayerNorm(nn.Module):
def __init__(self, hidden_size, variance_epsilon=1e-12):
super(LayerNorm, self).__init__()
self.gamma = nn.Parameter(torch.ones(hidden_size))
self.beta = nn.Parameter(torch.zeros(hidden_size))
self.variance_epsilon = variance_epsilon
def forward(self, x):
u = x.mean(-1, keepdim=True)
s = (x - u).pow(2).mean(-1, keepdim=True)
x = (x - u) / torch.sqrt(s + self.variance_epsilon)
return self.gamma * x + self.beta
class ScaleNorm(nn.Module):
"""ScaleNorm"""
"""All g’s in SCALE NORM are initialized to sqrt(d)"""
def __init__(self, scale, eps=1e-05):
super(ScaleNorm, self).__init__()
self.scale = nn.Parameter(torch.tensor(math.sqrt(scale)))
self.eps = eps
def forward(self, x):
norm = self.scale / torch.norm(x, dim=-1, keepdim=True).clamp(min=
self.eps)
return x * norm
class GeneratorNew(nn.Module):
"""Define standard linear + softmax generation step."""
def __init__(self, d_model, aggregation_type='mean', n_output=1,
n_layers=1, leaky_relu_slope=0.01, dropout=0.0, scale_norm=False):
super(GeneratorNew, self).__init__()
if n_layers == 1:
self.proj = nn.Linear(d_model, n_output)
else:
self.proj = []
for i in range(n_layers - 1):
self.proj.append(nn.Linear(d_model, d_model))
self.proj.append(nn.LeakyReLU(leaky_relu_slope))
self.proj.append(ScaleNorm(d_model) if scale_norm else
LayerNorm(d_model))
self.proj.append(nn.Dropout(dropout))
self.proj.append(nn.Linear(d_model, n_output))
self.proj = torch.nn.Sequential(*self.proj)
self.aggregation_type = aggregation_type
def forward(self, input_0, input_1):
primals_3 = self.proj.weight
primals_4 = self.proj.bias
primals_1 = input_0
primals_2 = input_1
output = call([primals_1, primals_2, primals_3, primals_4])
return output[0]
|
Jh-SYSU/MolRep
|
Generator
| false
| 13,884
|
[
"MIT"
] | 57
|
b2c802d18d41d7db26c19c6dd644098f945e48a1
|
https://github.com/Jh-SYSU/MolRep/tree/b2c802d18d41d7db26c19c6dd644098f945e48a1
|
CQAttention
|
import torch
import torch.nn as nn
import torch.utils.data
import torch.backends.cudnn
def mask_logits(inputs, mask, mask_value=-1e+30):
mask = mask.type(torch.float32)
return inputs + (1.0 - mask) * mask_value
class Conv1D(nn.Module):
def __init__(self, in_dim, out_dim, kernel_size=1, stride=1, padding=0,
bias=True):
super(Conv1D, self).__init__()
self.conv1d = nn.Conv1d(in_channels=in_dim, out_channels=out_dim,
kernel_size=kernel_size, padding=padding, stride=stride, bias=bias)
def forward(self, x):
x = x.transpose(1, 2)
x = self.conv1d(x)
return x.transpose(1, 2)
class CQAttention(nn.Module):
def __init__(self, dim, drop_rate=0.0):
super(CQAttention, self).__init__()
w4C = torch.empty(dim, 1)
w4Q = torch.empty(dim, 1)
w4mlu = torch.empty(1, 1, dim)
nn.init.xavier_uniform_(w4C)
nn.init.xavier_uniform_(w4Q)
nn.init.xavier_uniform_(w4mlu)
self.w4C = nn.Parameter(w4C, requires_grad=True)
self.w4Q = nn.Parameter(w4Q, requires_grad=True)
self.w4mlu = nn.Parameter(w4mlu, requires_grad=True)
self.dropout = nn.Dropout(p=drop_rate)
self.cqa_linear = Conv1D(in_dim=4 * dim, out_dim=dim, kernel_size=1,
stride=1, padding=0, bias=True)
def forward(self, context, query, c_mask, q_mask):
score = self.trilinear_attention(context, query)
score_ = nn.Softmax(dim=2)(mask_logits(score, q_mask.unsqueeze(1)))
score_t = nn.Softmax(dim=1)(mask_logits(score, c_mask.unsqueeze(2)))
score_t = score_t.transpose(1, 2)
c2q = torch.matmul(score_, query)
q2c = torch.matmul(torch.matmul(score_, score_t), context)
output = torch.cat([context, c2q, torch.mul(context, c2q), torch.
mul(context, q2c)], dim=2)
output = self.cqa_linear(output)
return output
def trilinear_attention(self, context, query):
_batch_size, c_seq_len, _dim = context.shape
_batch_size, q_seq_len, _dim = query.shape
context = self.dropout(context)
query = self.dropout(query)
subres0 = torch.matmul(context, self.w4C).expand([-1, -1, q_seq_len])
subres1 = torch.matmul(query, self.w4Q).transpose(1, 2).expand([-1,
c_seq_len, -1])
subres2 = torch.matmul(context * self.w4mlu, query.transpose(1, 2))
res = subres0 + subres1 + subres2
return res
def get_inputs():
return [torch.rand([4, 4, 4]), torch.rand([4, 4, 4]), torch.rand([4, 4]
), torch.rand([4, 4])]
def get_init_inputs():
return [[], {'dim': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import math as tl_math
import torch.nn as nn
import torch.utils.data
import torch.backends.cudnn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_mul_0(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK: tl.
constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr1 + x0, xmask, eviction_policy='evict_last')
tmp2 = tmp0 * tmp1
tl.store(out_ptr0 + x2, tmp2, xmask)
@triton.jit
def triton_poi_fused__softmax_add_mul_rsub_1(in_ptr0, in_ptr1, in_ptr2,
in_ptr3, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr1 + 4 * x1, xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr2 + 4 * x2, xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr3 + 4 * x1, xmask, eviction_policy='evict_last')
tmp11 = tl.load(in_ptr1 + (1 + 4 * x1), xmask, eviction_policy='evict_last'
)
tmp13 = tl.load(in_ptr2 + (1 + 4 * x2), xmask, eviction_policy='evict_last'
)
tmp15 = tl.load(in_ptr3 + (1 + 4 * x1), xmask, eviction_policy='evict_last'
)
tmp20 = tl.load(in_ptr1 + (2 + 4 * x1), xmask, eviction_policy='evict_last'
)
tmp22 = tl.load(in_ptr2 + (2 + 4 * x2), xmask, eviction_policy='evict_last'
)
tmp24 = tl.load(in_ptr3 + (2 + 4 * x1), xmask, eviction_policy='evict_last'
)
tmp29 = tl.load(in_ptr1 + (3 + 4 * x1), xmask, eviction_policy='evict_last'
)
tmp31 = tl.load(in_ptr2 + (3 + 4 * x2), xmask, eviction_policy='evict_last'
)
tmp33 = tl.load(in_ptr3 + (3 + 4 * x1), xmask, eviction_policy='evict_last'
)
tmp2 = tmp0 + tmp1
tmp4 = tmp2 + tmp3
tmp6 = 1.0
tmp7 = tmp6 - tmp5
tmp8 = -1e+30
tmp9 = tmp7 * tmp8
tmp10 = tmp4 + tmp9
tmp12 = tmp0 + tmp11
tmp14 = tmp12 + tmp13
tmp16 = tmp6 - tmp15
tmp17 = tmp16 * tmp8
tmp18 = tmp14 + tmp17
tmp19 = triton_helpers.maximum(tmp10, tmp18)
tmp21 = tmp0 + tmp20
tmp23 = tmp21 + tmp22
tmp25 = tmp6 - tmp24
tmp26 = tmp25 * tmp8
tmp27 = tmp23 + tmp26
tmp28 = triton_helpers.maximum(tmp19, tmp27)
tmp30 = tmp0 + tmp29
tmp32 = tmp30 + tmp31
tmp34 = tmp6 - tmp33
tmp35 = tmp34 * tmp8
tmp36 = tmp32 + tmp35
tmp37 = triton_helpers.maximum(tmp28, tmp36)
tl.store(out_ptr0 + x2, tmp37, xmask)
@triton.jit
def triton_poi_fused__softmax_add_mul_rsub_2(in_ptr0, in_ptr1, in_ptr2,
in_ptr3, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 4
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_ptr0 + 4 * x1, xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + x2, xmask)
tmp3 = tl.load(in_ptr2 + (x0 + 16 * x1), xmask)
tmp5 = tl.load(in_ptr3 + 4 * x1, xmask, eviction_policy='evict_last')
tmp11 = tl.load(in_ptr0 + (1 + 4 * x1), xmask, eviction_policy='evict_last'
)
tmp13 = tl.load(in_ptr2 + (4 + x0 + 16 * x1), xmask)
tmp15 = tl.load(in_ptr3 + (1 + 4 * x1), xmask, eviction_policy='evict_last'
)
tmp20 = tl.load(in_ptr0 + (2 + 4 * x1), xmask, eviction_policy='evict_last'
)
tmp22 = tl.load(in_ptr2 + (8 + x0 + 16 * x1), xmask)
tmp24 = tl.load(in_ptr3 + (2 + 4 * x1), xmask, eviction_policy='evict_last'
)
tmp29 = tl.load(in_ptr0 + (3 + 4 * x1), xmask, eviction_policy='evict_last'
)
tmp31 = tl.load(in_ptr2 + (12 + x0 + 16 * x1), xmask)
tmp33 = tl.load(in_ptr3 + (3 + 4 * x1), xmask, eviction_policy='evict_last'
)
tmp2 = tmp0 + tmp1
tmp4 = tmp2 + tmp3
tmp6 = 1.0
tmp7 = tmp6 - tmp5
tmp8 = -1e+30
tmp9 = tmp7 * tmp8
tmp10 = tmp4 + tmp9
tmp12 = tmp11 + tmp1
tmp14 = tmp12 + tmp13
tmp16 = tmp6 - tmp15
tmp17 = tmp16 * tmp8
tmp18 = tmp14 + tmp17
tmp19 = triton_helpers.maximum(tmp10, tmp18)
tmp21 = tmp20 + tmp1
tmp23 = tmp21 + tmp22
tmp25 = tmp6 - tmp24
tmp26 = tmp25 * tmp8
tmp27 = tmp23 + tmp26
tmp28 = triton_helpers.maximum(tmp19, tmp27)
tmp30 = tmp29 + tmp1
tmp32 = tmp30 + tmp31
tmp34 = tmp6 - tmp33
tmp35 = tmp34 * tmp8
tmp36 = tmp32 + tmp35
tmp37 = triton_helpers.maximum(tmp28, tmp36)
tl.store(out_ptr0 + x2, tmp37, xmask)
@triton.jit
def triton_poi_fused__softmax_add_mul_rsub_3(in_ptr0, in_ptr1, in_ptr2,
in_ptr3, in_ptr4, in_ptr5, in_ptr6, out_ptr0, out_ptr1, xnumel, XBLOCK:
tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex // 4
x0 = xindex % 4
x2 = xindex // 16
x4 = xindex
tmp0 = tl.load(in_ptr0 + x3, xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (x0 + 4 * x2), xmask, eviction_policy='evict_last'
)
tmp3 = tl.load(in_ptr2 + x4, xmask)
tmp5 = tl.load(in_ptr3 + (x0 + 4 * x2), xmask, eviction_policy='evict_last'
)
tmp11 = tl.load(in_ptr4 + x3, xmask, eviction_policy='evict_last')
tmp14 = tl.load(in_ptr5 + x3, xmask, eviction_policy='evict_last')
tmp18 = tl.load(in_ptr6 + (x0 + 4 * x2), xmask, eviction_policy=
'evict_last')
tmp2 = tmp0 + tmp1
tmp4 = tmp2 + tmp3
tmp6 = 1.0
tmp7 = tmp6 - tmp5
tmp8 = -1e+30
tmp9 = tmp7 * tmp8
tmp10 = tmp4 + tmp9
tmp12 = tmp10 - tmp11
tmp13 = tl_math.exp(tmp12)
tmp15 = tmp6 - tmp14
tmp16 = tmp15 * tmp8
tmp17 = tmp4 + tmp16
tmp19 = tmp17 - tmp18
tmp20 = tl_math.exp(tmp19)
tl.store(out_ptr0 + x4, tmp13, xmask)
tl.store(out_ptr1 + x4, tmp20, xmask)
@triton.jit
def triton_poi_fused__softmax_4(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + 4 * x1, xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + 4 * x1), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (2 + 4 * x1), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (3 + 4 * x1), xmask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp5 = tmp3 + tmp4
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tl.store(out_ptr0 + x2, tmp8, xmask)
@triton.jit
def triton_poi_fused__softmax_5(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x0 = xindex % 4
x2 = xindex // 16
tmp0 = tl.load(in_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + (x0 + 16 * x2), xmask, eviction_policy=
'evict_last')
tmp2 = tl.load(in_ptr0 + (4 + x0 + 16 * x2), xmask, eviction_policy=
'evict_last')
tmp4 = tl.load(in_ptr0 + (8 + x0 + 16 * x2), xmask, eviction_policy=
'evict_last')
tmp6 = tl.load(in_ptr0 + (12 + x0 + 16 * x2), xmask, eviction_policy=
'evict_last')
tmp3 = tmp1 + tmp2
tmp5 = tmp3 + tmp4
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tl.store(out_ptr0 + x3, tmp8, xmask)
@triton.jit
def triton_poi_fused_cat_6(in_ptr0, in_ptr1, in_ptr2, out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 16
x1 = xindex // 16
x2 = xindex
tmp0 = x0
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (4 * x1 + x0), tmp4 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp6 = tmp0 >= tmp3
tmp7 = tl.full([1], 8, tl.int64)
tmp8 = tmp0 < tmp7
tmp9 = tmp6 & tmp8
tmp10 = tl.load(in_ptr1 + (4 * x1 + (-4 + x0)), tmp9 & xmask,
eviction_policy='evict_last', other=0.0)
tmp11 = tmp0 >= tmp7
tmp12 = tl.full([1], 12, tl.int64)
tmp13 = tmp0 < tmp12
tmp14 = tmp11 & tmp13
tmp15 = tl.load(in_ptr0 + (4 * x1 + (-8 + x0)), tmp14 & xmask,
eviction_policy='evict_last', other=0.0)
tmp16 = tl.load(in_ptr1 + (4 * x1 + (-8 + x0)), tmp14 & xmask,
eviction_policy='evict_last', other=0.0)
tmp17 = tmp15 * tmp16
tmp18 = tl.full(tmp17.shape, 0.0, tmp17.dtype)
tmp19 = tl.where(tmp14, tmp17, tmp18)
tmp20 = tmp0 >= tmp12
tl.full([1], 16, tl.int64)
tmp23 = tl.load(in_ptr0 + (4 * x1 + (-12 + x0)), tmp20 & xmask,
eviction_policy='evict_last', other=0.0)
tmp24 = tl.load(in_ptr2 + (4 * x1 + (-12 + x0)), tmp20 & xmask,
eviction_policy='evict_last', other=0.0)
tmp25 = tmp23 * tmp24
tmp26 = tl.full(tmp25.shape, 0.0, tmp25.dtype)
tmp27 = tl.where(tmp20, tmp25, tmp26)
tmp28 = tl.where(tmp14, tmp19, tmp27)
tmp29 = tl.where(tmp9, tmp10, tmp28)
tmp30 = tl.where(tmp4, tmp5, tmp29)
tl.store(out_ptr0 + x2, tmp30, xmask)
@triton.jit
def triton_poi_fused_convolution_7(in_ptr0, out_ptr0, ynumel, xnumel,
YBLOCK: tl.constexpr, XBLOCK: tl.constexpr):
ynumel = 64
xnumel = 4
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 16
y1 = yindex // 16
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + 16 * x2 + 64 * y1), xmask & ymask,
eviction_policy='evict_last')
tl.store(out_ptr0 + (x2 + 4 * y3), tmp0, xmask & ymask)
@triton.jit
def triton_poi_fused_convolution_8(in_out_ptr0, in_ptr0, xnumel, XBLOCK: tl
.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = xindex // 4 % 4
tmp0 = tl.load(in_out_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + x1, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(in_out_ptr0 + x3, tmp2, xmask)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7, primals_8, primals_9) = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_2, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_3, (4, 1), (1, 1))
assert_size_stride(primals_4, (4, 1), (1, 1))
assert_size_stride(primals_5, (1, 1, 4), (4, 4, 1))
assert_size_stride(primals_6, (4, 4), (4, 1))
assert_size_stride(primals_7, (4, 4), (4, 1))
assert_size_stride(primals_8, (4, 16, 1), (16, 1, 1))
assert_size_stride(primals_9, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((16, 1), (1, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_1, (16, 4), (4, 1), 0),
primals_3, out=buf0)
del primals_3
buf1 = empty_strided_cuda((16, 1), (1, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_2, (16, 4), (4, 1), 0),
primals_4, out=buf1)
del primals_4
buf2 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_mul_0[grid(64)](primals_1, primals_5, buf2, 64,
XBLOCK=64, num_warps=1, num_stages=1)
del primals_5
buf3 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
extern_kernels.bmm(buf2, reinterpret_tensor(primals_2, (4, 4, 4), (
16, 1, 4), 0), out=buf3)
buf4 = empty_strided_cuda((4, 4, 1), (4, 1, 16), torch.float32)
triton_poi_fused__softmax_add_mul_rsub_1[grid(16)](buf0, buf1, buf3,
primals_6, buf4, 16, XBLOCK=16, num_warps=1, num_stages=1)
buf7 = empty_strided_cuda((4, 1, 4), (4, 16, 1), torch.float32)
triton_poi_fused__softmax_add_mul_rsub_2[grid(16)](buf0, buf1, buf3,
primals_7, buf7, 16, XBLOCK=16, num_warps=1, num_stages=1)
buf5 = buf2
del buf2
buf8 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
triton_poi_fused__softmax_add_mul_rsub_3[grid(64)](buf0, buf1, buf3,
primals_6, buf4, primals_7, buf7, buf5, buf8, 64, XBLOCK=64,
num_warps=1, num_stages=1)
del buf0
del buf1
del buf4
del buf7
del primals_6
del primals_7
buf6 = buf3
del buf3
triton_poi_fused__softmax_4[grid(64)](buf5, buf6, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf9 = buf5
del buf5
triton_poi_fused__softmax_5[grid(64)](buf8, buf9, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf10 = buf8
del buf8
extern_kernels.bmm(buf6, primals_2, out=buf10)
buf11 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
extern_kernels.bmm(buf6, reinterpret_tensor(buf9, (4, 4, 4), (16, 1,
4), 0), out=buf11)
buf12 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
extern_kernels.bmm(buf11, primals_1, out=buf12)
del buf11
buf13 = empty_strided_cuda((4, 4, 16), (64, 16, 1), torch.float32)
triton_poi_fused_cat_6[grid(256)](primals_1, buf10, buf12, buf13,
256, XBLOCK=128, num_warps=4, num_stages=1)
del buf10
del buf12
buf14 = empty_strided_cuda((4, 16, 4), (64, 4, 1), torch.float32)
triton_poi_fused_convolution_7[grid(64, 4)](buf13, buf14, 64, 4,
XBLOCK=4, YBLOCK=32, num_warps=4, num_stages=1)
buf15 = extern_kernels.convolution(buf14, primals_8, stride=(1,),
padding=(0,), dilation=(1,), transposed=False, output_padding=(
0,), groups=1, bias=None)
assert_size_stride(buf15, (4, 4, 4), (16, 4, 1))
del buf14
buf16 = buf15
del buf15
triton_poi_fused_convolution_8[grid(64)](buf16, primals_9, 64,
XBLOCK=64, num_warps=1, num_stages=1)
del primals_9
return reinterpret_tensor(buf16, (4, 4, 4), (16, 1, 4), 0
), primals_1, primals_2, primals_8, buf6, buf9, reinterpret_tensor(
buf13, (4, 16, 4), (64, 1, 16), 0)
def mask_logits(inputs, mask, mask_value=-1e+30):
mask = mask.type(torch.float32)
return inputs + (1.0 - mask) * mask_value
class Conv1D(nn.Module):
def __init__(self, in_dim, out_dim, kernel_size=1, stride=1, padding=0,
bias=True):
super(Conv1D, self).__init__()
self.conv1d = nn.Conv1d(in_channels=in_dim, out_channels=out_dim,
kernel_size=kernel_size, padding=padding, stride=stride, bias=bias)
def forward(self, x):
x = x.transpose(1, 2)
x = self.conv1d(x)
return x.transpose(1, 2)
class CQAttentionNew(nn.Module):
def __init__(self, dim, drop_rate=0.0):
super(CQAttentionNew, self).__init__()
w4C = torch.empty(dim, 1)
w4Q = torch.empty(dim, 1)
w4mlu = torch.empty(1, 1, dim)
nn.init.xavier_uniform_(w4C)
nn.init.xavier_uniform_(w4Q)
nn.init.xavier_uniform_(w4mlu)
self.w4C = nn.Parameter(w4C, requires_grad=True)
self.w4Q = nn.Parameter(w4Q, requires_grad=True)
self.w4mlu = nn.Parameter(w4mlu, requires_grad=True)
self.dropout = nn.Dropout(p=drop_rate)
self.cqa_linear = Conv1D(in_dim=4 * dim, out_dim=dim, kernel_size=1,
stride=1, padding=0, bias=True)
def trilinear_attention(self, context, query):
_batch_size, c_seq_len, _dim = context.shape
_batch_size, q_seq_len, _dim = query.shape
context = self.dropout(context)
query = self.dropout(query)
subres0 = torch.matmul(context, self.w4C).expand([-1, -1, q_seq_len])
subres1 = torch.matmul(query, self.w4Q).transpose(1, 2).expand([-1,
c_seq_len, -1])
subres2 = torch.matmul(context * self.w4mlu, query.transpose(1, 2))
res = subres0 + subres1 + subres2
return res
def forward(self, input_0, input_1, input_2, input_3):
primals_3 = self.w4C
primals_4 = self.w4Q
primals_5 = self.w4mlu
primals_8 = self.cqa_linear.conv1d.weight
primals_9 = self.cqa_linear.conv1d.bias
primals_1 = input_0
primals_2 = input_1
primals_6 = input_2
primals_7 = input_3
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7, primals_8, primals_9])
return output[0]
|
IsaacChanghau/VSLNet
|
CQAttention
| false
| 13,885
|
[
"MIT"
] | 62
|
3793c625f2e251a5f19a0d59f0c83b12e386f808
|
https://github.com/IsaacChanghau/VSLNet/tree/3793c625f2e251a5f19a0d59f0c83b12e386f808
|
FusionLayer
|
import torch
from torch import nn
from torch.nn import init
class FusionLayer(nn.Module):
def __init__(self, nums=6):
super(FusionLayer, self).__init__()
self.weights = nn.Parameter(torch.randn(nums))
self.nums = nums
self._reset_parameters()
def _reset_parameters(self):
init.constant_(self.weights, 1 / self.nums)
def forward(self, x):
for i in range(self.nums):
out = self.weights[i] * x[i] if i == 0 else out + self.weights[i
] * x[i]
return out
def get_inputs():
return [torch.rand([6, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch import nn
from torch.nn import init
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_add_mul_0(in_out_ptr0, in_ptr0, in_ptr1, xnumel,
XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + 0)
tmp1 = tl.broadcast_to(tmp0, [XBLOCK])
tmp2 = tl.load(in_ptr1 + x0, xmask)
tmp4 = tl.load(in_ptr0 + 1)
tmp5 = tl.broadcast_to(tmp4, [XBLOCK])
tmp6 = tl.load(in_ptr1 + (64 + x0), xmask)
tmp9 = tl.load(in_ptr0 + 2)
tmp10 = tl.broadcast_to(tmp9, [XBLOCK])
tmp11 = tl.load(in_ptr1 + (128 + x0), xmask)
tmp14 = tl.load(in_ptr0 + 3)
tmp15 = tl.broadcast_to(tmp14, [XBLOCK])
tmp16 = tl.load(in_ptr1 + (192 + x0), xmask)
tmp19 = tl.load(in_ptr0 + 4)
tmp20 = tl.broadcast_to(tmp19, [XBLOCK])
tmp21 = tl.load(in_ptr1 + (256 + x0), xmask)
tmp24 = tl.load(in_ptr0 + 5)
tmp25 = tl.broadcast_to(tmp24, [XBLOCK])
tmp26 = tl.load(in_ptr1 + (320 + x0), xmask)
tmp3 = tmp1 * tmp2
tmp7 = tmp5 * tmp6
tmp8 = tmp3 + tmp7
tmp12 = tmp10 * tmp11
tmp13 = tmp8 + tmp12
tmp17 = tmp15 * tmp16
tmp18 = tmp13 + tmp17
tmp22 = tmp20 * tmp21
tmp23 = tmp18 + tmp22
tmp27 = tmp25 * tmp26
tmp28 = tmp23 + tmp27
tl.store(in_out_ptr0 + x0, tmp28, xmask)
def call(args):
primals_1, primals_2 = args
args.clear()
assert_size_stride(primals_1, (6,), (1,))
assert_size_stride(primals_2, (6, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
buf1 = buf0
del buf0
get_raw_stream(0)
triton_poi_fused_add_mul_0[grid(64)](buf1, primals_1, primals_2, 64,
XBLOCK=64, num_warps=1, num_stages=1)
del primals_1
return buf1, reinterpret_tensor(primals_2, (4, 4, 4), (16, 4, 1), 0
), reinterpret_tensor(primals_2, (4, 4, 4), (16, 4, 1), 64
), reinterpret_tensor(primals_2, (4, 4, 4), (16, 4, 1), 128
), reinterpret_tensor(primals_2, (4, 4, 4), (16, 4, 1), 192
), reinterpret_tensor(primals_2, (4, 4, 4), (16, 4, 1), 256
), reinterpret_tensor(primals_2, (4, 4, 4), (16, 4, 1), 320)
class FusionLayerNew(nn.Module):
def __init__(self, nums=6):
super(FusionLayerNew, self).__init__()
self.weights = nn.Parameter(torch.randn(nums))
self.nums = nums
self._reset_parameters()
def _reset_parameters(self):
init.constant_(self.weights, 1 / self.nums)
def forward(self, input_0):
primals_1 = self.weights
primals_2 = input_0
output = call([primals_1, primals_2])
return output[0]
|
JasonLin1998/DSS-pytorch
|
FusionLayer
| false
| 13,886
|
[
"MIT"
] | 188
|
f249541bf7e5e479e050b562dd6024d6219f36f4
|
https://github.com/JasonLin1998/DSS-pytorch/tree/f249541bf7e5e479e050b562dd6024d6219f36f4
|
ConvToVector
|
import torch
import torch.nn as nn
import torch.nn.functional as F
class ConvToVector(nn.Module):
def __init__(self, in_channels, padding=1):
super(ConvToVector, self).__init__()
self.in_channels = in_channels
self.conv1 = nn.Conv2d(in_channels, 3, kernel_size=3, padding=padding)
self.conv2 = nn.Conv2d(3, 6, kernel_size=3, padding=padding)
self.conv3 = nn.Conv2d(6, 12, kernel_size=3, padding=padding)
self.conv8 = nn.Conv2d(12, 6, kernel_size=3, padding=0)
self.conv9 = nn.Conv2d(6, 3, kernel_size=3, padding=0)
self.conv10 = nn.Conv2d(3, 1, kernel_size=3, padding=0)
def forward(self, x):
x = F.relu(self.conv1(x))
x = F.relu(self.conv2(x))
x = F.relu(self.conv3(x))
x = F.relu(self.conv8(x))
x = F.relu(self.conv9(x))
x = self.conv10(x)
return x
def get_inputs():
return [torch.rand([4, 4, 64, 64])]
def get_init_inputs():
return [[], {'in_channels': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
@triton.jit
def triton_poi_fused_convolution_relu_0(in_out_ptr0, in_ptr0, xnumel,
XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x3 = xindex
x1 = xindex // 4096 % 3
tmp0 = tl.load(in_out_ptr0 + x3, None)
tmp1 = tl.load(in_ptr0 + x1, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + x3, tmp4, None)
@triton.jit
def triton_poi_fused_convolution_relu_1(in_out_ptr0, in_ptr0, xnumel,
XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x3 = xindex
x1 = xindex // 4096 % 6
tmp0 = tl.load(in_out_ptr0 + x3, None)
tmp1 = tl.load(in_ptr0 + x1, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + x3, tmp4, None)
@triton.jit
def triton_poi_fused_convolution_relu_2(in_out_ptr0, in_ptr0, xnumel,
XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x3 = xindex
x1 = xindex // 4096 % 12
tmp0 = tl.load(in_out_ptr0 + x3, None)
tmp1 = tl.load(in_ptr0 + x1, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + x3, tmp4, None)
@triton.jit
def triton_poi_fused_convolution_relu_3(in_out_ptr0, in_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 92256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = xindex // 3844 % 6
tmp0 = tl.load(in_out_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + x1, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + x3, tmp4, xmask)
@triton.jit
def triton_poi_fused_convolution_relu_4(in_out_ptr0, in_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 43200
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = xindex // 3600 % 3
tmp0 = tl.load(in_out_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + x1, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + x3, tmp4, xmask)
@triton.jit
def triton_poi_fused_convolution_5(in_out_ptr0, in_ptr0, xnumel, XBLOCK: tl
.constexpr):
xnumel = 13456
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_out_ptr0 + x0, xmask)
tmp1 = tl.load(in_ptr0 + 0)
tmp2 = tl.broadcast_to(tmp1, [XBLOCK])
tmp3 = tmp0 + tmp2
tl.store(in_out_ptr0 + x0, tmp3, xmask)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7, primals_8, primals_9, primals_10, primals_11, primals_12,
primals_13) = args
args.clear()
assert_size_stride(primals_1, (3, 4, 3, 3), (36, 9, 3, 1))
assert_size_stride(primals_2, (3,), (1,))
assert_size_stride(primals_3, (4, 4, 64, 64), (16384, 4096, 64, 1))
assert_size_stride(primals_4, (6, 3, 3, 3), (27, 9, 3, 1))
assert_size_stride(primals_5, (6,), (1,))
assert_size_stride(primals_6, (12, 6, 3, 3), (54, 9, 3, 1))
assert_size_stride(primals_7, (12,), (1,))
assert_size_stride(primals_8, (6, 12, 3, 3), (108, 9, 3, 1))
assert_size_stride(primals_9, (6,), (1,))
assert_size_stride(primals_10, (3, 6, 3, 3), (54, 9, 3, 1))
assert_size_stride(primals_11, (3,), (1,))
assert_size_stride(primals_12, (1, 3, 3, 3), (27, 9, 3, 1))
assert_size_stride(primals_13, (1,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(primals_3, primals_1, stride=(1,
1), padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 3, 64, 64), (12288, 4096, 64, 1))
buf1 = buf0
del buf0
get_raw_stream(0)
triton_poi_fused_convolution_relu_0[grid(49152)](buf1, primals_2,
49152, XBLOCK=512, num_warps=4, num_stages=1)
del primals_2
buf2 = extern_kernels.convolution(buf1, primals_4, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf2, (4, 6, 64, 64), (24576, 4096, 64, 1))
buf3 = buf2
del buf2
triton_poi_fused_convolution_relu_1[grid(98304)](buf3, primals_5,
98304, XBLOCK=512, num_warps=8, num_stages=1)
del primals_5
buf4 = extern_kernels.convolution(buf3, primals_6, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf4, (4, 12, 64, 64), (49152, 4096, 64, 1))
buf5 = buf4
del buf4
triton_poi_fused_convolution_relu_2[grid(196608)](buf5, primals_7,
196608, XBLOCK=1024, num_warps=4, num_stages=1)
del primals_7
buf6 = extern_kernels.convolution(buf5, primals_8, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf6, (4, 6, 62, 62), (23064, 3844, 62, 1))
buf7 = buf6
del buf6
triton_poi_fused_convolution_relu_3[grid(92256)](buf7, primals_9,
92256, XBLOCK=1024, num_warps=4, num_stages=1)
del primals_9
buf8 = extern_kernels.convolution(buf7, primals_10, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf8, (4, 3, 60, 60), (10800, 3600, 60, 1))
buf9 = buf8
del buf8
triton_poi_fused_convolution_relu_4[grid(43200)](buf9, primals_11,
43200, XBLOCK=512, num_warps=4, num_stages=1)
del primals_11
buf10 = extern_kernels.convolution(buf9, primals_12, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf10, (4, 1, 58, 58), (3364, 3364, 58, 1))
buf11 = buf10
del buf10
triton_poi_fused_convolution_5[grid(13456)](buf11, primals_13,
13456, XBLOCK=256, num_warps=4, num_stages=1)
del primals_13
return (buf11, primals_1, primals_3, primals_4, primals_6, primals_8,
primals_10, primals_12, buf1, buf3, buf5, buf7, buf9)
class ConvToVectorNew(nn.Module):
def __init__(self, in_channels, padding=1):
super(ConvToVectorNew, self).__init__()
self.in_channels = in_channels
self.conv1 = nn.Conv2d(in_channels, 3, kernel_size=3, padding=padding)
self.conv2 = nn.Conv2d(3, 6, kernel_size=3, padding=padding)
self.conv3 = nn.Conv2d(6, 12, kernel_size=3, padding=padding)
self.conv8 = nn.Conv2d(12, 6, kernel_size=3, padding=0)
self.conv9 = nn.Conv2d(6, 3, kernel_size=3, padding=0)
self.conv10 = nn.Conv2d(3, 1, kernel_size=3, padding=0)
def forward(self, input_0):
primals_1 = self.conv1.weight
primals_2 = self.conv1.bias
primals_4 = self.conv2.weight
primals_5 = self.conv2.bias
primals_6 = self.conv3.weight
primals_7 = self.conv3.bias
primals_8 = self.conv8.weight
primals_9 = self.conv8.bias
primals_10 = self.conv9.weight
primals_11 = self.conv9.bias
primals_12 = self.conv10.weight
primals_13 = self.conv10.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7, primals_8, primals_9,
primals_10, primals_11, primals_12, primals_13])
return output[0]
|
JannerM/spatial-reasoning
|
ConvToVector
| false
| 13,887
|
[
"MIT"
] | 54
|
e163003a33177e41ca02d5feefee3fdfca5ba154
|
https://github.com/JannerM/spatial-reasoning/tree/e163003a33177e41ca02d5feefee3fdfca5ba154
|
MultiHeadAttention
|
import torch
from torch import nn
import torch.nn.functional as F
import torch.utils.data
class MultiHeadAttention(nn.Module):
"""
input:
query --- [N, T_q, query_dim]
key --- [N, T_k, key_dim]
output:
out --- [N, T_q, num_units]
"""
def __init__(self, query_dim, key_dim, num_units, num_heads):
super().__init__()
self.num_units = num_units
self.num_heads = num_heads
self.key_dim = key_dim
self.W_query = nn.Linear(in_features=query_dim, out_features=
num_units, bias=False)
self.W_key = nn.Linear(in_features=key_dim, out_features=num_units,
bias=False)
self.W_value = nn.Linear(in_features=key_dim, out_features=
num_units, bias=False)
def forward(self, query, key):
querys = self.W_query(query)
keys = self.W_key(key)
values = self.W_value(key)
split_size = self.num_units // self.num_heads
querys = torch.stack(torch.split(querys, split_size, dim=2), dim=0)
keys = torch.stack(torch.split(keys, split_size, dim=2), dim=0)
values = torch.stack(torch.split(values, split_size, dim=2), dim=0)
scores = torch.matmul(querys, keys.transpose(2, 3))
scores = scores / self.key_dim ** 0.5
scores = F.softmax(scores, dim=3)
out = torch.matmul(scores, values)
out = torch.cat(torch.split(out, 1, dim=0), dim=3).squeeze(0)
return out
def get_inputs():
return [torch.rand([4, 4, 4]), torch.rand([4, 4, 4])]
def get_init_inputs():
return [[], {'query_dim': 4, 'key_dim': 4, 'num_units': 4, 'num_heads': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import math as tl_math
from torch import nn
import torch.utils.data
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex // 4
x0 = xindex % 4
x1 = xindex // 4 % 4
x2 = xindex // 16
x4 = xindex
tmp0 = x3
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (4 * x0 + 16 * (x1 + 4 * x2)), tmp4 & xmask,
eviction_policy='evict_last', other=0.0)
tmp6 = tmp0 >= tmp3
tmp7 = tl.full([1], 8, tl.int64)
tmp8 = tmp0 < tmp7
tmp9 = tmp6 & tmp8
tmp10 = tl.load(in_ptr0 + (1 + 4 * x0 + 16 * (-4 + x1 + 4 * x2)), tmp9 &
xmask, eviction_policy='evict_last', other=0.0)
tmp11 = tmp0 >= tmp7
tmp12 = tl.full([1], 12, tl.int64)
tmp13 = tmp0 < tmp12
tmp14 = tmp11 & tmp13
tmp15 = tl.load(in_ptr0 + (2 + 4 * x0 + 16 * (-8 + x1 + 4 * x2)), tmp14 &
xmask, eviction_policy='evict_last', other=0.0)
tmp16 = tmp0 >= tmp12
tl.full([1], 16, tl.int64)
tmp19 = tl.load(in_ptr0 + (3 + 4 * x0 + 16 * (-12 + x1 + 4 * x2)),
tmp16 & xmask, eviction_policy='evict_last', other=0.0)
tmp20 = tl.where(tmp14, tmp15, tmp19)
tmp21 = tl.where(tmp9, tmp10, tmp20)
tmp22 = tl.where(tmp4, tmp5, tmp21)
tmp23 = 0.7071067811865476
tmp24 = tmp22 * tmp23
tl.store(out_ptr0 + x4, tmp24, xmask)
@triton.jit
def triton_poi_fused_1(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + 4 * x1, xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + 4 * x1), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (2 + 4 * x1), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (3 + 4 * x1), xmask, eviction_policy='evict_last')
tmp3 = triton_helpers.maximum(tmp1, tmp2)
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp0 - tmp7
tmp9 = tl_math.exp(tmp8)
tl.store(out_ptr0 + x2, tmp9, xmask)
@triton.jit
def triton_poi_fused_2(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 4
x2 = xindex
tmp0 = tl.load(in_ptr0 + 4 * x1, xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (1 + 4 * x1), xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr0 + (2 + 4 * x1), xmask, eviction_policy='evict_last'
)
tmp18 = tl.load(in_ptr0 + (3 + 4 * x1), xmask, eviction_policy='evict_last'
)
tmp25 = tl.load(in_ptr1 + x2, xmask)
tmp26 = tl.load(in_ptr1 + 4 * x1, xmask, eviction_policy='evict_last')
tmp27 = tl.load(in_ptr1 + (1 + 4 * x1), xmask, eviction_policy='evict_last'
)
tmp29 = tl.load(in_ptr1 + (2 + 4 * x1), xmask, eviction_policy='evict_last'
)
tmp31 = tl.load(in_ptr1 + (3 + 4 * x1), xmask, eviction_policy='evict_last'
)
tmp1 = float('-inf')
tmp2 = tmp0 == tmp1
tmp3 = tmp2 == 0
tmp4 = tmp3.to(tl.int64)
tmp5 = tmp4 != 0
tmp7 = tmp6 == tmp1
tmp8 = tmp7 == 0
tmp9 = tmp8.to(tl.int64)
tmp10 = tmp9 != 0
tmp11 = tmp5 | tmp10
tmp13 = tmp12 == tmp1
tmp14 = tmp13 == 0
tmp15 = tmp14.to(tl.int64)
tmp16 = tmp15 != 0
tmp17 = tmp11 | tmp16
tmp19 = tmp18 == tmp1
tmp20 = tmp19 == 0
tmp21 = tmp20.to(tl.int64)
tmp22 = tmp21 != 0
tmp23 = tmp17 | tmp22
tmp24 = tmp23 == 0
tmp28 = tmp26 + tmp27
tmp30 = tmp28 + tmp29
tmp32 = tmp30 + tmp31
tmp33 = tmp25 / tmp32
tmp34 = 0.0
tmp35 = tl.where(tmp24, tmp34, tmp33)
tl.store(out_ptr0 + x2, tmp35, xmask)
@triton.jit
def triton_poi_fused_stack_3(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 4
x0 = xindex % 4
x2 = xindex
tmp0 = x1
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (4 * x0 + 16 * x1), tmp4 & xmask,
eviction_policy='evict_last', other=0.0)
tmp6 = tmp0 >= tmp3
tmp7 = tl.full([1], 8, tl.int64)
tmp8 = tmp0 < tmp7
tmp9 = tmp6 & tmp8
tmp10 = tl.load(in_ptr0 + (1 + 4 * x0 + 16 * (-4 + x1)), tmp9 & xmask,
eviction_policy='evict_last', other=0.0)
tmp11 = tmp0 >= tmp7
tmp12 = tl.full([1], 12, tl.int64)
tmp13 = tmp0 < tmp12
tmp14 = tmp11 & tmp13
tmp15 = tl.load(in_ptr0 + (2 + 4 * x0 + 16 * (-8 + x1)), tmp14 & xmask,
eviction_policy='evict_last', other=0.0)
tmp16 = tmp0 >= tmp12
tl.full([1], 16, tl.int64)
tmp19 = tl.load(in_ptr0 + (3 + 4 * x0 + 16 * (-12 + x1)), tmp16 & xmask,
eviction_policy='evict_last', other=0.0)
tmp20 = tl.where(tmp14, tmp15, tmp19)
tmp21 = tl.where(tmp9, tmp10, tmp20)
tmp22 = tl.where(tmp4, tmp5, tmp21)
tl.store(out_ptr0 + x2, tmp22, xmask)
@triton.jit
def triton_poi_fused_cat_4(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 4
x1 = xindex // 4
x2 = xindex
tmp0 = x0
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 1, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + x1, tmp4 & xmask, eviction_policy='evict_last',
other=0.0)
tmp6 = tmp0 >= tmp3
tmp7 = tl.full([1], 2, tl.int64)
tmp8 = tmp0 < tmp7
tmp9 = tmp6 & tmp8
tmp10 = tl.load(in_ptr0 + (16 + x1), tmp9 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp11 = tmp0 >= tmp7
tmp12 = tl.full([1], 3, tl.int64)
tmp13 = tmp0 < tmp12
tmp14 = tmp11 & tmp13
tmp15 = tl.load(in_ptr0 + (32 + x1), tmp14 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp16 = tmp0 >= tmp12
tl.full([1], 4, tl.int64)
tmp19 = tl.load(in_ptr0 + (48 + x1), tmp16 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp20 = tl.where(tmp14, tmp15, tmp19)
tmp21 = tl.where(tmp9, tmp10, tmp20)
tmp22 = tl.where(tmp4, tmp5, tmp21)
tl.store(out_ptr0 + x2, tmp22, xmask)
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5 = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_3, (4, 4), (4, 1))
assert_size_stride(primals_4, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_5, (4, 4), (4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_2, (16, 4), (4, 1), 0),
reinterpret_tensor(primals_1, (4, 4), (1, 4), 0), out=buf0)
del primals_1
buf1 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_4, (16, 4), (4, 1), 0),
reinterpret_tensor(primals_3, (4, 4), (1, 4), 0), out=buf1)
del primals_3
buf2 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_4, (16, 4), (4, 1), 0),
reinterpret_tensor(primals_5, (4, 4), (1, 4), 0), out=buf2)
del primals_5
buf3 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_0[grid(64)](buf0, buf3, 64, XBLOCK=64, num_warps=1,
num_stages=1)
buf4 = reinterpret_tensor(buf0, (4, 4, 1, 4), (16, 4, 4, 1), 0)
del buf0
triton_poi_fused_0[grid(64)](buf1, buf4, 64, XBLOCK=64, num_warps=1,
num_stages=1)
buf5 = empty_strided_cuda((16, 4, 4), (16, 4, 1), torch.float32)
extern_kernels.bmm(reinterpret_tensor(buf3, (16, 4, 1), (4, 1, 0),
0), reinterpret_tensor(buf4, (16, 1, 4), (4, 0, 1), 0), out=buf5)
buf6 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused_1[grid(256)](buf5, buf6, 256, XBLOCK=128,
num_warps=4, num_stages=1)
buf7 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused_2[grid(256)](buf5, buf6, buf7, 256, XBLOCK=256,
num_warps=4, num_stages=1)
del buf5
del buf6
buf8 = reinterpret_tensor(buf1, (16, 4, 1), (4, 1, 1), 0)
del buf1
triton_poi_fused_stack_3[grid(64)](buf2, buf8, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf9 = reinterpret_tensor(buf2, (16, 4, 1), (4, 1, 1), 0)
del buf2
extern_kernels.bmm(reinterpret_tensor(buf7, (16, 4, 4), (16, 4, 1),
0), reinterpret_tensor(buf8, (16, 4, 1), (4, 1, 0), 0), out=buf9)
buf10 = empty_strided_cuda((1, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused_cat_4[grid(64)](buf9, buf10, 64, XBLOCK=64,
num_warps=1, num_stages=1)
del buf9
return reinterpret_tensor(buf10, (4, 4, 4), (16, 4, 1), 0
), reinterpret_tensor(primals_2, (16, 4), (4, 1), 0
), reinterpret_tensor(primals_4, (16, 4), (4, 1), 0
), buf7, reinterpret_tensor(buf8, (16, 1, 4), (4, 1, 1), 0
), reinterpret_tensor(buf3, (16, 1, 4), (4, 1, 1), 0
), reinterpret_tensor(buf4, (16, 4, 1), (4, 1, 4), 0)
class MultiHeadAttentionNew(nn.Module):
"""
input:
query --- [N, T_q, query_dim]
key --- [N, T_k, key_dim]
output:
out --- [N, T_q, num_units]
"""
def __init__(self, query_dim, key_dim, num_units, num_heads):
super().__init__()
self.num_units = num_units
self.num_heads = num_heads
self.key_dim = key_dim
self.W_query = nn.Linear(in_features=query_dim, out_features=
num_units, bias=False)
self.W_key = nn.Linear(in_features=key_dim, out_features=num_units,
bias=False)
self.W_value = nn.Linear(in_features=key_dim, out_features=
num_units, bias=False)
def forward(self, input_0, input_1):
primals_1 = self.W_query.weight
primals_3 = self.W_key.weight
primals_5 = self.W_value.weight
primals_2 = input_0
primals_4 = input_1
output = call([primals_1, primals_2, primals_3, primals_4, primals_5])
return output[0]
|
Jesse3692/ttskit
|
MultiHeadAttention
| false
| 13,888
|
[
"MIT"
] | 151
|
aa424cf46f5fbe67dc06e67d00c1d46c31a9974b
|
https://github.com/Jesse3692/ttskit/tree/aa424cf46f5fbe67dc06e67d00c1d46c31a9974b
|
DacBlock
|
import torch
from torch import nn
class DacBlock(nn.Module):
def __init__(self, channel):
super(DacBlock, self).__init__()
self.dilate1 = nn.Conv2d(channel, channel, kernel_size=3, dilation=
1, padding=1)
self.dilate2 = nn.Conv2d(channel, channel, kernel_size=3, dilation=
3, padding=3)
self.dilate3 = nn.Conv2d(channel, channel, kernel_size=3, dilation=
5, padding=5)
self.conv1x1 = nn.Conv2d(channel, channel, kernel_size=1, dilation=
1, padding=0)
self.relu = nn.ReLU(inplace=True)
for m in self.modules():
if isinstance(m, nn.Conv2d) or isinstance(m, nn.ConvTranspose2d):
if m.bias is not None:
m.bias.data.zero_()
def forward(self, x):
dilate1_out = self.relu(self.dilate1(x))
dilate2_out = self.relu(self.conv1x1(self.dilate2(x)))
dilate3_out = self.relu(self.conv1x1(self.dilate2(self.dilate1(x))))
dilate4_out = self.relu(self.conv1x1(self.dilate3(self.dilate2(self
.dilate1(x)))))
out = x + dilate1_out + dilate2_out + dilate3_out + dilate4_out
return out
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'channel': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch import nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_convolution_0(in_out_ptr0, in_ptr0, xnumel, XBLOCK: tl
.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = xindex // 16 % 4
tmp0 = tl.load(in_out_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + x1, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(in_out_ptr0 + x3, tmp2, xmask)
@triton.jit
def triton_poi_fused_convolution_1(in_out_ptr0, in_out_ptr1, in_ptr0,
xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = xindex // 16 % 4
tmp0 = tl.load(in_out_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + x1, xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_out_ptr1 + x3, xmask)
tmp2 = tmp0 + tmp1
tmp4 = tmp3 + tmp1
tl.store(in_out_ptr0 + x3, tmp2, xmask)
tl.store(in_out_ptr1 + x3, tmp4, xmask)
@triton.jit
def triton_poi_fused_add_convolution_relu_threshold_backward_2(in_ptr0,
in_ptr1, in_ptr2, in_ptr3, in_ptr4, in_ptr5, out_ptr0, out_ptr1,
out_ptr2, out_ptr3, xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = xindex // 16 % 4
tmp0 = tl.load(in_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr1 + x3, xmask)
tmp5 = tl.load(in_ptr2 + x3, xmask)
tmp6 = tl.load(in_ptr3 + x1, xmask, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr4 + x3, xmask)
tmp14 = tl.load(in_ptr5 + x3, xmask)
tmp2 = tl.full([1], 0, tl.int32)
tmp3 = triton_helpers.maximum(tmp2, tmp1)
tmp4 = tmp0 + tmp3
tmp7 = tmp5 + tmp6
tmp8 = triton_helpers.maximum(tmp2, tmp7)
tmp9 = tmp4 + tmp8
tmp11 = tmp10 + tmp6
tmp12 = triton_helpers.maximum(tmp2, tmp11)
tmp13 = tmp9 + tmp12
tmp15 = tmp14 + tmp6
tmp16 = triton_helpers.maximum(tmp2, tmp15)
tmp17 = tmp13 + tmp16
tmp18 = 0.0
tmp19 = tmp16 <= tmp18
tmp20 = tmp12 <= tmp18
tmp21 = tmp8 <= tmp18
tl.store(out_ptr0 + x3, tmp17, xmask)
tl.store(out_ptr1 + x3, tmp19, xmask)
tl.store(out_ptr2 + x3, tmp20, xmask)
tl.store(out_ptr3 + x3, tmp21, xmask)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7, primals_8, primals_9) = args
args.clear()
assert_size_stride(primals_1, (4, 4, 3, 3), (36, 9, 3, 1))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (4, 4, 3, 3), (36, 9, 3, 1))
assert_size_stride(primals_5, (4,), (1,))
assert_size_stride(primals_6, (4, 4, 1, 1), (4, 1, 1, 1))
assert_size_stride(primals_7, (4,), (1,))
assert_size_stride(primals_8, (4, 4, 3, 3), (36, 9, 3, 1))
assert_size_stride(primals_9, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(primals_3, primals_1, stride=(1,
1), padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 4, 4, 4), (64, 16, 4, 1))
buf1 = buf0
del buf0
get_raw_stream(0)
triton_poi_fused_convolution_0[grid(256)](buf1, primals_2, 256,
XBLOCK=256, num_warps=4, num_stages=1)
del primals_2
buf2 = extern_kernels.convolution(primals_3, primals_4, stride=(1,
1), padding=(3, 3), dilation=(3, 3), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf2, (4, 4, 4, 4), (64, 16, 4, 1))
buf5 = extern_kernels.convolution(buf1, primals_4, stride=(1, 1),
padding=(3, 3), dilation=(3, 3), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf5, (4, 4, 4, 4), (64, 16, 4, 1))
buf3 = buf2
del buf2
buf6 = buf5
del buf5
triton_poi_fused_convolution_1[grid(256)](buf3, buf6, primals_5,
256, XBLOCK=128, num_warps=4, num_stages=1)
del primals_5
buf4 = extern_kernels.convolution(buf3, primals_6, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf4, (4, 4, 4, 4), (64, 16, 4, 1))
buf7 = extern_kernels.convolution(buf6, primals_6, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf7, (4, 4, 4, 4), (64, 16, 4, 1))
buf8 = extern_kernels.convolution(buf6, primals_8, stride=(1, 1),
padding=(5, 5), dilation=(5, 5), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf8, (4, 4, 4, 4), (64, 16, 4, 1))
buf9 = buf8
del buf8
triton_poi_fused_convolution_0[grid(256)](buf9, primals_9, 256,
XBLOCK=256, num_warps=4, num_stages=1)
del primals_9
buf10 = extern_kernels.convolution(buf9, primals_6, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf10, (4, 4, 4, 4), (64, 16, 4, 1))
buf11 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
buf12 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
buf13 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
buf14 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
triton_poi_fused_add_convolution_relu_threshold_backward_2[grid(256)](
primals_3, buf1, buf4, primals_7, buf7, buf10, buf11, buf12,
buf13, buf14, 256, XBLOCK=128, num_warps=4, num_stages=1)
del buf10
del buf4
del buf7
del primals_7
return (buf11, primals_1, primals_3, primals_4, primals_6, primals_8,
buf1, buf3, buf6, buf9, buf12, buf13, buf14)
class DacBlockNew(nn.Module):
def __init__(self, channel):
super(DacBlockNew, self).__init__()
self.dilate1 = nn.Conv2d(channel, channel, kernel_size=3, dilation=
1, padding=1)
self.dilate2 = nn.Conv2d(channel, channel, kernel_size=3, dilation=
3, padding=3)
self.dilate3 = nn.Conv2d(channel, channel, kernel_size=3, dilation=
5, padding=5)
self.conv1x1 = nn.Conv2d(channel, channel, kernel_size=1, dilation=
1, padding=0)
self.relu = nn.ReLU(inplace=True)
for m in self.modules():
if isinstance(m, nn.Conv2d) or isinstance(m, nn.ConvTranspose2d):
if m.bias is not None:
m.bias.data.zero_()
def forward(self, input_0):
primals_1 = self.dilate1.weight
primals_2 = self.dilate1.bias
primals_4 = self.dilate2.weight
primals_5 = self.dilate2.bias
primals_8 = self.dilate3.weight
primals_7 = self.dilate3.bias
primals_6 = self.conv1x1.weight
primals_9 = self.conv1x1.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7, primals_8, primals_9])
return output[0]
|
JiYuanFeng/MCTrans
|
DacBlock
| false
| 13,889
|
[
"Apache-2.0"
] | 84
|
9b8b5677eef584b423d5e1630680a4b667cbe823
|
https://github.com/JiYuanFeng/MCTrans/tree/9b8b5677eef584b423d5e1630680a4b667cbe823
|
MultiHeadAttentionBlock
|
import math
import torch
import torch.nn as nn
import torch.utils.data
import torch.backends.cudnn
def mask_logits(inputs, mask, mask_value=-1e+30):
mask = mask.type(torch.float32)
return inputs + (1.0 - mask) * mask_value
class Conv1D(nn.Module):
def __init__(self, in_dim, out_dim, kernel_size=1, stride=1, padding=0,
bias=True):
super(Conv1D, self).__init__()
self.conv1d = nn.Conv1d(in_channels=in_dim, out_channels=out_dim,
kernel_size=kernel_size, padding=padding, stride=stride, bias=bias)
def forward(self, x):
x = x.transpose(1, 2)
x = self.conv1d(x)
return x.transpose(1, 2)
class MultiHeadAttentionBlock(nn.Module):
def __init__(self, dim, num_heads, drop_rate):
super(MultiHeadAttentionBlock, self).__init__()
assert dim % num_heads == 0, 'The channels (%d) is not a multiple of attention heads (%d)' % (
dim, num_heads)
self.head_size, self.num_heads, self.dim = int(dim / num_heads
), num_heads, dim
self.dropout = nn.Dropout(p=drop_rate)
self.query = Conv1D(in_dim=dim, out_dim=dim, kernel_size=1, stride=
1, padding=0, bias=True)
self.key = Conv1D(in_dim=dim, out_dim=dim, kernel_size=1, stride=1,
padding=0, bias=True)
self.value = Conv1D(in_dim=dim, out_dim=dim, kernel_size=1, stride=
1, padding=0, bias=True)
self.layer_norm1 = nn.LayerNorm(dim, eps=1e-06)
self.layer_norm2 = nn.LayerNorm(dim, eps=1e-06)
self.out_layer = Conv1D(in_dim=dim, out_dim=dim, kernel_size=1,
stride=1, padding=0, bias=True)
def transpose_for_scores(self, x):
new_x_shape = x.size()[:-1] + (self.num_heads, self.head_size)
x = x.view(*new_x_shape)
return x.permute(0, 2, 1, 3)
@staticmethod
def combine_last_two_dim(x):
old_shape = list(x.size())
new_shape = old_shape[:-2] + [old_shape[-2] * old_shape[-1]]
return x.reshape(shape=new_shape)
def forward(self, x, mask=None):
output = self.layer_norm1(x)
output = self.dropout(output)
query = self.transpose_for_scores(self.query(output))
key = self.transpose_for_scores(self.key(output))
value = self.transpose_for_scores(self.value(output))
attention_scores = torch.matmul(query, key.transpose(-1, -2))
attention_scores = attention_scores / math.sqrt(self.head_size)
if mask is not None:
mask = mask.unsqueeze(1).unsqueeze(2)
attention_scores = mask_logits(attention_scores, mask)
attention_probs = nn.Softmax(dim=-1)(attention_scores)
attention_probs = self.dropout(attention_probs)
value = torch.matmul(attention_probs, value)
value = self.combine_last_two_dim(value.permute(0, 2, 1, 3))
output = self.dropout(value)
residual = output + x
output = self.layer_norm2(residual)
output = self.dropout(output)
output = self.out_layer(output)
output = self.dropout(output) + residual
return output
def get_inputs():
return [torch.rand([4, 4, 4])]
def get_init_inputs():
return [[], {'dim': 4, 'num_heads': 4, 'drop_rate': 0.5}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
import torch.nn as nn
import torch.utils.data
import torch.backends.cudnn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_native_layer_norm_0(in_ptr0, out_ptr0, out_ptr1,
xnumel, XBLOCK: tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + 4 * x0, xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr0 + (1 + 4 * x0), xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (2 + 4 * x0), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (3 + 4 * x0), xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp4 = tmp2 + tmp3
tmp6 = tmp4 + tmp5
tmp7 = 4.0
tmp8 = tmp6 / tmp7
tmp9 = tmp0 - tmp8
tmp10 = tmp9 * tmp9
tmp11 = tmp1 - tmp8
tmp12 = tmp11 * tmp11
tmp13 = tmp10 + tmp12
tmp14 = tmp3 - tmp8
tmp15 = tmp14 * tmp14
tmp16 = tmp13 + tmp15
tmp17 = tmp5 - tmp8
tmp18 = tmp17 * tmp17
tmp19 = tmp16 + tmp18
tmp20 = tmp19 / tmp7
tmp21 = 1e-06
tmp22 = tmp20 + tmp21
tmp23 = libdevice.rsqrt(tmp22)
tl.store(out_ptr0 + x0, tmp8, xmask)
tl.store(out_ptr1 + x0, tmp23, xmask)
@triton.jit
def triton_poi_fused_native_layer_norm_1(in_ptr0, in_ptr1, in_ptr2, in_ptr3,
in_ptr4, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
x0 = xindex % 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr1 + x1, xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr2 + x1, xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr3 + x0, xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr4 + x0, xmask, eviction_policy='evict_last')
tmp2 = tmp0 - tmp1
tmp4 = tmp2 * tmp3
tmp6 = tmp4 * tmp5
tmp8 = tmp6 + tmp7
tl.store(out_ptr0 + x2, tmp8, xmask)
@triton.jit
def triton_poi_fused_convolution_2(in_ptr0, out_ptr0, out_ptr1, out_ptr2,
ynumel, xnumel, YBLOCK: tl.constexpr, XBLOCK: tl.constexpr):
ynumel = 16
xnumel = 4
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 4
y1 = yindex // 4
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + 4 * x2 + 16 * y1), xmask & ymask,
eviction_policy='evict_last')
tl.store(out_ptr0 + (x2 + 4 * y3), tmp0, xmask & ymask)
tl.store(out_ptr1 + (x2 + 4 * y3), tmp0, xmask & ymask)
tl.store(out_ptr2 + (x2 + 4 * y3), tmp0, xmask & ymask)
@triton.jit
def triton_poi_fused_3(in_out_ptr0, in_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = xindex // 4 % 4
tmp0 = tl.load(in_out_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + x1, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = 1.0
tmp4 = tmp2 * tmp3
tl.store(in_out_ptr0 + x3, tmp4, xmask)
@triton.jit
def triton_poi_fused_4(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + 4 * x1, xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + 4 * x1), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (2 + 4 * x1), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (3 + 4 * x1), xmask, eviction_policy='evict_last')
tmp3 = triton_helpers.maximum(tmp1, tmp2)
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp0 - tmp7
tmp9 = tl_math.exp(tmp8)
tl.store(out_ptr0 + x2, tmp9, xmask)
@triton.jit
def triton_poi_fused_5(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 4
x2 = xindex
tmp0 = tl.load(in_ptr0 + 4 * x1, xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (1 + 4 * x1), xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr0 + (2 + 4 * x1), xmask, eviction_policy='evict_last'
)
tmp18 = tl.load(in_ptr0 + (3 + 4 * x1), xmask, eviction_policy='evict_last'
)
tmp25 = tl.load(in_ptr1 + x2, xmask)
tmp26 = tl.load(in_ptr1 + 4 * x1, xmask, eviction_policy='evict_last')
tmp27 = tl.load(in_ptr1 + (1 + 4 * x1), xmask, eviction_policy='evict_last'
)
tmp29 = tl.load(in_ptr1 + (2 + 4 * x1), xmask, eviction_policy='evict_last'
)
tmp31 = tl.load(in_ptr1 + (3 + 4 * x1), xmask, eviction_policy='evict_last'
)
tmp1 = float('-inf')
tmp2 = tmp0 == tmp1
tmp3 = tmp2 == 0
tmp4 = tmp3.to(tl.int64)
tmp5 = tmp4 != 0
tmp7 = tmp6 == tmp1
tmp8 = tmp7 == 0
tmp9 = tmp8.to(tl.int64)
tmp10 = tmp9 != 0
tmp11 = tmp5 | tmp10
tmp13 = tmp12 == tmp1
tmp14 = tmp13 == 0
tmp15 = tmp14.to(tl.int64)
tmp16 = tmp15 != 0
tmp17 = tmp11 | tmp16
tmp19 = tmp18 == tmp1
tmp20 = tmp19 == 0
tmp21 = tmp20.to(tl.int64)
tmp22 = tmp21 != 0
tmp23 = tmp17 | tmp22
tmp24 = tmp23 == 0
tmp28 = tmp26 + tmp27
tmp30 = tmp28 + tmp29
tmp32 = tmp30 + tmp31
tmp33 = tmp25 / tmp32
tmp34 = 0.0
tmp35 = tl.where(tmp24, tmp34, tmp33)
tl.store(out_ptr0 + x2, tmp35, xmask)
@triton.jit
def triton_poi_fused_convolution_6(in_out_ptr0, in_ptr0, xnumel, XBLOCK: tl
.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = xindex // 4 % 4
tmp0 = tl.load(in_out_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + x1, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(in_out_ptr0 + x3, tmp2, xmask)
@triton.jit
def triton_poi_fused_add_native_layer_norm_7(in_ptr0, in_ptr1, out_ptr0,
out_ptr1, xnumel, XBLOCK: tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 4
x1 = xindex // 4
x2 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + 16 * x1), xmask)
tmp1 = tl.load(in_ptr1 + 4 * x2, xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (4 + x0 + 16 * x1), xmask)
tmp4 = tl.load(in_ptr1 + (1 + 4 * x2), xmask, eviction_policy='evict_last')
tmp7 = tl.load(in_ptr0 + (8 + x0 + 16 * x1), xmask)
tmp8 = tl.load(in_ptr1 + (2 + 4 * x2), xmask, eviction_policy='evict_last')
tmp11 = tl.load(in_ptr0 + (12 + x0 + 16 * x1), xmask)
tmp12 = tl.load(in_ptr1 + (3 + 4 * x2), xmask, eviction_policy='evict_last'
)
tmp2 = tmp0 + tmp1
tmp5 = tmp3 + tmp4
tmp6 = tmp2 + tmp5
tmp9 = tmp7 + tmp8
tmp10 = tmp6 + tmp9
tmp13 = tmp11 + tmp12
tmp14 = tmp10 + tmp13
tmp15 = 4.0
tmp16 = tmp14 / tmp15
tmp17 = tmp2 - tmp16
tmp18 = tmp17 * tmp17
tmp19 = tmp5 - tmp16
tmp20 = tmp19 * tmp19
tmp21 = tmp18 + tmp20
tmp22 = tmp9 - tmp16
tmp23 = tmp22 * tmp22
tmp24 = tmp21 + tmp23
tmp25 = tmp13 - tmp16
tmp26 = tmp25 * tmp25
tmp27 = tmp24 + tmp26
tmp28 = tmp27 / tmp15
tl.store(out_ptr0 + x2, tmp16, xmask)
tl.store(out_ptr1 + x2, tmp28, xmask)
@triton.jit
def triton_poi_fused_add_native_layer_norm_8(in_ptr0, in_ptr1, in_ptr2,
in_ptr3, in_ptr4, in_ptr5, out_ptr0, ynumel, xnumel, YBLOCK: tl.
constexpr, XBLOCK: tl.constexpr):
ynumel = 16
xnumel = 4
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 4
y1 = yindex // 4
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + 4 * x2 + 16 * y1), xmask & ymask,
eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (x2 + 4 * y3), xmask & ymask, eviction_policy=
'evict_last')
tmp3 = tl.load(in_ptr2 + y3, ymask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr3 + y3, ymask, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr4 + x2, xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr5 + x2, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp4 = tmp2 - tmp3
tmp6 = 1e-06
tmp7 = tmp5 + tmp6
tmp8 = libdevice.rsqrt(tmp7)
tmp9 = tmp4 * tmp8
tmp11 = tmp9 * tmp10
tmp13 = tmp11 + tmp12
tl.store(out_ptr0 + (x2 + 4 * y3), tmp13, xmask & ymask)
@triton.jit
def triton_poi_fused_convolution_9(in_ptr0, out_ptr0, ynumel, xnumel,
YBLOCK: tl.constexpr, XBLOCK: tl.constexpr):
ynumel = 16
xnumel = 4
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 4
y1 = yindex // 4
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + 4 * x2 + 16 * y1), xmask & ymask,
eviction_policy='evict_last')
tl.store(out_ptr0 + (x2 + 4 * y3), tmp0, xmask & ymask)
@triton.jit
def triton_poi_fused_add_10(in_out_ptr0, in_ptr0, in_ptr1, in_ptr2, ynumel,
xnumel, YBLOCK: tl.constexpr, XBLOCK: tl.constexpr):
ynumel = 16
xnumel = 4
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y3 = yindex
y0 = yindex % 4
y1 = yindex // 4
tmp0 = tl.load(in_out_ptr0 + (x2 + 4 * y3), xmask & ymask,
eviction_policy='evict_last')
tmp1 = tl.load(in_ptr0 + y0, ymask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr1 + (x2 + 4 * y3), xmask & ymask, eviction_policy=
'evict_last')
tmp4 = tl.load(in_ptr2 + (y0 + 4 * x2 + 16 * y1), xmask & ymask,
eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp5 = tmp3 + tmp4
tmp6 = tmp2 + tmp5
tl.debug_barrier()
tl.store(in_out_ptr0 + (x2 + 4 * y3), tmp6, xmask & ymask)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7, primals_8, primals_9, primals_10, primals_11, primals_12,
primals_13) = args
args.clear()
assert_size_stride(primals_1, (4,), (1,))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_4, (4, 4, 1), (4, 1, 1))
assert_size_stride(primals_5, (4,), (1,))
assert_size_stride(primals_6, (4, 4, 1), (4, 1, 1))
assert_size_stride(primals_7, (4,), (1,))
assert_size_stride(primals_8, (4, 4, 1), (4, 1, 1))
assert_size_stride(primals_9, (4,), (1,))
assert_size_stride(primals_10, (4,), (1,))
assert_size_stride(primals_11, (4,), (1,))
assert_size_stride(primals_12, (4, 4, 1), (4, 1, 1))
assert_size_stride(primals_13, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 1), (4, 1, 16), torch.float32)
buf1 = empty_strided_cuda((4, 4, 1), (4, 1, 16), torch.float32)
get_raw_stream(0)
triton_poi_fused_native_layer_norm_0[grid(16)](primals_3, buf0,
buf1, 16, XBLOCK=16, num_warps=1, num_stages=1)
buf2 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
triton_poi_fused_native_layer_norm_1[grid(64)](primals_3, buf0,
buf1, primals_1, primals_2, buf2, 64, XBLOCK=64, num_warps=1,
num_stages=1)
del primals_1
del primals_2
buf3 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
buf5 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
buf7 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
triton_poi_fused_convolution_2[grid(16, 4)](buf2, buf3, buf5, buf7,
16, 4, XBLOCK=4, YBLOCK=8, num_warps=1, num_stages=1)
buf4 = extern_kernels.convolution(buf3, primals_4, stride=(1,),
padding=(0,), dilation=(1,), transposed=False, output_padding=(
0,), groups=1, bias=None)
assert_size_stride(buf4, (4, 4, 4), (16, 4, 1))
buf6 = extern_kernels.convolution(buf5, primals_6, stride=(1,),
padding=(0,), dilation=(1,), transposed=False, output_padding=(
0,), groups=1, bias=None)
assert_size_stride(buf6, (4, 4, 4), (16, 4, 1))
buf8 = extern_kernels.convolution(buf7, primals_8, stride=(1,),
padding=(0,), dilation=(1,), transposed=False, output_padding=(
0,), groups=1, bias=None)
assert_size_stride(buf8, (4, 4, 4), (16, 4, 1))
buf9 = reinterpret_tensor(buf4, (4, 4, 4, 1), (16, 4, 1, 1), 0)
del buf4
triton_poi_fused_3[grid(64)](buf9, primals_5, 64, XBLOCK=64,
num_warps=1, num_stages=1)
del primals_5
buf10 = reinterpret_tensor(buf6, (4, 4, 1, 4), (16, 4, 4, 1), 0)
del buf6
triton_poi_fused_3[grid(64)](buf10, primals_7, 64, XBLOCK=64,
num_warps=1, num_stages=1)
del primals_7
buf11 = empty_strided_cuda((16, 4, 4), (16, 4, 1), torch.float32)
extern_kernels.bmm(reinterpret_tensor(buf9, (16, 4, 1), (4, 1, 0),
0), reinterpret_tensor(buf10, (16, 1, 4), (4, 0, 1), 0), out=buf11)
buf12 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused_4[grid(256)](buf11, buf12, 256, XBLOCK=128,
num_warps=4, num_stages=1)
buf13 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused_5[grid(256)](buf11, buf12, buf13, 256, XBLOCK=128,
num_warps=4, num_stages=1)
del buf11
del buf12
buf14 = buf8
del buf8
triton_poi_fused_convolution_6[grid(64)](buf14, primals_9, 64,
XBLOCK=64, num_warps=1, num_stages=1)
del primals_9
buf15 = reinterpret_tensor(buf7, (16, 4, 1), (4, 1, 1), 0)
del buf7
extern_kernels.bmm(reinterpret_tensor(buf13, (16, 4, 4), (16, 4, 1),
0), reinterpret_tensor(buf14, (16, 4, 1), (4, 1, 0), 0), out=buf15)
buf16 = buf1
del buf1
buf17 = buf0
del buf0
triton_poi_fused_add_native_layer_norm_7[grid(16)](buf15, primals_3,
buf16, buf17, 16, XBLOCK=16, num_warps=1, num_stages=1)
buf18 = buf5
del buf5
triton_poi_fused_add_native_layer_norm_8[grid(16, 4)](buf15,
primals_3, buf16, buf17, primals_10, primals_11, buf18, 16, 4,
XBLOCK=2, YBLOCK=16, num_warps=1, num_stages=1)
del buf16
del buf17
del primals_11
buf19 = buf3
del buf3
triton_poi_fused_convolution_9[grid(16, 4)](buf18, buf19, 16, 4,
XBLOCK=4, YBLOCK=16, num_warps=1, num_stages=1)
buf20 = extern_kernels.convolution(buf19, primals_12, stride=(1,),
padding=(0,), dilation=(1,), transposed=False, output_padding=(
0,), groups=1, bias=None)
assert_size_stride(buf20, (4, 4, 4), (16, 4, 1))
del buf19
buf21 = reinterpret_tensor(buf20, (4, 4, 4), (16, 1, 4), 0)
del buf20
triton_poi_fused_add_10[grid(16, 4)](buf21, primals_13, buf15,
primals_3, 16, 4, XBLOCK=4, YBLOCK=16, num_warps=1, num_stages=1)
del primals_13
return (buf21, primals_3, primals_4, primals_6, primals_8, primals_10,
primals_12, reinterpret_tensor(buf2, (4, 4, 4), (16, 1, 4), 0),
buf13, buf15, reinterpret_tensor(buf14, (16, 1, 4), (4, 4, 1), 0),
reinterpret_tensor(buf9, (16, 1, 4), (4, 1, 1), 0),
reinterpret_tensor(buf10, (16, 4, 1), (4, 1, 4), 0),
reinterpret_tensor(buf18, (4, 4, 4), (16, 1, 4), 0))
def mask_logits(inputs, mask, mask_value=-1e+30):
mask = mask.type(torch.float32)
return inputs + (1.0 - mask) * mask_value
class Conv1D(nn.Module):
def __init__(self, in_dim, out_dim, kernel_size=1, stride=1, padding=0,
bias=True):
super(Conv1D, self).__init__()
self.conv1d = nn.Conv1d(in_channels=in_dim, out_channels=out_dim,
kernel_size=kernel_size, padding=padding, stride=stride, bias=bias)
def forward(self, x):
x = x.transpose(1, 2)
x = self.conv1d(x)
return x.transpose(1, 2)
class MultiHeadAttentionBlockNew(nn.Module):
def __init__(self, dim, num_heads, drop_rate):
super(MultiHeadAttentionBlockNew, self).__init__()
assert dim % num_heads == 0, 'The channels (%d) is not a multiple of attention heads (%d)' % (
dim, num_heads)
self.head_size, self.num_heads, self.dim = int(dim / num_heads
), num_heads, dim
self.dropout = nn.Dropout(p=drop_rate)
self.query = Conv1D(in_dim=dim, out_dim=dim, kernel_size=1, stride=
1, padding=0, bias=True)
self.key = Conv1D(in_dim=dim, out_dim=dim, kernel_size=1, stride=1,
padding=0, bias=True)
self.value = Conv1D(in_dim=dim, out_dim=dim, kernel_size=1, stride=
1, padding=0, bias=True)
self.layer_norm1 = nn.LayerNorm(dim, eps=1e-06)
self.layer_norm2 = nn.LayerNorm(dim, eps=1e-06)
self.out_layer = Conv1D(in_dim=dim, out_dim=dim, kernel_size=1,
stride=1, padding=0, bias=True)
def transpose_for_scores(self, x):
new_x_shape = x.size()[:-1] + (self.num_heads, self.head_size)
x = x.view(*new_x_shape)
return x.permute(0, 2, 1, 3)
@staticmethod
def combine_last_two_dim(x):
old_shape = list(x.size())
new_shape = old_shape[:-2] + [old_shape[-2] * old_shape[-1]]
return x.reshape(shape=new_shape)
def forward(self, input_0):
primals_4 = self.query.conv1d.weight
primals_1 = self.query.conv1d.bias
primals_6 = self.key.conv1d.weight
primals_2 = self.key.conv1d.bias
primals_8 = self.value.conv1d.weight
primals_5 = self.value.conv1d.bias
primals_7 = self.layer_norm1.weight
primals_9 = self.layer_norm1.bias
primals_10 = self.layer_norm2.weight
primals_11 = self.layer_norm2.bias
primals_12 = self.out_layer.conv1d.weight
primals_13 = self.out_layer.conv1d.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7, primals_8, primals_9,
primals_10, primals_11, primals_12, primals_13])
return output[0]
|
IsaacChanghau/VSLNet
|
MultiHeadAttentionBlock
| false
| 13,890
|
[
"MIT"
] | 62
|
3793c625f2e251a5f19a0d59f0c83b12e386f808
|
https://github.com/IsaacChanghau/VSLNet/tree/3793c625f2e251a5f19a0d59f0c83b12e386f808
|
WassersteinLoss
|
from torch.nn import Module
import functools
import torch
import torch.utils.data
import torch.nn as nn
from torchvision.models import *
import torch.nn.init
class WassersteinLoss(Module):
"""For WGAN."""
def forward(self, real, fake):
return real.mean() - fake.mean()
class PrePostInitMeta(type):
"""A metaclass that calls optional `__pre_init__` and `__post_init__` methods"""
def __new__(cls, name, bases, dct):
x = super().__new__(cls, name, bases, dct)
old_init = x.__init__
def _pass(self):
pass
@functools.wraps(old_init)
def _init(self, *args, **kwargs):
self.__pre_init__()
old_init(self, *args, **kwargs)
self.__post_init__()
x.__init__ = _init
if not hasattr(x, '__pre_init__'):
x.__pre_init__ = _pass
if not hasattr(x, '__post_init__'):
x.__post_init__ = _pass
return x
class Module(nn.Module, metaclass=PrePostInitMeta):
"""Same as `nn.Module`, but no need for subclasses to call `super().__init__`"""
def __pre_init__(self):
super().__init__()
def __init__(self):
pass
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch.nn import Module
import functools
import torch.utils.data
import torch.nn as nn
from torchvision.models import *
import torch.nn.init
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_per_fused_mean_sub_0(in_out_ptr0, in_ptr0, in_ptr1, xnumel, rnumel):
XBLOCK: tl.constexpr = 1
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
tl.full([1], xoffset, tl.int32)
tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
tl.full([RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp4 = tl.load(in_ptr1 + r0, None)
tmp1 = tl.broadcast_to(tmp0, [RBLOCK])
tmp3 = triton_helpers.promote_to_tensor(tl.sum(tmp1, 0))
tmp5 = tl.broadcast_to(tmp4, [RBLOCK])
tmp7 = triton_helpers.promote_to_tensor(tl.sum(tmp5, 0))
tmp8 = 256.0
tmp9 = tmp3 / tmp8
tmp10 = tmp7 / tmp8
tmp11 = tmp9 - tmp10
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([1], 0, tl.int32), tmp11, None)
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((), (), torch.float32)
buf2 = buf0
del buf0
get_raw_stream(0)
triton_per_fused_mean_sub_0[grid(1)](buf2, arg0_1, arg1_1, 1, 256,
num_warps=2, num_stages=1)
del arg0_1
del arg1_1
return buf2,
class WassersteinLossNew(Module):
"""For WGAN."""
def forward(self, input_0, input_1):
arg0_1 = input_0
arg1_1 = input_1
output = call([arg0_1, arg1_1])
return output[0]
class PrePostInitMeta(type):
"""A metaclass that calls optional `__pre_init__` and `__post_init__` methods"""
def __new__(cls, name, bases, dct):
x = super().__new__(cls, name, bases, dct)
old_init = x.__init__
def _pass(self):
pass
@functools.wraps(old_init)
def _init(self, *args, **kwargs):
self.__pre_init__()
old_init(self, *args, **kwargs)
self.__post_init__()
x.__init__ = _init
if not hasattr(x, '__pre_init__'):
x.__pre_init__ = _pass
if not hasattr(x, '__post_init__'):
x.__post_init__ = _pass
return x
class Module(nn.Module, metaclass=PrePostInitMeta):
"""Same as `nn.Module`, but no need for subclasses to call `super().__init__`"""
def __pre_init__(self):
super().__init__()
def __init__(self):
pass
|
JiahuaWU/fastai
|
WassersteinLoss
| false
| 13,892
|
[
"Apache-2.0"
] | 59
|
13a2df812d875abf0558004283392ab40d9bdea1
|
https://github.com/JiahuaWU/fastai/tree/13a2df812d875abf0558004283392ab40d9bdea1
|
FocalLoss
|
import torch
import torch.nn as nn
import torch.nn.functional as F
def reduce_loss(loss, reduction):
"""Reduce loss as specified.
Args:
loss (Tensor): Elementwise loss tensor.
reduction (str): Options are "none", "mean" and "sum".
Return:
Tensor: Reduced loss tensor.
"""
reduction_enum = F._Reduction.get_enum(reduction)
if reduction_enum == 0:
return loss
elif reduction_enum == 1:
return loss.mean()
elif reduction_enum == 2:
return loss.sum()
def weight_reduce_loss(loss, weight=None, reduction='mean', avg_factor=None):
"""Apply element-wise weight and reduce loss.
Args:
loss (Tensor): Element-wise loss.
weight (Tensor): Element-wise weights.
reduction (str): Same as built-in losses of PyTorch.
avg_factor (float): Average factor when computing the mean of losses.
Returns:
Tensor: Processed loss values.
"""
if weight is not None:
loss = loss * weight
if avg_factor is None:
loss = reduce_loss(loss, reduction)
elif reduction == 'mean':
loss = loss.sum() / avg_factor
elif reduction != 'none':
raise ValueError('avg_factor can not be used with reduction="sum"')
return loss
def convert_to_one_hot(targets: 'torch.Tensor', classes) ->torch.Tensor:
"""This function converts target class indices to one-hot vectors, given
the number of classes.
Args:
targets (Tensor): The ground truth label of the prediction
with shape (N, 1)
classes (int): the number of classes.
Returns:
Tensor: Processed loss values.
"""
assert torch.max(targets).item(
) < classes, 'Class Index must be less than number of classes'
one_hot_targets = torch.zeros((targets.shape[0], classes), dtype=torch.
long, device=targets.device)
one_hot_targets.scatter_(1, targets.long(), 1)
return one_hot_targets
def sigmoid_focal_loss(pred, target, weight=None, gamma=2.0, alpha=0.25,
reduction='mean', avg_factor=None):
"""Sigmoid focal loss.
Args:
pred (torch.Tensor): The prediction with shape (N, \\*).
target (torch.Tensor): The ground truth label of the prediction with
shape (N, \\*).
weight (torch.Tensor, optional): Sample-wise loss weight with shape
(N, ). Defaults to None.
gamma (float): The gamma for calculating the modulating factor.
Defaults to 2.0.
alpha (float): A balanced form for Focal Loss. Defaults to 0.25.
reduction (str): The method used to reduce the loss.
Options are "none", "mean" and "sum". If reduction is 'none' ,
loss is same shape as pred and label. Defaults to 'mean'.
avg_factor (int, optional): Average factor that is used to average
the loss. Defaults to None.
Returns:
torch.Tensor: Loss.
"""
assert pred.shape == target.shape, 'pred and target should be in the same shape.'
pred_sigmoid = pred.sigmoid()
target = target.type_as(pred)
pt = (1 - pred_sigmoid) * target + pred_sigmoid * (1 - target)
focal_weight = (alpha * target + (1 - alpha) * (1 - target)) * pt.pow(gamma
)
loss = F.binary_cross_entropy_with_logits(pred, target, reduction='none'
) * focal_weight
if weight is not None:
assert weight.dim() == 1
weight = weight.float()
if pred.dim() > 1:
weight = weight.reshape(-1, 1)
loss = weight_reduce_loss(loss, weight, reduction, avg_factor)
return loss
class FocalLoss(nn.Module):
"""Focal loss.
Args:
gamma (float): Focusing parameter in focal loss.
Defaults to 2.0.
alpha (float): The parameter in balanced form of focal
loss. Defaults to 0.25.
reduction (str): The method used to reduce the loss into
a scalar. Options are "none" and "mean". Defaults to 'mean'.
loss_weight (float): Weight of loss. Defaults to 1.0.
"""
def __init__(self, gamma=2.0, alpha=0.25, reduction='mean', loss_weight=1.0
):
super(FocalLoss, self).__init__()
self.gamma = gamma
self.alpha = alpha
self.reduction = reduction
self.loss_weight = loss_weight
def forward(self, pred, target, weight=None, avg_factor=None,
reduction_override=None):
"""Sigmoid focal loss.
Args:
pred (torch.Tensor): The prediction with shape (N, \\*).
target (torch.Tensor): The ground truth label of the prediction
with shape (N, \\*), N or (N,1).
weight (torch.Tensor, optional): Sample-wise loss weight with shape
(N, \\*). Defaults to None.
avg_factor (int, optional): Average factor that is used to average
the loss. Defaults to None.
reduction_override (str, optional): The method used to reduce the
loss into a scalar. Options are "none", "mean" and "sum".
Defaults to None.
Returns:
torch.Tensor: Loss.
"""
assert reduction_override in (None, 'none', 'mean', 'sum')
reduction = (reduction_override if reduction_override else self.
reduction)
if target.dim() == 1 or target.dim() == 2 and target.shape[1] == 1:
target = convert_to_one_hot(target.view(-1, 1), pred.shape[-1])
loss_cls = self.loss_weight * sigmoid_focal_loss(pred, target,
weight, gamma=self.gamma, alpha=self.alpha, reduction=reduction,
avg_factor=avg_factor)
return loss_cls
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
import torch.nn as nn
import torch.nn.functional as F
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_per_fused_add_binary_cross_entropy_with_logits_mean_mul_pow_rsub_sigmoid_0(
in_out_ptr0, in_ptr0, in_ptr1, xnumel, rnumel):
XBLOCK: tl.constexpr = 1
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
tl.full([1], xoffset, tl.int32)
tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
tl.full([RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp3 = tl.load(in_ptr1 + r0, None)
tmp1 = 1.0
tmp2 = tmp1 - tmp0
tmp4 = tmp2 * tmp3
tmp5 = 0.0
tmp6 = triton_helpers.minimum(tmp5, tmp3)
tmp7 = tl_math.abs(tmp3)
tmp8 = -tmp7
tmp9 = tl_math.exp(tmp8)
tmp10 = libdevice.log1p(tmp9)
tmp11 = tmp6 - tmp10
tmp12 = tmp4 - tmp11
tmp13 = 0.25
tmp14 = tmp0 * tmp13
tmp15 = 0.75
tmp16 = tmp2 * tmp15
tmp17 = tmp14 + tmp16
tmp18 = tl.sigmoid(tmp3)
tmp19 = tmp1 - tmp18
tmp20 = tmp19 * tmp0
tmp21 = tmp18 * tmp2
tmp22 = tmp20 + tmp21
tmp23 = tmp22 * tmp22
tmp24 = tmp17 * tmp23
tmp25 = tmp12 * tmp24
tmp26 = tl.broadcast_to(tmp25, [RBLOCK])
tmp28 = triton_helpers.promote_to_tensor(tl.sum(tmp26, 0))
tmp29 = 256.0
tmp30 = tmp28 / tmp29
tmp31 = tmp30 * tmp1
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([1], 0, tl.int32), tmp31, None)
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((), (), torch.float32)
buf1 = buf0
del buf0
get_raw_stream(0)
triton_per_fused_add_binary_cross_entropy_with_logits_mean_mul_pow_rsub_sigmoid_0[
grid(1)](buf1, arg0_1, arg1_1, 1, 256, num_warps=2, num_stages=1)
del arg0_1
del arg1_1
return buf1,
def reduce_loss(loss, reduction):
"""Reduce loss as specified.
Args:
loss (Tensor): Elementwise loss tensor.
reduction (str): Options are "none", "mean" and "sum".
Return:
Tensor: Reduced loss tensor.
"""
reduction_enum = F._Reduction.get_enum(reduction)
if reduction_enum == 0:
return loss
elif reduction_enum == 1:
return loss.mean()
elif reduction_enum == 2:
return loss.sum()
def weight_reduce_loss(loss, weight=None, reduction='mean', avg_factor=None):
"""Apply element-wise weight and reduce loss.
Args:
loss (Tensor): Element-wise loss.
weight (Tensor): Element-wise weights.
reduction (str): Same as built-in losses of PyTorch.
avg_factor (float): Average factor when computing the mean of losses.
Returns:
Tensor: Processed loss values.
"""
if weight is not None:
loss = loss * weight
if avg_factor is None:
loss = reduce_loss(loss, reduction)
elif reduction == 'mean':
loss = loss.sum() / avg_factor
elif reduction != 'none':
raise ValueError('avg_factor can not be used with reduction="sum"')
return loss
def convert_to_one_hot(targets: 'torch.Tensor', classes) ->torch.Tensor:
"""This function converts target class indices to one-hot vectors, given
the number of classes.
Args:
targets (Tensor): The ground truth label of the prediction
with shape (N, 1)
classes (int): the number of classes.
Returns:
Tensor: Processed loss values.
"""
assert torch.max(targets).item(
) < classes, 'Class Index must be less than number of classes'
one_hot_targets = torch.zeros((targets.shape[0], classes), dtype=torch.
long, device=targets.device)
one_hot_targets.scatter_(1, targets.long(), 1)
return one_hot_targets
def sigmoid_focal_loss(pred, target, weight=None, gamma=2.0, alpha=0.25,
reduction='mean', avg_factor=None):
"""Sigmoid focal loss.
Args:
pred (torch.Tensor): The prediction with shape (N, \\*).
target (torch.Tensor): The ground truth label of the prediction with
shape (N, \\*).
weight (torch.Tensor, optional): Sample-wise loss weight with shape
(N, ). Defaults to None.
gamma (float): The gamma for calculating the modulating factor.
Defaults to 2.0.
alpha (float): A balanced form for Focal Loss. Defaults to 0.25.
reduction (str): The method used to reduce the loss.
Options are "none", "mean" and "sum". If reduction is 'none' ,
loss is same shape as pred and label. Defaults to 'mean'.
avg_factor (int, optional): Average factor that is used to average
the loss. Defaults to None.
Returns:
torch.Tensor: Loss.
"""
assert pred.shape == target.shape, 'pred and target should be in the same shape.'
pred_sigmoid = pred.sigmoid()
target = target.type_as(pred)
pt = (1 - pred_sigmoid) * target + pred_sigmoid * (1 - target)
focal_weight = (alpha * target + (1 - alpha) * (1 - target)) * pt.pow(gamma
)
loss = F.binary_cross_entropy_with_logits(pred, target, reduction='none'
) * focal_weight
if weight is not None:
assert weight.dim() == 1
weight = weight.float()
if pred.dim() > 1:
weight = weight.reshape(-1, 1)
loss = weight_reduce_loss(loss, weight, reduction, avg_factor)
return loss
class FocalLossNew(nn.Module):
"""Focal loss.
Args:
gamma (float): Focusing parameter in focal loss.
Defaults to 2.0.
alpha (float): The parameter in balanced form of focal
loss. Defaults to 0.25.
reduction (str): The method used to reduce the loss into
a scalar. Options are "none" and "mean". Defaults to 'mean'.
loss_weight (float): Weight of loss. Defaults to 1.0.
"""
def __init__(self, gamma=2.0, alpha=0.25, reduction='mean', loss_weight=1.0
):
super(FocalLossNew, self).__init__()
self.gamma = gamma
self.alpha = alpha
self.reduction = reduction
self.loss_weight = loss_weight
def forward(self, input_0, input_1):
arg0_1 = input_0
arg1_1 = input_1
output = call([arg0_1, arg1_1])
return output[0]
|
JiYuanFeng/mmclassification
|
FocalLoss
| false
| 13,893
|
[
"Apache-2.0"
] | 1,190
|
b337ef1f11b85148cca4b6fb0c4da3f8cc2eede6
|
https://github.com/JiYuanFeng/mmclassification/tree/b337ef1f11b85148cca4b6fb0c4da3f8cc2eede6
|
CrossEntropy2D
|
import torch
import torch.nn as nn
class CrossEntropy2D(nn.Module):
"""
2D Cross-entropy loss implemented as negative log likelihood
"""
def __init__(self, weight=None, reduction='none'):
super(CrossEntropy2D, self).__init__()
self.nll_loss = nn.CrossEntropyLoss(weight=weight, reduction=reduction)
def forward(self, inputs, targets):
return self.nll_loss(inputs, targets)
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import math as tl_math
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused__log_softmax_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.
constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x0 = xindex % 16
x2 = xindex // 64
tmp0 = tl.load(in_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + (x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp2 = tl.load(in_ptr0 + (16 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp4 = tl.load(in_ptr0 + (32 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp6 = tl.load(in_ptr0 + (48 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp3 = triton_helpers.maximum(tmp1, tmp2)
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp0 - tmp7
tl.store(out_ptr0 + x3, tmp8, xmask)
@triton.jit
def triton_poi_fused__log_softmax_mul_neg_sum_1(in_ptr0, in_ptr1, out_ptr0,
xnumel, XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 16
x1 = xindex // 16
x2 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + 64 * x1), xmask)
tmp2 = tl.load(in_ptr0 + (16 + x0 + 64 * x1), xmask)
tmp5 = tl.load(in_ptr0 + (32 + x0 + 64 * x1), xmask)
tmp8 = tl.load(in_ptr0 + (48 + x0 + 64 * x1), xmask)
tmp13 = tl.load(in_ptr1 + (x0 + 64 * x1), xmask)
tmp16 = tl.load(in_ptr1 + (16 + x0 + 64 * x1), xmask)
tmp20 = tl.load(in_ptr1 + (32 + x0 + 64 * x1), xmask)
tmp24 = tl.load(in_ptr1 + (48 + x0 + 64 * x1), xmask)
tmp1 = tl_math.exp(tmp0)
tmp3 = tl_math.exp(tmp2)
tmp4 = tmp1 + tmp3
tmp6 = tl_math.exp(tmp5)
tmp7 = tmp4 + tmp6
tmp9 = tl_math.exp(tmp8)
tmp10 = tmp7 + tmp9
tmp11 = tl_math.log(tmp10)
tmp12 = tmp0 - tmp11
tmp14 = tmp12 * tmp13
tmp15 = tmp2 - tmp11
tmp17 = tmp15 * tmp16
tmp18 = tmp14 + tmp17
tmp19 = tmp5 - tmp11
tmp21 = tmp19 * tmp20
tmp22 = tmp18 + tmp21
tmp23 = tmp8 - tmp11
tmp25 = tmp23 * tmp24
tmp26 = tmp22 + tmp25
tmp27 = -tmp26
tl.store(out_ptr0 + x2, tmp27, xmask)
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused__log_softmax_0[grid(256)](arg1_1, buf0, 256,
XBLOCK=128, num_warps=4, num_stages=1)
del arg1_1
buf1 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
triton_poi_fused__log_softmax_mul_neg_sum_1[grid(64)](buf0, arg0_1,
buf1, 64, XBLOCK=64, num_warps=1, num_stages=1)
del arg0_1
del buf0
return buf1,
class CrossEntropy2DNew(nn.Module):
"""
2D Cross-entropy loss implemented as negative log likelihood
"""
def __init__(self, weight=None, reduction='none'):
super(CrossEntropy2DNew, self).__init__()
self.nll_loss = nn.CrossEntropyLoss(weight=weight, reduction=reduction)
def forward(self, input_0, input_1):
arg0_1 = input_0
arg1_1 = input_1
output = call([arg0_1, arg1_1])
return output[0]
|
Jinboasltw/FastSurfer
|
CrossEntropy2D
| false
| 13,894
|
[
"Apache-2.0"
] | 257
|
3c0330c459c221b85428d3ec2e95f5196aee3129
|
https://github.com/Jinboasltw/FastSurfer/tree/3c0330c459c221b85428d3ec2e95f5196aee3129
|
MaxPoolPad
|
import torch
import torch.utils.data
import torch.nn as nn
from torchvision.models import *
import torch.nn.init
class MaxPoolPad(nn.Module):
def __init__(self):
super(MaxPoolPad, self).__init__()
self.pad = nn.ZeroPad2d((1, 0, 1, 0))
self.pool = nn.MaxPool2d(3, stride=2, padding=1)
def forward(self, x):
x = self.pad(x)
x = self.pool(x)
x = x[:, :, 1:, 1:]
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
import torch.utils.data
import torch.nn as nn
from torchvision.models import *
import torch.nn.init
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_constant_pad_nd_max_pool2d_with_indices_0(in_ptr0,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 144
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 3 % 3
x0 = xindex % 3
x2 = xindex // 9
x4 = xindex
tmp0 = -1 + 2 * x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 5, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tmp2 & tmp4
tmp6 = -1 + 2 * x0
tmp7 = tmp6 >= tmp1
tmp8 = tmp6 < tmp3
tmp9 = tmp7 & tmp8
tmp10 = tmp5 & tmp9
tmp11 = -2 + 2 * x1
tmp12 = tmp11 >= tmp1
tmp13 = -2 + 2 * x0
tmp14 = tmp13 >= tmp1
tmp15 = tmp12 & tmp14
tmp16 = tmp15 & tmp10
tmp17 = tl.load(in_ptr0 + (-10 + 2 * x0 + 8 * x1 + 16 * x2), tmp16 &
xmask, eviction_policy='evict_last', other=0.0)
tmp18 = tl.full(tmp17.shape, float('-inf'), tmp17.dtype)
tmp19 = tl.where(tmp10, tmp17, tmp18)
tmp20 = 2 * x0
tmp21 = tmp20 >= tmp1
tmp22 = tmp20 < tmp3
tmp23 = tmp21 & tmp22
tmp24 = tmp5 & tmp23
tmp25 = tmp12 & tmp7
tmp26 = tmp25 & tmp24
tmp27 = tl.load(in_ptr0 + (-9 + 2 * x0 + 8 * x1 + 16 * x2), tmp26 &
xmask, eviction_policy='evict_last', other=0.0)
tmp28 = tl.full(tmp27.shape, float('-inf'), tmp27.dtype)
tmp29 = tl.where(tmp24, tmp27, tmp28)
tmp30 = triton_helpers.maximum(tmp29, tmp19)
tmp31 = 1 + 2 * x0
tmp32 = tmp31 >= tmp1
tmp33 = tmp31 < tmp3
tmp34 = tmp32 & tmp33
tmp35 = tmp5 & tmp34
tmp36 = tmp12 & tmp21
tmp37 = tmp36 & tmp35
tmp38 = tl.load(in_ptr0 + (-8 + 2 * x0 + 8 * x1 + 16 * x2), tmp37 &
xmask, eviction_policy='evict_last', other=0.0)
tmp39 = tl.full(tmp38.shape, float('-inf'), tmp38.dtype)
tmp40 = tl.where(tmp35, tmp38, tmp39)
tmp41 = triton_helpers.maximum(tmp40, tmp30)
tmp42 = 2 * x1
tmp43 = tmp42 >= tmp1
tmp44 = tmp42 < tmp3
tmp45 = tmp43 & tmp44
tmp46 = tmp45 & tmp9
tmp47 = tmp2 & tmp14
tmp48 = tmp47 & tmp46
tmp49 = tl.load(in_ptr0 + (-6 + 2 * x0 + 8 * x1 + 16 * x2), tmp48 &
xmask, eviction_policy='evict_last', other=0.0)
tmp50 = tl.full(tmp49.shape, float('-inf'), tmp49.dtype)
tmp51 = tl.where(tmp46, tmp49, tmp50)
tmp52 = triton_helpers.maximum(tmp51, tmp41)
tmp53 = tmp45 & tmp23
tmp54 = tmp2 & tmp7
tmp55 = tmp54 & tmp53
tmp56 = tl.load(in_ptr0 + (-5 + 2 * x0 + 8 * x1 + 16 * x2), tmp55 &
xmask, eviction_policy='evict_last', other=0.0)
tmp57 = tl.full(tmp56.shape, float('-inf'), tmp56.dtype)
tmp58 = tl.where(tmp53, tmp56, tmp57)
tmp59 = triton_helpers.maximum(tmp58, tmp52)
tmp60 = tmp45 & tmp34
tmp61 = tmp2 & tmp21
tmp62 = tmp61 & tmp60
tmp63 = tl.load(in_ptr0 + (-4 + 2 * x0 + 8 * x1 + 16 * x2), tmp62 &
xmask, eviction_policy='evict_last', other=0.0)
tmp64 = tl.full(tmp63.shape, float('-inf'), tmp63.dtype)
tmp65 = tl.where(tmp60, tmp63, tmp64)
tmp66 = triton_helpers.maximum(tmp65, tmp59)
tmp67 = 1 + 2 * x1
tmp68 = tmp67 >= tmp1
tmp69 = tmp67 < tmp3
tmp70 = tmp68 & tmp69
tmp71 = tmp70 & tmp9
tmp72 = tmp43 & tmp14
tmp73 = tmp72 & tmp71
tmp74 = tl.load(in_ptr0 + (-2 + 2 * x0 + 8 * x1 + 16 * x2), tmp73 &
xmask, eviction_policy='evict_last', other=0.0)
tmp75 = tl.full(tmp74.shape, float('-inf'), tmp74.dtype)
tmp76 = tl.where(tmp71, tmp74, tmp75)
tmp77 = triton_helpers.maximum(tmp76, tmp66)
tmp78 = tmp70 & tmp23
tmp79 = tmp43 & tmp7
tmp80 = tmp79 & tmp78
tmp81 = tl.load(in_ptr0 + (-1 + 2 * x0 + 8 * x1 + 16 * x2), tmp80 &
xmask, eviction_policy='evict_last', other=0.0)
tmp82 = tl.full(tmp81.shape, float('-inf'), tmp81.dtype)
tmp83 = tl.where(tmp78, tmp81, tmp82)
tmp84 = triton_helpers.maximum(tmp83, tmp77)
tmp85 = tmp70 & tmp34
tmp86 = tmp43 & tmp21
tmp87 = tmp86 & tmp85
tmp88 = tl.load(in_ptr0 + (2 * x0 + 8 * x1 + 16 * x2), tmp87 & xmask,
eviction_policy='evict_last', other=0.0)
tmp89 = tl.full(tmp88.shape, float('-inf'), tmp88.dtype)
tmp90 = tl.where(tmp85, tmp88, tmp89)
tmp91 = triton_helpers.maximum(tmp90, tmp84)
tl.store(out_ptr0 + x4, tmp91, xmask)
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 3, 3), (36, 9, 3, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_constant_pad_nd_max_pool2d_with_indices_0[grid(144)](
arg0_1, buf0, 144, XBLOCK=128, num_warps=4, num_stages=1)
del arg0_1
return reinterpret_tensor(buf0, (4, 4, 2, 2), (36, 9, 3, 1), 4),
class MaxPoolPadNew(nn.Module):
def __init__(self):
super(MaxPoolPadNew, self).__init__()
self.pad = nn.ZeroPad2d((1, 0, 1, 0))
self.pool = nn.MaxPool2d(3, stride=2, padding=1)
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
|
JiahuaWU/fastai
|
MaxPoolPad
| false
| 13,895
|
[
"Apache-2.0"
] | 59
|
13a2df812d875abf0558004283392ab40d9bdea1
|
https://github.com/JiahuaWU/fastai/tree/13a2df812d875abf0558004283392ab40d9bdea1
|
SoftQNetwork
|
import torch
import torch.nn as nn
import torch.nn.functional as F
class SoftQNetwork(nn.Module):
def __init__(self, num_inputs, num_actions, hidden_size, init_w=0.003):
super(SoftQNetwork, self).__init__()
self.linear1 = nn.Linear(num_inputs + num_actions, hidden_size)
self.linear2 = nn.Linear(hidden_size, hidden_size)
self.linear3 = nn.Linear(hidden_size, hidden_size)
self.linear4 = nn.Linear(hidden_size, 1)
self.linear4.weight.data.uniform_(-init_w, init_w)
self.linear4.bias.data.uniform_(-init_w, init_w)
def forward(self, state, action):
x = torch.cat([state, action], 1)
x = F.relu(self.linear1(x))
x = F.relu(self.linear2(x))
x = F.relu(self.linear3(x))
x = self.linear4(x)
return x
def get_inputs():
return [torch.rand([4, 4]), torch.rand([4, 4])]
def get_init_inputs():
return [[], {'num_inputs': 4, 'num_actions': 4, 'hidden_size': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_cat_0(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK: tl.
constexpr):
xnumel = 32
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 8
x1 = xindex // 8
x2 = xindex
tmp0 = x0
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (4 * x1 + x0), tmp4 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp6 = tmp0 >= tmp3
tl.full([1], 8, tl.int64)
tmp9 = tl.load(in_ptr1 + (4 * x1 + (-4 + x0)), tmp6 & xmask,
eviction_policy='evict_last', other=0.0)
tmp10 = tl.where(tmp4, tmp5, tmp9)
tl.store(out_ptr0 + x2, tmp10, xmask)
@triton.jit
def triton_poi_fused_relu_1(in_out_ptr0, in_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_out_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + x2, tmp4, xmask)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7, primals_8, primals_9, primals_10) = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4, 4), (4, 1))
assert_size_stride(primals_3, (4, 8), (8, 1))
assert_size_stride(primals_4, (4,), (1,))
assert_size_stride(primals_5, (4, 4), (4, 1))
assert_size_stride(primals_6, (4,), (1,))
assert_size_stride(primals_7, (4, 4), (4, 1))
assert_size_stride(primals_8, (4,), (1,))
assert_size_stride(primals_9, (1, 4), (4, 1))
assert_size_stride(primals_10, (1,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 8), (8, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_cat_0[grid(32)](primals_1, primals_2, buf0, 32,
XBLOCK=32, num_warps=1, num_stages=1)
del primals_1
del primals_2
buf1 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.mm(buf0, reinterpret_tensor(primals_3, (8, 4), (1, 8
), 0), out=buf1)
del primals_3
buf2 = buf1
del buf1
triton_poi_fused_relu_1[grid(16)](buf2, primals_4, 16, XBLOCK=16,
num_warps=1, num_stages=1)
del primals_4
buf3 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.mm(buf2, reinterpret_tensor(primals_5, (4, 4), (1, 4
), 0), out=buf3)
buf4 = buf3
del buf3
triton_poi_fused_relu_1[grid(16)](buf4, primals_6, 16, XBLOCK=16,
num_warps=1, num_stages=1)
del primals_6
buf5 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.mm(buf4, reinterpret_tensor(primals_7, (4, 4), (1, 4
), 0), out=buf5)
buf6 = buf5
del buf5
triton_poi_fused_relu_1[grid(16)](buf6, primals_8, 16, XBLOCK=16,
num_warps=1, num_stages=1)
del primals_8
buf8 = empty_strided_cuda((4, 1), (1, 1), torch.float32)
extern_kernels.addmm(primals_10, buf6, reinterpret_tensor(primals_9,
(4, 1), (1, 4), 0), alpha=1, beta=1, out=buf8)
del primals_10
return buf8, buf0, buf2, buf4, buf6, primals_9, primals_7, primals_5
class SoftQNetworkNew(nn.Module):
def __init__(self, num_inputs, num_actions, hidden_size, init_w=0.003):
super(SoftQNetworkNew, self).__init__()
self.linear1 = nn.Linear(num_inputs + num_actions, hidden_size)
self.linear2 = nn.Linear(hidden_size, hidden_size)
self.linear3 = nn.Linear(hidden_size, hidden_size)
self.linear4 = nn.Linear(hidden_size, 1)
self.linear4.weight.data.uniform_(-init_w, init_w)
self.linear4.bias.data.uniform_(-init_w, init_w)
def forward(self, input_0, input_1):
primals_3 = self.linear1.weight
primals_4 = self.linear1.bias
primals_1 = self.linear2.weight
primals_6 = self.linear2.bias
primals_2 = self.linear3.weight
primals_8 = self.linear3.bias
primals_9 = self.linear4.weight
primals_10 = self.linear4.bias
primals_5 = input_0
primals_7 = input_1
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7, primals_8, primals_9, primals_10])
return output[0]
|
JieRen98/Popular-RL-Algorithms
|
SoftQNetwork
| false
| 13,896
|
[
"Apache-2.0"
] | 273
|
7f2bb74a51cf9cbde92a6ccfa42e97dc129dd145
|
https://github.com/JieRen98/Popular-RL-Algorithms/tree/7f2bb74a51cf9cbde92a6ccfa42e97dc129dd145
|
Attention
|
import torch
import torch.nn as nn
class Attention(nn.Module):
""" Applies attention mechanism on the `context` using the `query`.
**Thank you** to IBM for their initial implementation of :class:`Attention`. Here is
their `License
<https://github.com/IBM/pytorch-seq2seq/blob/master/LICENSE>`__.
Args:
dimensions (int): Dimensionality of the query and context.
attention_type (str, optional): How to compute the attention score:
* dot: :math:`score(H_j,q) = H_j^T q`
* general: :math:`score(H_j, q) = H_j^T W_a q`
Example:
>>> attention = Attention(256)
>>> query = torch.randn(5, 1, 256)
>>> context = torch.randn(5, 5, 256)
>>> output, weights = attention(query, context)
>>> output.size()
torch.Size([5, 1, 256])
>>> weights.size()
torch.Size([5, 1, 5])
"""
def __init__(self, dimensions, attention_type='general'):
super(Attention, self).__init__()
if attention_type not in ['dot', 'general']:
raise ValueError('Invalid attention type selected.')
self.attention_type = attention_type
if self.attention_type == 'general':
self.linear_in = nn.Linear(dimensions, dimensions, bias=False)
self.linear_out = nn.Linear(dimensions * 2, dimensions, bias=False)
self.softmax = nn.Softmax(dim=-1)
self.tanh = nn.Tanh()
def forward(self, query, context):
"""
Args:
query (:class:`torch.FloatTensor` [batch size, output length, dimensions]): Sequence of
queries to query the context.
context (:class:`torch.FloatTensor` [batch size, query length, dimensions]): Data
overwhich to apply the attention mechanism.
Returns:
:class:`tuple` with `output` and `weights`:
* **output** (:class:`torch.LongTensor` [batch size, output length, dimensions]):
Tensor containing the attended features.
* **weights** (:class:`torch.FloatTensor` [batch size, output length, query length]):
Tensor containing attention weights.
"""
batch_size, output_len, dimensions = query.size()
query_len = context.size(1)
if self.attention_type == 'general':
query = query.view(batch_size * output_len, dimensions)
query = self.linear_in(query)
query = query.view(batch_size, output_len, dimensions)
attention_scores = torch.bmm(query, context.transpose(1, 2).
contiguous())
attention_scores = attention_scores.view(batch_size * output_len,
query_len)
attention_weights = self.softmax(attention_scores)
attention_weights = attention_weights.view(batch_size, output_len,
query_len)
mix = torch.bmm(attention_weights, context)
combined = torch.cat((mix, query), dim=2)
combined = combined.view(batch_size * output_len, 2 * dimensions)
output = self.linear_out(combined).view(batch_size, output_len,
dimensions)
output = self.tanh(output)
return output, attention_weights
def get_inputs():
return [torch.rand([4, 4, 4]), torch.rand([4, 4, 4])]
def get_init_inputs():
return [[], {'dimensions': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_clone_transpose_0(in_ptr0, out_ptr0, out_ptr1, ynumel,
xnumel, YBLOCK: tl.constexpr, XBLOCK: tl.constexpr):
ynumel = 16
xnumel = 4
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x1 = xindex
y0 = yindex
y2 = yindex % 4
y3 = yindex // 4
tmp0 = tl.load(in_ptr0 + (x1 + 4 * y0), xmask & ymask, eviction_policy=
'evict_last')
tl.store(out_ptr0 + (x1 + 4 * y0), tmp0, xmask & ymask)
tl.store(out_ptr1 + (y2 + 4 * x1 + 16 * y3), tmp0, xmask & ymask)
@triton.jit
def triton_poi_fused__softmax_1(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + 4 * x1, xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + 4 * x1), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (2 + 4 * x1), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (3 + 4 * x1), xmask, eviction_policy='evict_last')
tmp3 = triton_helpers.maximum(tmp1, tmp2)
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp0 - tmp7
tmp9 = tl_math.exp(tmp8)
tl.store(out_ptr0 + x2, tmp9, xmask)
@triton.jit
def triton_poi_fused__softmax_2(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + 4 * x1, xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + 4 * x1), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (2 + 4 * x1), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (3 + 4 * x1), xmask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp5 = tmp3 + tmp4
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tl.store(out_ptr0 + x2, tmp8, xmask)
@triton.jit
def triton_poi_fused_cat_3(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK: tl.
constexpr):
xnumel = 128
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 8
x1 = xindex // 8
x2 = xindex
tmp0 = x0
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (4 * x1 + x0), tmp4 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp6 = tmp0 >= tmp3
tl.full([1], 8, tl.int64)
tmp9 = tl.load(in_ptr1 + (4 * x1 + (-4 + x0)), tmp6 & xmask,
eviction_policy='evict_last', other=0.0)
tmp10 = tl.where(tmp4, tmp5, tmp9)
tl.store(out_ptr0 + x2, tmp10, xmask)
@triton.jit
def triton_poi_fused_tanh_4(in_out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_out_ptr0 + x0, xmask)
tmp1 = libdevice.tanh(tmp0)
tl.store(in_out_ptr0 + x0, tmp1, xmask)
def call(args):
primals_1, primals_2, primals_3, primals_4 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_2, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_3, (4, 4), (4, 1))
assert_size_stride(primals_4, (4, 8), (8, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_1, (16, 4), (4, 1), 0),
reinterpret_tensor(primals_3, (4, 4), (1, 4), 0), out=buf0)
del primals_3
buf1 = empty_strided_cuda((4, 4, 4), (16, 1, 4), torch.float32)
buf9 = empty_strided_cuda((4, 4, 4), (16, 1, 4), torch.float32)
get_raw_stream(0)
triton_poi_fused_clone_transpose_0[grid(16, 4)](primals_2, buf1,
buf9, 16, 4, XBLOCK=4, YBLOCK=16, num_warps=1, num_stages=1)
buf2 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
extern_kernels.bmm(reinterpret_tensor(buf0, (4, 4, 4), (16, 4, 1),
0), buf1, out=buf2)
buf3 = reinterpret_tensor(buf1, (16, 4), (4, 1), 0)
del buf1
triton_poi_fused__softmax_1[grid(64)](buf2, buf3, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf4 = reinterpret_tensor(buf2, (16, 4), (4, 1), 0)
del buf2
triton_poi_fused__softmax_2[grid(64)](buf3, buf4, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf5 = reinterpret_tensor(buf3, (4, 4, 4), (16, 4, 1), 0)
del buf3
extern_kernels.bmm(reinterpret_tensor(buf4, (4, 4, 4), (16, 4, 1),
0), primals_2, out=buf5)
buf6 = empty_strided_cuda((4, 4, 8), (32, 8, 1), torch.float32)
triton_poi_fused_cat_3[grid(128)](buf5, buf0, buf6, 128, XBLOCK=128,
num_warps=4, num_stages=1)
del buf0
buf7 = reinterpret_tensor(buf5, (16, 4), (4, 1), 0)
del buf5
extern_kernels.mm(reinterpret_tensor(buf6, (16, 8), (8, 1), 0),
reinterpret_tensor(primals_4, (8, 4), (1, 8), 0), out=buf7)
buf8 = reinterpret_tensor(buf7, (4, 4, 4), (16, 4, 1), 0)
del buf7
triton_poi_fused_tanh_4[grid(64)](buf8, 64, XBLOCK=64, num_warps=1,
num_stages=1)
return buf8, reinterpret_tensor(buf4, (4, 4, 4), (16, 4, 1), 0
), reinterpret_tensor(primals_1, (16, 4), (4, 1), 0
), reinterpret_tensor(primals_2, (4, 4, 4), (16, 1, 4), 0
), buf4, reinterpret_tensor(buf6, (16, 8), (8, 1), 0
), buf8, primals_4, buf9
class AttentionNew(nn.Module):
""" Applies attention mechanism on the `context` using the `query`.
**Thank you** to IBM for their initial implementation of :class:`Attention`. Here is
their `License
<https://github.com/IBM/pytorch-seq2seq/blob/master/LICENSE>`__.
Args:
dimensions (int): Dimensionality of the query and context.
attention_type (str, optional): How to compute the attention score:
* dot: :math:`score(H_j,q) = H_j^T q`
* general: :math:`score(H_j, q) = H_j^T W_a q`
Example:
>>> attention = Attention(256)
>>> query = torch.randn(5, 1, 256)
>>> context = torch.randn(5, 5, 256)
>>> output, weights = attention(query, context)
>>> output.size()
torch.Size([5, 1, 256])
>>> weights.size()
torch.Size([5, 1, 5])
"""
def __init__(self, dimensions, attention_type='general'):
super(AttentionNew, self).__init__()
if attention_type not in ['dot', 'general']:
raise ValueError('Invalid attention type selected.')
self.attention_type = attention_type
if self.attention_type == 'general':
self.linear_in = nn.Linear(dimensions, dimensions, bias=False)
self.linear_out = nn.Linear(dimensions * 2, dimensions, bias=False)
self.softmax = nn.Softmax(dim=-1)
self.tanh = nn.Tanh()
def forward(self, input_0, input_1):
primals_3 = self.linear_in.weight
primals_4 = self.linear_out.weight
primals_1 = input_0
primals_2 = input_1
output = call([primals_1, primals_2, primals_3, primals_4])
return output[0], output[1]
|
JiaqiLiu/PyTorch-NLP
|
Attention
| false
| 13,897
|
[
"BSD-3-Clause"
] | 2,125
|
71d2ce1e8b8da5ab4e7732d1ebf971150986e6c8
|
https://github.com/JiaqiLiu/PyTorch-NLP/tree/71d2ce1e8b8da5ab4e7732d1ebf971150986e6c8
|
CharbonnierLoss
|
import functools
import torch
import torch.nn as nn
from torch.nn import functional as F
def reduce_loss(loss, reduction):
"""Reduce loss as specified.
Args:
loss (Tensor): Elementwise loss tensor.
reduction (str): Options are "none", "mean" and "sum".
Returns:
Tensor: Reduced loss tensor.
"""
reduction_enum = F._Reduction.get_enum(reduction)
if reduction_enum == 0:
return loss
if reduction_enum == 1:
return loss.mean()
return loss.sum()
def mask_reduce_loss(loss, weight=None, reduction='mean', sample_wise=False):
"""Apply element-wise weight and reduce loss.
Args:
loss (Tensor): Element-wise loss.
weight (Tensor): Element-wise weights. Default: None.
reduction (str): Same as built-in losses of PyTorch. Options are
"none", "mean" and "sum". Default: 'mean'.
sample_wise (bool): Whether calculate the loss sample-wise. This
argument only takes effect when `reduction` is 'mean' and `weight`
(argument of `forward()`) is not None. It will first reduces loss
with 'mean' per-sample, and then it means over all the samples.
Default: False.
Returns:
Tensor: Processed loss values.
"""
if weight is not None:
assert weight.dim() == loss.dim()
assert weight.size(1) == 1 or weight.size(1) == loss.size(1)
loss = loss * weight
if weight is None or reduction == 'sum':
loss = reduce_loss(loss, reduction)
elif reduction == 'mean':
if weight.size(1) == 1:
weight = weight.expand_as(loss)
eps = 1e-12
if sample_wise:
weight = weight.sum(dim=[1, 2, 3], keepdim=True)
loss = (loss / (weight + eps)).sum() / weight.size(0)
else:
loss = loss.sum() / (weight.sum() + eps)
return loss
def masked_loss(loss_func):
"""Create a masked version of a given loss function.
To use this decorator, the loss function must have the signature like
`loss_func(pred, target, **kwargs)`. The function only needs to compute
element-wise loss without any reduction. This decorator will add weight
and reduction arguments to the function. The decorated function will have
the signature like `loss_func(pred, target, weight=None, reduction='mean',
avg_factor=None, **kwargs)`.
:Example:
>>> import torch
>>> @masked_loss
>>> def l1_loss(pred, target):
>>> return (pred - target).abs()
>>> pred = torch.Tensor([0, 2, 3])
>>> target = torch.Tensor([1, 1, 1])
>>> weight = torch.Tensor([1, 0, 1])
>>> l1_loss(pred, target)
tensor(1.3333)
>>> l1_loss(pred, target, weight)
tensor(1.5000)
>>> l1_loss(pred, target, reduction='none')
tensor([1., 1., 2.])
>>> l1_loss(pred, target, weight, reduction='sum')
tensor(3.)
"""
@functools.wraps(loss_func)
def wrapper(pred, target, weight=None, reduction='mean', sample_wise=
False, **kwargs):
loss = loss_func(pred, target, **kwargs)
loss = mask_reduce_loss(loss, weight, reduction, sample_wise)
return loss
return wrapper
@masked_loss
def charbonnier_loss(pred, target, eps=1e-12):
"""Charbonnier loss.
Args:
pred (Tensor): Prediction Tensor with shape (n, c, h, w).
target ([type]): Target Tensor with shape (n, c, h, w).
Returns:
Tensor: Calculated Charbonnier loss.
"""
return torch.sqrt((pred - target) ** 2 + eps)
class CharbonnierLoss(nn.Module):
"""Charbonnier loss (one variant of Robust L1Loss, a differentiable
variant of L1Loss).
Described in "Deep Laplacian Pyramid Networks for Fast and Accurate
Super-Resolution".
Args:
loss_weight (float): Loss weight for L1 loss. Default: 1.0.
reduction (str): Specifies the reduction to apply to the output.
Supported choices are 'none' | 'mean' | 'sum'. Default: 'mean'.
sample_wise (bool): Whether calculate the loss sample-wise. This
argument only takes effect when `reduction` is 'mean' and `weight`
(argument of `forward()`) is not None. It will first reduces loss
with 'mean' per-sample, and then it means over all the samples.
Default: False.
eps (float): A value used to control the curvature near zero.
Default: 1e-12.
"""
def __init__(self, loss_weight=1.0, reduction='mean', sample_wise=False,
eps=1e-12):
super().__init__()
if reduction not in ['none', 'mean', 'sum']:
raise ValueError(
f'Unsupported reduction mode: {reduction}. Supported ones are: {_reduction_modes}'
)
self.loss_weight = loss_weight
self.reduction = reduction
self.sample_wise = sample_wise
self.eps = eps
def forward(self, pred, target, weight=None, **kwargs):
"""Forward Function.
Args:
pred (Tensor): of shape (N, C, H, W). Predicted tensor.
target (Tensor): of shape (N, C, H, W). Ground truth tensor.
weight (Tensor, optional): of shape (N, C, H, W). Element-wise
weights. Default: None.
"""
return self.loss_weight * charbonnier_loss(pred, target, weight,
eps=self.eps, reduction=self.reduction, sample_wise=self.
sample_wise)
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import libdevice
import functools
import torch.nn as nn
from torch.nn import functional as F
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_per_fused_add_mean_mul_pow_sqrt_sub_0(in_out_ptr0, in_ptr0,
in_ptr1, xnumel, rnumel):
XBLOCK: tl.constexpr = 1
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
tl.full([1], xoffset, tl.int32)
tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
tl.full([RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp1 = tl.load(in_ptr1 + r0, None)
tmp2 = tmp0 - tmp1
tmp3 = tmp2 * tmp2
tmp4 = 1e-12
tmp5 = tmp3 + tmp4
tmp6 = libdevice.sqrt(tmp5)
tmp7 = tl.broadcast_to(tmp6, [RBLOCK])
tmp9 = triton_helpers.promote_to_tensor(tl.sum(tmp7, 0))
tmp10 = 256.0
tmp11 = tmp9 / tmp10
tmp12 = 1.0
tmp13 = tmp11 * tmp12
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([1], 0, tl.int32), tmp13, None)
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((), (), torch.float32)
buf1 = buf0
del buf0
get_raw_stream(0)
triton_per_fused_add_mean_mul_pow_sqrt_sub_0[grid(1)](buf1, arg0_1,
arg1_1, 1, 256, num_warps=2, num_stages=1)
del arg0_1
del arg1_1
return buf1,
def reduce_loss(loss, reduction):
"""Reduce loss as specified.
Args:
loss (Tensor): Elementwise loss tensor.
reduction (str): Options are "none", "mean" and "sum".
Returns:
Tensor: Reduced loss tensor.
"""
reduction_enum = F._Reduction.get_enum(reduction)
if reduction_enum == 0:
return loss
if reduction_enum == 1:
return loss.mean()
return loss.sum()
def mask_reduce_loss(loss, weight=None, reduction='mean', sample_wise=False):
"""Apply element-wise weight and reduce loss.
Args:
loss (Tensor): Element-wise loss.
weight (Tensor): Element-wise weights. Default: None.
reduction (str): Same as built-in losses of PyTorch. Options are
"none", "mean" and "sum". Default: 'mean'.
sample_wise (bool): Whether calculate the loss sample-wise. This
argument only takes effect when `reduction` is 'mean' and `weight`
(argument of `forward()`) is not None. It will first reduces loss
with 'mean' per-sample, and then it means over all the samples.
Default: False.
Returns:
Tensor: Processed loss values.
"""
if weight is not None:
assert weight.dim() == loss.dim()
assert weight.size(1) == 1 or weight.size(1) == loss.size(1)
loss = loss * weight
if weight is None or reduction == 'sum':
loss = reduce_loss(loss, reduction)
elif reduction == 'mean':
if weight.size(1) == 1:
weight = weight.expand_as(loss)
eps = 1e-12
if sample_wise:
weight = weight.sum(dim=[1, 2, 3], keepdim=True)
loss = (loss / (weight + eps)).sum() / weight.size(0)
else:
loss = loss.sum() / (weight.sum() + eps)
return loss
def masked_loss(loss_func):
"""Create a masked version of a given loss function.
To use this decorator, the loss function must have the signature like
`loss_func(pred, target, **kwargs)`. The function only needs to compute
element-wise loss without any reduction. This decorator will add weight
and reduction arguments to the function. The decorated function will have
the signature like `loss_func(pred, target, weight=None, reduction='mean',
avg_factor=None, **kwargs)`.
:Example:
>>> import torch
>>> @masked_loss
>>> def l1_loss(pred, target):
>>> return (pred - target).abs()
>>> pred = torch.Tensor([0, 2, 3])
>>> target = torch.Tensor([1, 1, 1])
>>> weight = torch.Tensor([1, 0, 1])
>>> l1_loss(pred, target)
tensor(1.3333)
>>> l1_loss(pred, target, weight)
tensor(1.5000)
>>> l1_loss(pred, target, reduction='none')
tensor([1., 1., 2.])
>>> l1_loss(pred, target, weight, reduction='sum')
tensor(3.)
"""
@functools.wraps(loss_func)
def wrapper(pred, target, weight=None, reduction='mean', sample_wise=
False, **kwargs):
loss = loss_func(pred, target, **kwargs)
loss = mask_reduce_loss(loss, weight, reduction, sample_wise)
return loss
return wrapper
@masked_loss
def charbonnier_loss(pred, target, eps=1e-12):
"""Charbonnier loss.
Args:
pred (Tensor): Prediction Tensor with shape (n, c, h, w).
target ([type]): Target Tensor with shape (n, c, h, w).
Returns:
Tensor: Calculated Charbonnier loss.
"""
return torch.sqrt((pred - target) ** 2 + eps)
class CharbonnierLossNew(nn.Module):
"""Charbonnier loss (one variant of Robust L1Loss, a differentiable
variant of L1Loss).
Described in "Deep Laplacian Pyramid Networks for Fast and Accurate
Super-Resolution".
Args:
loss_weight (float): Loss weight for L1 loss. Default: 1.0.
reduction (str): Specifies the reduction to apply to the output.
Supported choices are 'none' | 'mean' | 'sum'. Default: 'mean'.
sample_wise (bool): Whether calculate the loss sample-wise. This
argument only takes effect when `reduction` is 'mean' and `weight`
(argument of `forward()`) is not None. It will first reduces loss
with 'mean' per-sample, and then it means over all the samples.
Default: False.
eps (float): A value used to control the curvature near zero.
Default: 1e-12.
"""
def __init__(self, loss_weight=1.0, reduction='mean', sample_wise=False,
eps=1e-12):
super().__init__()
if reduction not in ['none', 'mean', 'sum']:
raise ValueError(
f'Unsupported reduction mode: {reduction}. Supported ones are: {_reduction_modes}'
)
self.loss_weight = loss_weight
self.reduction = reduction
self.sample_wise = sample_wise
self.eps = eps
def forward(self, input_0, input_1):
arg0_1 = input_0
arg1_1 = input_1
output = call([arg0_1, arg1_1])
return output[0]
|
Juggernaut93/mmediting
|
CharbonnierLoss
| false
| 13,898
|
[
"Apache-2.0"
] | 1,884
|
8ef46ace29756dd2df1d92f2f73a33646e33e007
|
https://github.com/Juggernaut93/mmediting/tree/8ef46ace29756dd2df1d92f2f73a33646e33e007
|
CharbonnierCompLoss
|
import functools
import torch
import torch.nn as nn
from torch.nn import functional as F
def reduce_loss(loss, reduction):
"""Reduce loss as specified.
Args:
loss (Tensor): Elementwise loss tensor.
reduction (str): Options are "none", "mean" and "sum".
Returns:
Tensor: Reduced loss tensor.
"""
reduction_enum = F._Reduction.get_enum(reduction)
if reduction_enum == 0:
return loss
if reduction_enum == 1:
return loss.mean()
return loss.sum()
def mask_reduce_loss(loss, weight=None, reduction='mean', sample_wise=False):
"""Apply element-wise weight and reduce loss.
Args:
loss (Tensor): Element-wise loss.
weight (Tensor): Element-wise weights. Default: None.
reduction (str): Same as built-in losses of PyTorch. Options are
"none", "mean" and "sum". Default: 'mean'.
sample_wise (bool): Whether calculate the loss sample-wise. This
argument only takes effect when `reduction` is 'mean' and `weight`
(argument of `forward()`) is not None. It will first reduces loss
with 'mean' per-sample, and then it means over all the samples.
Default: False.
Returns:
Tensor: Processed loss values.
"""
if weight is not None:
assert weight.dim() == loss.dim()
assert weight.size(1) == 1 or weight.size(1) == loss.size(1)
loss = loss * weight
if weight is None or reduction == 'sum':
loss = reduce_loss(loss, reduction)
elif reduction == 'mean':
if weight.size(1) == 1:
weight = weight.expand_as(loss)
eps = 1e-12
if sample_wise:
weight = weight.sum(dim=[1, 2, 3], keepdim=True)
loss = (loss / (weight + eps)).sum() / weight.size(0)
else:
loss = loss.sum() / (weight.sum() + eps)
return loss
def masked_loss(loss_func):
"""Create a masked version of a given loss function.
To use this decorator, the loss function must have the signature like
`loss_func(pred, target, **kwargs)`. The function only needs to compute
element-wise loss without any reduction. This decorator will add weight
and reduction arguments to the function. The decorated function will have
the signature like `loss_func(pred, target, weight=None, reduction='mean',
avg_factor=None, **kwargs)`.
:Example:
>>> import torch
>>> @masked_loss
>>> def l1_loss(pred, target):
>>> return (pred - target).abs()
>>> pred = torch.Tensor([0, 2, 3])
>>> target = torch.Tensor([1, 1, 1])
>>> weight = torch.Tensor([1, 0, 1])
>>> l1_loss(pred, target)
tensor(1.3333)
>>> l1_loss(pred, target, weight)
tensor(1.5000)
>>> l1_loss(pred, target, reduction='none')
tensor([1., 1., 2.])
>>> l1_loss(pred, target, weight, reduction='sum')
tensor(3.)
"""
@functools.wraps(loss_func)
def wrapper(pred, target, weight=None, reduction='mean', sample_wise=
False, **kwargs):
loss = loss_func(pred, target, **kwargs)
loss = mask_reduce_loss(loss, weight, reduction, sample_wise)
return loss
return wrapper
@masked_loss
def charbonnier_loss(pred, target, eps=1e-12):
"""Charbonnier loss.
Args:
pred (Tensor): Prediction Tensor with shape (n, c, h, w).
target ([type]): Target Tensor with shape (n, c, h, w).
Returns:
Tensor: Calculated Charbonnier loss.
"""
return torch.sqrt((pred - target) ** 2 + eps)
class CharbonnierCompLoss(nn.Module):
"""Charbonnier composition loss.
Args:
loss_weight (float): Loss weight for L1 loss. Default: 1.0.
reduction (str): Specifies the reduction to apply to the output.
Supported choices are 'none' | 'mean' | 'sum'. Default: 'mean'.
sample_wise (bool): Whether calculate the loss sample-wise. This
argument only takes effect when `reduction` is 'mean' and `weight`
(argument of `forward()`) is not None. It will first reduces loss
with 'mean' per-sample, and then it means over all the samples.
Default: False.
eps (float): A value used to control the curvature near zero.
Default: 1e-12.
"""
def __init__(self, loss_weight=1.0, reduction='mean', sample_wise=False,
eps=1e-12):
super().__init__()
if reduction not in ['none', 'mean', 'sum']:
raise ValueError(
f'Unsupported reduction mode: {reduction}. Supported ones are: {_reduction_modes}'
)
self.loss_weight = loss_weight
self.reduction = reduction
self.sample_wise = sample_wise
self.eps = eps
def forward(self, pred_alpha, fg, bg, ori_merged, weight=None, **kwargs):
"""
Args:
pred_alpha (Tensor): of shape (N, 1, H, W). Predicted alpha matte.
fg (Tensor): of shape (N, 3, H, W). Tensor of foreground object.
bg (Tensor): of shape (N, 3, H, W). Tensor of background object.
ori_merged (Tensor): of shape (N, 3, H, W). Tensor of origin merged
image before normalized by ImageNet mean and std.
weight (Tensor, optional): of shape (N, 1, H, W). It is an
indicating matrix: weight[trimap == 128] = 1. Default: None.
"""
pred_merged = pred_alpha * fg + (1.0 - pred_alpha) * bg
if weight is not None:
weight = weight.expand(-1, 3, -1, -1)
return self.loss_weight * charbonnier_loss(pred_merged, ori_merged,
weight, eps=self.eps, reduction=self.reduction, sample_wise=
self.sample_wise)
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4]), torch.rand(
[4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import libdevice
import functools
import torch.nn as nn
from torch.nn import functional as F
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_per_fused_add_mean_mul_pow_rsub_sqrt_sub_0(in_out_ptr0, in_ptr0,
in_ptr1, in_ptr2, in_ptr3, xnumel, rnumel):
XBLOCK: tl.constexpr = 1
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
tl.full([1], xoffset, tl.int32)
tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
tl.full([RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp1 = tl.load(in_ptr1 + r0, None)
tmp5 = tl.load(in_ptr2 + r0, None)
tmp8 = tl.load(in_ptr3 + r0, None)
tmp2 = tmp0 * tmp1
tmp3 = 1.0
tmp4 = tmp3 - tmp0
tmp6 = tmp4 * tmp5
tmp7 = tmp2 + tmp6
tmp9 = tmp7 - tmp8
tmp10 = tmp9 * tmp9
tmp11 = 1e-12
tmp12 = tmp10 + tmp11
tmp13 = libdevice.sqrt(tmp12)
tmp14 = tl.broadcast_to(tmp13, [RBLOCK])
tmp16 = triton_helpers.promote_to_tensor(tl.sum(tmp14, 0))
tmp17 = 256.0
tmp18 = tmp16 / tmp17
tmp19 = tmp18 * tmp3
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([1], 0, tl.int32), tmp19, None)
def call(args):
arg0_1, arg1_1, arg2_1, arg3_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg2_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg3_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((), (), torch.float32)
buf1 = buf0
del buf0
get_raw_stream(0)
triton_per_fused_add_mean_mul_pow_rsub_sqrt_sub_0[grid(1)](buf1,
arg0_1, arg1_1, arg2_1, arg3_1, 1, 256, num_warps=2, num_stages=1)
del arg0_1
del arg1_1
del arg2_1
del arg3_1
return buf1,
def reduce_loss(loss, reduction):
"""Reduce loss as specified.
Args:
loss (Tensor): Elementwise loss tensor.
reduction (str): Options are "none", "mean" and "sum".
Returns:
Tensor: Reduced loss tensor.
"""
reduction_enum = F._Reduction.get_enum(reduction)
if reduction_enum == 0:
return loss
if reduction_enum == 1:
return loss.mean()
return loss.sum()
def mask_reduce_loss(loss, weight=None, reduction='mean', sample_wise=False):
"""Apply element-wise weight and reduce loss.
Args:
loss (Tensor): Element-wise loss.
weight (Tensor): Element-wise weights. Default: None.
reduction (str): Same as built-in losses of PyTorch. Options are
"none", "mean" and "sum". Default: 'mean'.
sample_wise (bool): Whether calculate the loss sample-wise. This
argument only takes effect when `reduction` is 'mean' and `weight`
(argument of `forward()`) is not None. It will first reduces loss
with 'mean' per-sample, and then it means over all the samples.
Default: False.
Returns:
Tensor: Processed loss values.
"""
if weight is not None:
assert weight.dim() == loss.dim()
assert weight.size(1) == 1 or weight.size(1) == loss.size(1)
loss = loss * weight
if weight is None or reduction == 'sum':
loss = reduce_loss(loss, reduction)
elif reduction == 'mean':
if weight.size(1) == 1:
weight = weight.expand_as(loss)
eps = 1e-12
if sample_wise:
weight = weight.sum(dim=[1, 2, 3], keepdim=True)
loss = (loss / (weight + eps)).sum() / weight.size(0)
else:
loss = loss.sum() / (weight.sum() + eps)
return loss
def masked_loss(loss_func):
"""Create a masked version of a given loss function.
To use this decorator, the loss function must have the signature like
`loss_func(pred, target, **kwargs)`. The function only needs to compute
element-wise loss without any reduction. This decorator will add weight
and reduction arguments to the function. The decorated function will have
the signature like `loss_func(pred, target, weight=None, reduction='mean',
avg_factor=None, **kwargs)`.
:Example:
>>> import torch
>>> @masked_loss
>>> def l1_loss(pred, target):
>>> return (pred - target).abs()
>>> pred = torch.Tensor([0, 2, 3])
>>> target = torch.Tensor([1, 1, 1])
>>> weight = torch.Tensor([1, 0, 1])
>>> l1_loss(pred, target)
tensor(1.3333)
>>> l1_loss(pred, target, weight)
tensor(1.5000)
>>> l1_loss(pred, target, reduction='none')
tensor([1., 1., 2.])
>>> l1_loss(pred, target, weight, reduction='sum')
tensor(3.)
"""
@functools.wraps(loss_func)
def wrapper(pred, target, weight=None, reduction='mean', sample_wise=
False, **kwargs):
loss = loss_func(pred, target, **kwargs)
loss = mask_reduce_loss(loss, weight, reduction, sample_wise)
return loss
return wrapper
@masked_loss
def charbonnier_loss(pred, target, eps=1e-12):
"""Charbonnier loss.
Args:
pred (Tensor): Prediction Tensor with shape (n, c, h, w).
target ([type]): Target Tensor with shape (n, c, h, w).
Returns:
Tensor: Calculated Charbonnier loss.
"""
return torch.sqrt((pred - target) ** 2 + eps)
class CharbonnierCompLossNew(nn.Module):
"""Charbonnier composition loss.
Args:
loss_weight (float): Loss weight for L1 loss. Default: 1.0.
reduction (str): Specifies the reduction to apply to the output.
Supported choices are 'none' | 'mean' | 'sum'. Default: 'mean'.
sample_wise (bool): Whether calculate the loss sample-wise. This
argument only takes effect when `reduction` is 'mean' and `weight`
(argument of `forward()`) is not None. It will first reduces loss
with 'mean' per-sample, and then it means over all the samples.
Default: False.
eps (float): A value used to control the curvature near zero.
Default: 1e-12.
"""
def __init__(self, loss_weight=1.0, reduction='mean', sample_wise=False,
eps=1e-12):
super().__init__()
if reduction not in ['none', 'mean', 'sum']:
raise ValueError(
f'Unsupported reduction mode: {reduction}. Supported ones are: {_reduction_modes}'
)
self.loss_weight = loss_weight
self.reduction = reduction
self.sample_wise = sample_wise
self.eps = eps
def forward(self, input_0, input_1, input_2, input_3):
arg0_1 = input_0
arg1_1 = input_1
arg2_1 = input_2
arg3_1 = input_3
output = call([arg0_1, arg1_1, arg2_1, arg3_1])
return output[0]
|
Juggernaut93/mmediting
|
CharbonnierCompLoss
| false
| 13,899
|
[
"Apache-2.0"
] | 1,884
|
8ef46ace29756dd2df1d92f2f73a33646e33e007
|
https://github.com/Juggernaut93/mmediting/tree/8ef46ace29756dd2df1d92f2f73a33646e33e007
|
L1CompositionLoss
|
import functools
import torch
import torch.nn as nn
from torch.nn import functional as F
def reduce_loss(loss, reduction):
"""Reduce loss as specified.
Args:
loss (Tensor): Elementwise loss tensor.
reduction (str): Options are "none", "mean" and "sum".
Returns:
Tensor: Reduced loss tensor.
"""
reduction_enum = F._Reduction.get_enum(reduction)
if reduction_enum == 0:
return loss
if reduction_enum == 1:
return loss.mean()
return loss.sum()
def mask_reduce_loss(loss, weight=None, reduction='mean', sample_wise=False):
"""Apply element-wise weight and reduce loss.
Args:
loss (Tensor): Element-wise loss.
weight (Tensor): Element-wise weights. Default: None.
reduction (str): Same as built-in losses of PyTorch. Options are
"none", "mean" and "sum". Default: 'mean'.
sample_wise (bool): Whether calculate the loss sample-wise. This
argument only takes effect when `reduction` is 'mean' and `weight`
(argument of `forward()`) is not None. It will first reduces loss
with 'mean' per-sample, and then it means over all the samples.
Default: False.
Returns:
Tensor: Processed loss values.
"""
if weight is not None:
assert weight.dim() == loss.dim()
assert weight.size(1) == 1 or weight.size(1) == loss.size(1)
loss = loss * weight
if weight is None or reduction == 'sum':
loss = reduce_loss(loss, reduction)
elif reduction == 'mean':
if weight.size(1) == 1:
weight = weight.expand_as(loss)
eps = 1e-12
if sample_wise:
weight = weight.sum(dim=[1, 2, 3], keepdim=True)
loss = (loss / (weight + eps)).sum() / weight.size(0)
else:
loss = loss.sum() / (weight.sum() + eps)
return loss
def masked_loss(loss_func):
"""Create a masked version of a given loss function.
To use this decorator, the loss function must have the signature like
`loss_func(pred, target, **kwargs)`. The function only needs to compute
element-wise loss without any reduction. This decorator will add weight
and reduction arguments to the function. The decorated function will have
the signature like `loss_func(pred, target, weight=None, reduction='mean',
avg_factor=None, **kwargs)`.
:Example:
>>> import torch
>>> @masked_loss
>>> def l1_loss(pred, target):
>>> return (pred - target).abs()
>>> pred = torch.Tensor([0, 2, 3])
>>> target = torch.Tensor([1, 1, 1])
>>> weight = torch.Tensor([1, 0, 1])
>>> l1_loss(pred, target)
tensor(1.3333)
>>> l1_loss(pred, target, weight)
tensor(1.5000)
>>> l1_loss(pred, target, reduction='none')
tensor([1., 1., 2.])
>>> l1_loss(pred, target, weight, reduction='sum')
tensor(3.)
"""
@functools.wraps(loss_func)
def wrapper(pred, target, weight=None, reduction='mean', sample_wise=
False, **kwargs):
loss = loss_func(pred, target, **kwargs)
loss = mask_reduce_loss(loss, weight, reduction, sample_wise)
return loss
return wrapper
@masked_loss
def l1_loss(pred, target):
"""L1 loss.
Args:
pred (Tensor): Prediction Tensor with shape (n, c, h, w).
target ([type]): Target Tensor with shape (n, c, h, w).
Returns:
Tensor: Calculated L1 loss.
"""
return F.l1_loss(pred, target, reduction='none')
class L1CompositionLoss(nn.Module):
"""L1 composition loss.
Args:
loss_weight (float): Loss weight for L1 loss. Default: 1.0.
reduction (str): Specifies the reduction to apply to the output.
Supported choices are 'none' | 'mean' | 'sum'. Default: 'mean'.
sample_wise (bool): Whether calculate the loss sample-wise. This
argument only takes effect when `reduction` is 'mean' and `weight`
(argument of `forward()`) is not None. It will first reduces loss
with 'mean' per-sample, and then it means over all the samples.
Default: False.
"""
def __init__(self, loss_weight=1.0, reduction='mean', sample_wise=False):
super().__init__()
if reduction not in ['none', 'mean', 'sum']:
raise ValueError(
f'Unsupported reduction mode: {reduction}. Supported ones are: {_reduction_modes}'
)
self.loss_weight = loss_weight
self.reduction = reduction
self.sample_wise = sample_wise
def forward(self, pred_alpha, fg, bg, ori_merged, weight=None, **kwargs):
"""
Args:
pred_alpha (Tensor): of shape (N, 1, H, W). Predicted alpha matte.
fg (Tensor): of shape (N, 3, H, W). Tensor of foreground object.
bg (Tensor): of shape (N, 3, H, W). Tensor of background object.
ori_merged (Tensor): of shape (N, 3, H, W). Tensor of origin merged
image before normalized by ImageNet mean and std.
weight (Tensor, optional): of shape (N, 1, H, W). It is an
indicating matrix: weight[trimap == 128] = 1. Default: None.
"""
pred_merged = pred_alpha * fg + (1.0 - pred_alpha) * bg
if weight is not None:
weight = weight.expand(-1, 3, -1, -1)
return self.loss_weight * l1_loss(pred_merged, ori_merged, weight,
reduction=self.reduction, sample_wise=self.sample_wise)
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4]), torch.rand(
[4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import math as tl_math
import functools
import torch.nn as nn
from torch.nn import functional as F
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_per_fused_abs_add_mean_mul_rsub_sub_0(in_out_ptr0, in_ptr0,
in_ptr1, in_ptr2, in_ptr3, xnumel, rnumel):
XBLOCK: tl.constexpr = 1
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
tl.full([1], xoffset, tl.int32)
tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
tl.full([RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp1 = tl.load(in_ptr1 + r0, None)
tmp5 = tl.load(in_ptr2 + r0, None)
tmp8 = tl.load(in_ptr3 + r0, None)
tmp2 = tmp0 * tmp1
tmp3 = 1.0
tmp4 = tmp3 - tmp0
tmp6 = tmp4 * tmp5
tmp7 = tmp2 + tmp6
tmp9 = tmp7 - tmp8
tmp10 = tl_math.abs(tmp9)
tmp11 = tl.broadcast_to(tmp10, [RBLOCK])
tmp13 = triton_helpers.promote_to_tensor(tl.sum(tmp11, 0))
tmp14 = 256.0
tmp15 = tmp13 / tmp14
tmp16 = tmp15 * tmp3
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([1], 0, tl.int32), tmp16, None)
def call(args):
arg0_1, arg1_1, arg2_1, arg3_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg2_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg3_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((), (), torch.float32)
buf1 = buf0
del buf0
get_raw_stream(0)
triton_per_fused_abs_add_mean_mul_rsub_sub_0[grid(1)](buf1, arg0_1,
arg1_1, arg2_1, arg3_1, 1, 256, num_warps=2, num_stages=1)
del arg0_1
del arg1_1
del arg2_1
del arg3_1
return buf1,
def reduce_loss(loss, reduction):
"""Reduce loss as specified.
Args:
loss (Tensor): Elementwise loss tensor.
reduction (str): Options are "none", "mean" and "sum".
Returns:
Tensor: Reduced loss tensor.
"""
reduction_enum = F._Reduction.get_enum(reduction)
if reduction_enum == 0:
return loss
if reduction_enum == 1:
return loss.mean()
return loss.sum()
def mask_reduce_loss(loss, weight=None, reduction='mean', sample_wise=False):
"""Apply element-wise weight and reduce loss.
Args:
loss (Tensor): Element-wise loss.
weight (Tensor): Element-wise weights. Default: None.
reduction (str): Same as built-in losses of PyTorch. Options are
"none", "mean" and "sum". Default: 'mean'.
sample_wise (bool): Whether calculate the loss sample-wise. This
argument only takes effect when `reduction` is 'mean' and `weight`
(argument of `forward()`) is not None. It will first reduces loss
with 'mean' per-sample, and then it means over all the samples.
Default: False.
Returns:
Tensor: Processed loss values.
"""
if weight is not None:
assert weight.dim() == loss.dim()
assert weight.size(1) == 1 or weight.size(1) == loss.size(1)
loss = loss * weight
if weight is None or reduction == 'sum':
loss = reduce_loss(loss, reduction)
elif reduction == 'mean':
if weight.size(1) == 1:
weight = weight.expand_as(loss)
eps = 1e-12
if sample_wise:
weight = weight.sum(dim=[1, 2, 3], keepdim=True)
loss = (loss / (weight + eps)).sum() / weight.size(0)
else:
loss = loss.sum() / (weight.sum() + eps)
return loss
def masked_loss(loss_func):
"""Create a masked version of a given loss function.
To use this decorator, the loss function must have the signature like
`loss_func(pred, target, **kwargs)`. The function only needs to compute
element-wise loss without any reduction. This decorator will add weight
and reduction arguments to the function. The decorated function will have
the signature like `loss_func(pred, target, weight=None, reduction='mean',
avg_factor=None, **kwargs)`.
:Example:
>>> import torch
>>> @masked_loss
>>> def l1_loss(pred, target):
>>> return (pred - target).abs()
>>> pred = torch.Tensor([0, 2, 3])
>>> target = torch.Tensor([1, 1, 1])
>>> weight = torch.Tensor([1, 0, 1])
>>> l1_loss(pred, target)
tensor(1.3333)
>>> l1_loss(pred, target, weight)
tensor(1.5000)
>>> l1_loss(pred, target, reduction='none')
tensor([1., 1., 2.])
>>> l1_loss(pred, target, weight, reduction='sum')
tensor(3.)
"""
@functools.wraps(loss_func)
def wrapper(pred, target, weight=None, reduction='mean', sample_wise=
False, **kwargs):
loss = loss_func(pred, target, **kwargs)
loss = mask_reduce_loss(loss, weight, reduction, sample_wise)
return loss
return wrapper
@masked_loss
def l1_loss(pred, target):
"""L1 loss.
Args:
pred (Tensor): Prediction Tensor with shape (n, c, h, w).
target ([type]): Target Tensor with shape (n, c, h, w).
Returns:
Tensor: Calculated L1 loss.
"""
return F.l1_loss(pred, target, reduction='none')
class L1CompositionLossNew(nn.Module):
"""L1 composition loss.
Args:
loss_weight (float): Loss weight for L1 loss. Default: 1.0.
reduction (str): Specifies the reduction to apply to the output.
Supported choices are 'none' | 'mean' | 'sum'. Default: 'mean'.
sample_wise (bool): Whether calculate the loss sample-wise. This
argument only takes effect when `reduction` is 'mean' and `weight`
(argument of `forward()`) is not None. It will first reduces loss
with 'mean' per-sample, and then it means over all the samples.
Default: False.
"""
def __init__(self, loss_weight=1.0, reduction='mean', sample_wise=False):
super().__init__()
if reduction not in ['none', 'mean', 'sum']:
raise ValueError(
f'Unsupported reduction mode: {reduction}. Supported ones are: {_reduction_modes}'
)
self.loss_weight = loss_weight
self.reduction = reduction
self.sample_wise = sample_wise
def forward(self, input_0, input_1, input_2, input_3):
arg0_1 = input_0
arg1_1 = input_1
arg2_1 = input_2
arg3_1 = input_3
output = call([arg0_1, arg1_1, arg2_1, arg3_1])
return output[0]
|
Juggernaut93/mmediting
|
L1CompositionLoss
| false
| 13,900
|
[
"Apache-2.0"
] | 1,884
|
8ef46ace29756dd2df1d92f2f73a33646e33e007
|
https://github.com/Juggernaut93/mmediting/tree/8ef46ace29756dd2df1d92f2f73a33646e33e007
|
FocalLoss
|
import torch
import torch.nn as nn
import torch.nn.parallel
import torch.optim
import torch.utils.data
import torch.nn.functional as F
def focal_loss(input_values, gamma):
"""Computes the focal loss"""
p = torch.exp(-input_values)
loss = (1 - p) ** gamma * input_values
return loss.mean()
class FocalLoss(nn.Module):
def __init__(self, weight=None, gamma=0.0):
super(FocalLoss, self).__init__()
assert gamma >= 0
self.gamma = gamma
self.weight = weight
def forward(self, input, target):
return focal_loss(F.cross_entropy(input, target, reduction='none',
weight=self.weight), self.gamma)
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import math as tl_math
import torch.nn as nn
import torch.nn.parallel
import torch.optim
import torch.utils.data
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused__log_softmax_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.
constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x0 = xindex % 16
x2 = xindex // 64
tmp0 = tl.load(in_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + (x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp2 = tl.load(in_ptr0 + (16 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp4 = tl.load(in_ptr0 + (32 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp6 = tl.load(in_ptr0 + (48 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp3 = triton_helpers.maximum(tmp1, tmp2)
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp0 - tmp7
tl.store(out_ptr0 + x3, tmp8, xmask)
@triton.jit
def triton_per_fused__log_softmax_exp_mean_mul_neg_pow_rsub_sum_1(in_out_ptr0,
in_ptr0, in_ptr1, xnumel, rnumel, XBLOCK: tl.constexpr):
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex % 16
r1 = rindex // 16
tmp0 = tl.load(in_ptr0 + (r0 + 64 * r1), None)
tmp2 = tl.load(in_ptr0 + (16 + r0 + 64 * r1), None)
tmp5 = tl.load(in_ptr0 + (32 + r0 + 64 * r1), None)
tmp8 = tl.load(in_ptr0 + (48 + r0 + 64 * r1), None)
tmp13 = tl.load(in_ptr1 + (r0 + 64 * r1), None)
tmp16 = tl.load(in_ptr1 + (16 + r0 + 64 * r1), None)
tmp20 = tl.load(in_ptr1 + (32 + r0 + 64 * r1), None)
tmp24 = tl.load(in_ptr1 + (48 + r0 + 64 * r1), None)
tmp1 = tl_math.exp(tmp0)
tmp3 = tl_math.exp(tmp2)
tmp4 = tmp1 + tmp3
tmp6 = tl_math.exp(tmp5)
tmp7 = tmp4 + tmp6
tmp9 = tl_math.exp(tmp8)
tmp10 = tmp7 + tmp9
tmp11 = tl_math.log(tmp10)
tmp12 = tmp0 - tmp11
tmp14 = tmp12 * tmp13
tmp15 = tmp2 - tmp11
tmp17 = tmp15 * tmp16
tmp18 = tmp14 + tmp17
tmp19 = tmp5 - tmp11
tmp21 = tmp19 * tmp20
tmp22 = tmp18 + tmp21
tmp23 = tmp8 - tmp11
tmp25 = tmp23 * tmp24
tmp26 = tmp22 + tmp25
tmp27 = -tmp26
tmp28 = -tmp27
tmp29 = tl_math.exp(tmp28)
tmp30 = 1.0
tmp30 - tmp29
tmp32 = tmp30 * tmp27
tmp33 = tl.broadcast_to(tmp32, [XBLOCK, RBLOCK])
tmp35 = tl.sum(tmp33, 1)[:, None]
tmp36 = 64.0
tmp37 = tmp35 / tmp36
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([XBLOCK, 1], 0, tl.int32), tmp37, None)
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused__log_softmax_0[grid(256)](arg1_1, buf0, 256,
XBLOCK=128, num_warps=4, num_stages=1)
del arg1_1
buf2 = empty_strided_cuda((), (), torch.float32)
buf3 = buf2
del buf2
triton_per_fused__log_softmax_exp_mean_mul_neg_pow_rsub_sum_1[grid(1)](
buf3, buf0, arg0_1, 1, 64, XBLOCK=1, num_warps=2, num_stages=1)
del arg0_1
del buf0
return buf3,
def focal_loss(input_values, gamma):
"""Computes the focal loss"""
p = torch.exp(-input_values)
loss = (1 - p) ** gamma * input_values
return loss.mean()
class FocalLossNew(nn.Module):
def __init__(self, weight=None, gamma=0.0):
super(FocalLossNew, self).__init__()
assert gamma >= 0
self.gamma = gamma
self.weight = weight
def forward(self, input_0, input_1):
arg0_1 = input_0
arg1_1 = input_1
output = call([arg0_1, arg1_1])
return output[0]
|
Jianf-Wang/RSG
|
FocalLoss
| false
| 13,901
|
[
"MIT"
] | 108
|
3c5074511455428d81af89e1621493dcdb5db6ce
|
https://github.com/Jianf-Wang/RSG/tree/3c5074511455428d81af89e1621493dcdb5db6ce
|
NormedLinear
|
import torch
import torch.nn as nn
import torch.nn.parallel
import torch.optim
import torch.utils.data
import torch.nn.functional as F
from torch.nn import Parameter
class NormedLinear(nn.Module):
def __init__(self, in_features, out_features):
super(NormedLinear, self).__init__()
self.weight = Parameter(torch.Tensor(in_features, out_features))
self.weight.data.uniform_(-1, 1).renorm_(2, 1, 1e-05).mul_(100000.0)
def forward(self, x):
out = F.normalize(x, dim=1).mm(F.normalize(self.weight, dim=0))
return out
def get_inputs():
return [torch.rand([4, 4])]
def get_init_inputs():
return [[], {'in_features': 4, 'out_features': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import libdevice
import torch.nn as nn
import torch.nn.parallel
import torch.optim
import torch.utils.data
from torch.nn import Parameter
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_div_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + 4 * x1, xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (1 + 4 * x1), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (2 + 4 * x1), xmask, eviction_policy='evict_last')
tmp9 = tl.load(in_ptr0 + (3 + 4 * x1), xmask, eviction_policy='evict_last')
tmp2 = tmp1 * tmp1
tmp4 = tmp3 * tmp3
tmp5 = tmp2 + tmp4
tmp7 = tmp6 * tmp6
tmp8 = tmp5 + tmp7
tmp10 = tmp9 * tmp9
tmp11 = tmp8 + tmp10
tmp12 = libdevice.sqrt(tmp11)
tmp13 = 1e-12
tmp14 = triton_helpers.maximum(tmp12, tmp13)
tmp15 = tmp0 / tmp14
tl.store(out_ptr0 + x2, tmp15, xmask)
@triton.jit
def triton_poi_fused_div_1(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (4 + x0), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (8 + x0), xmask, eviction_policy='evict_last')
tmp9 = tl.load(in_ptr0 + (12 + x0), xmask, eviction_policy='evict_last')
tmp2 = tmp1 * tmp1
tmp4 = tmp3 * tmp3
tmp5 = tmp2 + tmp4
tmp7 = tmp6 * tmp6
tmp8 = tmp5 + tmp7
tmp10 = tmp9 * tmp9
tmp11 = tmp8 + tmp10
tmp12 = libdevice.sqrt(tmp11)
tmp13 = 1e-12
tmp14 = triton_helpers.maximum(tmp12, tmp13)
tmp15 = tmp0 / tmp14
tl.store(out_ptr0 + x2, tmp15, xmask)
def call(args):
primals_1, primals_2 = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4, 4), (4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_div_0[grid(16)](primals_1, buf0, 16, XBLOCK=16,
num_warps=1, num_stages=1)
del primals_1
buf1 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
triton_poi_fused_div_1[grid(16)](primals_2, buf1, 16, XBLOCK=16,
num_warps=1, num_stages=1)
buf2 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.mm(buf0, buf1, out=buf2)
del buf1
return buf2, primals_2, reinterpret_tensor(buf0, (4, 4), (1, 4), 0)
class NormedLinearNew(nn.Module):
def __init__(self, in_features, out_features):
super(NormedLinearNew, self).__init__()
self.weight = Parameter(torch.Tensor(in_features, out_features))
self.weight.data.uniform_(-1, 1).renorm_(2, 1, 1e-05).mul_(100000.0)
def forward(self, input_0):
primals_1 = self.weight
primals_2 = input_0
output = call([primals_1, primals_2])
return output[0]
|
Jianf-Wang/RSG
|
NormedLinear
| false
| 13,902
|
[
"MIT"
] | 108
|
3c5074511455428d81af89e1621493dcdb5db6ce
|
https://github.com/Jianf-Wang/RSG/tree/3c5074511455428d81af89e1621493dcdb5db6ce
|
ComponentConditionBlock
|
import torch
import torch.nn as nn
import torch.utils.data.distributed
class ComponentConditionBlock(nn.Module):
def __init__(self, in_shape, n_comps):
super().__init__()
self.in_shape = in_shape
self.bias = nn.Parameter(torch.zeros(n_comps, in_shape[0], 1, 1),
requires_grad=True)
def forward(self, x, comp_id):
b = self.bias[comp_id]
out = x + b
return out
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.ones([4], dtype=torch.int64)]
def get_init_inputs():
return [[], {'in_shape': [4, 4], 'n_comps': 4}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
import torch.nn as nn
import torch.utils.data.distributed
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_add_index_0(in_ptr0, in_ptr1, in_ptr2, out_ptr0,
xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x2 = xindex // 64
x1 = xindex // 16 % 4
tmp0 = tl.load(in_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr1 + x2, xmask, eviction_policy='evict_last')
tmp2 = tl.full([XBLOCK], 4, tl.int32)
tmp3 = tmp1 + tmp2
tmp4 = tmp1 < 0
tmp5 = tl.where(tmp4, tmp3, tmp1)
tl.device_assert((0 <= tmp5) & (tmp5 < 4) | ~xmask,
'index out of bounds: 0 <= tmp5 < 4')
tmp7 = tl.load(in_ptr2 + (x1 + 4 * tmp5), xmask, eviction_policy=
'evict_last')
tmp8 = tmp0 + tmp7
tl.store(out_ptr0 + x3, tmp8, xmask)
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 1, 1), (4, 1, 1, 1))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_add_index_0[grid(256)](primals_3, primals_2,
primals_1, buf0, 256, XBLOCK=256, num_warps=4, num_stages=1)
del primals_1
del primals_3
return buf0, primals_2
class ComponentConditionBlockNew(nn.Module):
def __init__(self, in_shape, n_comps):
super().__init__()
self.in_shape = in_shape
self.bias = nn.Parameter(torch.zeros(n_comps, in_shape[0], 1, 1),
requires_grad=True)
def forward(self, input_0, input_1):
primals_1 = self.bias
primals_3 = input_0
primals_2 = input_1
output = call([primals_1, primals_2, primals_3])
return output[0]
|
Johnson-yue/lffont
|
ComponentConditionBlock
| false
| 13,903
|
[
"MIT"
] | 98
|
f31f5a1cd6a075449a0f18aaafd945d373121e15
|
https://github.com/Johnson-yue/lffont/tree/f31f5a1cd6a075449a0f18aaafd945d373121e15
|
TwoLayerNet
|
import torch
from torch import Tensor
import torch.nn as nn
import torch.nn.functional as F
class TwoLayerNet(nn.Module):
def __init__(self, D_in: 'int', H: 'int', D_out: 'int') ->None:
"""
In the constructor we instantiate two nn.Linear modules and assign them as
member variables.
D_in: input dimension
H: dimension of hidden layer
D_out: output dimension
"""
super(TwoLayerNet, self).__init__()
self.linear1 = nn.Linear(D_in, H)
self.linear2 = nn.Linear(H, D_out)
def forward(self, x) ->Tensor:
"""
In the forward function we accept a Variable of input data and we must
return a Variable of output data. We can use Modules defined in the
constructor as well as arbitrary operators on Variables.
"""
h_relu = F.relu(self.linear1(x))
y_pred = self.linear2(h_relu)
return F.log_softmax(y_pred, 1)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'D_in': 4, 'H': 4, 'D_out': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import math as tl_math
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_relu_threshold_backward_0(in_out_ptr0, in_ptr0,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_out_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(in_out_ptr0 + x2, tmp4, xmask)
tl.store(out_ptr0 + x2, tmp6, xmask)
@triton.jit
def triton_poi_fused__log_softmax_1(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.
constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x0 = xindex % 16
x2 = xindex // 64
tmp0 = tl.load(in_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + (x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp2 = tl.load(in_ptr0 + (16 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp4 = tl.load(in_ptr0 + (32 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp6 = tl.load(in_ptr0 + (48 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp3 = triton_helpers.maximum(tmp1, tmp2)
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp0 - tmp7
tl.store(out_ptr0 + x3, tmp8, xmask)
@triton.jit
def triton_poi_fused__log_softmax_2(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.
constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x0 = xindex % 16
x2 = xindex // 64
tmp0 = tl.load(in_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + (x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp3 = tl.load(in_ptr0 + (16 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp6 = tl.load(in_ptr0 + (32 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp9 = tl.load(in_ptr0 + (48 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp2 = tl_math.exp(tmp1)
tmp4 = tl_math.exp(tmp3)
tmp5 = tmp2 + tmp4
tmp7 = tl_math.exp(tmp6)
tmp8 = tmp5 + tmp7
tmp10 = tl_math.exp(tmp9)
tmp11 = tmp8 + tmp10
tmp12 = tl_math.log(tmp11)
tmp13 = tmp0 - tmp12
tl.store(out_ptr0 + x3, tmp13, xmask)
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5 = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (4, 4), (4, 1))
assert_size_stride(primals_5, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_3, (64, 4), (4, 1), 0),
reinterpret_tensor(primals_1, (4, 4), (1, 4), 0), out=buf0)
del primals_1
buf1 = reinterpret_tensor(buf0, (4, 4, 4, 4), (64, 16, 4, 1), 0)
del buf0
buf5 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
get_raw_stream(0)
triton_poi_fused_relu_threshold_backward_0[grid(256)](buf1,
primals_2, buf5, 256, XBLOCK=256, num_warps=4, num_stages=1)
del primals_2
buf2 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_5, reinterpret_tensor(buf1, (64, 4), (
4, 1), 0), reinterpret_tensor(primals_4, (4, 4), (1, 4), 0),
alpha=1, beta=1, out=buf2)
del primals_5
buf3 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused__log_softmax_1[grid(256)](buf2, buf3, 256, XBLOCK=
256, num_warps=4, num_stages=1)
buf4 = reinterpret_tensor(buf2, (4, 4, 4, 4), (64, 16, 4, 1), 0)
del buf2
triton_poi_fused__log_softmax_2[grid(256)](buf3, buf4, 256, XBLOCK=
256, num_warps=4, num_stages=1)
del buf3
return buf4, reinterpret_tensor(primals_3, (64, 4), (4, 1), 0
), reinterpret_tensor(buf1, (64, 4), (4, 1), 0), buf4, primals_4, buf5
class TwoLayerNetNew(nn.Module):
def __init__(self, D_in: 'int', H: 'int', D_out: 'int') ->None:
"""
In the constructor we instantiate two nn.Linear modules and assign them as
member variables.
D_in: input dimension
H: dimension of hidden layer
D_out: output dimension
"""
super(TwoLayerNetNew, self).__init__()
self.linear1 = nn.Linear(D_in, H)
self.linear2 = nn.Linear(H, D_out)
def forward(self, input_0):
primals_1 = self.linear1.weight
primals_2 = self.linear1.bias
primals_4 = self.linear2.weight
primals_5 = self.linear2.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4, primals_5])
return output[0]
|
JohnlNguyen/FLSim
|
TwoLayerNet
| false
| 13,904
|
[
"BSD-3-Clause"
] | 79
|
a5ed7c0b84499cd9dbc5fe95f8bcb4ba8ab5a5cb
|
https://github.com/JohnlNguyen/FLSim/tree/a5ed7c0b84499cd9dbc5fe95f8bcb4ba8ab5a5cb
|
Get_gradient_nopadding
|
import torch
import torch.nn as nn
import torch.nn.functional as F
class Get_gradient_nopadding(nn.Module):
def __init__(self):
super(Get_gradient_nopadding, self).__init__()
kernel_v = [[0, -1, 0], [0, 0, 0], [0, 1, 0]]
kernel_h = [[0, 0, 0], [-1, 0, 1], [0, 0, 0]]
kernel_h = torch.FloatTensor(kernel_h).unsqueeze(0).unsqueeze(0)
kernel_v = torch.FloatTensor(kernel_v).unsqueeze(0).unsqueeze(0)
self.weight_h = nn.Parameter(data=kernel_h, requires_grad=False)
self.weight_v = nn.Parameter(data=kernel_v, requires_grad=False)
def forward(self, x):
x_list = []
for i in range(x.shape[1]):
x_i = x[:, i]
x_i_v = F.conv2d(x_i.unsqueeze(1), self.weight_v, padding=1)
x_i_h = F.conv2d(x_i.unsqueeze(1), self.weight_h, padding=1)
x_i = torch.sqrt(torch.pow(x_i_v, 2) + torch.pow(x_i_h, 2) + 1e-06)
x_list.append(x_i)
x = torch.cat(x_list, dim=1)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_cat_0(in_ptr0, in_ptr1, in_ptr2, in_ptr3, in_ptr4,
in_ptr5, in_ptr6, in_ptr7, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 16 % 4
x0 = xindex % 16
x2 = xindex // 64
x3 = xindex
tmp0 = x1
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 1, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (x0 + 16 * x2), tmp4 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp6 = tmp5 * tmp5
tmp7 = tl.load(in_ptr1 + (x0 + 16 * x2), tmp4 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp8 = tmp7 * tmp7
tmp9 = tmp6 + tmp8
tmp10 = 1e-06
tmp11 = tmp9 + tmp10
tmp12 = libdevice.sqrt(tmp11)
tmp13 = tl.full(tmp12.shape, 0.0, tmp12.dtype)
tmp14 = tl.where(tmp4, tmp12, tmp13)
tmp15 = tmp0 >= tmp3
tmp16 = tl.full([1], 2, tl.int64)
tmp17 = tmp0 < tmp16
tmp18 = tmp15 & tmp17
tmp19 = tl.load(in_ptr2 + (x0 + 16 * x2), tmp18 & xmask,
eviction_policy='evict_last', other=0.0)
tmp20 = tmp19 * tmp19
tmp21 = tl.load(in_ptr3 + (x0 + 16 * x2), tmp18 & xmask,
eviction_policy='evict_last', other=0.0)
tmp22 = tmp21 * tmp21
tmp23 = tmp20 + tmp22
tmp24 = tmp23 + tmp10
tmp25 = libdevice.sqrt(tmp24)
tmp26 = tl.full(tmp25.shape, 0.0, tmp25.dtype)
tmp27 = tl.where(tmp18, tmp25, tmp26)
tmp28 = tmp0 >= tmp16
tmp29 = tl.full([1], 3, tl.int64)
tmp30 = tmp0 < tmp29
tmp31 = tmp28 & tmp30
tmp32 = tl.load(in_ptr4 + (x0 + 16 * x2), tmp31 & xmask,
eviction_policy='evict_last', other=0.0)
tmp33 = tmp32 * tmp32
tmp34 = tl.load(in_ptr5 + (x0 + 16 * x2), tmp31 & xmask,
eviction_policy='evict_last', other=0.0)
tmp35 = tmp34 * tmp34
tmp36 = tmp33 + tmp35
tmp37 = tmp36 + tmp10
tmp38 = libdevice.sqrt(tmp37)
tmp39 = tl.full(tmp38.shape, 0.0, tmp38.dtype)
tmp40 = tl.where(tmp31, tmp38, tmp39)
tmp41 = tmp0 >= tmp29
tl.full([1], 4, tl.int64)
tmp44 = tl.load(in_ptr6 + (x0 + 16 * x2), tmp41 & xmask,
eviction_policy='evict_last', other=0.0)
tmp45 = tmp44 * tmp44
tmp46 = tl.load(in_ptr7 + (x0 + 16 * x2), tmp41 & xmask,
eviction_policy='evict_last', other=0.0)
tmp47 = tmp46 * tmp46
tmp48 = tmp45 + tmp47
tmp49 = tmp48 + tmp10
tmp50 = libdevice.sqrt(tmp49)
tmp51 = tl.full(tmp50.shape, 0.0, tmp50.dtype)
tmp52 = tl.where(tmp41, tmp50, tmp51)
tmp53 = tl.where(tmp31, tmp40, tmp52)
tmp54 = tl.where(tmp18, tmp27, tmp53)
tmp55 = tl.where(tmp4, tmp14, tmp54)
tl.store(out_ptr0 + x3, tmp55, xmask)
def call(args):
arg0_1, arg1_1, arg2_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (1, 1, 3, 3), (9, 9, 3, 1))
assert_size_stride(arg2_1, (1, 1, 3, 3), (9, 9, 3, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(reinterpret_tensor(arg0_1, (4, 1,
4, 4), (64, 0, 4, 1), 0), arg1_1, stride=(1, 1), padding=(1, 1),
dilation=(1, 1), transposed=False, output_padding=(0, 0),
groups=1, bias=None)
assert_size_stride(buf0, (4, 1, 4, 4), (16, 16, 4, 1))
buf1 = extern_kernels.convolution(reinterpret_tensor(arg0_1, (4, 1,
4, 4), (64, 0, 4, 1), 0), arg2_1, stride=(1, 1), padding=(1, 1),
dilation=(1, 1), transposed=False, output_padding=(0, 0),
groups=1, bias=None)
assert_size_stride(buf1, (4, 1, 4, 4), (16, 16, 4, 1))
buf2 = extern_kernels.convolution(reinterpret_tensor(arg0_1, (4, 1,
4, 4), (64, 0, 4, 1), 16), arg1_1, stride=(1, 1), padding=(1, 1
), dilation=(1, 1), transposed=False, output_padding=(0, 0),
groups=1, bias=None)
assert_size_stride(buf2, (4, 1, 4, 4), (16, 16, 4, 1))
buf3 = extern_kernels.convolution(reinterpret_tensor(arg0_1, (4, 1,
4, 4), (64, 0, 4, 1), 16), arg2_1, stride=(1, 1), padding=(1, 1
), dilation=(1, 1), transposed=False, output_padding=(0, 0),
groups=1, bias=None)
assert_size_stride(buf3, (4, 1, 4, 4), (16, 16, 4, 1))
buf4 = extern_kernels.convolution(reinterpret_tensor(arg0_1, (4, 1,
4, 4), (64, 0, 4, 1), 32), arg1_1, stride=(1, 1), padding=(1, 1
), dilation=(1, 1), transposed=False, output_padding=(0, 0),
groups=1, bias=None)
assert_size_stride(buf4, (4, 1, 4, 4), (16, 16, 4, 1))
buf5 = extern_kernels.convolution(reinterpret_tensor(arg0_1, (4, 1,
4, 4), (64, 0, 4, 1), 32), arg2_1, stride=(1, 1), padding=(1, 1
), dilation=(1, 1), transposed=False, output_padding=(0, 0),
groups=1, bias=None)
assert_size_stride(buf5, (4, 1, 4, 4), (16, 16, 4, 1))
buf6 = extern_kernels.convolution(reinterpret_tensor(arg0_1, (4, 1,
4, 4), (64, 0, 4, 1), 48), arg1_1, stride=(1, 1), padding=(1, 1
), dilation=(1, 1), transposed=False, output_padding=(0, 0),
groups=1, bias=None)
assert_size_stride(buf6, (4, 1, 4, 4), (16, 16, 4, 1))
del arg1_1
buf7 = extern_kernels.convolution(reinterpret_tensor(arg0_1, (4, 1,
4, 4), (64, 0, 4, 1), 48), arg2_1, stride=(1, 1), padding=(1, 1
), dilation=(1, 1), transposed=False, output_padding=(0, 0),
groups=1, bias=None)
assert_size_stride(buf7, (4, 1, 4, 4), (16, 16, 4, 1))
del arg0_1
del arg2_1
buf8 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_cat_0[grid(256)](buf0, buf1, buf2, buf3, buf4,
buf5, buf6, buf7, buf8, 256, XBLOCK=128, num_warps=4, num_stages=1)
del buf0
del buf1
del buf2
del buf3
del buf4
del buf5
del buf6
del buf7
return buf8,
class Get_gradient_nopaddingNew(nn.Module):
def __init__(self):
super(Get_gradient_nopaddingNew, self).__init__()
kernel_v = [[0, -1, 0], [0, 0, 0], [0, 1, 0]]
kernel_h = [[0, 0, 0], [-1, 0, 1], [0, 0, 0]]
kernel_h = torch.FloatTensor(kernel_h).unsqueeze(0).unsqueeze(0)
kernel_v = torch.FloatTensor(kernel_v).unsqueeze(0).unsqueeze(0)
self.weight_h = nn.Parameter(data=kernel_h, requires_grad=False)
self.weight_v = nn.Parameter(data=kernel_v, requires_grad=False)
def forward(self, input_0):
arg1_1 = self.weight_h
arg2_1 = self.weight_v
arg0_1 = input_0
output = call([arg0_1, arg1_1, arg2_1])
return output[0]
|
JoeyBallentine/ESRGAN
|
Get_gradient_nopadding
| false
| 13,905
|
[
"Apache-2.0"
] | 95
|
9000b43e3acf8709626f45951bb91ace1d983359
|
https://github.com/JoeyBallentine/ESRGAN/tree/9000b43e3acf8709626f45951bb91ace1d983359
|
LinearRegression
|
import torch
import torch.nn as nn
class LinearRegression(nn.Module):
def __init__(self):
super().__init__()
self.a = nn.Parameter(torch.randn(1, requires_grad=True, dtype=
torch.float))
self.b = nn.Parameter(torch.randn(1, requires_grad=True, dtype=
torch.float))
def forward(self, x):
return self.a + self.b * x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_add_mul_0(in_ptr0, in_ptr1, in_ptr2, out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + 0)
tmp1 = tl.broadcast_to(tmp0, [XBLOCK])
tmp2 = tl.load(in_ptr1 + 0)
tmp3 = tl.broadcast_to(tmp2, [XBLOCK])
tmp4 = tl.load(in_ptr2 + x0, xmask)
tmp5 = tmp3 * tmp4
tmp6 = tmp1 + tmp5
tl.store(out_ptr0 + x0, tmp6, xmask)
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (1,), (1,))
assert_size_stride(primals_2, (1,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_add_mul_0[grid(256)](primals_1, primals_2,
primals_3, buf0, 256, XBLOCK=128, num_warps=4, num_stages=1)
del primals_1
del primals_2
return buf0, primals_3
class LinearRegressionNew(nn.Module):
def __init__(self):
super().__init__()
self.a = nn.Parameter(torch.randn(1, requires_grad=True, dtype=
torch.float))
self.b = nn.Parameter(torch.randn(1, requires_grad=True, dtype=
torch.float))
def forward(self, input_0):
primals_1 = self.a
primals_2 = self.b
primals_3 = input_0
output = call([primals_1, primals_2, primals_3])
return output[0]
|
JohnlNguyen/FLSim
|
LinearRegression
| false
| 13,906
|
[
"BSD-3-Clause"
] | 79
|
a5ed7c0b84499cd9dbc5fe95f8bcb4ba8ab5a5cb
|
https://github.com/JohnlNguyen/FLSim/tree/a5ed7c0b84499cd9dbc5fe95f8bcb4ba8ab5a5cb
|
ModMBStddevLayer
|
import torch
import torch.nn as nn
class ModMBStddevLayer(nn.Module):
"""Modified MiniBatch Stddev Layer.
This layer is modified from ``MiniBatchStddevLayer`` used in PGGAN. In
StyleGAN2, the authors add a new feature, `channel_groups`, into this
layer.
"""
def __init__(self, group_size=4, channel_groups=1, sync_groups=None,
eps=1e-08):
super(ModMBStddevLayer, self).__init__()
self.group_size = group_size
self.eps = eps
self.channel_groups = channel_groups
self.sync_groups = group_size if sync_groups is None else sync_groups
def forward(self, x):
assert x.shape[0] <= self.group_size or x.shape[0
] % self.group_size == 0, f'Batch size be smaller than or equal to group size. Otherwise, batch size should be divisible by the group size.But got batch size {x.shape[0]}, group size {self.group_size}'
assert x.shape[1
] % self.channel_groups == 0, f'"channel_groups" must be divided by the feature channels. channel_groups: {self.channel_groups}, feature channels: {x.shape[1]}'
n, c, h, w = x.shape
group_size = min(n, self.group_size)
y = torch.reshape(x, (group_size, -1, self.channel_groups, c //
self.channel_groups, h, w))
y = torch.var(y, dim=0, unbiased=False)
y = torch.sqrt(y + self.eps)
y = y.mean(dim=(2, 3, 4), keepdim=True).squeeze(2)
y = y.repeat(group_size, 1, h, w)
return torch.cat([x, y], dim=1)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_per_fused_add_mean_repeat_sqrt_var_0(in_ptr0, out_ptr1, xnumel,
rnumel, XBLOCK: tl.constexpr):
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
r1 = rindex % 16
r2 = rindex // 16
tmp0 = tl.load(in_ptr0 + r0, None)
tmp1 = tl.load(in_ptr0 + (64 + r0), None)
tmp3 = tl.load(in_ptr0 + (128 + r0), None)
tmp5 = tl.load(in_ptr0 + (192 + r0), None)
tmp2 = tmp0 + tmp1
tmp4 = tmp2 + tmp3
tmp6 = tmp4 + tmp5
tmp7 = 4.0
tmp8 = tmp6 / tmp7
tmp9 = tmp0 - tmp8
tmp10 = tmp9 * tmp9
tmp11 = tmp1 - tmp8
tmp12 = tmp11 * tmp11
tmp13 = tmp10 + tmp12
tmp14 = tmp3 - tmp8
tmp15 = tmp14 * tmp14
tmp16 = tmp13 + tmp15
tmp17 = tmp5 - tmp8
tmp18 = tmp17 * tmp17
tmp19 = tmp16 + tmp18
tmp20 = tmp19 / tmp7
tmp21 = 1e-08
tmp22 = tmp20 + tmp21
tmp23 = libdevice.sqrt(tmp22)
tmp24 = tl.broadcast_to(tmp23, [XBLOCK, RBLOCK])
tmp26 = tl.sum(tmp24, 1)[:, None]
tmp27 = 64.0
tmp28 = tmp26 / tmp27
tl.store(out_ptr1 + tl.broadcast_to(r1 + 80 * r2, [XBLOCK, RBLOCK]),
tmp28, None)
@triton.jit
def triton_poi_fused_cat_1(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 64
x1 = xindex // 64
tmp0 = tl.load(in_ptr0 + x2, xmask)
tl.store(out_ptr0 + (x0 + 80 * x1), tmp0, xmask)
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf3 = empty_strided_cuda((4, 5, 4, 4), (80, 16, 4, 1), torch.float32)
buf2 = reinterpret_tensor(buf3, (4, 1, 4, 4), (80, 16, 4, 1), 64)
get_raw_stream(0)
triton_per_fused_add_mean_repeat_sqrt_var_0[grid(1)](arg0_1, buf2,
1, 64, XBLOCK=1, num_warps=2, num_stages=1)
buf1 = reinterpret_tensor(buf3, (4, 4, 4, 4), (80, 16, 4, 1), 0)
triton_poi_fused_cat_1[grid(256)](arg0_1, buf1, 256, XBLOCK=256,
num_warps=4, num_stages=1)
del arg0_1
return buf3,
class ModMBStddevLayerNew(nn.Module):
"""Modified MiniBatch Stddev Layer.
This layer is modified from ``MiniBatchStddevLayer`` used in PGGAN. In
StyleGAN2, the authors add a new feature, `channel_groups`, into this
layer.
"""
def __init__(self, group_size=4, channel_groups=1, sync_groups=None,
eps=1e-08):
super(ModMBStddevLayerNew, self).__init__()
self.group_size = group_size
self.eps = eps
self.channel_groups = channel_groups
self.sync_groups = group_size if sync_groups is None else sync_groups
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
|
Juggernaut93/mmediting
|
ModMBStddevLayer
| false
| 13,907
|
[
"Apache-2.0"
] | 1,884
|
8ef46ace29756dd2df1d92f2f73a33646e33e007
|
https://github.com/Juggernaut93/mmediting/tree/8ef46ace29756dd2df1d92f2f73a33646e33e007
|
ValueNetwork
|
import torch
import torch.nn as nn
import torch.nn.functional as F
class ValueNetwork(nn.Module):
def __init__(self, state_dim, hidden_dim, init_w=0.003):
super(ValueNetwork, self).__init__()
self.linear1 = nn.Linear(state_dim, hidden_dim)
self.linear2 = nn.Linear(hidden_dim, hidden_dim)
self.linear3 = nn.Linear(hidden_dim, hidden_dim)
self.linear4 = nn.Linear(hidden_dim, 1)
self.linear4.weight.data.uniform_(-init_w, init_w)
self.linear4.bias.data.uniform_(-init_w, init_w)
def forward(self, state):
x = F.relu(self.linear1(state))
x = F.relu(self.linear2(x))
x = F.relu(self.linear3(x))
x = self.linear4(x)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'state_dim': 4, 'hidden_dim': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_relu_threshold_backward_0(in_out_ptr0, in_ptr0,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_out_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(in_out_ptr0 + x2, tmp4, xmask)
tl.store(out_ptr0 + x2, tmp6, xmask)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7, primals_8, primals_9) = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (4, 4), (4, 1))
assert_size_stride(primals_5, (4,), (1,))
assert_size_stride(primals_6, (4, 4), (4, 1))
assert_size_stride(primals_7, (4,), (1,))
assert_size_stride(primals_8, (1, 4), (4, 1))
assert_size_stride(primals_9, (1,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_3, (64, 4), (4, 1), 0),
reinterpret_tensor(primals_1, (4, 4), (1, 4), 0), out=buf0)
del primals_1
buf1 = reinterpret_tensor(buf0, (4, 4, 4, 4), (64, 16, 4, 1), 0)
del buf0
buf10 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
get_raw_stream(0)
triton_poi_fused_relu_threshold_backward_0[grid(256)](buf1,
primals_2, buf10, 256, XBLOCK=256, num_warps=4, num_stages=1)
del primals_2
buf2 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf1, (64, 4), (4, 1), 0),
reinterpret_tensor(primals_4, (4, 4), (1, 4), 0), out=buf2)
buf3 = reinterpret_tensor(buf2, (4, 4, 4, 4), (64, 16, 4, 1), 0)
del buf2
buf9 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
triton_poi_fused_relu_threshold_backward_0[grid(256)](buf3,
primals_5, buf9, 256, XBLOCK=256, num_warps=4, num_stages=1)
del primals_5
buf4 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf3, (64, 4), (4, 1), 0),
reinterpret_tensor(primals_6, (4, 4), (1, 4), 0), out=buf4)
buf5 = reinterpret_tensor(buf4, (4, 4, 4, 4), (64, 16, 4, 1), 0)
del buf4
buf8 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
triton_poi_fused_relu_threshold_backward_0[grid(256)](buf5,
primals_7, buf8, 256, XBLOCK=256, num_warps=4, num_stages=1)
del primals_7
buf7 = empty_strided_cuda((64, 1), (1, 1), torch.float32)
extern_kernels.addmm(primals_9, reinterpret_tensor(buf5, (64, 4), (
4, 1), 0), reinterpret_tensor(primals_8, (4, 1), (1, 4), 0),
alpha=1, beta=1, out=buf7)
del primals_9
return reinterpret_tensor(buf7, (4, 4, 4, 1), (16, 4, 1, 1), 0
), reinterpret_tensor(primals_3, (64, 4), (4, 1), 0
), reinterpret_tensor(buf1, (64, 4), (4, 1), 0), reinterpret_tensor(
buf3, (64, 4), (4, 1), 0), reinterpret_tensor(buf5, (64, 4), (4, 1), 0
), primals_8, buf8, primals_6, buf9, primals_4, buf10
class ValueNetworkNew(nn.Module):
def __init__(self, state_dim, hidden_dim, init_w=0.003):
super(ValueNetworkNew, self).__init__()
self.linear1 = nn.Linear(state_dim, hidden_dim)
self.linear2 = nn.Linear(hidden_dim, hidden_dim)
self.linear3 = nn.Linear(hidden_dim, hidden_dim)
self.linear4 = nn.Linear(hidden_dim, 1)
self.linear4.weight.data.uniform_(-init_w, init_w)
self.linear4.bias.data.uniform_(-init_w, init_w)
def forward(self, input_0):
primals_1 = self.linear1.weight
primals_2 = self.linear1.bias
primals_4 = self.linear2.weight
primals_5 = self.linear2.bias
primals_6 = self.linear3.weight
primals_7 = self.linear3.bias
primals_8 = self.linear4.weight
primals_9 = self.linear4.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7, primals_8, primals_9])
return output[0]
|
JieRen98/Popular-RL-Algorithms
|
ValueNetwork
| false
| 13,908
|
[
"Apache-2.0"
] | 273
|
7f2bb74a51cf9cbde92a6ccfa42e97dc129dd145
|
https://github.com/JieRen98/Popular-RL-Algorithms/tree/7f2bb74a51cf9cbde92a6ccfa42e97dc129dd145
|
MSECompositionLoss
|
import functools
import torch
import torch.nn as nn
from torch.nn import functional as F
def reduce_loss(loss, reduction):
"""Reduce loss as specified.
Args:
loss (Tensor): Elementwise loss tensor.
reduction (str): Options are "none", "mean" and "sum".
Returns:
Tensor: Reduced loss tensor.
"""
reduction_enum = F._Reduction.get_enum(reduction)
if reduction_enum == 0:
return loss
if reduction_enum == 1:
return loss.mean()
return loss.sum()
def mask_reduce_loss(loss, weight=None, reduction='mean', sample_wise=False):
"""Apply element-wise weight and reduce loss.
Args:
loss (Tensor): Element-wise loss.
weight (Tensor): Element-wise weights. Default: None.
reduction (str): Same as built-in losses of PyTorch. Options are
"none", "mean" and "sum". Default: 'mean'.
sample_wise (bool): Whether calculate the loss sample-wise. This
argument only takes effect when `reduction` is 'mean' and `weight`
(argument of `forward()`) is not None. It will first reduces loss
with 'mean' per-sample, and then it means over all the samples.
Default: False.
Returns:
Tensor: Processed loss values.
"""
if weight is not None:
assert weight.dim() == loss.dim()
assert weight.size(1) == 1 or weight.size(1) == loss.size(1)
loss = loss * weight
if weight is None or reduction == 'sum':
loss = reduce_loss(loss, reduction)
elif reduction == 'mean':
if weight.size(1) == 1:
weight = weight.expand_as(loss)
eps = 1e-12
if sample_wise:
weight = weight.sum(dim=[1, 2, 3], keepdim=True)
loss = (loss / (weight + eps)).sum() / weight.size(0)
else:
loss = loss.sum() / (weight.sum() + eps)
return loss
def masked_loss(loss_func):
"""Create a masked version of a given loss function.
To use this decorator, the loss function must have the signature like
`loss_func(pred, target, **kwargs)`. The function only needs to compute
element-wise loss without any reduction. This decorator will add weight
and reduction arguments to the function. The decorated function will have
the signature like `loss_func(pred, target, weight=None, reduction='mean',
avg_factor=None, **kwargs)`.
:Example:
>>> import torch
>>> @masked_loss
>>> def l1_loss(pred, target):
>>> return (pred - target).abs()
>>> pred = torch.Tensor([0, 2, 3])
>>> target = torch.Tensor([1, 1, 1])
>>> weight = torch.Tensor([1, 0, 1])
>>> l1_loss(pred, target)
tensor(1.3333)
>>> l1_loss(pred, target, weight)
tensor(1.5000)
>>> l1_loss(pred, target, reduction='none')
tensor([1., 1., 2.])
>>> l1_loss(pred, target, weight, reduction='sum')
tensor(3.)
"""
@functools.wraps(loss_func)
def wrapper(pred, target, weight=None, reduction='mean', sample_wise=
False, **kwargs):
loss = loss_func(pred, target, **kwargs)
loss = mask_reduce_loss(loss, weight, reduction, sample_wise)
return loss
return wrapper
@masked_loss
def mse_loss(pred, target):
"""MSE loss.
Args:
pred (Tensor): Prediction Tensor with shape (n, c, h, w).
target ([type]): Target Tensor with shape (n, c, h, w).
Returns:
Tensor: Calculated MSE loss.
"""
return F.mse_loss(pred, target, reduction='none')
class MSECompositionLoss(nn.Module):
"""MSE (L2) composition loss.
Args:
loss_weight (float): Loss weight for MSE loss. Default: 1.0.
reduction (str): Specifies the reduction to apply to the output.
Supported choices are 'none' | 'mean' | 'sum'. Default: 'mean'.
sample_wise (bool): Whether calculate the loss sample-wise. This
argument only takes effect when `reduction` is 'mean' and `weight`
(argument of `forward()`) is not None. It will first reduces loss
with 'mean' per-sample, and then it means over all the samples.
Default: False.
"""
def __init__(self, loss_weight=1.0, reduction='mean', sample_wise=False):
super().__init__()
if reduction not in ['none', 'mean', 'sum']:
raise ValueError(
f'Unsupported reduction mode: {reduction}. Supported ones are: {_reduction_modes}'
)
self.loss_weight = loss_weight
self.reduction = reduction
self.sample_wise = sample_wise
def forward(self, pred_alpha, fg, bg, ori_merged, weight=None, **kwargs):
"""
Args:
pred_alpha (Tensor): of shape (N, 1, H, W). Predicted alpha matte.
fg (Tensor): of shape (N, 3, H, W). Tensor of foreground object.
bg (Tensor): of shape (N, 3, H, W). Tensor of background object.
ori_merged (Tensor): of shape (N, 3, H, W). Tensor of origin merged
image before normalized by ImageNet mean and std.
weight (Tensor, optional): of shape (N, 1, H, W). It is an
indicating matrix: weight[trimap == 128] = 1. Default: None.
"""
pred_merged = pred_alpha * fg + (1.0 - pred_alpha) * bg
if weight is not None:
weight = weight.expand(-1, 3, -1, -1)
return self.loss_weight * mse_loss(pred_merged, ori_merged, weight,
reduction=self.reduction, sample_wise=self.sample_wise)
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4]), torch.rand(
[4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
import functools
import torch.nn as nn
from torch.nn import functional as F
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_per_fused_add_mean_mse_loss_mul_rsub_0(in_out_ptr0, in_ptr0,
in_ptr1, in_ptr2, in_ptr3, xnumel, rnumel):
XBLOCK: tl.constexpr = 1
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
tl.full([1], xoffset, tl.int32)
tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
tl.full([RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp1 = tl.load(in_ptr1 + r0, None)
tmp5 = tl.load(in_ptr2 + r0, None)
tmp8 = tl.load(in_ptr3 + r0, None)
tmp2 = tmp0 * tmp1
tmp3 = 1.0
tmp4 = tmp3 - tmp0
tmp6 = tmp4 * tmp5
tmp7 = tmp2 + tmp6
tmp9 = tmp7 - tmp8
tmp10 = tmp9 * tmp9
tmp11 = tl.broadcast_to(tmp10, [RBLOCK])
tmp13 = triton_helpers.promote_to_tensor(tl.sum(tmp11, 0))
tmp14 = 256.0
tmp15 = tmp13 / tmp14
tmp16 = tmp15 * tmp3
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([1], 0, tl.int32), tmp16, None)
def call(args):
arg0_1, arg1_1, arg2_1, arg3_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg2_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg3_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((), (), torch.float32)
buf1 = buf0
del buf0
get_raw_stream(0)
triton_per_fused_add_mean_mse_loss_mul_rsub_0[grid(1)](buf1, arg0_1,
arg1_1, arg2_1, arg3_1, 1, 256, num_warps=2, num_stages=1)
del arg0_1
del arg1_1
del arg2_1
del arg3_1
return buf1,
def reduce_loss(loss, reduction):
"""Reduce loss as specified.
Args:
loss (Tensor): Elementwise loss tensor.
reduction (str): Options are "none", "mean" and "sum".
Returns:
Tensor: Reduced loss tensor.
"""
reduction_enum = F._Reduction.get_enum(reduction)
if reduction_enum == 0:
return loss
if reduction_enum == 1:
return loss.mean()
return loss.sum()
def mask_reduce_loss(loss, weight=None, reduction='mean', sample_wise=False):
"""Apply element-wise weight and reduce loss.
Args:
loss (Tensor): Element-wise loss.
weight (Tensor): Element-wise weights. Default: None.
reduction (str): Same as built-in losses of PyTorch. Options are
"none", "mean" and "sum". Default: 'mean'.
sample_wise (bool): Whether calculate the loss sample-wise. This
argument only takes effect when `reduction` is 'mean' and `weight`
(argument of `forward()`) is not None. It will first reduces loss
with 'mean' per-sample, and then it means over all the samples.
Default: False.
Returns:
Tensor: Processed loss values.
"""
if weight is not None:
assert weight.dim() == loss.dim()
assert weight.size(1) == 1 or weight.size(1) == loss.size(1)
loss = loss * weight
if weight is None or reduction == 'sum':
loss = reduce_loss(loss, reduction)
elif reduction == 'mean':
if weight.size(1) == 1:
weight = weight.expand_as(loss)
eps = 1e-12
if sample_wise:
weight = weight.sum(dim=[1, 2, 3], keepdim=True)
loss = (loss / (weight + eps)).sum() / weight.size(0)
else:
loss = loss.sum() / (weight.sum() + eps)
return loss
def masked_loss(loss_func):
"""Create a masked version of a given loss function.
To use this decorator, the loss function must have the signature like
`loss_func(pred, target, **kwargs)`. The function only needs to compute
element-wise loss without any reduction. This decorator will add weight
and reduction arguments to the function. The decorated function will have
the signature like `loss_func(pred, target, weight=None, reduction='mean',
avg_factor=None, **kwargs)`.
:Example:
>>> import torch
>>> @masked_loss
>>> def l1_loss(pred, target):
>>> return (pred - target).abs()
>>> pred = torch.Tensor([0, 2, 3])
>>> target = torch.Tensor([1, 1, 1])
>>> weight = torch.Tensor([1, 0, 1])
>>> l1_loss(pred, target)
tensor(1.3333)
>>> l1_loss(pred, target, weight)
tensor(1.5000)
>>> l1_loss(pred, target, reduction='none')
tensor([1., 1., 2.])
>>> l1_loss(pred, target, weight, reduction='sum')
tensor(3.)
"""
@functools.wraps(loss_func)
def wrapper(pred, target, weight=None, reduction='mean', sample_wise=
False, **kwargs):
loss = loss_func(pred, target, **kwargs)
loss = mask_reduce_loss(loss, weight, reduction, sample_wise)
return loss
return wrapper
@masked_loss
def mse_loss(pred, target):
"""MSE loss.
Args:
pred (Tensor): Prediction Tensor with shape (n, c, h, w).
target ([type]): Target Tensor with shape (n, c, h, w).
Returns:
Tensor: Calculated MSE loss.
"""
return F.mse_loss(pred, target, reduction='none')
class MSECompositionLossNew(nn.Module):
"""MSE (L2) composition loss.
Args:
loss_weight (float): Loss weight for MSE loss. Default: 1.0.
reduction (str): Specifies the reduction to apply to the output.
Supported choices are 'none' | 'mean' | 'sum'. Default: 'mean'.
sample_wise (bool): Whether calculate the loss sample-wise. This
argument only takes effect when `reduction` is 'mean' and `weight`
(argument of `forward()`) is not None. It will first reduces loss
with 'mean' per-sample, and then it means over all the samples.
Default: False.
"""
def __init__(self, loss_weight=1.0, reduction='mean', sample_wise=False):
super().__init__()
if reduction not in ['none', 'mean', 'sum']:
raise ValueError(
f'Unsupported reduction mode: {reduction}. Supported ones are: {_reduction_modes}'
)
self.loss_weight = loss_weight
self.reduction = reduction
self.sample_wise = sample_wise
def forward(self, input_0, input_1, input_2, input_3):
arg0_1 = input_0
arg1_1 = input_1
arg2_1 = input_2
arg3_1 = input_3
output = call([arg0_1, arg1_1, arg2_1, arg3_1])
return output[0]
|
Juggernaut93/mmediting
|
MSECompositionLoss
| false
| 13,909
|
[
"Apache-2.0"
] | 1,884
|
8ef46ace29756dd2df1d92f2f73a33646e33e007
|
https://github.com/Juggernaut93/mmediting/tree/8ef46ace29756dd2df1d92f2f73a33646e33e007
|
PlainRefiner
|
import torch
import torch.nn as nn
class PlainRefiner(nn.Module):
"""Simple refiner from Deep Image Matting.
Args:
conv_channels (int): Number of channels produced by the three main
convolutional layer.
loss_refine (dict): Config of the loss of the refiner. Default: None.
pretrained (str): Name of pretrained model. Default: None.
"""
def __init__(self, conv_channels=64, pretrained=None):
super().__init__()
assert pretrained is None, 'pretrained not supported yet'
self.refine_conv1 = nn.Conv2d(4, conv_channels, kernel_size=3,
padding=1)
self.refine_conv2 = nn.Conv2d(conv_channels, conv_channels,
kernel_size=3, padding=1)
self.refine_conv3 = nn.Conv2d(conv_channels, conv_channels,
kernel_size=3, padding=1)
self.refine_pred = nn.Conv2d(conv_channels, 1, kernel_size=3, padding=1
)
self.relu = nn.ReLU(inplace=True)
def init_weights(self):
for m in self.modules():
if isinstance(m, nn.Conv2d):
xavier_init(m)
def forward(self, x, raw_alpha):
"""Forward function.
Args:
x (Tensor): The input feature map of refiner.
raw_alpha (Tensor): The raw predicted alpha matte.
Returns:
Tensor: The refined alpha matte.
"""
out = self.relu(self.refine_conv1(x))
out = self.relu(self.refine_conv2(out))
out = self.relu(self.refine_conv3(out))
raw_refine = self.refine_pred(out)
pred_refine = torch.sigmoid(raw_alpha + raw_refine)
return pred_refine
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_convolution_relu_0(in_out_ptr0, in_ptr0, xnumel,
XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x3 = xindex
x1 = xindex // 16 % 64
tmp0 = tl.load(in_out_ptr0 + x3, None)
tmp1 = tl.load(in_ptr0 + x1, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + x3, tmp4, None)
@triton.jit
def triton_poi_fused_add_convolution_sigmoid_1(in_ptr0, in_ptr1, in_ptr2,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x0 = xindex % 16
x2 = xindex // 64
tmp0 = tl.load(in_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr1 + (x0 + 16 * x2), xmask, eviction_policy=
'evict_last')
tmp2 = tl.load(in_ptr2 + 0)
tmp3 = tl.broadcast_to(tmp2, [XBLOCK])
tmp4 = tmp1 + tmp3
tmp5 = tmp0 + tmp4
tmp6 = tl.sigmoid(tmp5)
tl.store(out_ptr0 + x3, tmp6, xmask)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7, primals_8, primals_9, primals_10) = args
args.clear()
assert_size_stride(primals_1, (64, 4, 3, 3), (36, 9, 3, 1))
assert_size_stride(primals_2, (64,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_5, (64,), (1,))
assert_size_stride(primals_6, (64, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_7, (64,), (1,))
assert_size_stride(primals_8, (1, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_9, (1,), (1,))
assert_size_stride(primals_10, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(primals_3, primals_1, stride=(1,
1), padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 64, 4, 4), (1024, 16, 4, 1))
buf1 = buf0
del buf0
get_raw_stream(0)
triton_poi_fused_convolution_relu_0[grid(4096)](buf1, primals_2,
4096, XBLOCK=256, num_warps=4, num_stages=1)
del primals_2
buf2 = extern_kernels.convolution(buf1, primals_4, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf2, (4, 64, 4, 4), (1024, 16, 4, 1))
buf3 = buf2
del buf2
triton_poi_fused_convolution_relu_0[grid(4096)](buf3, primals_5,
4096, XBLOCK=256, num_warps=4, num_stages=1)
del primals_5
buf4 = extern_kernels.convolution(buf3, primals_6, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf4, (4, 64, 4, 4), (1024, 16, 4, 1))
buf5 = buf4
del buf4
triton_poi_fused_convolution_relu_0[grid(4096)](buf5, primals_7,
4096, XBLOCK=256, num_warps=4, num_stages=1)
del primals_7
buf6 = extern_kernels.convolution(buf5, primals_8, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf6, (4, 1, 4, 4), (16, 16, 4, 1))
buf7 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused_add_convolution_sigmoid_1[grid(256)](primals_10,
buf6, primals_9, buf7, 256, XBLOCK=256, num_warps=4, num_stages=1)
del buf6
del primals_10
del primals_9
return (buf7, primals_1, primals_3, primals_4, primals_6, primals_8,
buf1, buf3, buf5, buf7)
class PlainRefinerNew(nn.Module):
"""Simple refiner from Deep Image Matting.
Args:
conv_channels (int): Number of channels produced by the three main
convolutional layer.
loss_refine (dict): Config of the loss of the refiner. Default: None.
pretrained (str): Name of pretrained model. Default: None.
"""
def __init__(self, conv_channels=64, pretrained=None):
super().__init__()
assert pretrained is None, 'pretrained not supported yet'
self.refine_conv1 = nn.Conv2d(4, conv_channels, kernel_size=3,
padding=1)
self.refine_conv2 = nn.Conv2d(conv_channels, conv_channels,
kernel_size=3, padding=1)
self.refine_conv3 = nn.Conv2d(conv_channels, conv_channels,
kernel_size=3, padding=1)
self.refine_pred = nn.Conv2d(conv_channels, 1, kernel_size=3, padding=1
)
self.relu = nn.ReLU(inplace=True)
def init_weights(self):
for m in self.modules():
if isinstance(m, nn.Conv2d):
xavier_init(m)
def forward(self, input_0, input_1):
primals_1 = self.refine_conv1.weight
primals_2 = self.refine_conv1.bias
primals_4 = self.refine_conv2.weight
primals_5 = self.refine_conv2.bias
primals_6 = self.refine_conv3.weight
primals_7 = self.refine_conv3.bias
primals_8 = self.refine_pred.weight
primals_9 = self.refine_pred.bias
primals_3 = input_0
primals_10 = input_1
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7, primals_8, primals_9, primals_10])
return output[0]
|
Juggernaut93/mmediting
|
PlainRefiner
| false
| 13,910
|
[
"Apache-2.0"
] | 1,884
|
8ef46ace29756dd2df1d92f2f73a33646e33e007
|
https://github.com/Juggernaut93/mmediting/tree/8ef46ace29756dd2df1d92f2f73a33646e33e007
|
Transformer
|
import torch
from torch import nn
import torch.nn.functional as F
import torch.utils.data
class Transformer(nn.Module):
def __init__(self, in_channels, out_channels):
super(Transformer, self).__init__()
self.T_sigma = nn.Linear(in_channels, out_channels)
self.T_gamma = nn.Linear(in_channels, out_channels)
def forward(self, sigma, gamma):
sigma_out = self.T_sigma(sigma)
gamma_out = self.T_gamma(gamma)
return F.softplus(sigma_out), F.softplus(gamma_out)
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_channels': 4, 'out_channels': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch import nn
import torch.utils.data
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_softplus_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = 20.0
tmp2 = tmp0 > tmp1
tmp3 = tl_math.exp(tmp0)
tmp4 = libdevice.log1p(tmp3)
tmp5 = tl.where(tmp2, tmp0, tmp4)
tl.store(out_ptr0 + x0, tmp5, xmask)
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6 = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (4, 4), (4, 1))
assert_size_stride(primals_5, (4,), (1,))
assert_size_stride(primals_6, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_2, reinterpret_tensor(primals_3, (64,
4), (4, 1), 0), reinterpret_tensor(primals_1, (4, 4), (1, 4), 0
), alpha=1, beta=1, out=buf0)
del primals_1
del primals_2
buf1 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_5, reinterpret_tensor(primals_6, (64,
4), (4, 1), 0), reinterpret_tensor(primals_4, (4, 4), (1, 4), 0
), alpha=1, beta=1, out=buf1)
del primals_4
del primals_5
buf2 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_softplus_0[grid(256)](buf0, buf2, 256, XBLOCK=256,
num_warps=4, num_stages=1)
buf3 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused_softplus_0[grid(256)](buf1, buf3, 256, XBLOCK=256,
num_warps=4, num_stages=1)
return buf2, buf3, reinterpret_tensor(primals_3, (64, 4), (4, 1), 0
), buf0, reinterpret_tensor(primals_6, (64, 4), (4, 1), 0), buf1
class TransformerNew(nn.Module):
def __init__(self, in_channels, out_channels):
super(TransformerNew, self).__init__()
self.T_sigma = nn.Linear(in_channels, out_channels)
self.T_gamma = nn.Linear(in_channels, out_channels)
def forward(self, input_0, input_1):
primals_1 = self.T_sigma.weight
primals_2 = self.T_sigma.bias
primals_4 = self.T_gamma.weight
primals_5 = self.T_gamma.bias
primals_3 = input_0
primals_6 = input_1
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6])
return output[0], output[1]
|
JunLi-Galios/PGGAN
|
Transformer
| false
| 13,911
|
[
"Apache-2.0"
] | 58
|
b8bd3dc44c71a985315fb82070e911378cf210db
|
https://github.com/JunLi-Galios/PGGAN/tree/b8bd3dc44c71a985315fb82070e911378cf210db
|
ReLUHyperSolver
|
import torch
import torch.nn as nn
class ReLUHyperSolver(nn.Module):
def __init__(self, in_dim, out_dim, hidden_dim=32):
super().__init__()
self.fc1 = nn.Linear(in_dim, hidden_dim)
self.fc2 = nn.Linear(hidden_dim, hidden_dim)
self.fc3 = nn.Linear(hidden_dim, out_dim)
self.a1 = nn.ReLU()
self.a2 = nn.ReLU()
def forward(self, x):
x = self.a1(self.fc1(x))
x = self.a2(self.fc2(x))
x = self.fc3(x)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_dim': 4, 'out_dim': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_relu_threshold_backward_0(in_out_ptr0, in_ptr0,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x2 = xindex
x0 = xindex % 32
tmp0 = tl.load(in_out_ptr0 + x2, None)
tmp1 = tl.load(in_ptr0 + x0, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(in_out_ptr0 + x2, tmp4, None)
tl.store(out_ptr0 + x2, tmp6, None)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7) = args
args.clear()
assert_size_stride(primals_1, (32, 4), (4, 1))
assert_size_stride(primals_2, (32,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (32, 32), (32, 1))
assert_size_stride(primals_5, (32,), (1,))
assert_size_stride(primals_6, (4, 32), (32, 1))
assert_size_stride(primals_7, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 32), (32, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_3, (64, 4), (4, 1), 0),
reinterpret_tensor(primals_1, (4, 32), (1, 4), 0), out=buf0)
del primals_1
buf1 = reinterpret_tensor(buf0, (4, 4, 4, 32), (512, 128, 32, 1), 0)
del buf0
buf6 = empty_strided_cuda((4, 4, 4, 32), (512, 128, 32, 1), torch.bool)
get_raw_stream(0)
triton_poi_fused_relu_threshold_backward_0[grid(2048)](buf1,
primals_2, buf6, 2048, XBLOCK=256, num_warps=4, num_stages=1)
del primals_2
buf2 = empty_strided_cuda((64, 32), (32, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf1, (64, 32), (32, 1), 0),
reinterpret_tensor(primals_4, (32, 32), (1, 32), 0), out=buf2)
buf3 = reinterpret_tensor(buf2, (4, 4, 4, 32), (512, 128, 32, 1), 0)
del buf2
buf5 = empty_strided_cuda((4, 4, 4, 32), (512, 128, 32, 1), torch.bool)
triton_poi_fused_relu_threshold_backward_0[grid(2048)](buf3,
primals_5, buf5, 2048, XBLOCK=256, num_warps=4, num_stages=1)
del primals_5
buf4 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_7, reinterpret_tensor(buf3, (64, 32),
(32, 1), 0), reinterpret_tensor(primals_6, (32, 4), (1, 32), 0),
alpha=1, beta=1, out=buf4)
del primals_7
return reinterpret_tensor(buf4, (4, 4, 4, 4), (64, 16, 4, 1), 0
), reinterpret_tensor(primals_3, (64, 4), (4, 1), 0
), reinterpret_tensor(buf1, (64, 32), (32, 1), 0), reinterpret_tensor(
buf3, (64, 32), (32, 1), 0), primals_6, buf5, primals_4, buf6
class ReLUHyperSolverNew(nn.Module):
def __init__(self, in_dim, out_dim, hidden_dim=32):
super().__init__()
self.fc1 = nn.Linear(in_dim, hidden_dim)
self.fc2 = nn.Linear(hidden_dim, hidden_dim)
self.fc3 = nn.Linear(hidden_dim, out_dim)
self.a1 = nn.ReLU()
self.a2 = nn.ReLU()
def forward(self, input_0):
primals_1 = self.fc1.weight
primals_2 = self.fc1.bias
primals_4 = self.fc2.weight
primals_5 = self.fc2.bias
primals_6 = self.fc3.weight
primals_7 = self.fc3.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7])
return output[0]
|
Juju-botu/diffeqml-research
|
ReLUHyperSolver
| false
| 13,912
|
[
"Apache-2.0"
] | 49
|
aa796c87447e5299ec4f25a07fc4d032afb1f63e
|
https://github.com/Juju-botu/diffeqml-research/tree/aa796c87447e5299ec4f25a07fc4d032afb1f63e
|
DilatedModel
|
import torch
from torch import nn
import torch.nn.functional as F
class DilatedModel(nn.Module):
def __init__(self, k=16):
super(DilatedModel, self).__init__()
self.conv1 = nn.Conv2d(1, k, 3, stride=1, dilation=1, padding=1)
self.conv2 = nn.Conv2d(k, k, 3, stride=1, dilation=1, padding=1)
self.conv3 = nn.Conv2d(k, k, 3, stride=1, dilation=2, padding=2)
self.conv4 = nn.Conv2d(k, k, 3, stride=1, dilation=4, padding=4)
self.conv5 = nn.Conv2d(k, k, 3, stride=1, dilation=8, padding=8)
self.conv6 = nn.Conv2d(k, k, 3, stride=1, dilation=16, padding=16)
self.conv7 = nn.Conv2d(k, k, 3, stride=1, dilation=1, padding=1)
self.conv8 = nn.Conv2d(k, 1, 1, stride=1, dilation=1, padding=0)
def forward(self, x):
h = x
h = F.relu(self.conv1(h))
assert h.shape[2:] == x.shape[2:]
h = F.relu(self.conv2(h))
assert h.shape[2:] == x.shape[2:]
h = F.relu(self.conv3(h))
assert h.shape[2:] == x.shape[2:]
h = F.relu(self.conv3(h))
assert h.shape[2:] == x.shape[2:]
h = F.relu(self.conv4(h))
assert h.shape[2:] == x.shape[2:]
h = F.relu(self.conv5(h))
assert h.shape[2:] == x.shape[2:]
h = F.relu(self.conv6(h))
assert h.shape[2:] == x.shape[2:]
h = F.relu(self.conv7(h))
assert h.shape[2:] == x.shape[2:]
h = F.relu(self.conv8(h))
assert h.shape == x.shape
return h
def get_inputs():
return [torch.rand([4, 1, 64, 64])]
def get_init_inputs():
return [[], {}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch import nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_convolution_relu_0(in_out_ptr0, in_ptr0, xnumel,
XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x3 = xindex
x1 = xindex // 4096 % 16
tmp0 = tl.load(in_out_ptr0 + x3, None)
tmp1 = tl.load(in_ptr0 + x1, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + x3, tmp4, None)
@triton.jit
def triton_poi_fused_convolution_relu_threshold_backward_1(in_out_ptr0,
in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x0 = xindex
tmp0 = tl.load(in_out_ptr0 + x0, None)
tmp1 = tl.load(in_ptr0 + 0)
tmp2 = tl.broadcast_to(tmp1, [XBLOCK])
tmp3 = tmp0 + tmp2
tmp4 = tl.full([1], 0, tl.int32)
tmp5 = triton_helpers.maximum(tmp4, tmp3)
tmp6 = 0.0
tmp7 = tmp5 <= tmp6
tl.store(in_out_ptr0 + x0, tmp5, None)
tl.store(out_ptr0 + x0, tmp7, None)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7, primals_8, primals_9, primals_10, primals_11, primals_12,
primals_13, primals_14, primals_15, primals_16, primals_17) = args
args.clear()
assert_size_stride(primals_1, (4, 1, 64, 64), (4096, 4096, 64, 1))
assert_size_stride(primals_2, (16, 1, 3, 3), (9, 9, 3, 1))
assert_size_stride(primals_3, (16,), (1,))
assert_size_stride(primals_4, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_5, (16,), (1,))
assert_size_stride(primals_6, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_7, (16,), (1,))
assert_size_stride(primals_8, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_9, (16,), (1,))
assert_size_stride(primals_10, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_11, (16,), (1,))
assert_size_stride(primals_12, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_13, (16,), (1,))
assert_size_stride(primals_14, (16, 16, 3, 3), (144, 9, 3, 1))
assert_size_stride(primals_15, (16,), (1,))
assert_size_stride(primals_16, (1, 16, 1, 1), (16, 1, 1, 1))
assert_size_stride(primals_17, (1,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(primals_1, primals_2, stride=(1,
1), padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf1 = buf0
del buf0
get_raw_stream(0)
triton_poi_fused_convolution_relu_0[grid(262144)](buf1, primals_3,
262144, XBLOCK=1024, num_warps=4, num_stages=1)
del primals_3
buf2 = extern_kernels.convolution(buf1, primals_4, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf2, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf3 = buf2
del buf2
triton_poi_fused_convolution_relu_0[grid(262144)](buf3, primals_5,
262144, XBLOCK=1024, num_warps=4, num_stages=1)
del primals_5
buf4 = extern_kernels.convolution(buf3, primals_6, stride=(1, 1),
padding=(2, 2), dilation=(2, 2), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf4, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf5 = buf4
del buf4
triton_poi_fused_convolution_relu_0[grid(262144)](buf5, primals_7,
262144, XBLOCK=1024, num_warps=4, num_stages=1)
buf6 = extern_kernels.convolution(buf5, primals_6, stride=(1, 1),
padding=(2, 2), dilation=(2, 2), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf6, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf7 = buf6
del buf6
triton_poi_fused_convolution_relu_0[grid(262144)](buf7, primals_7,
262144, XBLOCK=1024, num_warps=4, num_stages=1)
del primals_7
buf8 = extern_kernels.convolution(buf7, primals_8, stride=(1, 1),
padding=(4, 4), dilation=(4, 4), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf8, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf9 = buf8
del buf8
triton_poi_fused_convolution_relu_0[grid(262144)](buf9, primals_9,
262144, XBLOCK=1024, num_warps=4, num_stages=1)
del primals_9
buf10 = extern_kernels.convolution(buf9, primals_10, stride=(1, 1),
padding=(8, 8), dilation=(8, 8), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf10, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf11 = buf10
del buf10
triton_poi_fused_convolution_relu_0[grid(262144)](buf11, primals_11,
262144, XBLOCK=1024, num_warps=4, num_stages=1)
del primals_11
buf12 = extern_kernels.convolution(buf11, primals_12, stride=(1, 1),
padding=(16, 16), dilation=(16, 16), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf12, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf13 = buf12
del buf12
triton_poi_fused_convolution_relu_0[grid(262144)](buf13, primals_13,
262144, XBLOCK=1024, num_warps=4, num_stages=1)
del primals_13
buf14 = extern_kernels.convolution(buf13, primals_14, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf14, (4, 16, 64, 64), (65536, 4096, 64, 1))
buf15 = buf14
del buf14
triton_poi_fused_convolution_relu_0[grid(262144)](buf15, primals_15,
262144, XBLOCK=1024, num_warps=4, num_stages=1)
del primals_15
buf16 = extern_kernels.convolution(buf15, primals_16, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf16, (4, 1, 64, 64), (4096, 4096, 64, 1))
buf17 = buf16
del buf16
buf18 = empty_strided_cuda((4, 1, 64, 64), (4096, 4096, 64, 1),
torch.bool)
triton_poi_fused_convolution_relu_threshold_backward_1[grid(16384)](
buf17, primals_17, buf18, 16384, XBLOCK=128, num_warps=4,
num_stages=1)
del primals_17
return (buf17, primals_1, primals_2, primals_4, primals_6, primals_8,
primals_10, primals_12, primals_14, primals_16, buf1, buf3, buf5,
buf7, buf9, buf11, buf13, buf15, buf18)
class DilatedModelNew(nn.Module):
def __init__(self, k=16):
super(DilatedModelNew, self).__init__()
self.conv1 = nn.Conv2d(1, k, 3, stride=1, dilation=1, padding=1)
self.conv2 = nn.Conv2d(k, k, 3, stride=1, dilation=1, padding=1)
self.conv3 = nn.Conv2d(k, k, 3, stride=1, dilation=2, padding=2)
self.conv4 = nn.Conv2d(k, k, 3, stride=1, dilation=4, padding=4)
self.conv5 = nn.Conv2d(k, k, 3, stride=1, dilation=8, padding=8)
self.conv6 = nn.Conv2d(k, k, 3, stride=1, dilation=16, padding=16)
self.conv7 = nn.Conv2d(k, k, 3, stride=1, dilation=1, padding=1)
self.conv8 = nn.Conv2d(k, 1, 1, stride=1, dilation=1, padding=0)
def forward(self, input_0):
primals_2 = self.conv1.weight
primals_3 = self.conv1.bias
primals_4 = self.conv2.weight
primals_5 = self.conv2.bias
primals_6 = self.conv3.weight
primals_7 = self.conv3.bias
primals_8 = self.conv4.weight
primals_9 = self.conv4.bias
primals_10 = self.conv5.weight
primals_11 = self.conv5.bias
primals_12 = self.conv6.weight
primals_13 = self.conv6.bias
primals_14 = self.conv7.weight
primals_15 = self.conv7.bias
primals_16 = self.conv8.weight
primals_17 = self.conv8.bias
primals_1 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7, primals_8, primals_9,
primals_10, primals_11, primals_12, primals_13, primals_14,
primals_15, primals_16, primals_17])
return output[0]
|
JulianYu123456/icnn
|
DilatedModel
| false
| 13,913
|
[
"Apache-2.0"
] | 258
|
0aaf4b5cd13d71d98b0d05f367e1f71657ea6eb8
|
https://github.com/JulianYu123456/icnn/tree/0aaf4b5cd13d71d98b0d05f367e1f71657ea6eb8
|
PolicyNetwork
|
import torch
import numpy as np
import torch.nn as nn
import torch.nn.functional as F
from torch.distributions import Normal
class PolicyNetwork(nn.Module):
def __init__(self, num_inputs, num_actions, hidden_size, action_range=
1.0, init_w=0.003, log_std_min=-20, log_std_max=2):
super(PolicyNetwork, self).__init__()
self.log_std_min = log_std_min
self.log_std_max = log_std_max
self.linear1 = nn.Linear(num_inputs, hidden_size)
self.linear2 = nn.Linear(hidden_size, hidden_size)
self.linear3 = nn.Linear(hidden_size, hidden_size)
self.linear4 = nn.Linear(hidden_size, hidden_size)
self.mean_linear = nn.Linear(hidden_size, num_actions)
self.mean_linear.weight.data.uniform_(-init_w, init_w)
self.mean_linear.bias.data.uniform_(-init_w, init_w)
self.log_std_linear = nn.Linear(hidden_size, num_actions)
self.log_std_linear.weight.data.uniform_(-init_w, init_w)
self.log_std_linear.bias.data.uniform_(-init_w, init_w)
self.action_range = action_range
self.num_actions = num_actions
def forward(self, state):
x = F.relu(self.linear1(state))
x = F.relu(self.linear2(x))
x = F.relu(self.linear3(x))
x = F.relu(self.linear4(x))
mean = F.tanh(self.mean_linear(x))
log_std = self.log_std_linear(x)
log_std = torch.clamp(log_std, self.log_std_min, self.log_std_max)
return mean, log_std
def evaluate(self, state, deterministic, eval_noise_scale, epsilon=1e-06):
"""
generate action with state as input wrt the policy network, for calculating gradients
"""
mean, log_std = self.forward(state)
std = log_std.exp()
normal = Normal(0, 1)
z = normal.sample()
action_0 = torch.tanh(mean + std * z)
action = (self.action_range * mean if deterministic else self.
action_range * action_0)
log_prob = Normal(mean, std).log_prob(mean + std * z) - torch.log(
1.0 - action_0.pow(2) + epsilon) - np.log(self.action_range)
log_prob = log_prob.sum(dim=1, keepdim=True)
""" add noise """
eval_noise_clip = 2 * eval_noise_scale
noise = normal.sample(action.shape) * eval_noise_scale
noise = torch.clamp(noise, -eval_noise_clip, eval_noise_clip)
action = action + noise
return action, log_prob, z, mean, log_std
def get_action(self, state, deterministic, explore_noise_scale):
"""
generate action for interaction with env
"""
state = torch.FloatTensor(state).unsqueeze(0)
mean, log_std = self.forward(state)
std = log_std.exp()
normal = Normal(0, 1)
z = normal.sample()
action = mean.detach().cpu().numpy()[0
] if deterministic else torch.tanh(mean + std * z).detach().cpu(
).numpy()[0]
""" add noise """
noise = normal.sample(action.shape) * explore_noise_scale
action = self.action_range * action + noise.numpy()
return action
def sample_action(self):
a = torch.FloatTensor(self.num_actions).uniform_(-1, 1)
return self.action_range * a.numpy()
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'num_inputs': 4, 'num_actions': 4, 'hidden_size': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import libdevice
import numpy as np
import torch.nn as nn
from torch.distributions import Normal
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_relu_threshold_backward_0(in_out_ptr0, in_ptr0,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_out_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(in_out_ptr0 + x2, tmp4, xmask)
tl.store(out_ptr0 + x2, tmp6, xmask)
@triton.jit
def triton_poi_fused_tanh_1(in_out_ptr0, in_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_out_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = libdevice.tanh(tmp2)
tl.store(in_out_ptr0 + x2, tmp3, xmask)
@triton.jit
def triton_poi_fused_clamp_ge_le_logical_and_2(in_ptr0, in_ptr1, out_ptr0,
out_ptr1, xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr1 + x0, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = -20.0
tmp4 = triton_helpers.maximum(tmp2, tmp3)
tmp5 = 2.0
tmp6 = triton_helpers.minimum(tmp4, tmp5)
tmp7 = tmp2 >= tmp3
tmp8 = tmp2 <= tmp5
tmp9 = tmp7 & tmp8
tl.store(out_ptr0 + x2, tmp6, xmask)
tl.store(out_ptr1 + x2, tmp9, xmask)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7, primals_8, primals_9, primals_10, primals_11, primals_12,
primals_13) = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (4, 4), (4, 1))
assert_size_stride(primals_5, (4,), (1,))
assert_size_stride(primals_6, (4, 4), (4, 1))
assert_size_stride(primals_7, (4,), (1,))
assert_size_stride(primals_8, (4, 4), (4, 1))
assert_size_stride(primals_9, (4,), (1,))
assert_size_stride(primals_10, (4, 4), (4, 1))
assert_size_stride(primals_11, (4,), (1,))
assert_size_stride(primals_12, (4, 4), (4, 1))
assert_size_stride(primals_13, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_3, (64, 4), (4, 1), 0),
reinterpret_tensor(primals_1, (4, 4), (1, 4), 0), out=buf0)
del primals_1
buf1 = reinterpret_tensor(buf0, (4, 4, 4, 4), (64, 16, 4, 1), 0)
del buf0
buf16 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
get_raw_stream(0)
triton_poi_fused_relu_threshold_backward_0[grid(256)](buf1,
primals_2, buf16, 256, XBLOCK=256, num_warps=4, num_stages=1)
del primals_2
buf2 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf1, (64, 4), (4, 1), 0),
reinterpret_tensor(primals_4, (4, 4), (1, 4), 0), out=buf2)
buf3 = reinterpret_tensor(buf2, (4, 4, 4, 4), (64, 16, 4, 1), 0)
del buf2
buf15 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
triton_poi_fused_relu_threshold_backward_0[grid(256)](buf3,
primals_5, buf15, 256, XBLOCK=256, num_warps=4, num_stages=1)
del primals_5
buf4 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf3, (64, 4), (4, 1), 0),
reinterpret_tensor(primals_6, (4, 4), (1, 4), 0), out=buf4)
buf5 = reinterpret_tensor(buf4, (4, 4, 4, 4), (64, 16, 4, 1), 0)
del buf4
buf14 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
triton_poi_fused_relu_threshold_backward_0[grid(256)](buf5,
primals_7, buf14, 256, XBLOCK=256, num_warps=4, num_stages=1)
del primals_7
buf6 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf5, (64, 4), (4, 1), 0),
reinterpret_tensor(primals_8, (4, 4), (1, 4), 0), out=buf6)
buf7 = reinterpret_tensor(buf6, (4, 4, 4, 4), (64, 16, 4, 1), 0)
del buf6
buf13 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
triton_poi_fused_relu_threshold_backward_0[grid(256)](buf7,
primals_9, buf13, 256, XBLOCK=256, num_warps=4, num_stages=1)
del primals_9
buf8 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf7, (64, 4), (4, 1), 0),
reinterpret_tensor(primals_10, (4, 4), (1, 4), 0), out=buf8)
buf9 = reinterpret_tensor(buf8, (4, 4, 4, 4), (64, 16, 4, 1), 0)
del buf8
triton_poi_fused_tanh_1[grid(256)](buf9, primals_11, 256, XBLOCK=
256, num_warps=4, num_stages=1)
del primals_11
buf10 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf7, (64, 4), (4, 1), 0),
reinterpret_tensor(primals_12, (4, 4), (1, 4), 0), out=buf10)
buf11 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
buf12 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.bool)
triton_poi_fused_clamp_ge_le_logical_and_2[grid(256)](buf10,
primals_13, buf11, buf12, 256, XBLOCK=128, num_warps=4,
num_stages=1)
del buf10
del primals_13
return (buf9, buf11, reinterpret_tensor(primals_3, (64, 4), (4, 1), 0),
reinterpret_tensor(buf1, (64, 4), (4, 1), 0), reinterpret_tensor(
buf3, (64, 4), (4, 1), 0), reinterpret_tensor(buf5, (64, 4), (4, 1),
0), reinterpret_tensor(buf7, (64, 4), (4, 1), 0), buf9, buf12,
primals_12, primals_10, buf13, primals_8, buf14, primals_6, buf15,
primals_4, buf16)
class PolicyNetworkNew(nn.Module):
def __init__(self, num_inputs, num_actions, hidden_size, action_range=
1.0, init_w=0.003, log_std_min=-20, log_std_max=2):
super(PolicyNetworkNew, self).__init__()
self.log_std_min = log_std_min
self.log_std_max = log_std_max
self.linear1 = nn.Linear(num_inputs, hidden_size)
self.linear2 = nn.Linear(hidden_size, hidden_size)
self.linear3 = nn.Linear(hidden_size, hidden_size)
self.linear4 = nn.Linear(hidden_size, hidden_size)
self.mean_linear = nn.Linear(hidden_size, num_actions)
self.mean_linear.weight.data.uniform_(-init_w, init_w)
self.mean_linear.bias.data.uniform_(-init_w, init_w)
self.log_std_linear = nn.Linear(hidden_size, num_actions)
self.log_std_linear.weight.data.uniform_(-init_w, init_w)
self.log_std_linear.bias.data.uniform_(-init_w, init_w)
self.action_range = action_range
self.num_actions = num_actions
def evaluate(self, state, deterministic, eval_noise_scale, epsilon=1e-06):
"""
generate action with state as input wrt the policy network, for calculating gradients
"""
mean, log_std = self.forward(state)
std = log_std.exp()
normal = Normal(0, 1)
z = normal.sample()
action_0 = torch.tanh(mean + std * z)
action = (self.action_range * mean if deterministic else self.
action_range * action_0)
log_prob = Normal(mean, std).log_prob(mean + std * z) - torch.log(
1.0 - action_0.pow(2) + epsilon) - np.log(self.action_range)
log_prob = log_prob.sum(dim=1, keepdim=True)
""" add noise """
eval_noise_clip = 2 * eval_noise_scale
noise = normal.sample(action.shape) * eval_noise_scale
noise = torch.clamp(noise, -eval_noise_clip, eval_noise_clip)
action = action + noise
return action, log_prob, z, mean, log_std
def get_action(self, state, deterministic, explore_noise_scale):
"""
generate action for interaction with env
"""
state = torch.FloatTensor(state).unsqueeze(0)
mean, log_std = self.forward(state)
std = log_std.exp()
normal = Normal(0, 1)
z = normal.sample()
action = mean.detach().cpu().numpy()[0
] if deterministic else torch.tanh(mean + std * z).detach().cpu(
).numpy()[0]
""" add noise """
noise = normal.sample(action.shape) * explore_noise_scale
action = self.action_range * action + noise.numpy()
return action
def sample_action(self):
a = torch.FloatTensor(self.num_actions).uniform_(-1, 1)
return self.action_range * a.numpy()
def forward(self, input_0):
primals_1 = self.linear1.weight
primals_2 = self.linear1.bias
primals_4 = self.linear2.weight
primals_5 = self.linear2.bias
primals_6 = self.linear3.weight
primals_7 = self.linear3.bias
primals_8 = self.linear4.weight
primals_9 = self.linear4.bias
primals_10 = self.mean_linear.weight
primals_11 = self.mean_linear.bias
primals_12 = self.log_std_linear.weight
primals_13 = self.log_std_linear.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7, primals_8, primals_9,
primals_10, primals_11, primals_12, primals_13])
return output[0], output[1]
|
JieRen98/Popular-RL-Algorithms
|
PolicyNetwork
| false
| 13,914
|
[
"Apache-2.0"
] | 273
|
7f2bb74a51cf9cbde92a6ccfa42e97dc129dd145
|
https://github.com/JieRen98/Popular-RL-Algorithms/tree/7f2bb74a51cf9cbde92a6ccfa42e97dc129dd145
|
DiscShiftLoss
|
import torch
import torch.nn as nn
class DiscShiftLoss(nn.Module):
"""Disc shift loss.
Args:
loss_weight (float, optional): Loss weight. Defaults to 1.0.
"""
def __init__(self, loss_weight=0.1):
super().__init__()
self.loss_weight = loss_weight
def forward(self, x):
"""Forward function.
Args:
x (Tensor): Tensor with shape (n, c, h, w)
Returns:
Tensor: Loss.
"""
loss = torch.mean(x ** 2)
return loss * self.loss_weight
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_per_fused_mean_mul_pow_0(in_out_ptr0, in_ptr0, xnumel, rnumel):
XBLOCK: tl.constexpr = 1
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
tl.full([1], xoffset, tl.int32)
tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
tl.full([RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp1 = tmp0 * tmp0
tmp2 = tl.broadcast_to(tmp1, [RBLOCK])
tmp4 = triton_helpers.promote_to_tensor(tl.sum(tmp2, 0))
tmp5 = 256.0
tmp6 = tmp4 / tmp5
tmp7 = 0.1
tmp8 = tmp6 * tmp7
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([1], 0, tl.int32), tmp8, None)
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((), (), torch.float32)
buf1 = buf0
del buf0
get_raw_stream(0)
triton_per_fused_mean_mul_pow_0[grid(1)](buf1, arg0_1, 1, 256,
num_warps=2, num_stages=1)
del arg0_1
return buf1,
class DiscShiftLossNew(nn.Module):
"""Disc shift loss.
Args:
loss_weight (float, optional): Loss weight. Defaults to 1.0.
"""
def __init__(self, loss_weight=0.1):
super().__init__()
self.loss_weight = loss_weight
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
|
Juggernaut93/mmediting
|
DiscShiftLoss
| false
| 13,915
|
[
"Apache-2.0"
] | 1,884
|
8ef46ace29756dd2df1d92f2f73a33646e33e007
|
https://github.com/Juggernaut93/mmediting/tree/8ef46ace29756dd2df1d92f2f73a33646e33e007
|
EqualLinearActModule
|
import torch
import torch.nn as nn
from copy import deepcopy
from functools import partial
from torch.nn.init import _calculate_correct_fan
def equalized_lr(module, name='weight', gain=2 ** 0.5, mode='fan_in',
lr_mul=1.0):
"""Equalized Learning Rate.
This trick is proposed in:
Progressive Growing of GANs for Improved Quality, Stability, and Variation
The general idea is to dynamically rescale the weight in training instead
of in initializing so that the variance of the responses in each layer is
guaranteed with some statistical properties.
Note that this function is always combined with a convolution module which
is initialized with :math:`\\mathcal{N}(0, 1)`.
Args:
module (nn.Module): Module to be wrapped.
name (str | optional): The name of weights. Defaults to 'weight'.
mode (str, optional): The mode of computing ``fan`` which is the
same as ``kaiming_init`` in pytorch. You can choose one from
['fan_in', 'fan_out']. Defaults to 'fan_in'.
Returns:
nn.Module: Module that is registered with equalized lr hook.
"""
EqualizedLR.apply(module, name, gain=gain, mode=mode, lr_mul=lr_mul)
return module
class EqualizedLR:
"""Equalized Learning Rate.
This trick is proposed in:
Progressive Growing of GANs for Improved Quality, Stability, and Variation
The general idea is to dynamically rescale the weight in training instead
of in initializing so that the variance of the responses in each layer is
guaranteed with some statistical properties.
Note that this function is always combined with a convolution module which
is initialized with :math:`\\mathcal{N}(0, 1)`.
Args:
name (str | optional): The name of weights. Defaults to 'weight'.
mode (str, optional): The mode of computing ``fan`` which is the
same as ``kaiming_init`` in pytorch. You can choose one from
['fan_in', 'fan_out']. Defaults to 'fan_in'.
"""
def __init__(self, name='weight', gain=2 ** 0.5, mode='fan_in', lr_mul=1.0
):
self.name = name
self.mode = mode
self.gain = gain
self.lr_mul = lr_mul
def compute_weight(self, module):
"""Compute weight with equalized learning rate.
Args:
module (nn.Module): A module that is wrapped with equalized lr.
Returns:
torch.Tensor: Updated weight.
"""
weight = getattr(module, self.name + '_orig')
if weight.ndim == 5:
fan = _calculate_correct_fan(weight[0], self.mode)
else:
assert weight.ndim <= 4
fan = _calculate_correct_fan(weight, self.mode)
weight = weight * torch.tensor(self.gain, device=weight.device
) * torch.sqrt(torch.tensor(1.0 / fan, device=weight.device)
) * self.lr_mul
return weight
def __call__(self, module, inputs):
"""Standard interface for forward pre hooks."""
setattr(module, self.name, self.compute_weight(module))
@staticmethod
def apply(module, name, gain=2 ** 0.5, mode='fan_in', lr_mul=1.0):
"""Apply function.
This function is to register an equalized learning rate hook in an
``nn.Module``.
Args:
module (nn.Module): Module to be wrapped.
name (str | optional): The name of weights. Defaults to 'weight'.
mode (str, optional): The mode of computing ``fan`` which is the
same as ``kaiming_init`` in pytorch. You can choose one from
['fan_in', 'fan_out']. Defaults to 'fan_in'.
Returns:
nn.Module: Module that is registered with equalized lr hook.
"""
for _, hook in module._forward_pre_hooks.items():
if isinstance(hook, EqualizedLR):
raise RuntimeError(
f'Cannot register two equalized_lr hooks on the same parameter {name} in {module} module.'
)
fn = EqualizedLR(name, gain=gain, mode=mode, lr_mul=lr_mul)
weight = module._parameters[name]
delattr(module, name)
module.register_parameter(name + '_orig', weight)
setattr(module, name, weight.data)
module.register_forward_pre_hook(fn)
return fn
class EqualizedLRLinearModule(nn.Linear):
"""Equalized LR LinearModule.
In this module, we adopt equalized lr in ``nn.Linear``. The equalized
learning rate is proposed in:
Progressive Growing of GANs for Improved Quality, Stability, and Variation
Note that, the initialization of ``self.weight`` will be overwritten as
:math:`\\mathcal{N}(0, 1)`.
Args:
equalized_lr_cfg (dict | None, optional): Config for ``EqualizedLR``.
If ``None``, equalized learning rate is ignored. Defaults to
dict(mode='fan_in').
"""
def __init__(self, *args, equalized_lr_cfg=dict(mode='fan_in'), **kwargs):
super(EqualizedLRLinearModule, self).__init__(*args, **kwargs)
self.with_equlized_lr = equalized_lr_cfg is not None
if self.with_equlized_lr:
self.lr_mul = equalized_lr_cfg.get('lr_mul', 1.0)
else:
self.lr_mul = 1.0
if self.with_equlized_lr:
equalized_lr(self, **equalized_lr_cfg)
self._init_linear_weights()
def _init_linear_weights(self):
"""Initialize linear weights as described in PGGAN."""
nn.init.normal_(self.weight, 0, 1.0 / self.lr_mul)
if self.bias is not None:
nn.init.constant_(self.bias, 0.0)
class EqualLinearActModule(nn.Module):
"""Equalized LR Linear Module with Activation Layer.
Args:
nn ([type]): [description]
"""
def __init__(self, *args, equalized_lr_cfg=dict(gain=1.0, lr_mul=1.0),
bias=True, bias_init=0.0, act_cfg=None, **kwargs):
super(EqualLinearActModule, self).__init__()
self.with_activation = act_cfg is not None
self.linear = EqualizedLRLinearModule(*args, bias=False,
equalized_lr_cfg=equalized_lr_cfg, **kwargs)
if equalized_lr_cfg is not None:
self.lr_mul = equalized_lr_cfg.get('lr_mul', 1.0)
else:
self.lr_mul = 1.0
if bias:
self.bias = nn.Parameter(torch.zeros(self.linear.out_features).
fill_(bias_init))
else:
self.bias = None
if self.with_activation:
act_cfg = deepcopy(act_cfg)
if act_cfg['type'] == 'fused_bias':
self.act_type = act_cfg.pop('type')
assert self.bias is not None
self.activate = partial(fused_bias_leakyrelu, **act_cfg)
else:
self.act_type = 'normal'
self.activate = build_activation_layer(act_cfg)
else:
self.act_type = None
def forward(self, x):
if x.ndim >= 3:
x = x.reshape(x.size(0), -1)
x = self.linear(x)
if self.with_activation and self.act_type == 'fused_bias':
x = self.activate(x, self.bias * self.lr_mul)
elif self.bias is not None and self.with_activation:
x = self.activate(x + self.bias * self.lr_mul)
elif self.bias is not None:
x = x + self.bias * self.lr_mul
elif self.with_activation:
x = self.activate(x)
return x
def get_inputs():
return [torch.rand([4, 4])]
def get_init_inputs():
return [[], {'in_features': 4, 'out_features': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
import torch.nn as nn
from copy import deepcopy
from functools import partial
from torch.nn.init import _calculate_correct_fan
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_mul_sqrt_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = 0.5
tmp2 = tmp0 * tmp1
tmp3 = 1.0
tmp4 = tmp2 * tmp3
tl.store(out_ptr0 + x0, tmp4, xmask)
@triton.jit
def triton_poi_fused_mul_1(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = 1.0
tmp2 = tmp0 * tmp1
tl.store(out_ptr0 + x0, tmp2, xmask)
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4, 4), (4, 1))
assert_size_stride(primals_3, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_mul_sqrt_0[grid(16)](primals_2, buf0, 16, XBLOCK=
16, num_warps=1, num_stages=1)
del primals_2
buf1 = empty_strided_cuda((4,), (1,), torch.float32)
triton_poi_fused_mul_1[grid(4)](primals_3, buf1, 4, XBLOCK=4,
num_warps=1, num_stages=1)
del primals_3
buf2 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.addmm(buf1, primals_1, reinterpret_tensor(buf0, (4,
4), (1, 4), 0), alpha=1, beta=1, out=buf2)
del buf1
return buf2, buf0, primals_1
def equalized_lr(module, name='weight', gain=2 ** 0.5, mode='fan_in',
lr_mul=1.0):
"""Equalized Learning Rate.
This trick is proposed in:
Progressive Growing of GANs for Improved Quality, Stability, and Variation
The general idea is to dynamically rescale the weight in training instead
of in initializing so that the variance of the responses in each layer is
guaranteed with some statistical properties.
Note that this function is always combined with a convolution module which
is initialized with :math:`\\mathcal{N}(0, 1)`.
Args:
module (nn.Module): Module to be wrapped.
name (str | optional): The name of weights. Defaults to 'weight'.
mode (str, optional): The mode of computing ``fan`` which is the
same as ``kaiming_init`` in pytorch. You can choose one from
['fan_in', 'fan_out']. Defaults to 'fan_in'.
Returns:
nn.Module: Module that is registered with equalized lr hook.
"""
EqualizedLR.apply(module, name, gain=gain, mode=mode, lr_mul=lr_mul)
return module
class EqualizedLR:
"""Equalized Learning Rate.
This trick is proposed in:
Progressive Growing of GANs for Improved Quality, Stability, and Variation
The general idea is to dynamically rescale the weight in training instead
of in initializing so that the variance of the responses in each layer is
guaranteed with some statistical properties.
Note that this function is always combined with a convolution module which
is initialized with :math:`\\mathcal{N}(0, 1)`.
Args:
name (str | optional): The name of weights. Defaults to 'weight'.
mode (str, optional): The mode of computing ``fan`` which is the
same as ``kaiming_init`` in pytorch. You can choose one from
['fan_in', 'fan_out']. Defaults to 'fan_in'.
"""
def __init__(self, name='weight', gain=2 ** 0.5, mode='fan_in', lr_mul=1.0
):
self.name = name
self.mode = mode
self.gain = gain
self.lr_mul = lr_mul
def compute_weight(self, module):
"""Compute weight with equalized learning rate.
Args:
module (nn.Module): A module that is wrapped with equalized lr.
Returns:
torch.Tensor: Updated weight.
"""
weight = getattr(module, self.name + '_orig')
if weight.ndim == 5:
fan = _calculate_correct_fan(weight[0], self.mode)
else:
assert weight.ndim <= 4
fan = _calculate_correct_fan(weight, self.mode)
weight = weight * torch.tensor(self.gain, device=weight.device
) * torch.sqrt(torch.tensor(1.0 / fan, device=weight.device)
) * self.lr_mul
return weight
def __call__(self, module, inputs):
"""Standard interface for forward pre hooks."""
setattr(module, self.name, self.compute_weight(module))
@staticmethod
def apply(module, name, gain=2 ** 0.5, mode='fan_in', lr_mul=1.0):
"""Apply function.
This function is to register an equalized learning rate hook in an
``nn.Module``.
Args:
module (nn.Module): Module to be wrapped.
name (str | optional): The name of weights. Defaults to 'weight'.
mode (str, optional): The mode of computing ``fan`` which is the
same as ``kaiming_init`` in pytorch. You can choose one from
['fan_in', 'fan_out']. Defaults to 'fan_in'.
Returns:
nn.Module: Module that is registered with equalized lr hook.
"""
for _, hook in module._forward_pre_hooks.items():
if isinstance(hook, EqualizedLR):
raise RuntimeError(
f'Cannot register two equalized_lr hooks on the same parameter {name} in {module} module.'
)
fn = EqualizedLR(name, gain=gain, mode=mode, lr_mul=lr_mul)
weight = module._parameters[name]
delattr(module, name)
module.register_parameter(name + '_orig', weight)
setattr(module, name, weight.data)
module.register_forward_pre_hook(fn)
return fn
class EqualizedLRLinearModule(nn.Linear):
"""Equalized LR LinearModule.
In this module, we adopt equalized lr in ``nn.Linear``. The equalized
learning rate is proposed in:
Progressive Growing of GANs for Improved Quality, Stability, and Variation
Note that, the initialization of ``self.weight`` will be overwritten as
:math:`\\mathcal{N}(0, 1)`.
Args:
equalized_lr_cfg (dict | None, optional): Config for ``EqualizedLR``.
If ``None``, equalized learning rate is ignored. Defaults to
dict(mode='fan_in').
"""
def __init__(self, *args, equalized_lr_cfg=dict(mode='fan_in'), **kwargs):
super(EqualizedLRLinearModule, self).__init__(*args, **kwargs)
self.with_equlized_lr = equalized_lr_cfg is not None
if self.with_equlized_lr:
self.lr_mul = equalized_lr_cfg.get('lr_mul', 1.0)
else:
self.lr_mul = 1.0
if self.with_equlized_lr:
equalized_lr(self, **equalized_lr_cfg)
self._init_linear_weights()
def _init_linear_weights(self):
"""Initialize linear weights as described in PGGAN."""
nn.init.normal_(self.weight, 0, 1.0 / self.lr_mul)
if self.bias is not None:
nn.init.constant_(self.bias, 0.0)
class EqualLinearActModuleNew(nn.Module):
"""Equalized LR Linear Module with Activation Layer.
Args:
nn ([type]): [description]
"""
def __init__(self, *args, equalized_lr_cfg=dict(gain=1.0, lr_mul=1.0),
bias=True, bias_init=0.0, act_cfg=None, **kwargs):
super(EqualLinearActModuleNew, self).__init__()
self.with_activation = act_cfg is not None
self.linear = EqualizedLRLinearModule(*args, bias=False,
equalized_lr_cfg=equalized_lr_cfg, **kwargs)
if equalized_lr_cfg is not None:
self.lr_mul = equalized_lr_cfg.get('lr_mul', 1.0)
else:
self.lr_mul = 1.0
if bias:
self.bias = nn.Parameter(torch.zeros(self.linear.out_features).
fill_(bias_init))
else:
self.bias = None
if self.with_activation:
act_cfg = deepcopy(act_cfg)
if act_cfg['type'] == 'fused_bias':
self.act_type = act_cfg.pop('type')
assert self.bias is not None
self.activate = partial(fused_bias_leakyrelu, **act_cfg)
else:
self.act_type = 'normal'
self.activate = build_activation_layer(act_cfg)
else:
self.act_type = None
def forward(self, input_0):
primals_3 = self.bias
primals_1 = self.linear.weight_orig
primals_2 = input_0
output = call([primals_1, primals_2, primals_3])
return output[0]
|
Juggernaut93/mmediting
|
EqualLinearActModule
| false
| 13,916
|
[
"Apache-2.0"
] | 1,884
|
8ef46ace29756dd2df1d92f2f73a33646e33e007
|
https://github.com/Juggernaut93/mmediting/tree/8ef46ace29756dd2df1d92f2f73a33646e33e007
|
AvgPoolHead
|
import torch
import torch.nn as nn
import torch.optim
class AvgPoolHead(nn.Module):
def __init__(self, in_channels, out_channels, fea_map_size):
super(AvgPoolHead, self).__init__()
self.avgpool = nn.AvgPool2d(fea_map_size, stride=1)
self.fc = nn.Linear(in_channels, out_channels)
def forward(self, x):
x = self.avgpool(x)
x = x.view(x.size(0), -1)
x = self.fc(x)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_channels': 4, 'out_channels': 4, 'fea_map_size': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
import torch.nn as nn
import torch.optim
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_avg_pool2d_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.
constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + 16 * x0, xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr0 + (1 + 16 * x0), xmask, eviction_policy='evict_last'
)
tmp3 = tl.load(in_ptr0 + (2 + 16 * x0), xmask, eviction_policy='evict_last'
)
tmp5 = tl.load(in_ptr0 + (3 + 16 * x0), xmask, eviction_policy='evict_last'
)
tmp7 = tl.load(in_ptr0 + (4 + 16 * x0), xmask, eviction_policy='evict_last'
)
tmp9 = tl.load(in_ptr0 + (5 + 16 * x0), xmask, eviction_policy='evict_last'
)
tmp11 = tl.load(in_ptr0 + (6 + 16 * x0), xmask, eviction_policy=
'evict_last')
tmp13 = tl.load(in_ptr0 + (7 + 16 * x0), xmask, eviction_policy=
'evict_last')
tmp15 = tl.load(in_ptr0 + (8 + 16 * x0), xmask, eviction_policy=
'evict_last')
tmp17 = tl.load(in_ptr0 + (9 + 16 * x0), xmask, eviction_policy=
'evict_last')
tmp19 = tl.load(in_ptr0 + (10 + 16 * x0), xmask, eviction_policy=
'evict_last')
tmp21 = tl.load(in_ptr0 + (11 + 16 * x0), xmask, eviction_policy=
'evict_last')
tmp23 = tl.load(in_ptr0 + (12 + 16 * x0), xmask, eviction_policy=
'evict_last')
tmp25 = tl.load(in_ptr0 + (13 + 16 * x0), xmask, eviction_policy=
'evict_last')
tmp27 = tl.load(in_ptr0 + (14 + 16 * x0), xmask, eviction_policy=
'evict_last')
tmp29 = tl.load(in_ptr0 + (15 + 16 * x0), xmask, eviction_policy=
'evict_last')
tmp2 = tmp1 + tmp0
tmp4 = tmp3 + tmp2
tmp6 = tmp5 + tmp4
tmp8 = tmp7 + tmp6
tmp10 = tmp9 + tmp8
tmp12 = tmp11 + tmp10
tmp14 = tmp13 + tmp12
tmp16 = tmp15 + tmp14
tmp18 = tmp17 + tmp16
tmp20 = tmp19 + tmp18
tmp22 = tmp21 + tmp20
tmp24 = tmp23 + tmp22
tmp26 = tmp25 + tmp24
tmp28 = tmp27 + tmp26
tmp30 = tmp29 + tmp28
tmp31 = 0.0625
tmp32 = tmp30 * tmp31
tl.store(out_ptr0 + x0, tmp32, xmask)
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4, 4), (4, 1))
assert_size_stride(primals_3, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 1, 1), (4, 1, 1, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_avg_pool2d_0[grid(16)](primals_1, buf0, 16, XBLOCK
=16, num_warps=1, num_stages=1)
del primals_1
buf1 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_3, reinterpret_tensor(buf0, (4, 4), (4,
1), 0), reinterpret_tensor(primals_2, (4, 4), (1, 4), 0), alpha
=1, beta=1, out=buf1)
del primals_2
del primals_3
return buf1, reinterpret_tensor(buf0, (4, 4), (4, 1), 0)
class AvgPoolHeadNew(nn.Module):
def __init__(self, in_channels, out_channels, fea_map_size):
super(AvgPoolHeadNew, self).__init__()
self.avgpool = nn.AvgPool2d(fea_map_size, stride=1)
self.fc = nn.Linear(in_channels, out_channels)
def forward(self, input_0):
primals_2 = self.fc.weight
primals_3 = self.fc.bias
primals_1 = input_0
output = call([primals_1, primals_2, primals_3])
return output[0]
|
KGMSFT/integral-human-pose
|
AvgPoolHead
| false
| 13,917
|
[
"MIT"
] | 472
|
d3ad4117ed71c580d2ab17987e15f9b2c3318a3b
|
https://github.com/KGMSFT/integral-human-pose/tree/d3ad4117ed71c580d2ab17987e15f9b2c3318a3b
|
PositioningCost
|
import torch
import torch.nn as nn
class PositioningCost(nn.Module):
def __init__(self, target, Q=1, R=0, P=0):
super().__init__()
self.target = target
self.Q, self.R, self.P = Q, R, P
def forward(self, traj, u=None, mesh_p=None):
cost = 0.1 * torch.norm(traj[..., -1, :3] - self.target
) + 1 * torch.norm(traj[..., :3] - self.target
) + 0.01 * torch.norm(traj[..., 3:6]) + 0.01 * torch.norm(traj[
..., 6:9]) + 0.01 * torch.norm(traj[..., 9:12])
return cost
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'target': 4}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_per_fused_linalg_vector_norm_sub_0(in_ptr0, out_ptr0, xnumel,
rnumel, XBLOCK: tl.constexpr):
rnumel = 48
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
rmask = rindex < rnumel
r0 = rindex % 3
r1 = rindex // 3
tmp0 = tl.load(in_ptr0 + (12 + r0 + 16 * r1), rmask, other=0.0)
tmp1 = 4.0
tmp2 = tmp0 - tmp1
tmp3 = tmp2 * tmp2
tmp4 = tl.broadcast_to(tmp3, [XBLOCK, RBLOCK])
tmp6 = tl.where(rmask, tmp4, 0)
tmp7 = tl.sum(tmp6, 1)[:, None]
tl.store(out_ptr0 + tl.full([XBLOCK, 1], 0, tl.int32), tmp7, None)
@triton.jit
def triton_per_fused_linalg_vector_norm_sub_1(in_ptr0, out_ptr0, xnumel,
rnumel, XBLOCK: tl.constexpr):
rnumel = 192
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
rmask = rindex < rnumel
r0 = rindex % 3
r1 = rindex // 3
tmp0 = tl.load(in_ptr0 + (r0 + 4 * r1), rmask, other=0.0)
tmp1 = 4.0
tmp2 = tmp0 - tmp1
tmp3 = tmp2 * tmp2
tmp4 = tl.broadcast_to(tmp3, [XBLOCK, RBLOCK])
tmp6 = tl.where(rmask, tmp4, 0)
tmp7 = tl.sum(tmp6, 1)[:, None]
tl.store(out_ptr0 + tl.full([XBLOCK, 1], 0, tl.int32), tmp7, None)
@triton.jit
def triton_per_fused_add_linalg_vector_norm_mul_2(in_out_ptr0, in_ptr0,
in_ptr1, xnumel, rnumel, XBLOCK: tl.constexpr):
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + (3 + 4 * r0), None, eviction_policy='evict_last')
tmp5 = tl.load(in_out_ptr0 + 0)
tmp6 = tl.broadcast_to(tmp5, [XBLOCK, 1])
tmp10 = tl.load(in_ptr1 + 0)
tmp11 = tl.broadcast_to(tmp10, [XBLOCK, 1])
tmp1 = tmp0 * tmp0
tmp2 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp4 = tl.sum(tmp2, 1)[:, None]
tmp7 = libdevice.sqrt(tmp6)
tmp8 = 0.1
tmp9 = tmp7 * tmp8
tmp12 = libdevice.sqrt(tmp11)
tmp13 = 1.0
tmp14 = tmp12 * tmp13
tmp15 = tmp9 + tmp14
tmp16 = libdevice.sqrt(tmp4)
tmp17 = 0.01
tmp18 = tmp16 * tmp17
tmp19 = tmp15 + tmp18
tmp20 = 0.0
tmp21 = libdevice.sqrt(tmp20)
tmp22 = tmp21 * tmp17
tmp23 = tmp19 + tmp22
tmp24 = tmp23 + tmp22
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([XBLOCK, 1], 0, tl.int32), tmp24, None)
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((), (), torch.float32)
get_raw_stream(0)
triton_per_fused_linalg_vector_norm_sub_0[grid(1)](arg0_1, buf0, 1,
48, XBLOCK=1, num_warps=2, num_stages=1)
buf1 = empty_strided_cuda((), (), torch.float32)
triton_per_fused_linalg_vector_norm_sub_1[grid(1)](arg0_1, buf1, 1,
192, XBLOCK=1, num_warps=2, num_stages=1)
buf3 = buf0
del buf0
triton_per_fused_add_linalg_vector_norm_mul_2[grid(1)](buf3, arg0_1,
buf1, 1, 64, XBLOCK=1, num_warps=2, num_stages=1)
del arg0_1
del buf1
return buf3,
class PositioningCostNew(nn.Module):
def __init__(self, target, Q=1, R=0, P=0):
super().__init__()
self.target = target
self.Q, self.R, self.P = Q, R, P
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
|
Juju-botu/diffeqml-research
|
PositioningCost
| false
| 13,918
|
[
"Apache-2.0"
] | 49
|
aa796c87447e5299ec4f25a07fc4d032afb1f63e
|
https://github.com/Juju-botu/diffeqml-research/tree/aa796c87447e5299ec4f25a07fc4d032afb1f63e
|
TanhHyperSolver
|
import torch
import torch.nn as nn
class TanhHyperSolver(nn.Module):
def __init__(self, in_dim, out_dim, hidden_dim=32):
super().__init__()
self.fc1 = nn.Linear(in_dim, hidden_dim)
self.fc2 = nn.Linear(hidden_dim, hidden_dim)
self.fc3 = nn.Linear(hidden_dim, out_dim)
self.a1 = nn.Tanh()
self.a2 = nn.Tanh()
def forward(self, x):
x = self.a1(self.fc1(x))
x = self.a2(self.fc2(x))
x = self.fc3(x)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_dim': 4, 'out_dim': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_tanh_0(in_out_ptr0, in_ptr0, xnumel, XBLOCK: tl.constexpr
):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x2 = xindex
x0 = xindex % 32
tmp0 = tl.load(in_out_ptr0 + x2, None)
tmp1 = tl.load(in_ptr0 + x0, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = libdevice.tanh(tmp2)
tl.store(in_out_ptr0 + x2, tmp3, None)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7) = args
args.clear()
assert_size_stride(primals_1, (32, 4), (4, 1))
assert_size_stride(primals_2, (32,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (32, 32), (32, 1))
assert_size_stride(primals_5, (32,), (1,))
assert_size_stride(primals_6, (4, 32), (32, 1))
assert_size_stride(primals_7, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 32), (32, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_3, (64, 4), (4, 1), 0),
reinterpret_tensor(primals_1, (4, 32), (1, 4), 0), out=buf0)
del primals_1
buf1 = reinterpret_tensor(buf0, (4, 4, 4, 32), (512, 128, 32, 1), 0)
del buf0
get_raw_stream(0)
triton_poi_fused_tanh_0[grid(2048)](buf1, primals_2, 2048, XBLOCK=
256, num_warps=4, num_stages=1)
del primals_2
buf2 = empty_strided_cuda((64, 32), (32, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf1, (64, 32), (32, 1), 0),
reinterpret_tensor(primals_4, (32, 32), (1, 32), 0), out=buf2)
buf3 = reinterpret_tensor(buf2, (4, 4, 4, 32), (512, 128, 32, 1), 0)
del buf2
triton_poi_fused_tanh_0[grid(2048)](buf3, primals_5, 2048, XBLOCK=
256, num_warps=4, num_stages=1)
del primals_5
buf4 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_7, reinterpret_tensor(buf3, (64, 32),
(32, 1), 0), reinterpret_tensor(primals_6, (32, 4), (1, 32), 0),
alpha=1, beta=1, out=buf4)
del primals_7
return reinterpret_tensor(buf4, (4, 4, 4, 4), (64, 16, 4, 1), 0
), reinterpret_tensor(primals_3, (64, 4), (4, 1), 0
), buf1, buf3, primals_6, primals_4
class TanhHyperSolverNew(nn.Module):
def __init__(self, in_dim, out_dim, hidden_dim=32):
super().__init__()
self.fc1 = nn.Linear(in_dim, hidden_dim)
self.fc2 = nn.Linear(hidden_dim, hidden_dim)
self.fc3 = nn.Linear(hidden_dim, out_dim)
self.a1 = nn.Tanh()
self.a2 = nn.Tanh()
def forward(self, input_0):
primals_1 = self.fc1.weight
primals_2 = self.fc1.bias
primals_4 = self.fc2.weight
primals_5 = self.fc2.bias
primals_6 = self.fc3.weight
primals_7 = self.fc3.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7])
return output[0]
|
Juju-botu/diffeqml-research
|
TanhHyperSolver
| false
| 13,919
|
[
"Apache-2.0"
] | 49
|
aa796c87447e5299ec4f25a07fc4d032afb1f63e
|
https://github.com/Juju-botu/diffeqml-research/tree/aa796c87447e5299ec4f25a07fc4d032afb1f63e
|
NeuralArray
|
import torch
import torch.utils.data
import torch
import torch.nn as nn
class NeuralArray(nn.Module):
def __init__(self, dim, random_init=False):
super(NeuralArray, self).__init__()
self.dim = dim
if random_init:
self.register_parameter('data', torch.nn.Parameter(torch.randn(
self.dim, requires_grad=True)))
else:
self.register_parameter('data', torch.nn.Parameter(torch.zeros(
self.dim, requires_grad=True)))
def forward(self, id):
return self.data[id]
def regularizer_zero(self):
return torch.mean(torch.pow(self.data, 2.0))
def get_inputs():
return [torch.ones([4], dtype=torch.int64)]
def get_init_inputs():
return [[], {'dim': 4}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
import torch.utils.data
import torch
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_index_0(in_ptr0, in_ptr1, out_ptr0, xnumel, XBLOCK: tl
.constexpr):
xnumel = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = tl.full([XBLOCK], 4, tl.int32)
tmp2 = tmp0 + tmp1
tmp3 = tmp0 < 0
tmp4 = tl.where(tmp3, tmp2, tmp0)
tl.device_assert((0 <= tmp4) & (tmp4 < 4) | ~xmask,
'index out of bounds: 0 <= tmp4 < 4')
tmp6 = tl.load(in_ptr1 + tmp4, xmask, eviction_policy='evict_last')
tl.store(out_ptr0 + x0, tmp6, xmask)
def call(args):
primals_1, primals_2 = args
args.clear()
assert_size_stride(primals_1, (4,), (1,))
assert_size_stride(primals_2, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4,), (1,), torch.float32)
get_raw_stream(0)
triton_poi_fused_index_0[grid(4)](primals_2, primals_1, buf0, 4,
XBLOCK=4, num_warps=1, num_stages=1)
del primals_1
return buf0, primals_2
class NeuralArrayNew(nn.Module):
def __init__(self, dim, random_init=False):
super(NeuralArrayNew, self).__init__()
self.dim = dim
if random_init:
self.register_parameter('data', torch.nn.Parameter(torch.randn(
self.dim, requires_grad=True)))
else:
self.register_parameter('data', torch.nn.Parameter(torch.zeros(
self.dim, requires_grad=True)))
def regularizer_zero(self):
return torch.mean(torch.pow(self.data, 2.0))
def forward(self, input_0):
primals_1 = self.data
primals_2 = input_0
output = call([primals_1, primals_2])
return output[0]
|
JustusThies/NeuralTexGen
|
NeuralArray
| false
| 13,920
|
[
"BSD-3-Clause"
] | 49
|
008a6596cf54db3dab2d73f6248e243ca9a46e32
|
https://github.com/JustusThies/NeuralTexGen/tree/008a6596cf54db3dab2d73f6248e243ca9a46e32
|
Downsample
|
import torch
import torchvision.transforms.functional as F
import torch.nn as nn
import torch.nn.functional as F
class Downsample(nn.Module):
def __init__(self, in_ch=None, out_ch=None, with_conv=False, fir=False,
fir_kernel=(1, 3, 3, 1)):
super().__init__()
out_ch = out_ch if out_ch else in_ch
if not fir:
if with_conv:
self.Conv_0 = conv3x3(in_ch, out_ch, stride=2, padding=0)
elif with_conv:
self.Conv2d_0 = up_or_down_sampling.Conv2d(in_ch, out_ch,
kernel=3, down=True, resample_kernel=fir_kernel, use_bias=
True, kernel_init=default_init())
self.fir = fir
self.fir_kernel = fir_kernel
self.with_conv = with_conv
self.out_ch = out_ch
def forward(self, x):
_B, _C, _H, _W = x.shape
if not self.fir:
if self.with_conv:
x = F.pad(x, (0, 1, 0, 1))
x = self.Conv_0(x)
else:
x = F.avg_pool2d(x, 2, stride=2)
elif not self.with_conv:
x = up_or_down_sampling.downsample_2d(x, self.fir_kernel, factor=2)
else:
x = self.Conv2d_0(x)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_avg_pool2d_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.
constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 2
x1 = xindex // 2
x2 = xindex
tmp0 = tl.load(in_ptr0 + (2 * x0 + 8 * x1), xmask, eviction_policy=
'evict_last')
tmp1 = tl.load(in_ptr0 + (1 + 2 * x0 + 8 * x1), xmask, eviction_policy=
'evict_last')
tmp3 = tl.load(in_ptr0 + (4 + 2 * x0 + 8 * x1), xmask, eviction_policy=
'evict_last')
tmp5 = tl.load(in_ptr0 + (5 + 2 * x0 + 8 * x1), xmask, eviction_policy=
'evict_last')
tmp2 = tmp1 + tmp0
tmp4 = tmp3 + tmp2
tmp6 = tmp5 + tmp4
tmp7 = 0.25
tmp8 = tmp6 * tmp7
tl.store(out_ptr0 + x2, tmp8, xmask)
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 2, 2), (16, 4, 2, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_avg_pool2d_0[grid(64)](arg0_1, buf0, 64, XBLOCK=64,
num_warps=1, num_stages=1)
del arg0_1
return buf0,
class DownsampleNew(nn.Module):
def __init__(self, in_ch=None, out_ch=None, with_conv=False, fir=False,
fir_kernel=(1, 3, 3, 1)):
super().__init__()
out_ch = out_ch if out_ch else in_ch
if not fir:
if with_conv:
self.Conv_0 = conv3x3(in_ch, out_ch, stride=2, padding=0)
elif with_conv:
self.Conv2d_0 = up_or_down_sampling.Conv2d(in_ch, out_ch,
kernel=3, down=True, resample_kernel=fir_kernel, use_bias=
True, kernel_init=default_init())
self.fir = fir
self.fir_kernel = fir_kernel
self.with_conv = with_conv
self.out_ch = out_ch
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
|
DeepTitan/PNDM
|
Downsample
| false
| 13,921
|
[
"Apache-2.0"
] | 61
|
4037a4f40011c9a0d47b92303e64d47fcc7ed56a
|
https://github.com/DeepTitan/PNDM/tree/4037a4f40011c9a0d47b92303e64d47fcc7ed56a
|
WeightShareConv1d
|
import torch
import torch.nn as nn
import torch.nn
import torch.nn.functional
import torch.jit
import torch.nn.functional as F
import torch.utils.data
import torch.nn.utils
class VariationalHidDropout(nn.Module):
def __init__(self, dropout=0.0):
"""
Hidden-to-hidden (VD-based) dropout that applies the same mask at every time step and every layer of TrellisNet
:param dropout: The dropout rate (0 means no dropout is applied)
:param temporal: Whether the dropout mask is the same across the temporal dimension (or only the depth dimension)
"""
super(VariationalHidDropout, self).__init__()
self.dropout = dropout
self.mask = None
def reset_mask(self, x):
dropout = self.dropout
m = torch.zeros_like(x[:, :, :1]).bernoulli_(1 - dropout)
mask = m.requires_grad_(False) / (1 - dropout)
self.mask = mask
return mask
def forward(self, x):
if not self.training or self.dropout == 0:
return x
assert self.mask is not None, 'You need to reset mask before using VariationalHidDropout'
mask = self.mask.expand_as(x)
return mask * x
class WeightShareConv1d(nn.Module):
def __init__(self, n_hid, n_out, kernel_size, dropouth=0.0):
"""
The weight-tied 1D convolution used in TrellisNet.
:param n_hid: The dim of hidden input
:param n_out: The dim of the pre-activation (i.e. convolutional) output
:param kernel_size: The size of the convolutional kernel
:param dropouth: Hidden-to-hidden dropout
"""
super(WeightShareConv1d, self).__init__()
self.kernel_size = kernel_size
conv = nn.Conv1d(n_hid, n_out, kernel_size)
self.weight = conv.weight
self.bias = conv.bias
self.init_weights()
self.dict = dict()
self.drop = VariationalHidDropout(dropout=dropouth)
def init_weights(self):
bound = 0.01
self.weight.data.normal_(0, bound)
self.bias.data.normal_(0, bound)
def copy(self, func):
self.weight.data = func.weight.data.clone().detach()
self.bias.data = func.bias.data.clone().detach()
self.drop.mask = func.drop.mask.clone().detach()
def forward(self, x, dilation=1, hid=None):
k = self.kernel_size
padding = (k - 1) * dilation
x = F.pad(x, (padding, 0))
if hid is not None:
x[:, :, :padding] = hid.repeat(1, 1, padding)
res = F.conv1d(self.drop(x), self.weight, self.bias, dilation=dilation)
return res
def get_inputs():
return [torch.rand([4, 4])]
def get_init_inputs():
return [[], {'n_hid': 4, 'n_out': 4, 'kernel_size': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
import torch.nn as nn
import torch.nn
import torch.nn.functional
import torch.jit
import torch.utils.data
import torch.nn.utils
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_constant_pad_nd_0(in_ptr0, out_ptr0, xnumel, XBLOCK:
tl.constexpr):
xnumel = 28
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 7
x1 = xindex // 7
x2 = xindex
tmp0 = -3 + x0
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.load(in_ptr0 + (-3 + x0 + 4 * x1), tmp2 & xmask, other=0.0)
tl.store(out_ptr0 + x2, tmp3, xmask)
@triton.jit
def triton_poi_fused_convolution_1(in_out_ptr0, in_ptr0, xnumel, XBLOCK: tl
.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
tmp0 = tl.load(in_out_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + x1, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(in_out_ptr0 + x2, tmp2, xmask)
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_3, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 7), (7, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_constant_pad_nd_0[grid(28)](primals_1, buf0, 28,
XBLOCK=32, num_warps=1, num_stages=1)
del primals_1
buf1 = extern_kernels.convolution(reinterpret_tensor(buf0, (1, 4, 7
), (0, 7, 1), 0), primals_2, stride=(1,), padding=(0,),
dilation=(1,), transposed=False, output_padding=(0,), groups=1,
bias=None)
assert_size_stride(buf1, (1, 4, 4), (16, 4, 1))
buf2 = buf1
del buf1
triton_poi_fused_convolution_1[grid(16)](buf2, primals_3, 16,
XBLOCK=16, num_warps=1, num_stages=1)
del primals_3
return reinterpret_tensor(buf2, (4, 4), (4, 1), 0
), primals_2, reinterpret_tensor(buf0, (1, 4, 7), (28, 7, 1), 0)
class VariationalHidDropout(nn.Module):
def __init__(self, dropout=0.0):
"""
Hidden-to-hidden (VD-based) dropout that applies the same mask at every time step and every layer of TrellisNet
:param dropout: The dropout rate (0 means no dropout is applied)
:param temporal: Whether the dropout mask is the same across the temporal dimension (or only the depth dimension)
"""
super(VariationalHidDropout, self).__init__()
self.dropout = dropout
self.mask = None
def reset_mask(self, x):
dropout = self.dropout
m = torch.zeros_like(x[:, :, :1]).bernoulli_(1 - dropout)
mask = m.requires_grad_(False) / (1 - dropout)
self.mask = mask
return mask
def forward(self, x):
if not self.training or self.dropout == 0:
return x
assert self.mask is not None, 'You need to reset mask before using VariationalHidDropout'
mask = self.mask.expand_as(x)
return mask * x
class WeightShareConv1dNew(nn.Module):
def __init__(self, n_hid, n_out, kernel_size, dropouth=0.0):
"""
The weight-tied 1D convolution used in TrellisNet.
:param n_hid: The dim of hidden input
:param n_out: The dim of the pre-activation (i.e. convolutional) output
:param kernel_size: The size of the convolutional kernel
:param dropouth: Hidden-to-hidden dropout
"""
super(WeightShareConv1dNew, self).__init__()
self.kernel_size = kernel_size
conv = nn.Conv1d(n_hid, n_out, kernel_size)
self.weight = conv.weight
self.bias = conv.bias
self.init_weights()
self.dict = dict()
self.drop = VariationalHidDropout(dropout=dropouth)
def init_weights(self):
bound = 0.01
self.weight.data.normal_(0, bound)
self.bias.data.normal_(0, bound)
def copy(self, func):
self.weight.data = func.weight.data.clone().detach()
self.bias.data = func.bias.data.clone().detach()
self.drop.mask = func.drop.mask.clone().detach()
def forward(self, input_0):
primals_2 = self.weight
primals_3 = self.bias
primals_1 = input_0
output = call([primals_1, primals_2, primals_3])
return output[0]
|
JunLi-Galios/deq
|
WeightShareConv1d
| false
| 13,922
|
[
"MIT"
] | 548
|
80eb6b598357e8e01ad419126465fa3ed53b12c7
|
https://github.com/JunLi-Galios/deq/tree/80eb6b598357e8e01ad419126465fa3ed53b12c7
|
DropConnect
|
import torch
class DropConnect(torch.nn.Module):
def __init__(self, p):
super(DropConnect, self).__init__()
self.p = p
def forward(self, inputs):
batch_size = inputs.shape[0]
inputs.shape[2]
inputs.shape[3]
channel_size = inputs.shape[1]
keep_prob = 1 - self.p
random_tensor = keep_prob
random_tensor += torch.rand([batch_size, channel_size, 1, 1], dtype
=inputs.dtype, device=inputs.device)
binary_tensor = torch.floor(random_tensor)
output = inputs / keep_prob * binary_tensor
return output
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'p': 4}]
|
import torch
from torch import device
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_add_div_floor_mul_0(in_ptr0, in_ptr1, out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 16
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp3 = tl.load(in_ptr1 + x1, xmask, eviction_policy='evict_last')
tmp1 = -0.3333333333333333
tmp2 = tmp0 * tmp1
tmp4 = -3.0
tmp5 = tmp3 + tmp4
tmp6 = libdevice.floor(tmp5)
tmp7 = tmp2 * tmp6
tl.store(out_ptr0 + x2, tmp7, xmask)
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = torch.ops.aten.rand.default([4, 4, 1, 1], dtype=torch.
float32, device=device(type='cuda', index=0), pin_memory=False)
buf1 = buf0
del buf0
buf2 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_add_div_floor_mul_0[grid(256)](arg0_1, buf1, buf2,
256, XBLOCK=256, num_warps=4, num_stages=1)
del arg0_1
del buf1
return buf2,
class DropConnectNew(torch.nn.Module):
def __init__(self, p):
super(DropConnectNew, self).__init__()
self.p = p
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
|
KelvinYang0320/nas-without-training
|
DropConnect
| false
| 13,923
|
[
"MIT"
] | 385
|
5ed77a06726a73233a5a93b8f70a7172ce570029
|
https://github.com/KelvinYang0320/nas-without-training/tree/5ed77a06726a73233a5a93b8f70a7172ce570029
|
AuxiliaryConvolutions
|
import torch
from torch import nn
import torch.nn.functional as F
class AuxiliaryConvolutions(nn.Module):
"""
Additional convolutions to produce higher-level feature maps.
"""
def __init__(self):
super(AuxiliaryConvolutions, self).__init__()
self.conv8_1 = nn.Conv2d(1024, 256, kernel_size=1, padding=0)
self.conv8_2 = nn.Conv2d(256, 512, kernel_size=3, stride=2, padding=1)
self.conv9_1 = nn.Conv2d(512, 128, kernel_size=1, padding=0)
self.conv9_2 = nn.Conv2d(128, 256, kernel_size=3, stride=2, padding=1)
self.conv10_1 = nn.Conv2d(256, 128, kernel_size=1, padding=0)
self.conv10_2 = nn.Conv2d(128, 256, kernel_size=3, padding=0)
self.conv11_1 = nn.Conv2d(256, 128, kernel_size=1, padding=0)
self.conv11_2 = nn.Conv2d(128, 256, kernel_size=3, padding=0)
self.init_conv2d()
def init_conv2d(self):
"""
Initialize convolution parameters.
"""
for c in self.children():
if isinstance(c, nn.Conv2d):
nn.init.xavier_uniform_(c.weight)
nn.init.constant_(c.bias, 0.0)
def forward(self, conv7_feats):
"""
Forward propagation.
:param conv7_feats: lower-level conv7 feature map, a tensor of dimensions (N, 1024, 19, 19)
:return: higher-level feature maps conv8_2, conv9_2, conv10_2, and conv11_2
"""
out = F.relu(self.conv8_1(conv7_feats))
out = F.relu(self.conv8_2(out))
conv8_2_feats = out
out = F.relu(self.conv9_1(out))
out = F.relu(self.conv9_2(out))
conv9_2_feats = out
out = F.relu(self.conv10_1(out))
out = F.relu(self.conv10_2(out))
conv10_2_feats = out
out = F.relu(self.conv11_1(out))
conv11_2_feats = F.relu(self.conv11_2(out))
return conv8_2_feats, conv9_2_feats, conv10_2_feats, conv11_2_feats
def get_inputs():
return [torch.rand([4, 1024, 64, 64])]
def get_init_inputs():
return [[], {}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch import nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_0(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK: tl.
constexpr, XBLOCK: tl.constexpr):
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
tl.full([XBLOCK, YBLOCK], True, tl.int1)
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, YBLOCK], True, tl.int1)
x2 = xindex
y3 = yindex
y0 = yindex % 1024
y1 = yindex // 1024
tmp0 = tl.load(in_ptr0 + (x2 + 4096 * y3), None)
tl.store(out_ptr0 + (y0 + 1024 * x2 + 4194304 * y1), tmp0, None)
@triton.jit
def triton_poi_fused_1(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK: tl.
constexpr, XBLOCK: tl.constexpr):
xnumel = 9
yoffset = (tl.program_id(1) + tl.program_id(2) * tl.num_programs(1)
) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
tl.full([XBLOCK, YBLOCK], True, tl.int1)
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y3 = yindex
y0 = yindex % 256
y1 = yindex // 256
tmp0 = tl.load(in_ptr0 + (x2 + 9 * y3), xmask, eviction_policy='evict_last'
)
tl.store(out_ptr0 + (y0 + 256 * x2 + 2304 * y1), tmp0, xmask)
@triton.jit
def triton_poi_fused_2(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK: tl.
constexpr, XBLOCK: tl.constexpr):
xnumel = 9
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
tl.full([XBLOCK, YBLOCK], True, tl.int1)
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y3 = yindex
y0 = yindex % 128
y1 = yindex // 128
tmp0 = tl.load(in_ptr0 + (x2 + 9 * y3), xmask, eviction_policy='evict_last'
)
tl.store(out_ptr0 + (y0 + 128 * x2 + 1152 * y1), tmp0, xmask)
@triton.jit
def triton_poi_fused_convolution_relu_3(in_out_ptr0, in_ptr0, xnumel,
XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x2 = xindex
x0 = xindex % 256
tmp0 = tl.load(in_out_ptr0 + x2, None)
tmp1 = tl.load(in_ptr0 + x0, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + x2, tmp4, None)
@triton.jit
def triton_poi_fused_convolution_relu_4(in_ptr0, in_ptr1, out_ptr0,
out_ptr1, ynumel, xnumel, YBLOCK: tl.constexpr, XBLOCK: tl.constexpr):
xnumel = 1024
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
tl.full([XBLOCK, YBLOCK], True, tl.int1)
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 512
y1 = yindex // 512
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + 512 * x2 + 524288 * y1), xmask,
eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + y0, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1, 1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(out_ptr0 + (x2 + 1024 * y3), tmp4, xmask)
tl.store(out_ptr1 + (y0 + 512 * x2 + 524288 * y1), tmp4, xmask)
@triton.jit
def triton_poi_fused_convolution_relu_5(in_out_ptr0, in_ptr0, xnumel,
XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x2 = xindex
x0 = xindex % 128
tmp0 = tl.load(in_out_ptr0 + x2, None)
tmp1 = tl.load(in_ptr0 + x0, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + x2, tmp4, None)
@triton.jit
def triton_poi_fused_convolution_relu_6(in_ptr0, in_ptr1, out_ptr0,
out_ptr1, ynumel, xnumel, YBLOCK: tl.constexpr, XBLOCK: tl.constexpr):
xnumel = 256
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
tl.full([XBLOCK, YBLOCK], True, tl.int1)
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 256
y1 = yindex // 256
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + 256 * x2 + 65536 * y1), xmask,
eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + y0, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1, 1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(out_ptr0 + (x2 + 256 * y3), tmp4, xmask)
tl.store(out_ptr1 + (y0 + 256 * x2 + 65536 * y1), tmp4, xmask)
@triton.jit
def triton_poi_fused_convolution_relu_7(in_out_ptr0, in_ptr0, xnumel,
XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x2 = xindex
x0 = xindex % 128
tmp0 = tl.load(in_out_ptr0 + x2, None)
tmp1 = tl.load(in_ptr0 + x0, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + x2, tmp4, None)
@triton.jit
def triton_poi_fused_convolution_relu_8(in_ptr0, in_ptr1, out_ptr0,
out_ptr1, ynumel, xnumel, YBLOCK: tl.constexpr, XBLOCK: tl.constexpr):
xnumel = 196
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
tl.full([XBLOCK, YBLOCK], True, tl.int1)
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 256
y1 = yindex // 256
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + 256 * x2 + 50176 * y1), xmask,
eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + y0, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1, 1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(out_ptr0 + (x2 + 196 * y3), tmp4, xmask)
tl.store(out_ptr1 + (y0 + 256 * x2 + 50176 * y1), tmp4, xmask)
@triton.jit
def triton_poi_fused_convolution_relu_9(in_out_ptr0, in_ptr0, xnumel,
XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x2 = xindex
x0 = xindex % 128
tmp0 = tl.load(in_out_ptr0 + x2, None)
tmp1 = tl.load(in_ptr0 + x0, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + x2, tmp4, None)
@triton.jit
def triton_poi_fused_convolution_relu_threshold_backward_10(in_ptr0,
in_ptr1, out_ptr0, out_ptr1, ynumel, xnumel, YBLOCK: tl.constexpr,
XBLOCK: tl.constexpr):
xnumel = 144
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
tl.full([XBLOCK, YBLOCK], True, tl.int1)
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 256
y1 = yindex // 256
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + 256 * x2 + 36864 * y1), xmask,
eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + y0, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1, 1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(out_ptr0 + (x2 + 144 * y3), tmp4, xmask)
tl.store(out_ptr1 + (y0 + 256 * x2 + 36864 * y1), tmp6, xmask)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7, primals_8, primals_9, primals_10, primals_11, primals_12,
primals_13, primals_14, primals_15, primals_16, primals_17) = args
args.clear()
assert_size_stride(primals_1, (256, 1024, 1, 1), (1024, 1, 1, 1))
assert_size_stride(primals_2, (256,), (1,))
assert_size_stride(primals_3, (4, 1024, 64, 64), (4194304, 4096, 64, 1))
assert_size_stride(primals_4, (512, 256, 3, 3), (2304, 9, 3, 1))
assert_size_stride(primals_5, (512,), (1,))
assert_size_stride(primals_6, (128, 512, 1, 1), (512, 1, 1, 1))
assert_size_stride(primals_7, (128,), (1,))
assert_size_stride(primals_8, (256, 128, 3, 3), (1152, 9, 3, 1))
assert_size_stride(primals_9, (256,), (1,))
assert_size_stride(primals_10, (128, 256, 1, 1), (256, 1, 1, 1))
assert_size_stride(primals_11, (128,), (1,))
assert_size_stride(primals_12, (256, 128, 3, 3), (1152, 9, 3, 1))
assert_size_stride(primals_13, (256,), (1,))
assert_size_stride(primals_14, (128, 256, 1, 1), (256, 1, 1, 1))
assert_size_stride(primals_15, (128,), (1,))
assert_size_stride(primals_16, (256, 128, 3, 3), (1152, 9, 3, 1))
assert_size_stride(primals_17, (256,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 1024, 64, 64), (4194304, 1, 65536,
1024), torch.float32)
get_raw_stream(0)
triton_poi_fused_0[grid(4096, 4096)](primals_3, buf0, 4096, 4096,
XBLOCK=32, YBLOCK=32, num_warps=4, num_stages=1)
del primals_3
buf1 = empty_strided_cuda((512, 256, 3, 3), (2304, 1, 768, 256),
torch.float32)
triton_poi_fused_1[grid(131072, 9)](primals_4, buf1, 131072, 9,
XBLOCK=16, YBLOCK=64, num_warps=4, num_stages=1)
del primals_4
buf2 = empty_strided_cuda((256, 128, 3, 3), (1152, 1, 384, 128),
torch.float32)
triton_poi_fused_2[grid(32768, 9)](primals_8, buf2, 32768, 9,
XBLOCK=16, YBLOCK=64, num_warps=4, num_stages=1)
del primals_8
buf3 = empty_strided_cuda((256, 128, 3, 3), (1152, 1, 384, 128),
torch.float32)
triton_poi_fused_2[grid(32768, 9)](primals_12, buf3, 32768, 9,
XBLOCK=16, YBLOCK=64, num_warps=4, num_stages=1)
del primals_12
buf4 = empty_strided_cuda((256, 128, 3, 3), (1152, 1, 384, 128),
torch.float32)
triton_poi_fused_2[grid(32768, 9)](primals_16, buf4, 32768, 9,
XBLOCK=16, YBLOCK=64, num_warps=4, num_stages=1)
del primals_16
buf5 = extern_kernels.convolution(buf0, primals_1, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf5, (4, 256, 64, 64), (1048576, 1, 16384, 256))
buf6 = buf5
del buf5
triton_poi_fused_convolution_relu_3[grid(4194304)](buf6, primals_2,
4194304, XBLOCK=1024, num_warps=4, num_stages=1)
del primals_2
buf7 = extern_kernels.convolution(buf6, buf1, stride=(2, 2),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf7, (4, 512, 32, 32), (524288, 1, 16384, 512))
buf8 = empty_strided_cuda((4, 512, 32, 32), (524288, 1024, 32, 1),
torch.float32)
buf9 = empty_strided_cuda((4, 512, 32, 32), (524288, 1, 16384, 512),
torch.float32)
triton_poi_fused_convolution_relu_4[grid(2048, 1024)](buf7,
primals_5, buf8, buf9, 2048, 1024, XBLOCK=64, YBLOCK=64,
num_warps=8, num_stages=1)
del buf7
del primals_5
buf10 = extern_kernels.convolution(buf9, primals_6, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf10, (4, 128, 32, 32), (131072, 1, 4096, 128))
del buf9
buf11 = buf10
del buf10
triton_poi_fused_convolution_relu_5[grid(524288)](buf11, primals_7,
524288, XBLOCK=1024, num_warps=4, num_stages=1)
del primals_7
buf12 = extern_kernels.convolution(buf11, buf2, stride=(2, 2),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf12, (4, 256, 16, 16), (65536, 1, 4096, 256))
buf13 = empty_strided_cuda((4, 256, 16, 16), (65536, 256, 16, 1),
torch.float32)
buf14 = empty_strided_cuda((4, 256, 16, 16), (65536, 1, 4096, 256),
torch.float32)
triton_poi_fused_convolution_relu_6[grid(1024, 256)](buf12,
primals_9, buf13, buf14, 1024, 256, XBLOCK=32, YBLOCK=32,
num_warps=4, num_stages=1)
del buf12
del primals_9
buf15 = extern_kernels.convolution(buf14, primals_10, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf15, (4, 128, 16, 16), (32768, 1, 2048, 128))
del buf14
buf16 = buf15
del buf15
triton_poi_fused_convolution_relu_7[grid(131072)](buf16, primals_11,
131072, XBLOCK=512, num_warps=8, num_stages=1)
del primals_11
buf17 = extern_kernels.convolution(buf16, buf3, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf17, (4, 256, 14, 14), (50176, 1, 3584, 256))
buf18 = empty_strided_cuda((4, 256, 14, 14), (50176, 196, 14, 1),
torch.float32)
buf19 = empty_strided_cuda((4, 256, 14, 14), (50176, 1, 3584, 256),
torch.float32)
triton_poi_fused_convolution_relu_8[grid(1024, 196)](buf17,
primals_13, buf18, buf19, 1024, 196, XBLOCK=32, YBLOCK=32,
num_warps=4, num_stages=1)
del buf17
del primals_13
buf20 = extern_kernels.convolution(buf19, primals_14, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf20, (4, 128, 14, 14), (25088, 1, 1792, 128))
del buf19
buf21 = buf20
del buf20
triton_poi_fused_convolution_relu_9[grid(100352)](buf21, primals_15,
100352, XBLOCK=1024, num_warps=4, num_stages=1)
del primals_15
buf22 = extern_kernels.convolution(buf21, buf4, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf22, (4, 256, 12, 12), (36864, 1, 3072, 256))
buf23 = empty_strided_cuda((4, 256, 12, 12), (36864, 144, 12, 1),
torch.float32)
buf24 = empty_strided_cuda((4, 256, 12, 12), (36864, 1, 3072, 256),
torch.bool)
triton_poi_fused_convolution_relu_threshold_backward_10[grid(1024, 144)
](buf22, primals_17, buf23, buf24, 1024, 144, XBLOCK=64, YBLOCK
=64, num_warps=8, num_stages=1)
del buf22
del primals_17
return (buf8, buf13, buf18, buf23, primals_1, buf0, buf1, primals_6,
buf2, primals_10, buf3, primals_14, buf4, buf6, buf8, buf11, buf13,
buf16, buf18, buf21, buf24)
class AuxiliaryConvolutionsNew(nn.Module):
"""
Additional convolutions to produce higher-level feature maps.
"""
def __init__(self):
super(AuxiliaryConvolutionsNew, self).__init__()
self.conv8_1 = nn.Conv2d(1024, 256, kernel_size=1, padding=0)
self.conv8_2 = nn.Conv2d(256, 512, kernel_size=3, stride=2, padding=1)
self.conv9_1 = nn.Conv2d(512, 128, kernel_size=1, padding=0)
self.conv9_2 = nn.Conv2d(128, 256, kernel_size=3, stride=2, padding=1)
self.conv10_1 = nn.Conv2d(256, 128, kernel_size=1, padding=0)
self.conv10_2 = nn.Conv2d(128, 256, kernel_size=3, padding=0)
self.conv11_1 = nn.Conv2d(256, 128, kernel_size=1, padding=0)
self.conv11_2 = nn.Conv2d(128, 256, kernel_size=3, padding=0)
self.init_conv2d()
def init_conv2d(self):
"""
Initialize convolution parameters.
"""
for c in self.children():
if isinstance(c, nn.Conv2d):
nn.init.xavier_uniform_(c.weight)
nn.init.constant_(c.bias, 0.0)
def forward(self, input_0):
primals_1 = self.conv8_1.weight
primals_2 = self.conv8_1.bias
primals_4 = self.conv8_2.weight
primals_5 = self.conv8_2.bias
primals_6 = self.conv9_1.weight
primals_7 = self.conv9_1.bias
primals_8 = self.conv9_2.weight
primals_9 = self.conv9_2.bias
primals_10 = self.conv10_1.weight
primals_11 = self.conv10_1.bias
primals_12 = self.conv10_2.weight
primals_13 = self.conv10_2.bias
primals_14 = self.conv11_1.weight
primals_15 = self.conv11_1.bias
primals_16 = self.conv11_2.weight
primals_17 = self.conv11_2.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7, primals_8, primals_9,
primals_10, primals_11, primals_12, primals_13, primals_14,
primals_15, primals_16, primals_17])
return output[0], output[1], output[2], output[3]
|
HFAiLab/ffrecord
|
AuxiliaryConvolutions
| false
| 13,924
|
[
"MIT"
] | 47
|
e916dc715ffa38a304a673ade7c5aa1efff5936d
|
https://github.com/HFAiLab/ffrecord/tree/e916dc715ffa38a304a673ade7c5aa1efff5936d
|
Linear_Q
|
from torch.autograd import Function
import torch
import torch.utils.data.distributed
import torch.nn as nn
import torch.nn.functional as F
import torch.utils.data
def quantize(input, nbit):
return Quantizer.apply(input, nbit)
def dorefa_a(input, nbit_a):
return quantize(torch.clamp(0.1 * input, 0, 1), nbit_a)
def scale_sign(input):
return ScaleSigner.apply(input)
def dorefa_w(w, nbit_w):
if nbit_w == 1:
w = scale_sign(w)
else:
w = torch.tanh(w)
w = w / (2 * torch.max(torch.abs(w))) + 0.5
w = 2 * quantize(w, nbit_w) - 1
return w
class Quantizer(Function):
@staticmethod
def forward(ctx, input, nbit):
scale = 2 ** nbit - 1
return torch.round(input * scale) / scale
@staticmethod
def backward(ctx, grad_output):
return grad_output, None
class ScaleSigner(Function):
"""take a real value x, output sign(x)*E(|x|)"""
@staticmethod
def forward(ctx, input):
return torch.sign(input) * torch.mean(torch.abs(input))
@staticmethod
def backward(ctx, grad_output):
return grad_output
class Linear_Q(nn.Linear):
def __init__(self, in_features, out_features, bias=True, quan_name_w=
'dorefa', quan_name_a='dorefa', nbit_w=1, nbit_a=1):
super(Linear_Q, self).__init__(in_features, out_features, bias)
self.nbit_w = nbit_w
self.nbit_a = nbit_a
name_w_dict = {'dorefa': dorefa_w}
name_a_dict = {'dorefa': dorefa_a}
self.quan_w = name_w_dict[quan_name_w]
self.quan_a = name_a_dict[quan_name_a]
def forward(self, input):
if self.nbit_w < 32:
w = self.quan_w(self.weight, self.nbit_w)
else:
w = self.weight
if self.nbit_a < 32:
x = self.quan_a(input, self.nbit_a)
else:
x = input
output = F.linear(x, w, self.bias)
return output
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_features': 4, 'out_features': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch.autograd import Function
import torch.utils.data.distributed
import torch.nn as nn
import torch.utils.data
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_per_fused_abs_mean_mul_sign_0(in_ptr0, out_ptr1, xnumel, rnumel,
XBLOCK: tl.constexpr):
RBLOCK: tl.constexpr = 16
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp1 = tl_math.abs(tmp0)
tmp2 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp4 = tl.sum(tmp2, 1)[:, None]
tmp5 = tl.full([1, 1], 0, tl.int32)
tmp6 = tmp5 < tmp0
tmp7 = tmp6.to(tl.int8)
tmp8 = tmp0 < tmp5
tmp9 = tmp8.to(tl.int8)
tmp10 = tmp7 - tmp9
tmp11 = tmp10.to(tmp0.dtype)
tmp12 = 16.0
tmp13 = tmp4 / tmp12
tmp14 = tmp11 * tmp13
tl.store(out_ptr1 + tl.broadcast_to(r0, [XBLOCK, RBLOCK]), tmp14, None)
@triton.jit
def triton_poi_fused_clamp_div_mul_round_1(in_ptr0, out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = 0.1
tmp2 = tmp0 * tmp1
tmp3 = 0.0
tmp4 = triton_helpers.maximum(tmp2, tmp3)
tmp5 = 1.0
tmp6 = triton_helpers.minimum(tmp4, tmp5)
tmp7 = tmp6 * tmp5
tmp8 = libdevice.nearbyint(tmp7)
tmp9 = tmp8 * tmp5
tl.store(out_ptr0 + x0, tmp9, xmask)
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_3, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf2 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
get_raw_stream(0)
triton_per_fused_abs_mean_mul_sign_0[grid(1)](primals_1, buf2, 1,
16, XBLOCK=1, num_warps=2, num_stages=1)
del primals_1
buf1 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused_clamp_div_mul_round_1[grid(256)](primals_2, buf1,
256, XBLOCK=128, num_warps=4, num_stages=1)
del primals_2
buf3 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_3, reinterpret_tensor(buf1, (64, 4), (
4, 1), 0), reinterpret_tensor(buf2, (4, 4), (1, 4), 0), alpha=1,
beta=1, out=buf3)
del buf2
del primals_3
return reinterpret_tensor(buf3, (4, 4, 4, 4), (64, 16, 4, 1), 0
), reinterpret_tensor(buf1, (64, 4), (4, 1), 0)
def quantize(input, nbit):
return Quantizer.apply(input, nbit)
def dorefa_a(input, nbit_a):
return quantize(torch.clamp(0.1 * input, 0, 1), nbit_a)
def scale_sign(input):
return ScaleSigner.apply(input)
def dorefa_w(w, nbit_w):
if nbit_w == 1:
w = scale_sign(w)
else:
w = torch.tanh(w)
w = w / (2 * torch.max(torch.abs(w))) + 0.5
w = 2 * quantize(w, nbit_w) - 1
return w
class Quantizer(Function):
@staticmethod
def forward(ctx, input, nbit):
scale = 2 ** nbit - 1
return torch.round(input * scale) / scale
@staticmethod
def backward(ctx, grad_output):
return grad_output, None
class ScaleSigner(Function):
"""take a real value x, output sign(x)*E(|x|)"""
@staticmethod
def forward(ctx, input):
return torch.sign(input) * torch.mean(torch.abs(input))
@staticmethod
def backward(ctx, grad_output):
return grad_output
class Linear_QNew(nn.Linear):
def __init__(self, in_features, out_features, bias=True, quan_name_w=
'dorefa', quan_name_a='dorefa', nbit_w=1, nbit_a=1):
super(Linear_QNew, self).__init__(in_features, out_features, bias)
self.nbit_w = nbit_w
self.nbit_a = nbit_a
name_w_dict = {'dorefa': dorefa_w}
name_a_dict = {'dorefa': dorefa_a}
self.quan_w = name_w_dict[quan_name_w]
self.quan_a = name_a_dict[quan_name_a]
def forward(self, input_0):
primals_1 = self.weight
primals_3 = self.bias
primals_2 = input_0
output = call([primals_1, primals_2, primals_3])
return output[0]
|
Jzz24/pytorch_quantization
|
Linear_Q
| false
| 13,925
|
[
"MIT"
] | 71
|
0c2d93c8ce4f85dd2c34ea6f36c58d14db21bf8e
|
https://github.com/Jzz24/pytorch_quantization/tree/0c2d93c8ce4f85dd2c34ea6f36c58d14db21bf8e
|
TransformerNet
|
import torch
import numpy as np
import torch.nn as nn
class ConvLayer(torch.nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, stride):
super(ConvLayer, self).__init__()
reflection_padding = int(np.floor(kernel_size / 2))
self.reflection_pad = nn.ReflectionPad2d(reflection_padding)
self.conv2d = nn.Conv2d(in_channels, out_channels, kernel_size, stride)
def forward(self, x):
out = self.reflection_pad(x)
out = self.conv2d(out)
return out
class InstanceNormalization(torch.nn.Module):
"""InstanceNormalization
Improves convergence of neural-style.
ref: https://arxiv.org/pdf/1607.08022.pdf
"""
def __init__(self, dim, eps=1e-09):
super(InstanceNormalization, self).__init__()
self.scale = nn.Parameter(torch.FloatTensor(dim))
self.shift = nn.Parameter(torch.FloatTensor(dim))
self.eps = eps
self._reset_parameters()
def _reset_parameters(self):
self.scale.data.uniform_()
self.shift.data.zero_()
def forward(self, x):
n = x.size(2) * x.size(3)
t = x.view(x.size(0), x.size(1), n)
mean = torch.mean(t, 2).unsqueeze(2).unsqueeze(3).expand_as(x)
var = torch.var(t, 2).unsqueeze(2).unsqueeze(3).expand_as(x) * ((n -
1) / float(n))
scale_broadcast = self.scale.unsqueeze(1).unsqueeze(1).unsqueeze(0)
scale_broadcast = scale_broadcast.expand_as(x)
shift_broadcast = self.shift.unsqueeze(1).unsqueeze(1).unsqueeze(0)
shift_broadcast = shift_broadcast.expand_as(x)
out = (x - mean) / torch.sqrt(var + self.eps)
out = out * scale_broadcast + shift_broadcast
return out
class ResidualBlock(torch.nn.Module):
"""ResidualBlock
introduced in: https://arxiv.org/abs/1512.03385
recommended architecture: http://torch.ch/blog/2016/02/04/resnets.html
"""
def __init__(self, channels):
super(ResidualBlock, self).__init__()
self.conv1 = ConvLayer(channels, channels, kernel_size=3, stride=1)
self.in1 = InstanceNormalization(channels)
self.conv2 = ConvLayer(channels, channels, kernel_size=3, stride=1)
self.in2 = InstanceNormalization(channels)
self.relu = nn.ReLU()
def forward(self, x):
residual = x
out = self.relu(self.in1(self.conv1(x)))
out = self.in2(self.conv2(out))
out = out + residual
return out
class UpsampleConvLayer(torch.nn.Module):
"""UpsampleConvLayer
Upsamples the input and then does a convolution. This method gives better results
compared to ConvTranspose2d.
ref: http://distill.pub/2016/deconv-checkerboard/
"""
def __init__(self, in_channels, out_channels, kernel_size, stride,
upsample=None):
super(UpsampleConvLayer, self).__init__()
self.upsample = upsample
if upsample:
self.upsample_layer = torch.nn.Upsample(scale_factor=upsample)
reflection_padding = int(np.floor(kernel_size / 2))
self.reflection_pad = nn.ReflectionPad2d(reflection_padding)
self.conv2d = nn.Conv2d(in_channels, out_channels, kernel_size, stride)
def forward(self, x):
x_in = x
if self.upsample:
x_in = self.upsample_layer(x_in)
out = self.reflection_pad(x_in)
out = self.conv2d(out)
return out
class TransformerNet(torch.nn.Module):
def __init__(self):
super(TransformerNet, self).__init__()
self.conv1 = ConvLayer(3, 32, kernel_size=9, stride=1)
self.in1 = InstanceNormalization(32)
self.conv2 = ConvLayer(32, 64, kernel_size=3, stride=2)
self.in2 = InstanceNormalization(64)
self.conv3 = ConvLayer(64, 128, kernel_size=3, stride=2)
self.in3 = InstanceNormalization(128)
self.res1 = ResidualBlock(128)
self.res2 = ResidualBlock(128)
self.res3 = ResidualBlock(128)
self.res4 = ResidualBlock(128)
self.res5 = ResidualBlock(128)
self.deconv1 = UpsampleConvLayer(128, 64, kernel_size=3, stride=1,
upsample=2)
self.in4 = InstanceNormalization(64)
self.deconv2 = UpsampleConvLayer(64, 32, kernel_size=3, stride=1,
upsample=2)
self.in5 = InstanceNormalization(32)
self.deconv3 = ConvLayer(32, 3, kernel_size=9, stride=1)
self.relu = nn.ReLU()
def forward(self, X):
in_X = X
y = self.relu(self.in1(self.conv1(in_X)))
y = self.relu(self.in2(self.conv2(y)))
y = self.relu(self.in3(self.conv3(y)))
y = self.res1(y)
y = self.res2(y)
y = self.res3(y)
y = self.res4(y)
y = self.res5(y)
y = self.relu(self.in4(self.deconv1(y)))
y = self.relu(self.in5(self.deconv2(y)))
y = self.deconv3(y)
return y
def get_inputs():
return [torch.rand([4, 3, 64, 64])]
def get_init_inputs():
return [[], {}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
import numpy as np
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_reflection_pad2d_0(in_ptr0, out_ptr0, xnumel, XBLOCK:
tl.constexpr):
xnumel = 62208
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 72
x1 = xindex // 72 % 72
x2 = xindex // 5184
x3 = xindex
tmp0 = tl.load(in_ptr0 + (4095 + -1 * tl_math.abs(-63 + tl_math.abs(-4 +
x0)) + -64 * tl_math.abs(-63 + tl_math.abs(-4 + x1)) + 4096 * x2),
xmask, eviction_policy='evict_last')
tl.store(out_ptr0 + x3, tmp0, xmask)
@triton.jit
def triton_red_fused_convolution_mean_var_1(in_out_ptr0, in_out_ptr1,
in_out_ptr2, in_ptr0, xnumel, rnumel, XBLOCK: tl.constexpr, RBLOCK: tl.
constexpr):
xnumel = 128
rnumel = 4096
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rbase = tl.arange(0, RBLOCK)[None, :]
x3 = xindex
x0 = xindex % 32
tmp1 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
_tmp4 = tl.full([XBLOCK, RBLOCK], 0, tl.float32)
tmp6_mean = tl.zeros([XBLOCK, RBLOCK], tl.float32)
tmp6_m2 = tl.zeros([XBLOCK, RBLOCK], tl.float32)
tmp6_weight = tl.zeros([XBLOCK, RBLOCK], tl.float32)
for roffset in range(0, rnumel, RBLOCK):
rindex = roffset + rbase
rmask = rindex < rnumel
r2 = rindex
tmp0 = tl.load(in_out_ptr0 + (r2 + 4096 * x3), rmask & xmask,
eviction_policy='evict_first', other=0.0)
tmp2 = tmp0 + tmp1
tmp3 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp5 = _tmp4 + tmp3
_tmp4 = tl.where(rmask & xmask, tmp5, _tmp4)
tmp6_mean_next, tmp6_m2_next, tmp6_weight_next = (triton_helpers.
welford_reduce(tmp3, tmp6_mean, tmp6_m2, tmp6_weight, roffset == 0)
)
tmp6_mean = tl.where(rmask & xmask, tmp6_mean_next, tmp6_mean)
tmp6_m2 = tl.where(rmask & xmask, tmp6_m2_next, tmp6_m2)
tmp6_weight = tl.where(rmask & xmask, tmp6_weight_next, tmp6_weight)
tl.store(in_out_ptr0 + (r2 + 4096 * x3), tmp2, rmask & xmask)
tmp4 = tl.sum(_tmp4, 1)[:, None]
tmp6_tmp, tmp7_tmp, tmp8_tmp = triton_helpers.welford(tmp6_mean,
tmp6_m2, tmp6_weight, 1)
tmp6_tmp[:, None]
tmp7 = tmp7_tmp[:, None]
tmp8_tmp[:, None]
tmp9 = 4096.0
tmp10 = tmp4 / tmp9
tmp11 = 4095.0
tmp12 = tmp7 / tmp11
tl.debug_barrier()
tl.store(in_out_ptr1 + x3, tmp10, xmask)
tl.debug_barrier()
tl.store(in_out_ptr2 + x3, tmp12, xmask)
@triton.jit
def triton_poi_fused_add_div_mul_reflection_pad2d_relu_sqrt_sub_2(in_ptr0,
in_ptr1, in_ptr2, in_ptr3, in_ptr4, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 557568
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 66
x1 = xindex // 66 % 66
x4 = xindex // 4356
x2 = xindex // 4356 % 32
x6 = xindex
tmp0 = tl.load(in_ptr0 + (4095 + -1 * tl_math.abs(-63 + tl_math.abs(-1 +
x0)) + -64 * tl_math.abs(-63 + tl_math.abs(-1 + x1)) + 4096 * x4),
xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + x4, xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr2 + x4, xmask, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr3 + x2, xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr4 + x2, xmask, eviction_policy='evict_last')
tmp2 = tmp0 - tmp1
tmp4 = 0.999755859375
tmp5 = tmp3 * tmp4
tmp6 = 1e-09
tmp7 = tmp5 + tmp6
tmp8 = libdevice.sqrt(tmp7)
tmp9 = tmp2 / tmp8
tmp11 = tmp9 * tmp10
tmp13 = tmp11 + tmp12
tmp14 = tl.full([1], 0, tl.int32)
tmp15 = triton_helpers.maximum(tmp14, tmp13)
tl.store(out_ptr0 + x6, tmp15, xmask)
@triton.jit
def triton_per_fused_convolution_mean_var_3(in_out_ptr0, in_out_ptr1,
in_out_ptr2, in_ptr0, xnumel, rnumel):
XBLOCK: tl.constexpr = 1
RBLOCK: tl.constexpr = 1024
xoffset = tl.program_id(0) * XBLOCK
xindex = tl.full([1], xoffset, tl.int32)
tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
tl.full([RBLOCK], True, tl.int1)
r2 = rindex
x3 = xindex
x0 = xindex % 64
tmp0 = tl.load(in_out_ptr0 + (r2 + 1024 * x3), None)
tmp1 = tl.load(in_ptr0 + x0, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.broadcast_to(tmp2, [RBLOCK])
tmp5 = triton_helpers.promote_to_tensor(tl.sum(tmp3, 0))
tmp7 = tl.broadcast_to(tmp3, [RBLOCK])
tmp9 = triton_helpers.promote_to_tensor(tl.sum(tmp7, 0))
tmp10 = tl.full([1], 1024, tl.int32)
tmp11 = tmp10.to(tl.float32)
tmp12 = tmp9 / tmp11
tmp13 = tmp3 - tmp12
tmp14 = tmp13 * tmp13
tmp15 = tl.broadcast_to(tmp14, [RBLOCK])
tmp17 = triton_helpers.promote_to_tensor(tl.sum(tmp15, 0))
tmp18 = 1024.0
tmp19 = tmp5 / tmp18
tmp20 = 1023.0
tmp21 = tmp17 / tmp20
tl.store(in_out_ptr0 + (r2 + 1024 * x3), tmp2, None)
tl.debug_barrier()
tl.store(in_out_ptr1 + x3, tmp19, None)
tl.debug_barrier()
tl.store(in_out_ptr2 + x3, tmp21, None)
@triton.jit
def triton_poi_fused_add_div_mul_reflection_pad2d_relu_sqrt_sub_4(in_ptr0,
in_ptr1, in_ptr2, in_ptr3, in_ptr4, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 295936
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 34
x1 = xindex // 34 % 34
x4 = xindex // 1156
x2 = xindex // 1156 % 64
x6 = xindex
tmp0 = tl.load(in_ptr0 + (1023 + -1 * tl_math.abs(-31 + tl_math.abs(-1 +
x0)) + -32 * tl_math.abs(-31 + tl_math.abs(-1 + x1)) + 1024 * x4),
xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + x4, xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr2 + x4, xmask, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr3 + x2, xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr4 + x2, xmask, eviction_policy='evict_last')
tmp2 = tmp0 - tmp1
tmp4 = 0.9990234375
tmp5 = tmp3 * tmp4
tmp6 = 1e-09
tmp7 = tmp5 + tmp6
tmp8 = libdevice.sqrt(tmp7)
tmp9 = tmp2 / tmp8
tmp11 = tmp9 * tmp10
tmp13 = tmp11 + tmp12
tmp14 = tl.full([1], 0, tl.int32)
tmp15 = triton_helpers.maximum(tmp14, tmp13)
tl.store(out_ptr0 + x6, tmp15, xmask)
@triton.jit
def triton_per_fused_add_convolution_div_mean_mul_relu_sqrt_sub_var_5(
in_out_ptr0, in_out_ptr1, in_out_ptr2, in_ptr0, in_ptr1, in_ptr2,
out_ptr0, xnumel, rnumel):
XBLOCK: tl.constexpr = 1
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = tl.full([1], xoffset, tl.int32)
tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
tl.full([RBLOCK], True, tl.int1)
r2 = rindex
x3 = xindex
x0 = xindex % 128
tmp0 = tl.load(in_out_ptr0 + (r2 + 256 * x3), None)
tmp1 = tl.load(in_ptr0 + x0, None, eviction_policy='evict_last')
tmp29 = tl.load(in_ptr1 + x0, None, eviction_policy='evict_last')
tmp31 = tl.load(in_ptr2 + x0, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.broadcast_to(tmp2, [RBLOCK])
tmp5 = triton_helpers.promote_to_tensor(tl.sum(tmp3, 0))
tmp7 = tl.broadcast_to(tmp3, [RBLOCK])
tmp9 = triton_helpers.promote_to_tensor(tl.sum(tmp7, 0))
tmp10 = tl.full([1], 256, tl.int32)
tmp11 = tmp10.to(tl.float32)
tmp12 = tmp9 / tmp11
tmp13 = tmp3 - tmp12
tmp14 = tmp13 * tmp13
tmp15 = tl.broadcast_to(tmp14, [RBLOCK])
tmp17 = triton_helpers.promote_to_tensor(tl.sum(tmp15, 0))
tmp18 = 256.0
tmp19 = tmp5 / tmp18
tmp20 = 255.0
tmp21 = tmp17 / tmp20
tmp22 = tmp2 - tmp19
tmp23 = 0.99609375
tmp24 = tmp21 * tmp23
tmp25 = 1e-09
tmp26 = tmp24 + tmp25
tmp27 = libdevice.sqrt(tmp26)
tmp28 = tmp22 / tmp27
tmp30 = tmp28 * tmp29
tmp32 = tmp30 + tmp31
tmp33 = tl.full([1], 0, tl.int32)
tmp34 = triton_helpers.maximum(tmp33, tmp32)
tl.store(in_out_ptr0 + (r2 + 256 * x3), tmp2, None)
tl.debug_barrier()
tl.store(in_out_ptr1 + x3, tmp19, None)
tl.debug_barrier()
tl.store(in_out_ptr2 + x3, tmp21, None)
tl.store(out_ptr0 + (r2 + 256 * x3), tmp34, None)
@triton.jit
def triton_poi_fused_reflection_pad2d_6(in_ptr0, out_ptr0, xnumel, XBLOCK:
tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x0 = xindex % 18
x1 = xindex // 18 % 18
x2 = xindex // 324
x3 = xindex
tmp0 = tl.load(in_ptr0 + (255 + -1 * tl_math.abs(-15 + tl_math.abs(-1 +
x0)) + -16 * tl_math.abs(-15 + tl_math.abs(-1 + x1)) + 256 * x2),
None, eviction_policy='evict_last')
tl.store(out_ptr0 + x3, tmp0, None)
@triton.jit
def triton_per_fused_convolution_mean_var_7(in_out_ptr0, in_out_ptr1,
in_out_ptr2, in_ptr0, xnumel, rnumel):
XBLOCK: tl.constexpr = 1
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = tl.full([1], xoffset, tl.int32)
tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
tl.full([RBLOCK], True, tl.int1)
r2 = rindex
x3 = xindex
x0 = xindex % 128
tmp0 = tl.load(in_out_ptr0 + (r2 + 256 * x3), None)
tmp1 = tl.load(in_ptr0 + x0, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.broadcast_to(tmp2, [RBLOCK])
tmp5 = triton_helpers.promote_to_tensor(tl.sum(tmp3, 0))
tmp7 = tl.broadcast_to(tmp3, [RBLOCK])
tmp9 = triton_helpers.promote_to_tensor(tl.sum(tmp7, 0))
tmp10 = tl.full([1], 256, tl.int32)
tmp11 = tmp10.to(tl.float32)
tmp12 = tmp9 / tmp11
tmp13 = tmp3 - tmp12
tmp14 = tmp13 * tmp13
tmp15 = tl.broadcast_to(tmp14, [RBLOCK])
tmp17 = triton_helpers.promote_to_tensor(tl.sum(tmp15, 0))
tmp18 = 256.0
tmp19 = tmp5 / tmp18
tmp20 = 255.0
tmp21 = tmp17 / tmp20
tl.store(in_out_ptr0 + (r2 + 256 * x3), tmp2, None)
tl.debug_barrier()
tl.store(in_out_ptr1 + x3, tmp19, None)
tl.debug_barrier()
tl.store(in_out_ptr2 + x3, tmp21, None)
@triton.jit
def triton_poi_fused_add_div_mul_reflection_pad2d_relu_sqrt_sub_8(in_ptr0,
in_ptr1, in_ptr2, in_ptr3, in_ptr4, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x0 = xindex % 18
x1 = xindex // 18 % 18
x4 = xindex // 324
x2 = xindex // 324 % 128
x6 = xindex
tmp0 = tl.load(in_ptr0 + (255 + -1 * tl_math.abs(-15 + tl_math.abs(-1 +
x0)) + -16 * tl_math.abs(-15 + tl_math.abs(-1 + x1)) + 256 * x4),
None, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + x4, None, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr2 + x4, None, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr3 + x2, None, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr4 + x2, None, eviction_policy='evict_last')
tmp2 = tmp0 - tmp1
tmp4 = 0.99609375
tmp5 = tmp3 * tmp4
tmp6 = 1e-09
tmp7 = tmp5 + tmp6
tmp8 = libdevice.sqrt(tmp7)
tmp9 = tmp2 / tmp8
tmp11 = tmp9 * tmp10
tmp13 = tmp11 + tmp12
tmp14 = tl.full([1], 0, tl.int32)
tmp15 = triton_helpers.maximum(tmp14, tmp13)
tl.store(out_ptr0 + x6, tmp15, None)
@triton.jit
def triton_per_fused_add_convolution_div_mean_mul_sqrt_sub_var_9(in_out_ptr0,
in_out_ptr1, in_out_ptr2, in_out_ptr3, in_ptr0, in_ptr1, in_ptr2,
xnumel, rnumel):
XBLOCK: tl.constexpr = 1
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = tl.full([1], xoffset, tl.int32)
tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
tl.full([RBLOCK], True, tl.int1)
r2 = rindex
x3 = xindex
x0 = xindex % 128
tmp0 = tl.load(in_out_ptr0 + (r2 + 256 * x3), None)
tmp1 = tl.load(in_ptr0 + x0, None, eviction_policy='evict_last')
tmp29 = tl.load(in_ptr1 + x0, None, eviction_policy='evict_last')
tmp31 = tl.load(in_ptr2 + x0, None, eviction_policy='evict_last')
tmp33 = tl.load(in_out_ptr3 + (r2 + 256 * x3), None)
tmp2 = tmp0 + tmp1
tmp3 = tl.broadcast_to(tmp2, [RBLOCK])
tmp5 = triton_helpers.promote_to_tensor(tl.sum(tmp3, 0))
tmp7 = tl.broadcast_to(tmp3, [RBLOCK])
tmp9 = triton_helpers.promote_to_tensor(tl.sum(tmp7, 0))
tmp10 = tl.full([1], 256, tl.int32)
tmp11 = tmp10.to(tl.float32)
tmp12 = tmp9 / tmp11
tmp13 = tmp3 - tmp12
tmp14 = tmp13 * tmp13
tmp15 = tl.broadcast_to(tmp14, [RBLOCK])
tmp17 = triton_helpers.promote_to_tensor(tl.sum(tmp15, 0))
tmp18 = 256.0
tmp19 = tmp5 / tmp18
tmp20 = 255.0
tmp21 = tmp17 / tmp20
tmp22 = tmp2 - tmp19
tmp23 = 0.99609375
tmp24 = tmp21 * tmp23
tmp25 = 1e-09
tmp26 = tmp24 + tmp25
tmp27 = libdevice.sqrt(tmp26)
tmp28 = tmp22 / tmp27
tmp30 = tmp28 * tmp29
tmp32 = tmp30 + tmp31
tmp34 = tmp32 + tmp33
tl.store(in_out_ptr0 + (r2 + 256 * x3), tmp2, None)
tl.debug_barrier()
tl.store(in_out_ptr1 + x3, tmp19, None)
tl.debug_barrier()
tl.store(in_out_ptr2 + x3, tmp21, None)
tl.store(in_out_ptr3 + (r2 + 256 * x3), tmp34, None)
@triton.jit
def triton_poi_fused_arange_10(out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 32
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = x0
tl.store(out_ptr0 + x0, tmp0, xmask)
@triton.jit
def triton_poi_fused__to_copy_add_arange_mul_11(out_ptr0, xnumel, XBLOCK:
tl.constexpr):
xnumel = 32
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = x0
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.5
tmp3 = tmp1 * tmp2
tmp4 = tmp3.to(tl.int32)
tl.store(out_ptr0 + x0, tmp4, xmask)
@triton.jit
def triton_poi_fused__unsafe_index_add_div_mul_reflection_pad2d_sqrt_sub_12(
in_ptr0, in_ptr1, in_ptr2, in_ptr3, in_ptr4, in_ptr5, in_ptr6, out_ptr0,
xnumel, XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x1 = xindex // 34 % 34
x0 = xindex % 34
x4 = xindex // 1156
x2 = xindex // 1156 % 128
x7 = xindex
tmp0 = tl.load(in_ptr0 + (31 + -1 * tl_math.abs(-31 + tl_math.abs(-1 +
x1))), None, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (31 + -1 * tl_math.abs(-31 + tl_math.abs(-1 +
x0))), None, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr2 + x4, None, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr3 + x4, None, eviction_policy='evict_last')
tmp19 = tl.load(in_ptr4 + x2, None, eviction_policy='evict_last')
tmp21 = tl.load(in_ptr5 + x2, None, eviction_policy='evict_last')
tmp1 = tl.full([XBLOCK], 16, tl.int32)
tmp2 = tmp0 + tmp1
tmp3 = tmp0 < 0
tmp4 = tl.where(tmp3, tmp2, tmp0)
tmp6 = tmp5 + tmp1
tmp7 = tmp5 < 0
tmp8 = tl.where(tmp7, tmp6, tmp5)
tmp9 = tl.load(in_ptr1 + (tmp8 + 16 * tmp4 + 256 * x4), None,
eviction_policy='evict_last')
tmp11 = tmp9 - tmp10
tmp13 = 0.99609375
tmp14 = tmp12 * tmp13
tmp15 = 1e-09
tmp16 = tmp14 + tmp15
tmp17 = libdevice.sqrt(tmp16)
tmp18 = tmp11 / tmp17
tmp20 = tmp18 * tmp19
tmp22 = tmp20 + tmp21
tmp23 = tl.load(in_ptr6 + (tmp8 + 16 * tmp4 + 256 * x4), None,
eviction_policy='evict_last')
tmp24 = tmp22 + tmp23
tl.store(out_ptr0 + x7, tmp24, None)
@triton.jit
def triton_poi_fused_arange_13(out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = x0
tl.store(out_ptr0 + x0, tmp0, xmask)
@triton.jit
def triton_poi_fused__to_copy_add_arange_mul_14(out_ptr0, xnumel, XBLOCK:
tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = x0
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.5
tmp3 = tmp1 * tmp2
tmp4 = tmp3.to(tl.int32)
tl.store(out_ptr0 + x0, tmp4, xmask)
@triton.jit
def triton_poi_fused__unsafe_index_add_div_mul_reflection_pad2d_relu_sqrt_sub_15(
in_ptr0, in_ptr1, in_ptr2, in_ptr3, in_ptr4, in_ptr5, out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 1115136
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 66 % 66
x0 = xindex % 66
x4 = xindex // 4356
x2 = xindex // 4356 % 64
x7 = xindex
tmp0 = tl.load(in_ptr0 + (63 + -1 * tl_math.abs(-63 + tl_math.abs(-1 +
x1))), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (63 + -1 * tl_math.abs(-63 + tl_math.abs(-1 +
x0))), xmask, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr2 + x4, xmask, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr3 + x4, xmask, eviction_policy='evict_last')
tmp19 = tl.load(in_ptr4 + x2, xmask, eviction_policy='evict_last')
tmp21 = tl.load(in_ptr5 + x2, xmask, eviction_policy='evict_last')
tmp1 = tl.full([XBLOCK], 32, tl.int32)
tmp2 = tmp0 + tmp1
tmp3 = tmp0 < 0
tmp4 = tl.where(tmp3, tmp2, tmp0)
tmp6 = tmp5 + tmp1
tmp7 = tmp5 < 0
tmp8 = tl.where(tmp7, tmp6, tmp5)
tmp9 = tl.load(in_ptr1 + (tmp8 + 32 * tmp4 + 1024 * x4), xmask,
eviction_policy='evict_last')
tmp11 = tmp9 - tmp10
tmp13 = 0.9990234375
tmp14 = tmp12 * tmp13
tmp15 = 1e-09
tmp16 = tmp14 + tmp15
tmp17 = libdevice.sqrt(tmp16)
tmp18 = tmp11 / tmp17
tmp20 = tmp18 * tmp19
tmp22 = tmp20 + tmp21
tmp23 = tl.full([1], 0, tl.int32)
tmp24 = triton_helpers.maximum(tmp23, tmp22)
tl.store(out_ptr0 + x7, tmp24, xmask)
@triton.jit
def triton_poi_fused_add_div_mul_reflection_pad2d_relu_sqrt_sub_16(in_ptr0,
in_ptr1, in_ptr2, in_ptr3, in_ptr4, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x0 = xindex % 72
x1 = xindex // 72 % 72
x4 = xindex // 5184
x2 = xindex // 5184 % 32
x6 = xindex
tmp0 = tl.load(in_ptr0 + (4095 + -1 * tl_math.abs(-63 + tl_math.abs(-4 +
x0)) + -64 * tl_math.abs(-63 + tl_math.abs(-4 + x1)) + 4096 * x4),
None, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + x4, None, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr2 + x4, None, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr3 + x2, None, eviction_policy='evict_last')
tmp12 = tl.load(in_ptr4 + x2, None, eviction_policy='evict_last')
tmp2 = tmp0 - tmp1
tmp4 = 0.999755859375
tmp5 = tmp3 * tmp4
tmp6 = 1e-09
tmp7 = tmp5 + tmp6
tmp8 = libdevice.sqrt(tmp7)
tmp9 = tmp2 / tmp8
tmp11 = tmp9 * tmp10
tmp13 = tmp11 + tmp12
tmp14 = tl.full([1], 0, tl.int32)
tmp15 = triton_helpers.maximum(tmp14, tmp13)
tl.store(out_ptr0 + x6, tmp15, None)
@triton.jit
def triton_poi_fused_convolution_17(in_out_ptr0, in_ptr0, xnumel, XBLOCK:
tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x3 = xindex
x1 = xindex // 4096 % 3
tmp0 = tl.load(in_out_ptr0 + x3, None)
tmp1 = tl.load(in_ptr0 + x1, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(in_out_ptr0 + x3, tmp2, None)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7, primals_8, primals_9, primals_10, primals_11, primals_12,
primals_13, primals_14, primals_15, primals_16, primals_17,
primals_18, primals_19, primals_20, primals_21, primals_22,
primals_23, primals_24, primals_25, primals_26, primals_27,
primals_28, primals_29, primals_30, primals_31, primals_32,
primals_33, primals_34, primals_35, primals_36, primals_37,
primals_38, primals_39, primals_40, primals_41, primals_42,
primals_43, primals_44, primals_45, primals_46, primals_47,
primals_48, primals_49, primals_50, primals_51, primals_52,
primals_53, primals_54, primals_55, primals_56, primals_57,
primals_58, primals_59, primals_60, primals_61, primals_62, primals_63
) = args
args.clear()
assert_size_stride(primals_1, (4, 3, 64, 64), (12288, 4096, 64, 1))
assert_size_stride(primals_2, (32, 3, 9, 9), (243, 81, 9, 1))
assert_size_stride(primals_3, (32,), (1,))
assert_size_stride(primals_4, (32,), (1,))
assert_size_stride(primals_5, (32,), (1,))
assert_size_stride(primals_6, (64, 32, 3, 3), (288, 9, 3, 1))
assert_size_stride(primals_7, (64,), (1,))
assert_size_stride(primals_8, (64,), (1,))
assert_size_stride(primals_9, (64,), (1,))
assert_size_stride(primals_10, (128, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_11, (128,), (1,))
assert_size_stride(primals_12, (128,), (1,))
assert_size_stride(primals_13, (128,), (1,))
assert_size_stride(primals_14, (128, 128, 3, 3), (1152, 9, 3, 1))
assert_size_stride(primals_15, (128,), (1,))
assert_size_stride(primals_16, (128,), (1,))
assert_size_stride(primals_17, (128,), (1,))
assert_size_stride(primals_18, (128, 128, 3, 3), (1152, 9, 3, 1))
assert_size_stride(primals_19, (128,), (1,))
assert_size_stride(primals_20, (128,), (1,))
assert_size_stride(primals_21, (128,), (1,))
assert_size_stride(primals_22, (128, 128, 3, 3), (1152, 9, 3, 1))
assert_size_stride(primals_23, (128,), (1,))
assert_size_stride(primals_24, (128,), (1,))
assert_size_stride(primals_25, (128,), (1,))
assert_size_stride(primals_26, (128, 128, 3, 3), (1152, 9, 3, 1))
assert_size_stride(primals_27, (128,), (1,))
assert_size_stride(primals_28, (128,), (1,))
assert_size_stride(primals_29, (128,), (1,))
assert_size_stride(primals_30, (128, 128, 3, 3), (1152, 9, 3, 1))
assert_size_stride(primals_31, (128,), (1,))
assert_size_stride(primals_32, (128,), (1,))
assert_size_stride(primals_33, (128,), (1,))
assert_size_stride(primals_34, (128, 128, 3, 3), (1152, 9, 3, 1))
assert_size_stride(primals_35, (128,), (1,))
assert_size_stride(primals_36, (128,), (1,))
assert_size_stride(primals_37, (128,), (1,))
assert_size_stride(primals_38, (128, 128, 3, 3), (1152, 9, 3, 1))
assert_size_stride(primals_39, (128,), (1,))
assert_size_stride(primals_40, (128,), (1,))
assert_size_stride(primals_41, (128,), (1,))
assert_size_stride(primals_42, (128, 128, 3, 3), (1152, 9, 3, 1))
assert_size_stride(primals_43, (128,), (1,))
assert_size_stride(primals_44, (128,), (1,))
assert_size_stride(primals_45, (128,), (1,))
assert_size_stride(primals_46, (128, 128, 3, 3), (1152, 9, 3, 1))
assert_size_stride(primals_47, (128,), (1,))
assert_size_stride(primals_48, (128,), (1,))
assert_size_stride(primals_49, (128,), (1,))
assert_size_stride(primals_50, (128, 128, 3, 3), (1152, 9, 3, 1))
assert_size_stride(primals_51, (128,), (1,))
assert_size_stride(primals_52, (128,), (1,))
assert_size_stride(primals_53, (128,), (1,))
assert_size_stride(primals_54, (64, 128, 3, 3), (1152, 9, 3, 1))
assert_size_stride(primals_55, (64,), (1,))
assert_size_stride(primals_56, (64,), (1,))
assert_size_stride(primals_57, (64,), (1,))
assert_size_stride(primals_58, (32, 64, 3, 3), (576, 9, 3, 1))
assert_size_stride(primals_59, (32,), (1,))
assert_size_stride(primals_60, (32,), (1,))
assert_size_stride(primals_61, (32,), (1,))
assert_size_stride(primals_62, (3, 32, 9, 9), (2592, 81, 9, 1))
assert_size_stride(primals_63, (3,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 3, 72, 72), (15552, 5184, 72, 1),
torch.float32)
get_raw_stream(0)
triton_poi_fused_reflection_pad2d_0[grid(62208)](primals_1, buf0,
62208, XBLOCK=256, num_warps=4, num_stages=1)
del primals_1
buf1 = extern_kernels.convolution(buf0, primals_2, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf1, (4, 32, 64, 64), (131072, 4096, 64, 1))
buf2 = buf1
del buf1
buf3 = empty_strided_cuda((4, 32), (32, 1), torch.float32)
buf6 = empty_strided_cuda((4, 32), (32, 1), torch.float32)
buf4 = buf3
del buf3
buf8 = buf6
del buf6
triton_red_fused_convolution_mean_var_1[grid(128)](buf2, buf4, buf8,
primals_3, 128, 4096, XBLOCK=1, RBLOCK=2048, num_warps=16,
num_stages=1)
del primals_3
buf9 = empty_strided_cuda((4, 32, 66, 66), (139392, 4356, 66, 1),
torch.float32)
triton_poi_fused_add_div_mul_reflection_pad2d_relu_sqrt_sub_2[grid(
557568)](buf2, buf4, buf8, primals_4, primals_5, buf9, 557568,
XBLOCK=512, num_warps=8, num_stages=1)
buf10 = extern_kernels.convolution(buf9, primals_6, stride=(2, 2),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf10, (4, 64, 32, 32), (65536, 1024, 32, 1))
buf11 = buf10
del buf10
buf12 = empty_strided_cuda((4, 64), (64, 1), torch.float32)
buf15 = empty_strided_cuda((4, 64), (64, 1), torch.float32)
buf13 = buf12
del buf12
buf17 = buf15
del buf15
triton_per_fused_convolution_mean_var_3[grid(256)](buf11, buf13,
buf17, primals_7, 256, 1024, num_warps=8, num_stages=1)
del primals_7
buf18 = empty_strided_cuda((4, 64, 34, 34), (73984, 1156, 34, 1),
torch.float32)
triton_poi_fused_add_div_mul_reflection_pad2d_relu_sqrt_sub_4[grid(
295936)](buf11, buf13, buf17, primals_8, primals_9, buf18,
295936, XBLOCK=1024, num_warps=4, num_stages=1)
buf19 = extern_kernels.convolution(buf18, primals_10, stride=(2, 2),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf19, (4, 128, 16, 16), (32768, 256, 16, 1))
buf20 = buf19
del buf19
buf21 = empty_strided_cuda((4, 128), (128, 1), torch.float32)
buf24 = empty_strided_cuda((4, 128), (128, 1), torch.float32)
buf22 = buf21
del buf21
buf26 = buf24
del buf24
buf27 = empty_strided_cuda((4, 128, 16, 16), (32768, 256, 16, 1),
torch.float32)
triton_per_fused_add_convolution_div_mean_mul_relu_sqrt_sub_var_5[grid
(512)](buf20, buf22, buf26, primals_11, primals_12, primals_13,
buf27, 512, 256, num_warps=2, num_stages=1)
del primals_11
buf28 = empty_strided_cuda((4, 128, 18, 18), (41472, 324, 18, 1),
torch.float32)
triton_poi_fused_reflection_pad2d_6[grid(165888)](buf27, buf28,
165888, XBLOCK=512, num_warps=8, num_stages=1)
buf29 = extern_kernels.convolution(buf28, primals_14, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf29, (4, 128, 16, 16), (32768, 256, 16, 1))
buf30 = buf29
del buf29
buf31 = empty_strided_cuda((4, 128), (128, 1), torch.float32)
buf34 = empty_strided_cuda((4, 128), (128, 1), torch.float32)
buf32 = buf31
del buf31
buf36 = buf34
del buf34
triton_per_fused_convolution_mean_var_7[grid(512)](buf30, buf32,
buf36, primals_15, 512, 256, num_warps=2, num_stages=1)
del primals_15
buf37 = empty_strided_cuda((4, 128, 18, 18), (41472, 324, 18, 1),
torch.float32)
triton_poi_fused_add_div_mul_reflection_pad2d_relu_sqrt_sub_8[grid(
165888)](buf30, buf32, buf36, primals_16, primals_17, buf37,
165888, XBLOCK=512, num_warps=8, num_stages=1)
buf38 = extern_kernels.convolution(buf37, primals_18, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf38, (4, 128, 16, 16), (32768, 256, 16, 1))
buf39 = buf38
del buf38
buf40 = empty_strided_cuda((4, 128), (128, 1), torch.float32)
buf43 = empty_strided_cuda((4, 128), (128, 1), torch.float32)
buf41 = buf40
del buf40
buf45 = buf43
del buf43
buf46 = buf27
del buf27
triton_per_fused_add_convolution_div_mean_mul_sqrt_sub_var_9[grid(512)
](buf39, buf41, buf45, buf46, primals_19, primals_20,
primals_21, 512, 256, num_warps=2, num_stages=1)
del primals_19
del primals_21
buf47 = empty_strided_cuda((4, 128, 18, 18), (41472, 324, 18, 1),
torch.float32)
triton_poi_fused_reflection_pad2d_6[grid(165888)](buf46, buf47,
165888, XBLOCK=512, num_warps=8, num_stages=1)
buf48 = extern_kernels.convolution(buf47, primals_22, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf48, (4, 128, 16, 16), (32768, 256, 16, 1))
buf49 = buf48
del buf48
buf50 = empty_strided_cuda((4, 128), (128, 1), torch.float32)
buf53 = empty_strided_cuda((4, 128), (128, 1), torch.float32)
buf51 = buf50
del buf50
buf55 = buf53
del buf53
triton_per_fused_convolution_mean_var_7[grid(512)](buf49, buf51,
buf55, primals_23, 512, 256, num_warps=2, num_stages=1)
del primals_23
buf56 = empty_strided_cuda((4, 128, 18, 18), (41472, 324, 18, 1),
torch.float32)
triton_poi_fused_add_div_mul_reflection_pad2d_relu_sqrt_sub_8[grid(
165888)](buf49, buf51, buf55, primals_24, primals_25, buf56,
165888, XBLOCK=512, num_warps=8, num_stages=1)
buf57 = extern_kernels.convolution(buf56, primals_26, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf57, (4, 128, 16, 16), (32768, 256, 16, 1))
buf58 = buf57
del buf57
buf59 = empty_strided_cuda((4, 128), (128, 1), torch.float32)
buf62 = empty_strided_cuda((4, 128), (128, 1), torch.float32)
buf60 = buf59
del buf59
buf64 = buf62
del buf62
buf65 = buf46
del buf46
triton_per_fused_add_convolution_div_mean_mul_sqrt_sub_var_9[grid(512)
](buf58, buf60, buf64, buf65, primals_27, primals_28,
primals_29, 512, 256, num_warps=2, num_stages=1)
del primals_27
del primals_29
buf66 = empty_strided_cuda((4, 128, 18, 18), (41472, 324, 18, 1),
torch.float32)
triton_poi_fused_reflection_pad2d_6[grid(165888)](buf65, buf66,
165888, XBLOCK=512, num_warps=8, num_stages=1)
buf67 = extern_kernels.convolution(buf66, primals_30, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf67, (4, 128, 16, 16), (32768, 256, 16, 1))
buf68 = buf67
del buf67
buf69 = empty_strided_cuda((4, 128), (128, 1), torch.float32)
buf72 = empty_strided_cuda((4, 128), (128, 1), torch.float32)
buf70 = buf69
del buf69
buf74 = buf72
del buf72
triton_per_fused_convolution_mean_var_7[grid(512)](buf68, buf70,
buf74, primals_31, 512, 256, num_warps=2, num_stages=1)
del primals_31
buf75 = empty_strided_cuda((4, 128, 18, 18), (41472, 324, 18, 1),
torch.float32)
triton_poi_fused_add_div_mul_reflection_pad2d_relu_sqrt_sub_8[grid(
165888)](buf68, buf70, buf74, primals_32, primals_33, buf75,
165888, XBLOCK=512, num_warps=8, num_stages=1)
buf76 = extern_kernels.convolution(buf75, primals_34, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf76, (4, 128, 16, 16), (32768, 256, 16, 1))
buf77 = buf76
del buf76
buf78 = empty_strided_cuda((4, 128), (128, 1), torch.float32)
buf81 = empty_strided_cuda((4, 128), (128, 1), torch.float32)
buf79 = buf78
del buf78
buf83 = buf81
del buf81
buf84 = buf65
del buf65
triton_per_fused_add_convolution_div_mean_mul_sqrt_sub_var_9[grid(512)
](buf77, buf79, buf83, buf84, primals_35, primals_36,
primals_37, 512, 256, num_warps=2, num_stages=1)
del primals_35
del primals_37
buf85 = empty_strided_cuda((4, 128, 18, 18), (41472, 324, 18, 1),
torch.float32)
triton_poi_fused_reflection_pad2d_6[grid(165888)](buf84, buf85,
165888, XBLOCK=512, num_warps=8, num_stages=1)
buf86 = extern_kernels.convolution(buf85, primals_38, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf86, (4, 128, 16, 16), (32768, 256, 16, 1))
buf87 = buf86
del buf86
buf88 = empty_strided_cuda((4, 128), (128, 1), torch.float32)
buf91 = empty_strided_cuda((4, 128), (128, 1), torch.float32)
buf89 = buf88
del buf88
buf93 = buf91
del buf91
triton_per_fused_convolution_mean_var_7[grid(512)](buf87, buf89,
buf93, primals_39, 512, 256, num_warps=2, num_stages=1)
del primals_39
buf94 = empty_strided_cuda((4, 128, 18, 18), (41472, 324, 18, 1),
torch.float32)
triton_poi_fused_add_div_mul_reflection_pad2d_relu_sqrt_sub_8[grid(
165888)](buf87, buf89, buf93, primals_40, primals_41, buf94,
165888, XBLOCK=512, num_warps=8, num_stages=1)
buf95 = extern_kernels.convolution(buf94, primals_42, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf95, (4, 128, 16, 16), (32768, 256, 16, 1))
buf96 = buf95
del buf95
buf97 = empty_strided_cuda((4, 128), (128, 1), torch.float32)
buf100 = empty_strided_cuda((4, 128), (128, 1), torch.float32)
buf98 = buf97
del buf97
buf102 = buf100
del buf100
buf103 = buf84
del buf84
triton_per_fused_add_convolution_div_mean_mul_sqrt_sub_var_9[grid(512)
](buf96, buf98, buf102, buf103, primals_43, primals_44,
primals_45, 512, 256, num_warps=2, num_stages=1)
del primals_43
del primals_45
buf104 = empty_strided_cuda((4, 128, 18, 18), (41472, 324, 18, 1),
torch.float32)
triton_poi_fused_reflection_pad2d_6[grid(165888)](buf103, buf104,
165888, XBLOCK=512, num_warps=8, num_stages=1)
buf105 = extern_kernels.convolution(buf104, primals_46, stride=(1,
1), padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf105, (4, 128, 16, 16), (32768, 256, 16, 1))
buf106 = buf105
del buf105
buf107 = empty_strided_cuda((4, 128), (128, 1), torch.float32)
buf110 = empty_strided_cuda((4, 128), (128, 1), torch.float32)
buf108 = buf107
del buf107
buf112 = buf110
del buf110
triton_per_fused_convolution_mean_var_7[grid(512)](buf106, buf108,
buf112, primals_47, 512, 256, num_warps=2, num_stages=1)
del primals_47
buf113 = empty_strided_cuda((4, 128, 18, 18), (41472, 324, 18, 1),
torch.float32)
triton_poi_fused_add_div_mul_reflection_pad2d_relu_sqrt_sub_8[grid(
165888)](buf106, buf108, buf112, primals_48, primals_49, buf113,
165888, XBLOCK=512, num_warps=8, num_stages=1)
buf114 = extern_kernels.convolution(buf113, primals_50, stride=(1,
1), padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf114, (4, 128, 16, 16), (32768, 256, 16, 1))
buf115 = buf114
del buf114
buf116 = empty_strided_cuda((4, 128), (128, 1), torch.float32)
buf119 = empty_strided_cuda((4, 128), (128, 1), torch.float32)
buf117 = buf116
del buf116
buf121 = buf119
del buf119
triton_per_fused_convolution_mean_var_7[grid(512)](buf115, buf117,
buf121, primals_51, 512, 256, num_warps=2, num_stages=1)
del primals_51
buf122 = empty_strided_cuda((32,), (1,), torch.int64)
triton_poi_fused_arange_10[grid(32)](buf122, 32, XBLOCK=32,
num_warps=1, num_stages=1)
buf123 = empty_strided_cuda((32,), (1,), torch.int64)
triton_poi_fused__to_copy_add_arange_mul_11[grid(32)](buf123, 32,
XBLOCK=32, num_warps=1, num_stages=1)
buf124 = empty_strided_cuda((4, 128, 34, 34), (147968, 1156, 34, 1),
torch.float32)
triton_poi_fused__unsafe_index_add_div_mul_reflection_pad2d_sqrt_sub_12[
grid(591872)](buf123, buf115, buf117, buf121, primals_52,
primals_53, buf103, buf124, 591872, XBLOCK=512, num_warps=8,
num_stages=1)
del buf103
del primals_53
buf125 = extern_kernels.convolution(buf124, primals_54, stride=(1,
1), padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf125, (4, 64, 32, 32), (65536, 1024, 32, 1))
buf126 = buf125
del buf125
buf127 = empty_strided_cuda((4, 64), (64, 1), torch.float32)
buf130 = empty_strided_cuda((4, 64), (64, 1), torch.float32)
buf128 = buf127
del buf127
buf132 = buf130
del buf130
triton_per_fused_convolution_mean_var_3[grid(256)](buf126, buf128,
buf132, primals_55, 256, 1024, num_warps=8, num_stages=1)
del primals_55
buf133 = empty_strided_cuda((64,), (1,), torch.int64)
triton_poi_fused_arange_13[grid(64)](buf133, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf134 = empty_strided_cuda((64,), (1,), torch.int64)
triton_poi_fused__to_copy_add_arange_mul_14[grid(64)](buf134, 64,
XBLOCK=64, num_warps=1, num_stages=1)
buf135 = empty_strided_cuda((4, 64, 66, 66), (278784, 4356, 66, 1),
torch.float32)
triton_poi_fused__unsafe_index_add_div_mul_reflection_pad2d_relu_sqrt_sub_15[
grid(1115136)](buf134, buf126, buf128, buf132, primals_56,
primals_57, buf135, 1115136, XBLOCK=512, num_warps=8, num_stages=1)
buf136 = extern_kernels.convolution(buf135, primals_58, stride=(1,
1), padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf136, (4, 32, 64, 64), (131072, 4096, 64, 1))
buf137 = buf136
del buf136
buf138 = empty_strided_cuda((4, 32), (32, 1), torch.float32)
buf141 = empty_strided_cuda((4, 32), (32, 1), torch.float32)
buf139 = buf138
del buf138
buf143 = buf141
del buf141
triton_red_fused_convolution_mean_var_1[grid(128)](buf137, buf139,
buf143, primals_59, 128, 4096, XBLOCK=1, RBLOCK=2048, num_warps
=16, num_stages=1)
del primals_59
buf144 = empty_strided_cuda((4, 32, 72, 72), (165888, 5184, 72, 1),
torch.float32)
triton_poi_fused_add_div_mul_reflection_pad2d_relu_sqrt_sub_16[grid
(663552)](buf137, buf139, buf143, primals_60, primals_61,
buf144, 663552, XBLOCK=512, num_warps=8, num_stages=1)
buf145 = extern_kernels.convolution(buf144, primals_62, stride=(1,
1), padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf145, (4, 3, 64, 64), (12288, 4096, 64, 1))
buf146 = buf145
del buf145
triton_poi_fused_convolution_17[grid(49152)](buf146, primals_63,
49152, XBLOCK=256, num_warps=4, num_stages=1)
del primals_63
return (buf146, primals_2, primals_4, primals_5, primals_6, primals_8,
primals_9, primals_10, primals_12, primals_13, primals_14,
primals_16, primals_17, primals_18, primals_20, primals_22,
primals_24, primals_25, primals_26, primals_28, primals_30,
primals_32, primals_33, primals_34, primals_36, primals_38,
primals_40, primals_41, primals_42, primals_44, primals_46,
primals_48, primals_49, primals_50, primals_52, primals_54,
primals_56, primals_57, primals_58, primals_60, primals_61,
primals_62, buf0, buf2, reinterpret_tensor(buf4, (4, 32, 1, 1), (32,
1, 1, 1), 0), reinterpret_tensor(buf8, (4, 32, 1, 1), (32, 1, 1, 1),
0), buf9, buf11, reinterpret_tensor(buf13, (4, 64, 1, 1), (64, 1, 1,
1), 0), reinterpret_tensor(buf17, (4, 64, 1, 1), (64, 1, 1, 1), 0),
buf18, buf20, reinterpret_tensor(buf22, (4, 128, 1, 1), (128, 1, 1,
1), 0), reinterpret_tensor(buf26, (4, 128, 1, 1), (128, 1, 1, 1), 0
), buf28, buf30, reinterpret_tensor(buf32, (4, 128, 1, 1), (128, 1,
1, 1), 0), reinterpret_tensor(buf36, (4, 128, 1, 1), (128, 1, 1, 1),
0), buf37, buf39, reinterpret_tensor(buf41, (4, 128, 1, 1), (128, 1,
1, 1), 0), reinterpret_tensor(buf45, (4, 128, 1, 1), (128, 1, 1, 1),
0), buf47, buf49, reinterpret_tensor(buf51, (4, 128, 1, 1), (128, 1,
1, 1), 0), reinterpret_tensor(buf55, (4, 128, 1, 1), (128, 1, 1, 1),
0), buf56, buf58, reinterpret_tensor(buf60, (4, 128, 1, 1), (128, 1,
1, 1), 0), reinterpret_tensor(buf64, (4, 128, 1, 1), (128, 1, 1, 1),
0), buf66, buf68, reinterpret_tensor(buf70, (4, 128, 1, 1), (128, 1,
1, 1), 0), reinterpret_tensor(buf74, (4, 128, 1, 1), (128, 1, 1, 1),
0), buf75, buf77, reinterpret_tensor(buf79, (4, 128, 1, 1), (128, 1,
1, 1), 0), reinterpret_tensor(buf83, (4, 128, 1, 1), (128, 1, 1, 1),
0), buf85, buf87, reinterpret_tensor(buf89, (4, 128, 1, 1), (128, 1,
1, 1), 0), reinterpret_tensor(buf93, (4, 128, 1, 1), (128, 1, 1, 1),
0), buf94, buf96, reinterpret_tensor(buf98, (4, 128, 1, 1), (128, 1,
1, 1), 0), reinterpret_tensor(buf102, (4, 128, 1, 1), (128, 1, 1, 1
), 0), buf104, buf106, reinterpret_tensor(buf108, (4, 128, 1, 1), (
128, 1, 1, 1), 0), reinterpret_tensor(buf112, (4, 128, 1, 1), (128,
1, 1, 1), 0), buf113, buf115, reinterpret_tensor(buf117, (4, 128, 1,
1), (128, 1, 1, 1), 0), reinterpret_tensor(buf121, (4, 128, 1, 1),
(128, 1, 1, 1), 0), buf122, buf123, buf124, buf126,
reinterpret_tensor(buf128, (4, 64, 1, 1), (64, 1, 1, 1), 0),
reinterpret_tensor(buf132, (4, 64, 1, 1), (64, 1, 1, 1), 0), buf133,
buf134, buf135, buf137, reinterpret_tensor(buf139, (4, 32, 1, 1), (
32, 1, 1, 1), 0), reinterpret_tensor(buf143, (4, 32, 1, 1), (32, 1,
1, 1), 0), buf144)
class ConvLayer(torch.nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, stride):
super(ConvLayer, self).__init__()
reflection_padding = int(np.floor(kernel_size / 2))
self.reflection_pad = nn.ReflectionPad2d(reflection_padding)
self.conv2d = nn.Conv2d(in_channels, out_channels, kernel_size, stride)
def forward(self, x):
out = self.reflection_pad(x)
out = self.conv2d(out)
return out
class InstanceNormalization(torch.nn.Module):
"""InstanceNormalization
Improves convergence of neural-style.
ref: https://arxiv.org/pdf/1607.08022.pdf
"""
def __init__(self, dim, eps=1e-09):
super(InstanceNormalization, self).__init__()
self.scale = nn.Parameter(torch.FloatTensor(dim))
self.shift = nn.Parameter(torch.FloatTensor(dim))
self.eps = eps
self._reset_parameters()
def _reset_parameters(self):
self.scale.data.uniform_()
self.shift.data.zero_()
def forward(self, x):
n = x.size(2) * x.size(3)
t = x.view(x.size(0), x.size(1), n)
mean = torch.mean(t, 2).unsqueeze(2).unsqueeze(3).expand_as(x)
var = torch.var(t, 2).unsqueeze(2).unsqueeze(3).expand_as(x) * ((n -
1) / float(n))
scale_broadcast = self.scale.unsqueeze(1).unsqueeze(1).unsqueeze(0)
scale_broadcast = scale_broadcast.expand_as(x)
shift_broadcast = self.shift.unsqueeze(1).unsqueeze(1).unsqueeze(0)
shift_broadcast = shift_broadcast.expand_as(x)
out = (x - mean) / torch.sqrt(var + self.eps)
out = out * scale_broadcast + shift_broadcast
return out
class ResidualBlock(torch.nn.Module):
"""ResidualBlock
introduced in: https://arxiv.org/abs/1512.03385
recommended architecture: http://torch.ch/blog/2016/02/04/resnets.html
"""
def __init__(self, channels):
super(ResidualBlock, self).__init__()
self.conv1 = ConvLayer(channels, channels, kernel_size=3, stride=1)
self.in1 = InstanceNormalization(channels)
self.conv2 = ConvLayer(channels, channels, kernel_size=3, stride=1)
self.in2 = InstanceNormalization(channels)
self.relu = nn.ReLU()
def forward(self, x):
residual = x
out = self.relu(self.in1(self.conv1(x)))
out = self.in2(self.conv2(out))
out = out + residual
return out
class UpsampleConvLayer(torch.nn.Module):
"""UpsampleConvLayer
Upsamples the input and then does a convolution. This method gives better results
compared to ConvTranspose2d.
ref: http://distill.pub/2016/deconv-checkerboard/
"""
def __init__(self, in_channels, out_channels, kernel_size, stride,
upsample=None):
super(UpsampleConvLayer, self).__init__()
self.upsample = upsample
if upsample:
self.upsample_layer = torch.nn.Upsample(scale_factor=upsample)
reflection_padding = int(np.floor(kernel_size / 2))
self.reflection_pad = nn.ReflectionPad2d(reflection_padding)
self.conv2d = nn.Conv2d(in_channels, out_channels, kernel_size, stride)
def forward(self, x):
x_in = x
if self.upsample:
x_in = self.upsample_layer(x_in)
out = self.reflection_pad(x_in)
out = self.conv2d(out)
return out
class TransformerNetNew(torch.nn.Module):
def __init__(self):
super(TransformerNetNew, self).__init__()
self.conv1 = ConvLayer(3, 32, kernel_size=9, stride=1)
self.in1 = InstanceNormalization(32)
self.conv2 = ConvLayer(32, 64, kernel_size=3, stride=2)
self.in2 = InstanceNormalization(64)
self.conv3 = ConvLayer(64, 128, kernel_size=3, stride=2)
self.in3 = InstanceNormalization(128)
self.res1 = ResidualBlock(128)
self.res2 = ResidualBlock(128)
self.res3 = ResidualBlock(128)
self.res4 = ResidualBlock(128)
self.res5 = ResidualBlock(128)
self.deconv1 = UpsampleConvLayer(128, 64, kernel_size=3, stride=1,
upsample=2)
self.in4 = InstanceNormalization(64)
self.deconv2 = UpsampleConvLayer(64, 32, kernel_size=3, stride=1,
upsample=2)
self.in5 = InstanceNormalization(32)
self.deconv3 = ConvLayer(32, 3, kernel_size=9, stride=1)
self.relu = nn.ReLU()
def forward(self, input_0):
primals_2 = self.conv1.conv2d.weight
primals_3 = self.conv1.conv2d.bias
primals_4 = self.in1.scale
primals_5 = self.in1.shift
primals_6 = self.conv2.conv2d.weight
primals_7 = self.conv2.conv2d.bias
primals_8 = self.in2.scale
primals_9 = self.in2.shift
primals_10 = self.conv3.conv2d.weight
primals_11 = self.conv3.conv2d.bias
primals_12 = self.in3.scale
primals_13 = self.in3.shift
primals_14 = self.res1.conv1.conv2d.weight
primals_15 = self.res1.conv1.conv2d.bias
primals_16 = self.res1.in1.scale
primals_17 = self.res1.in1.shift
primals_18 = self.res1.conv2.conv2d.weight
primals_19 = self.res1.conv2.conv2d.bias
primals_20 = self.res1.in2.scale
primals_21 = self.res1.in2.shift
primals_22 = self.res2.conv1.conv2d.weight
primals_23 = self.res2.conv1.conv2d.bias
primals_24 = self.res2.in1.scale
primals_25 = self.res2.in1.shift
primals_26 = self.res2.conv2.conv2d.weight
primals_27 = self.res2.conv2.conv2d.bias
primals_28 = self.res2.in2.scale
primals_29 = self.res2.in2.shift
primals_30 = self.res3.conv1.conv2d.weight
primals_31 = self.res3.conv1.conv2d.bias
primals_32 = self.res3.in1.scale
primals_33 = self.res3.in1.shift
primals_34 = self.res3.conv2.conv2d.weight
primals_35 = self.res3.conv2.conv2d.bias
primals_36 = self.res3.in2.scale
primals_37 = self.res3.in2.shift
primals_38 = self.res4.conv1.conv2d.weight
primals_39 = self.res4.conv1.conv2d.bias
primals_40 = self.res4.in1.scale
primals_41 = self.res4.in1.shift
primals_42 = self.res4.conv2.conv2d.weight
primals_43 = self.res4.conv2.conv2d.bias
primals_44 = self.res4.in2.scale
primals_45 = self.res4.in2.shift
primals_46 = self.res5.conv1.conv2d.weight
primals_47 = self.res5.conv1.conv2d.bias
primals_48 = self.res5.in1.scale
primals_49 = self.res5.in1.shift
primals_50 = self.res5.conv2.conv2d.weight
primals_51 = self.res5.conv2.conv2d.bias
primals_52 = self.res5.in2.scale
primals_53 = self.res5.in2.shift
primals_54 = self.deconv1.conv2d.weight
primals_55 = self.deconv1.conv2d.bias
primals_56 = self.in4.scale
primals_57 = self.in4.shift
primals_58 = self.deconv2.conv2d.weight
primals_59 = self.deconv2.conv2d.bias
primals_60 = self.in5.scale
primals_61 = self.in5.shift
primals_62 = self.deconv3.conv2d.weight
primals_63 = self.deconv3.conv2d.bias
primals_1 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7, primals_8, primals_9,
primals_10, primals_11, primals_12, primals_13, primals_14,
primals_15, primals_16, primals_17, primals_18, primals_19,
primals_20, primals_21, primals_22, primals_23, primals_24,
primals_25, primals_26, primals_27, primals_28, primals_29,
primals_30, primals_31, primals_32, primals_33, primals_34,
primals_35, primals_36, primals_37, primals_38, primals_39,
primals_40, primals_41, primals_42, primals_43, primals_44,
primals_45, primals_46, primals_47, primals_48, primals_49,
primals_50, primals_51, primals_52, primals_53, primals_54,
primals_55, primals_56, primals_57, primals_58, primals_59,
primals_60, primals_61, primals_62, primals_63])
return output[0]
|
ImageProcessingCentraleLille2021/fast-neural-style
|
TransformerNet
| false
| 13,926
|
[
"MIT"
] | 350
|
e77456c35c2a49f90227119d158828a0964c7e13
|
https://github.com/ImageProcessingCentraleLille2021/fast-neural-style/tree/e77456c35c2a49f90227119d158828a0964c7e13
|
QNetwork
|
import torch
from torch import nn
class QNetwork(nn.Module):
def __init__(self, num_states, num_actions):
super().__init__()
self._num_states = num_states
self._num_actions = num_actions
self._fc1 = nn.Linear(self._num_states, 100)
self._relu1 = nn.ReLU(inplace=True)
self._fc2 = nn.Linear(100, 60)
self._relu2 = nn.ReLU(inplace=True)
self._fc_final = nn.Linear(60, self._num_actions)
nn.init.zeros_(self._fc1.bias)
nn.init.zeros_(self._fc2.bias)
nn.init.zeros_(self._fc_final.bias)
nn.init.uniform_(self._fc_final.weight, a=-1e-06, b=1e-06)
def forward(self, state):
h = self._relu1(self._fc1(state))
h = self._relu2(self._fc2(h))
q_values = self._fc_final(h)
return q_values
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'num_states': 4, 'num_actions': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch import nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_relu_threshold_backward_0(in_out_ptr0, in_ptr0,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 6400
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x4 = xindex
x0 = xindex % 100
x3 = xindex // 1600
x5 = xindex % 1600
tmp0 = tl.load(in_out_ptr0 + x4, xmask)
tmp1 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(in_out_ptr0 + x4, tmp4, xmask)
tl.store(out_ptr0 + (x5 + 1664 * x3), tmp6, xmask)
@triton.jit
def triton_poi_fused_view_1(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 6400
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 100
x1 = xindex // 100
x2 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + 100 * x1 + 400 * (x1 % 4 // 4) + 1600 *
((4 * (x1 // 4 % 4) + x1 % 4) // 16)), xmask)
tl.store(out_ptr0 + x2, tmp0, xmask)
@triton.jit
def triton_poi_fused_relu_threshold_backward_2(in_out_ptr0, in_ptr0,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 3840
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x4 = xindex
x0 = xindex % 60
tmp0 = tl.load(in_out_ptr0 + x4, xmask)
tmp1 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(in_out_ptr0 + x4, tmp4, xmask)
tl.store(out_ptr0 + x4, tmp6, xmask)
@triton.jit
def triton_poi_fused_view_3(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 3840
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 60
x1 = xindex // 60
x2 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + 60 * x1 + 240 * (x1 % 4 // 4) + 960 * ((
4 * (x1 // 4 % 4) + x1 % 4) // 16)), xmask)
tl.store(out_ptr0 + x2, tmp0, xmask)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7) = args
args.clear()
assert_size_stride(primals_1, (100, 4), (4, 1))
assert_size_stride(primals_2, (100,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (60, 100), (100, 1))
assert_size_stride(primals_5, (60,), (1,))
assert_size_stride(primals_6, (4, 60), (60, 1))
assert_size_stride(primals_7, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 100), (100, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_3, (64, 4), (4, 1), 0),
reinterpret_tensor(primals_1, (4, 100), (1, 4), 0), out=buf0)
del primals_1
buf1 = reinterpret_tensor(buf0, (4, 4, 4, 100), (1600, 400, 100, 1), 0)
del buf0
buf8 = empty_strided_cuda((4, 4, 4, 100), (1664, 400, 100, 1),
torch.bool)
get_raw_stream(0)
triton_poi_fused_relu_threshold_backward_0[grid(6400)](buf1,
primals_2, buf8, 6400, XBLOCK=128, num_warps=4, num_stages=1)
del primals_2
buf2 = empty_strided_cuda((64, 100), (100, 1), torch.float32)
triton_poi_fused_view_1[grid(6400)](buf1, buf2, 6400, XBLOCK=128,
num_warps=4, num_stages=1)
del buf1
buf3 = empty_strided_cuda((64, 60), (60, 1), torch.float32)
extern_kernels.mm(buf2, reinterpret_tensor(primals_4, (100, 60), (1,
100), 0), out=buf3)
buf4 = reinterpret_tensor(buf3, (4, 4, 4, 60), (960, 240, 60, 1), 0)
del buf3
buf7 = empty_strided_cuda((4, 4, 4, 60), (960, 240, 60, 1), torch.bool)
triton_poi_fused_relu_threshold_backward_2[grid(3840)](buf4,
primals_5, buf7, 3840, XBLOCK=128, num_warps=4, num_stages=1)
del primals_5
buf5 = empty_strided_cuda((64, 60), (60, 1), torch.float32)
triton_poi_fused_view_3[grid(3840)](buf4, buf5, 3840, XBLOCK=256,
num_warps=4, num_stages=1)
del buf4
buf6 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_7, buf5, reinterpret_tensor(primals_6,
(60, 4), (1, 60), 0), alpha=1, beta=1, out=buf6)
del primals_7
return reinterpret_tensor(buf6, (4, 4, 4, 4), (64, 16, 4, 1), 0
), reinterpret_tensor(primals_3, (64, 4), (4, 1), 0
), buf2, buf5, primals_6, buf7, primals_4, buf8
class QNetworkNew(nn.Module):
def __init__(self, num_states, num_actions):
super().__init__()
self._num_states = num_states
self._num_actions = num_actions
self._fc1 = nn.Linear(self._num_states, 100)
self._relu1 = nn.ReLU(inplace=True)
self._fc2 = nn.Linear(100, 60)
self._relu2 = nn.ReLU(inplace=True)
self._fc_final = nn.Linear(60, self._num_actions)
nn.init.zeros_(self._fc1.bias)
nn.init.zeros_(self._fc2.bias)
nn.init.zeros_(self._fc_final.bias)
nn.init.uniform_(self._fc_final.weight, a=-1e-06, b=1e-06)
def forward(self, input_0):
primals_1 = self._fc1.weight
primals_2 = self._fc1.bias
primals_4 = self._fc2.weight
primals_5 = self._fc2.bias
primals_6 = self._fc_final.weight
primals_7 = self._fc_final.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7])
return output[0]
|
JulianoLagana/deep-machine-learning
|
QNetwork
| false
| 13,927
|
[
"MIT"
] | 49
|
0135a84067be357c8bc3d3a4298b60dcaf7d53d5
|
https://github.com/JulianoLagana/deep-machine-learning/tree/0135a84067be357c8bc3d3a4298b60dcaf7d53d5
|
SRCNN
|
import logging
import torch
import torch.nn as nn
def get_root_logger(log_file=None, log_level=logging.INFO):
"""Get the root logger.
The logger will be initialized if it has not been initialized. By default a
StreamHandler will be added. If `log_file` is specified, a FileHandler will
also be added. The name of the root logger is the top-level package name,
e.g., "mmedit".
Args:
log_file (str | None): The log filename. If specified, a FileHandler
will be added to the root logger.
log_level (int): The root logger level. Note that only the process of
rank 0 is affected, while other processes will set the level to
"Error" and be silent most of the time.
Returns:
logging.Logger: The root logger.
"""
logger = get_logger(__name__.split('.')[0], log_file, log_level)
return logger
class SRCNN(nn.Module):
"""SRCNN network structure for image super resolution.
SRCNN has three conv layers. For each layer, we can define the
`in_channels`, `out_channels` and `kernel_size`.
The input image will first be upsampled with a bicubic upsampler, and then
super-resolved in the HR spatial size.
Paper: Learning a Deep Convolutional Network for Image Super-Resolution.
Args:
channels (tuple[int]): A tuple of channel numbers for each layer
including channels of input and output . Default: (3, 64, 32, 3).
kernel_sizes (tuple[int]): A tuple of kernel sizes for each conv layer.
Default: (9, 1, 5).
upscale_factor (int): Upsampling factor. Default: 4.
"""
def __init__(self, channels=(3, 64, 32, 3), kernel_sizes=(9, 1, 5),
upscale_factor=4):
super().__init__()
assert len(channels
) == 4, f'The length of channel tuple should be 4, but got {len(channels)}'
assert len(kernel_sizes
) == 3, f'The length of kernel tuple should be 3, but got {len(kernel_sizes)}'
self.upscale_factor = upscale_factor
self.img_upsampler = nn.Upsample(scale_factor=self.upscale_factor,
mode='bicubic', align_corners=False)
self.conv1 = nn.Conv2d(channels[0], channels[1], kernel_size=
kernel_sizes[0], padding=kernel_sizes[0] // 2)
self.conv2 = nn.Conv2d(channels[1], channels[2], kernel_size=
kernel_sizes[1], padding=kernel_sizes[1] // 2)
self.conv3 = nn.Conv2d(channels[2], channels[3], kernel_size=
kernel_sizes[2], padding=kernel_sizes[2] // 2)
self.relu = nn.ReLU()
def forward(self, x):
"""Forward function.
Args:
x (Tensor): Input tensor with shape (n, c, h, w).
Returns:
Tensor: Forward results.
"""
x = self.img_upsampler(x)
out = self.relu(self.conv1(x))
out = self.relu(self.conv2(out))
out = self.conv3(out)
return out
def init_weights(self, pretrained=None, strict=True):
"""Init weights for models.
Args:
pretrained (str, optional): Path for pretrained weights. If given
None, pretrained weights will not be loaded. Defaults to None.
strict (boo, optional): Whether strictly load the pretrained model.
Defaults to True.
"""
if isinstance(pretrained, str):
logger = get_root_logger()
load_checkpoint(self, pretrained, strict=strict, logger=logger)
elif pretrained is None:
pass
else:
raise TypeError(
f'"pretrained" must be a str or None. But received {type(pretrained)}.'
)
def get_inputs():
return [torch.rand([4, 3, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import libdevice
import logging
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused__to_copy__unsafe_index_add_arange_clamp_floor_mul_rsub_sub_0(
in_out_ptr1, in_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 3072
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 16 % 16
x0 = xindex % 16
x2 = xindex // 256
x3 = xindex
tmp0 = x1
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.5
tmp3 = tmp1 + tmp2
tmp4 = 0.25
tmp5 = tmp3 * tmp4
tmp6 = tmp5 - tmp2
tmp7 = libdevice.floor(tmp6)
tmp8 = tmp7.to(tl.int32)
tmp9 = tl.full([1], 1, tl.int64)
tmp10 = tmp8 - tmp9
tmp11 = tl.full([1], 0, tl.int64)
tmp12 = triton_helpers.maximum(tmp10, tmp11)
tmp13 = tl.full([1], 3, tl.int64)
tmp14 = triton_helpers.minimum(tmp12, tmp13)
tmp15 = x0
tmp16 = tmp15.to(tl.float32)
tmp17 = tmp16 + tmp2
tmp18 = tmp17 * tmp4
tmp19 = tmp18 - tmp2
tmp20 = libdevice.floor(tmp19)
tmp21 = tmp20.to(tl.int32)
tmp22 = tmp21 - tmp9
tmp23 = triton_helpers.maximum(tmp22, tmp11)
tmp24 = triton_helpers.minimum(tmp23, tmp13)
tmp25 = tl.load(in_ptr0 + (tmp24 + 4 * tmp14 + 16 * x2), xmask,
eviction_policy='evict_last')
tmp26 = tmp19 - tmp20
tmp27 = 0.0
tmp28 = triton_helpers.maximum(tmp26, tmp27)
tmp29 = 1.0
tmp30 = triton_helpers.minimum(tmp28, tmp29)
tmp31 = tmp30 + tmp29
tmp32 = -0.75
tmp33 = tmp31 * tmp32
tmp34 = -3.75
tmp35 = tmp33 - tmp34
tmp36 = tmp35 * tmp31
tmp37 = -6.0
tmp38 = tmp36 + tmp37
tmp39 = tmp38 * tmp31
tmp40 = -3.0
tmp41 = tmp39 - tmp40
tmp42 = tmp25 * tmp41
tmp43 = triton_helpers.maximum(tmp21, tmp11)
tmp44 = triton_helpers.minimum(tmp43, tmp13)
tmp45 = tl.load(in_ptr0 + (tmp44 + 4 * tmp14 + 16 * x2), xmask,
eviction_policy='evict_last')
tmp46 = 1.25
tmp47 = tmp30 * tmp46
tmp48 = 2.25
tmp49 = tmp47 - tmp48
tmp50 = tmp49 * tmp30
tmp51 = tmp50 * tmp30
tmp52 = tmp51 + tmp29
tmp53 = tmp45 * tmp52
tmp54 = tmp21 + tmp9
tmp55 = triton_helpers.maximum(tmp54, tmp11)
tmp56 = triton_helpers.minimum(tmp55, tmp13)
tmp57 = tl.load(in_ptr0 + (tmp56 + 4 * tmp14 + 16 * x2), xmask,
eviction_policy='evict_last')
tmp58 = tmp29 - tmp30
tmp59 = tmp58 * tmp46
tmp60 = tmp59 - tmp48
tmp61 = tmp60 * tmp58
tmp62 = tmp61 * tmp58
tmp63 = tmp62 + tmp29
tmp64 = tmp57 * tmp63
tmp65 = triton_helpers.maximum(tmp8, tmp11)
tmp66 = triton_helpers.minimum(tmp65, tmp13)
tmp67 = tl.load(in_ptr0 + (tmp24 + 4 * tmp66 + 16 * x2), xmask,
eviction_policy='evict_last')
tmp68 = tmp67 * tmp41
tmp69 = tl.full([1], 2, tl.int64)
tmp70 = tmp21 + tmp69
tmp71 = triton_helpers.maximum(tmp70, tmp11)
tmp72 = triton_helpers.minimum(tmp71, tmp13)
tmp73 = tl.load(in_ptr0 + (tmp72 + 4 * tmp14 + 16 * x2), xmask,
eviction_policy='evict_last')
tmp74 = 2.0
tmp75 = tmp74 - tmp30
tmp76 = tmp75 * tmp32
tmp77 = tmp76 - tmp34
tmp78 = tmp77 * tmp75
tmp79 = tmp78 + tmp37
tmp80 = tmp79 * tmp75
tmp81 = tmp80 - tmp40
tmp82 = tmp73 * tmp81
tmp83 = tl.load(in_ptr0 + (tmp44 + 4 * tmp66 + 16 * x2), xmask,
eviction_policy='evict_last')
tmp84 = tmp83 * tmp52
tmp85 = tl.load(in_ptr0 + (tmp56 + 4 * tmp66 + 16 * x2), xmask,
eviction_policy='evict_last')
tmp86 = tmp85 * tmp63
tmp87 = tmp8 + tmp9
tmp88 = triton_helpers.maximum(tmp87, tmp11)
tmp89 = triton_helpers.minimum(tmp88, tmp13)
tmp90 = tl.load(in_ptr0 + (tmp24 + 4 * tmp89 + 16 * x2), xmask,
eviction_policy='evict_last')
tmp91 = tmp90 * tmp41
tmp92 = tl.load(in_ptr0 + (tmp72 + 4 * tmp66 + 16 * x2), xmask,
eviction_policy='evict_last')
tmp93 = tmp92 * tmp81
tmp94 = tl.load(in_ptr0 + (tmp44 + 4 * tmp89 + 16 * x2), xmask,
eviction_policy='evict_last')
tmp95 = tmp94 * tmp52
tmp96 = tl.load(in_ptr0 + (tmp56 + 4 * tmp89 + 16 * x2), xmask,
eviction_policy='evict_last')
tmp97 = tmp96 * tmp63
tmp98 = tmp8 + tmp69
tmp99 = triton_helpers.maximum(tmp98, tmp11)
tmp100 = triton_helpers.minimum(tmp99, tmp13)
tmp101 = tl.load(in_ptr0 + (tmp24 + 4 * tmp100 + 16 * x2), xmask,
eviction_policy='evict_last')
tmp102 = tmp101 * tmp41
tmp103 = tl.load(in_ptr0 + (tmp72 + 4 * tmp89 + 16 * x2), xmask,
eviction_policy='evict_last')
tmp104 = tmp103 * tmp81
tmp105 = tl.load(in_ptr0 + (tmp44 + 4 * tmp100 + 16 * x2), xmask,
eviction_policy='evict_last')
tmp106 = tmp105 * tmp52
tmp107 = tl.load(in_ptr0 + (tmp56 + 4 * tmp100 + 16 * x2), xmask,
eviction_policy='evict_last')
tmp108 = tmp107 * tmp63
tmp109 = tl.load(in_ptr0 + (tmp72 + 4 * tmp100 + 16 * x2), xmask,
eviction_policy='evict_last')
tmp110 = tmp109 * tmp81
tmp111 = tmp42 + tmp53
tmp112 = tmp111 + tmp64
tmp113 = tmp112 + tmp82
tmp114 = tmp6 - tmp7
tmp115 = triton_helpers.maximum(tmp114, tmp27)
tmp116 = triton_helpers.minimum(tmp115, tmp29)
tmp117 = tmp116 + tmp29
tmp118 = tmp117 * tmp32
tmp119 = tmp118 - tmp34
tmp120 = tmp119 * tmp117
tmp121 = tmp120 + tmp37
tmp122 = tmp121 * tmp117
tmp123 = tmp122 - tmp40
tmp124 = tmp113 * tmp123
tmp125 = tmp68 + tmp84
tmp126 = tmp125 + tmp86
tmp127 = tmp126 + tmp93
tmp128 = tmp116 * tmp46
tmp129 = tmp128 - tmp48
tmp130 = tmp129 * tmp116
tmp131 = tmp130 * tmp116
tmp132 = tmp131 + tmp29
tmp133 = tmp127 * tmp132
tmp134 = tmp124 + tmp133
tmp135 = tmp91 + tmp95
tmp136 = tmp135 + tmp97
tmp137 = tmp136 + tmp104
tmp138 = tmp29 - tmp116
tmp139 = tmp138 * tmp46
tmp140 = tmp139 - tmp48
tmp141 = tmp140 * tmp138
tmp142 = tmp141 * tmp138
tmp143 = tmp142 + tmp29
tmp144 = tmp137 * tmp143
tmp145 = tmp134 + tmp144
tmp146 = tmp102 + tmp106
tmp147 = tmp146 + tmp108
tmp148 = tmp147 + tmp110
tmp149 = tmp74 - tmp116
tmp150 = tmp149 * tmp32
tmp151 = tmp150 - tmp34
tmp152 = tmp151 * tmp149
tmp153 = tmp152 + tmp37
tmp154 = tmp153 * tmp149
tmp155 = tmp154 - tmp40
tmp156 = tmp148 * tmp155
tmp157 = tmp145 + tmp156
tl.store(in_out_ptr1 + x3, tmp157, xmask)
@triton.jit
def triton_poi_fused_convolution_relu_1(in_out_ptr0, in_ptr0, xnumel,
XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x3 = xindex
x1 = xindex // 256 % 64
tmp0 = tl.load(in_out_ptr0 + x3, None)
tmp1 = tl.load(in_ptr0 + x1, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + x3, tmp4, None)
@triton.jit
def triton_poi_fused_convolution_relu_2(in_out_ptr0, in_ptr0, xnumel,
XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x3 = xindex
x1 = xindex // 256 % 32
tmp0 = tl.load(in_out_ptr0 + x3, None)
tmp1 = tl.load(in_ptr0 + x1, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + x3, tmp4, None)
@triton.jit
def triton_poi_fused_convolution_3(in_out_ptr0, in_ptr0, xnumel, XBLOCK: tl
.constexpr):
xnumel = 3072
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = xindex // 256 % 3
tmp0 = tl.load(in_out_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + x1, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(in_out_ptr0 + x3, tmp2, xmask)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7) = args
args.clear()
assert_size_stride(primals_1, (4, 3, 4, 4), (48, 16, 4, 1))
assert_size_stride(primals_2, (64, 3, 9, 9), (243, 81, 9, 1))
assert_size_stride(primals_3, (64,), (1,))
assert_size_stride(primals_4, (32, 64, 1, 1), (64, 1, 1, 1))
assert_size_stride(primals_5, (32,), (1,))
assert_size_stride(primals_6, (3, 32, 5, 5), (800, 25, 5, 1))
assert_size_stride(primals_7, (3,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf10 = empty_strided_cuda((4, 3, 16, 16), (768, 256, 16, 1), torch
.float32)
buf18 = buf10
del buf10
buf20 = buf18
del buf18
get_raw_stream(0)
triton_poi_fused__to_copy__unsafe_index_add_arange_clamp_floor_mul_rsub_sub_0[
grid(3072)](buf20, primals_1, 3072, XBLOCK=128, num_warps=4,
num_stages=1)
del primals_1
buf21 = extern_kernels.convolution(buf20, primals_2, stride=(1, 1),
padding=(4, 4), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf21, (4, 64, 16, 16), (16384, 256, 16, 1))
buf22 = buf21
del buf21
triton_poi_fused_convolution_relu_1[grid(65536)](buf22, primals_3,
65536, XBLOCK=256, num_warps=4, num_stages=1)
del primals_3
buf23 = extern_kernels.convolution(buf22, primals_4, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf23, (4, 32, 16, 16), (8192, 256, 16, 1))
buf24 = buf23
del buf23
triton_poi_fused_convolution_relu_2[grid(32768)](buf24, primals_5,
32768, XBLOCK=256, num_warps=4, num_stages=1)
del primals_5
buf25 = extern_kernels.convolution(buf24, primals_6, stride=(1, 1),
padding=(2, 2), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf25, (4, 3, 16, 16), (768, 256, 16, 1))
buf26 = buf25
del buf25
triton_poi_fused_convolution_3[grid(3072)](buf26, primals_7, 3072,
XBLOCK=256, num_warps=4, num_stages=1)
del primals_7
return buf26, primals_2, primals_4, primals_6, buf20, buf22, buf24
def get_root_logger(log_file=None, log_level=logging.INFO):
"""Get the root logger.
The logger will be initialized if it has not been initialized. By default a
StreamHandler will be added. If `log_file` is specified, a FileHandler will
also be added. The name of the root logger is the top-level package name,
e.g., "mmedit".
Args:
log_file (str | None): The log filename. If specified, a FileHandler
will be added to the root logger.
log_level (int): The root logger level. Note that only the process of
rank 0 is affected, while other processes will set the level to
"Error" and be silent most of the time.
Returns:
logging.Logger: The root logger.
"""
logger = get_logger(__name__.split('.')[0], log_file, log_level)
return logger
class SRCNNNew(nn.Module):
"""SRCNN network structure for image super resolution.
SRCNN has three conv layers. For each layer, we can define the
`in_channels`, `out_channels` and `kernel_size`.
The input image will first be upsampled with a bicubic upsampler, and then
super-resolved in the HR spatial size.
Paper: Learning a Deep Convolutional Network for Image Super-Resolution.
Args:
channels (tuple[int]): A tuple of channel numbers for each layer
including channels of input and output . Default: (3, 64, 32, 3).
kernel_sizes (tuple[int]): A tuple of kernel sizes for each conv layer.
Default: (9, 1, 5).
upscale_factor (int): Upsampling factor. Default: 4.
"""
def __init__(self, channels=(3, 64, 32, 3), kernel_sizes=(9, 1, 5),
upscale_factor=4):
super().__init__()
assert len(channels
) == 4, f'The length of channel tuple should be 4, but got {len(channels)}'
assert len(kernel_sizes
) == 3, f'The length of kernel tuple should be 3, but got {len(kernel_sizes)}'
self.upscale_factor = upscale_factor
self.img_upsampler = nn.Upsample(scale_factor=self.upscale_factor,
mode='bicubic', align_corners=False)
self.conv1 = nn.Conv2d(channels[0], channels[1], kernel_size=
kernel_sizes[0], padding=kernel_sizes[0] // 2)
self.conv2 = nn.Conv2d(channels[1], channels[2], kernel_size=
kernel_sizes[1], padding=kernel_sizes[1] // 2)
self.conv3 = nn.Conv2d(channels[2], channels[3], kernel_size=
kernel_sizes[2], padding=kernel_sizes[2] // 2)
self.relu = nn.ReLU()
def init_weights(self, pretrained=None, strict=True):
"""Init weights for models.
Args:
pretrained (str, optional): Path for pretrained weights. If given
None, pretrained weights will not be loaded. Defaults to None.
strict (boo, optional): Whether strictly load the pretrained model.
Defaults to True.
"""
if isinstance(pretrained, str):
logger = get_root_logger()
load_checkpoint(self, pretrained, strict=strict, logger=logger)
elif pretrained is None:
pass
else:
raise TypeError(
f'"pretrained" must be a str or None. But received {type(pretrained)}.'
)
def forward(self, input_0):
primals_2 = self.conv1.weight
primals_3 = self.conv1.bias
primals_4 = self.conv2.weight
primals_5 = self.conv2.bias
primals_6 = self.conv3.weight
primals_7 = self.conv3.bias
primals_1 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7])
return output[0]
|
Juggernaut93/mmediting
|
SRCNN
| false
| 13,928
|
[
"Apache-2.0"
] | 1,884
|
8ef46ace29756dd2df1d92f2f73a33646e33e007
|
https://github.com/Juggernaut93/mmediting/tree/8ef46ace29756dd2df1d92f2f73a33646e33e007
|
SnakeHyperSolver
|
import torch
import torch.nn as nn
from torch import sin
from torch import pow
from torch.nn import Parameter
from torch.distributions.exponential import Exponential
class Snake(nn.Module):
"""
Implementation of the serpentine-like sine-based periodic activation function
.. math::
Snake_a := x + rac{1}{a} sin^2(ax) = x - rac{1}{2a}cos{2ax} + rac{1}{2a}
Shape:
- Input: (N, *) where * means, any number of additional
dimensions
- Output: (N, *), same shape as the input
Parameters:
- a - trainable parameter
References:
- This activation function is from this paper by Liu Ziyin, Tilman Hartwig, Masahito Ueda:
https://arxiv.org/abs/2006.08195
Examples:
>>> a1 = snake(256)
>>> x = torch.randn(256)
>>> x = a1(x)
"""
def __init__(self, in_features, a=None, trainable=True):
"""
Initialization.
Args:
in_features: shape of the input
a: trainable parameter
trainable: sets `a` as a trainable parameter
`a` is initialized to 1 by default, higher values = higher-frequency,
5-50 is a good starting point if you already think your data is periodic,
consider starting lower e.g. 0.5 if you think not, but don't worry,
`a` will be trained along with the rest of your model.
"""
super(Snake, self).__init__()
self.in_features = in_features if isinstance(in_features, list) else [
in_features]
if a is not None:
self.a = Parameter(torch.ones(self.in_features) * a)
else:
m = Exponential(torch.tensor([0.1]))
self.a = Parameter(m.rsample(self.in_features).squeeze())
self.a.requiresGrad = trainable
def forward(self, x):
"""
Forward pass of the function.
Applies the function to the input elementwise.
Snake ∶= x + 1/a* sin^2 (xa)
"""
return x + 1.0 / self.a * pow(sin(x * self.a), 2)
class SnakeHyperSolver(nn.Module):
def __init__(self, in_dim, out_dim, hidden_dim=32):
super().__init__()
self.fc1 = nn.Linear(in_dim, hidden_dim)
self.fc2 = nn.Linear(hidden_dim, hidden_dim)
self.fc3 = nn.Linear(hidden_dim, out_dim)
self.a1 = Snake(hidden_dim)
self.a2 = Snake(hidden_dim)
def forward(self, x):
x = self.a1(self.fc1(x))
x = self.a2(self.fc2(x))
x = self.fc3(x)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_dim': 4, 'out_dim': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import math as tl_math
import torch.nn as nn
from torch import sin
from torch import pow
from torch.nn import Parameter
from torch.distributions.exponential import Exponential
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_add_mul_pow_reciprocal_sin_0(in_ptr0, in_ptr1,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x2 = xindex
x0 = xindex % 32
tmp0 = tl.load(in_ptr0 + x2, None)
tmp1 = tl.load(in_ptr1 + x0, None, eviction_policy='evict_last')
tmp2 = tl.full([1], 1, tl.int32)
tmp3 = tmp2 / tmp1
tmp4 = 1.0
tmp5 = tmp3 * tmp4
tmp6 = tmp0 * tmp1
tmp7 = tl_math.sin(tmp6)
tmp8 = tmp7 * tmp7
tmp9 = tmp5 * tmp8
tmp10 = tmp0 + tmp9
tl.store(out_ptr0 + x2, tmp10, None)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7, primals_8, primals_9) = args
args.clear()
assert_size_stride(primals_1, (32, 4), (4, 1))
assert_size_stride(primals_2, (32,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (32,), (1,))
assert_size_stride(primals_5, (32, 32), (32, 1))
assert_size_stride(primals_6, (32,), (1,))
assert_size_stride(primals_7, (32,), (1,))
assert_size_stride(primals_8, (4, 32), (32, 1))
assert_size_stride(primals_9, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 32), (32, 1), torch.float32)
extern_kernels.addmm(primals_2, reinterpret_tensor(primals_3, (64,
4), (4, 1), 0), reinterpret_tensor(primals_1, (4, 32), (1, 4),
0), alpha=1, beta=1, out=buf0)
del primals_1
del primals_2
buf1 = empty_strided_cuda((4, 4, 4, 32), (512, 128, 32, 1), torch.
float32)
get_raw_stream(0)
triton_poi_fused_add_mul_pow_reciprocal_sin_0[grid(2048)](buf0,
primals_4, buf1, 2048, XBLOCK=256, num_warps=4, num_stages=1)
buf2 = empty_strided_cuda((64, 32), (32, 1), torch.float32)
extern_kernels.addmm(primals_6, reinterpret_tensor(buf1, (64, 32),
(32, 1), 0), reinterpret_tensor(primals_5, (32, 32), (1, 32), 0
), alpha=1, beta=1, out=buf2)
del primals_6
buf3 = empty_strided_cuda((4, 4, 4, 32), (512, 128, 32, 1), torch.
float32)
triton_poi_fused_add_mul_pow_reciprocal_sin_0[grid(2048)](buf2,
primals_7, buf3, 2048, XBLOCK=256, num_warps=4, num_stages=1)
buf4 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_9, reinterpret_tensor(buf3, (64, 32),
(32, 1), 0), reinterpret_tensor(primals_8, (32, 4), (1, 32), 0),
alpha=1, beta=1, out=buf4)
del primals_9
return reinterpret_tensor(buf4, (4, 4, 4, 4), (64, 16, 4, 1), 0
), primals_4, primals_7, reinterpret_tensor(primals_3, (64, 4), (4,
1), 0), buf0, reinterpret_tensor(buf1, (64, 32), (32, 1), 0
), buf2, reinterpret_tensor(buf3, (64, 32), (32, 1), 0
), primals_8, primals_5
class Snake(nn.Module):
"""
Implementation of the serpentine-like sine-based periodic activation function
.. math::
Snake_a := x + rac{1}{a} sin^2(ax) = x - rac{1}{2a}cos{2ax} + rac{1}{2a}
Shape:
- Input: (N, *) where * means, any number of additional
dimensions
- Output: (N, *), same shape as the input
Parameters:
- a - trainable parameter
References:
- This activation function is from this paper by Liu Ziyin, Tilman Hartwig, Masahito Ueda:
https://arxiv.org/abs/2006.08195
Examples:
>>> a1 = snake(256)
>>> x = torch.randn(256)
>>> x = a1(x)
"""
def __init__(self, in_features, a=None, trainable=True):
"""
Initialization.
Args:
in_features: shape of the input
a: trainable parameter
trainable: sets `a` as a trainable parameter
`a` is initialized to 1 by default, higher values = higher-frequency,
5-50 is a good starting point if you already think your data is periodic,
consider starting lower e.g. 0.5 if you think not, but don't worry,
`a` will be trained along with the rest of your model.
"""
super(Snake, self).__init__()
self.in_features = in_features if isinstance(in_features, list) else [
in_features]
if a is not None:
self.a = Parameter(torch.ones(self.in_features) * a)
else:
m = Exponential(torch.tensor([0.1]))
self.a = Parameter(m.rsample(self.in_features).squeeze())
self.a.requiresGrad = trainable
def forward(self, x):
"""
Forward pass of the function.
Applies the function to the input elementwise.
Snake ∶= x + 1/a* sin^2 (xa)
"""
return x + 1.0 / self.a * pow(sin(x * self.a), 2)
class SnakeHyperSolverNew(nn.Module):
def __init__(self, in_dim, out_dim, hidden_dim=32):
super().__init__()
self.fc1 = nn.Linear(in_dim, hidden_dim)
self.fc2 = nn.Linear(hidden_dim, hidden_dim)
self.fc3 = nn.Linear(hidden_dim, out_dim)
self.a1 = Snake(hidden_dim)
self.a2 = Snake(hidden_dim)
def forward(self, input_0):
primals_1 = self.fc1.weight
primals_2 = self.fc1.bias
primals_5 = self.fc2.weight
primals_4 = self.fc2.bias
primals_8 = self.fc3.weight
primals_9 = self.fc3.bias
primals_6 = self.a1.a
primals_7 = self.a2.a
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7, primals_8, primals_9])
return output[0]
|
Juju-botu/diffeqml-research
|
SnakeHyperSolver
| false
| 13,929
|
[
"Apache-2.0"
] | 49
|
aa796c87447e5299ec4f25a07fc4d032afb1f63e
|
https://github.com/Juju-botu/diffeqml-research/tree/aa796c87447e5299ec4f25a07fc4d032afb1f63e
|
RLFeatPreprocessNet
|
import torch
from torch import nn
import torch.nn.parallel
class RLFeatPreprocessNet(nn.Module):
def __init__(self, feature_size, embed_size, box_info_size,
overlap_info_size, output_size):
super(RLFeatPreprocessNet, self).__init__()
self.feature_size = feature_size
self.embed_size = embed_size
self.box_info_size = box_info_size
self.overlap_info_size = overlap_info_size
self.output_size = output_size
self.resize_feat = nn.Linear(self.feature_size, int(output_size / 4))
self.resize_embed = nn.Linear(self.embed_size, int(output_size / 4))
self.resize_box = nn.Linear(self.box_info_size, int(output_size / 4))
self.resize_overlap = nn.Linear(self.overlap_info_size, int(
output_size / 4))
self.resize_feat.weight.data.normal_(0, 0.001)
self.resize_embed.weight.data.normal_(0, 0.01)
self.resize_box.weight.data.normal_(0, 1)
self.resize_overlap.weight.data.normal_(0, 1)
self.resize_feat.bias.data.zero_()
self.resize_embed.bias.data.zero_()
self.resize_box.bias.data.zero_()
self.resize_overlap.bias.data.zero_()
def forward(self, obj_feat, obj_embed, box_info, overlap_info):
resized_obj = self.resize_feat(obj_feat)
resized_embed = self.resize_embed(obj_embed)
resized_box = self.resize_box(box_info)
resized_overlap = self.resize_overlap(overlap_info)
output_feat = torch.cat((resized_obj, resized_embed, resized_box,
resized_overlap), 1)
return output_feat
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4]), torch.rand(
[4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'feature_size': 4, 'embed_size': 4, 'box_info_size': 4,
'overlap_info_size': 4, 'output_size': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch import nn
import torch.nn.parallel
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_cat_0(in_ptr0, in_ptr1, in_ptr2, in_ptr3, out_ptr0,
xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 4 % 16
x0 = xindex % 4
x2 = xindex // 64
x3 = xindex
tmp0 = x1
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (x0 + 4 * x1 + 16 * x2), tmp4 & xmask, other=0.0)
tmp6 = tmp0 >= tmp3
tmp7 = tl.full([1], 8, tl.int64)
tmp8 = tmp0 < tmp7
tmp9 = tmp6 & tmp8
tmp10 = tl.load(in_ptr1 + (x0 + 4 * (-4 + x1) + 16 * x2), tmp9 & xmask,
other=0.0)
tmp11 = tmp0 >= tmp7
tmp12 = tl.full([1], 12, tl.int64)
tmp13 = tmp0 < tmp12
tmp14 = tmp11 & tmp13
tmp15 = tl.load(in_ptr2 + (x0 + 4 * (-8 + x1) + 16 * x2), tmp14 & xmask,
other=0.0)
tmp16 = tmp0 >= tmp12
tl.full([1], 16, tl.int64)
tmp19 = tl.load(in_ptr3 + (x0 + 4 * (-12 + x1) + 16 * x2), tmp16 &
xmask, other=0.0)
tmp20 = tl.where(tmp14, tmp15, tmp19)
tmp21 = tl.where(tmp9, tmp10, tmp20)
tmp22 = tl.where(tmp4, tmp5, tmp21)
tl.store(out_ptr0 + x3, tmp22, xmask)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7, primals_8, primals_9, primals_10, primals_11, primals_12
) = args
args.clear()
assert_size_stride(primals_1, (1, 4), (4, 1))
assert_size_stride(primals_2, (1,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (1, 4), (4, 1))
assert_size_stride(primals_5, (1,), (1,))
assert_size_stride(primals_6, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_7, (1, 4), (4, 1))
assert_size_stride(primals_8, (1,), (1,))
assert_size_stride(primals_9, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_10, (1, 4), (4, 1))
assert_size_stride(primals_11, (1,), (1,))
assert_size_stride(primals_12, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf1 = empty_strided_cuda((64, 1), (1, 1), torch.float32)
extern_kernels.addmm(primals_2, reinterpret_tensor(primals_3, (64,
4), (4, 1), 0), reinterpret_tensor(primals_1, (4, 1), (1, 4), 0
), alpha=1, beta=1, out=buf1)
del primals_1
del primals_2
buf3 = empty_strided_cuda((64, 1), (1, 1), torch.float32)
extern_kernels.addmm(primals_5, reinterpret_tensor(primals_6, (64,
4), (4, 1), 0), reinterpret_tensor(primals_4, (4, 1), (1, 4), 0
), alpha=1, beta=1, out=buf3)
del primals_4
del primals_5
buf5 = empty_strided_cuda((64, 1), (1, 1), torch.float32)
extern_kernels.addmm(primals_8, reinterpret_tensor(primals_9, (64,
4), (4, 1), 0), reinterpret_tensor(primals_7, (4, 1), (1, 4), 0
), alpha=1, beta=1, out=buf5)
del primals_7
del primals_8
buf7 = empty_strided_cuda((64, 1), (1, 1), torch.float32)
extern_kernels.addmm(primals_11, reinterpret_tensor(primals_12, (64,
4), (4, 1), 0), reinterpret_tensor(primals_10, (4, 1), (1, 4),
0), alpha=1, beta=1, out=buf7)
del primals_10
del primals_11
buf8 = empty_strided_cuda((4, 16, 4, 1), (64, 4, 1, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_cat_0[grid(256)](buf1, buf3, buf5, buf7, buf8, 256,
XBLOCK=256, num_warps=4, num_stages=1)
del buf1
del buf3
del buf5
del buf7
return buf8, reinterpret_tensor(primals_3, (64, 4), (4, 1), 0
), reinterpret_tensor(primals_6, (64, 4), (4, 1), 0
), reinterpret_tensor(primals_9, (64, 4), (4, 1), 0
), reinterpret_tensor(primals_12, (64, 4), (4, 1), 0)
class RLFeatPreprocessNetNew(nn.Module):
def __init__(self, feature_size, embed_size, box_info_size,
overlap_info_size, output_size):
super(RLFeatPreprocessNetNew, self).__init__()
self.feature_size = feature_size
self.embed_size = embed_size
self.box_info_size = box_info_size
self.overlap_info_size = overlap_info_size
self.output_size = output_size
self.resize_feat = nn.Linear(self.feature_size, int(output_size / 4))
self.resize_embed = nn.Linear(self.embed_size, int(output_size / 4))
self.resize_box = nn.Linear(self.box_info_size, int(output_size / 4))
self.resize_overlap = nn.Linear(self.overlap_info_size, int(
output_size / 4))
self.resize_feat.weight.data.normal_(0, 0.001)
self.resize_embed.weight.data.normal_(0, 0.01)
self.resize_box.weight.data.normal_(0, 1)
self.resize_overlap.weight.data.normal_(0, 1)
self.resize_feat.bias.data.zero_()
self.resize_embed.bias.data.zero_()
self.resize_box.bias.data.zero_()
self.resize_overlap.bias.data.zero_()
def forward(self, input_0, input_1, input_2, input_3):
primals_1 = self.resize_feat.weight
primals_2 = self.resize_feat.bias
primals_4 = self.resize_embed.weight
primals_5 = self.resize_embed.bias
primals_7 = self.resize_box.weight
primals_8 = self.resize_box.bias
primals_10 = self.resize_overlap.weight
primals_11 = self.resize_overlap.bias
primals_3 = input_0
primals_6 = input_1
primals_9 = input_2
primals_12 = input_3
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7, primals_8, primals_9,
primals_10, primals_11, primals_12])
return output[0]
|
KaihuaTang/VCTree-Scene-Graph-Generation
|
RLFeatPreprocessNet
| false
| 13,930
|
[
"MIT"
] | 109
|
75bc30543dbb5a869acff65b2183efa7ee4ac35d
|
https://github.com/KaihuaTang/VCTree-Scene-Graph-Generation/tree/75bc30543dbb5a869acff65b2183efa7ee4ac35d
|
Softplus
|
import torch
import numpy as np
from torch.utils.data import Dataset as Dataset
import torch.nn as nn
import torch.utils.data
def activation_shifting(activation):
def shifted_activation(x):
return activation(x) - activation(torch.zeros_like(x))
return shifted_activation
def cauchy_softplus(x):
pi = np.pi
return (x * pi - torch.log(x ** 2 + 1) + 2 * x * torch.atan(x)) / (2 * pi)
def gaussian_softplus(x):
z = np.sqrt(np.pi / 2)
return (z * x * torch.erf(x / np.sqrt(2)) + torch.exp(-x ** 2 / 2) + z * x
) / (2 * z)
def gaussian_softplus2(x):
z = np.sqrt(np.pi / 2)
return (z * x * torch.erf(x / np.sqrt(2)) + torch.exp(-x ** 2 / 2) + z * x
) / z
def get_softplus(softplus_type='softplus', zero_softplus=False):
if softplus_type == 'softplus':
act = nn.functional.softplus
elif softplus_type == 'gaussian_softplus':
act = gaussian_softplus
elif softplus_type == 'gaussian_softplus2':
act = gaussian_softplus2
elif softplus_type == 'laplace_softplus':
act = gaussian_softplus
elif softplus_type == 'cauchy_softplus':
act = cauchy_softplus
else:
raise NotImplementedError(
f'softplus type {softplus_type} not supported.')
if zero_softplus:
act = activation_shifting(act)
return act
class Softplus(nn.Module):
def __init__(self, softplus_type='softplus', zero_softplus=False):
super(Softplus, self).__init__()
self.softplus_type = softplus_type
self.zero_softplus = zero_softplus
def forward(self, x):
return get_softplus(self.softplus_type, self.zero_softplus)(x)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
import numpy as np
from torch.utils.data import Dataset as Dataset
import torch.nn as nn
import torch.utils.data
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_softplus_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = 20.0
tmp2 = tmp0 > tmp1
tmp3 = tl_math.exp(tmp0)
tmp4 = libdevice.log1p(tmp3)
tmp5 = tl.where(tmp2, tmp0, tmp4)
tl.store(out_ptr0 + x0, tmp5, xmask)
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_softplus_0[grid(256)](arg0_1, buf0, 256, XBLOCK=
256, num_warps=4, num_stages=1)
del arg0_1
return buf0,
def activation_shifting(activation):
def shifted_activation(x):
return activation(x) - activation(torch.zeros_like(x))
return shifted_activation
def cauchy_softplus(x):
pi = np.pi
return (x * pi - torch.log(x ** 2 + 1) + 2 * x * torch.atan(x)) / (2 * pi)
def gaussian_softplus(x):
z = np.sqrt(np.pi / 2)
return (z * x * torch.erf(x / np.sqrt(2)) + torch.exp(-x ** 2 / 2) + z * x
) / (2 * z)
def gaussian_softplus2(x):
z = np.sqrt(np.pi / 2)
return (z * x * torch.erf(x / np.sqrt(2)) + torch.exp(-x ** 2 / 2) + z * x
) / z
def get_softplus(softplus_type='softplus', zero_softplus=False):
if softplus_type == 'softplus':
act = nn.functional.softplus
elif softplus_type == 'gaussian_softplus':
act = gaussian_softplus
elif softplus_type == 'gaussian_softplus2':
act = gaussian_softplus2
elif softplus_type == 'laplace_softplus':
act = gaussian_softplus
elif softplus_type == 'cauchy_softplus':
act = cauchy_softplus
else:
raise NotImplementedError(
f'softplus type {softplus_type} not supported.')
if zero_softplus:
act = activation_shifting(act)
return act
class SoftplusNew(nn.Module):
def __init__(self, softplus_type='softplus', zero_softplus=False):
super(SoftplusNew, self).__init__()
self.softplus_type = softplus_type
self.zero_softplus = zero_softplus
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
|
KelvinKan/CP-Flow
|
Softplus
| false
| 13,931
|
[
"MIT"
] | 64
|
d01303cb4ebeb5a0bbfca638ffaf5b7a8ec22fb1
|
https://github.com/KelvinKan/CP-Flow/tree/d01303cb4ebeb5a0bbfca638ffaf5b7a8ec22fb1
|
MaxPool3x3
|
import torch
import torch.nn as nn
class MaxPool3x3(nn.Module):
"""3x3 max pool with no subsampling."""
def __init__(self, in_channels, out_channels, kernel_size=3, stride=1,
padding=1):
super(MaxPool3x3, self).__init__()
self.maxpool = nn.MaxPool2d(kernel_size, stride, padding)
def forward(self, x):
x = self.maxpool(x)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_channels': 4, 'out_channels': 4}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_max_pool2d_with_indices_0(in_ptr0, out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 4 % 4
x0 = xindex % 4
x4 = xindex
tmp0 = -1 + x1
tmp1 = tl.full([1], 0, tl.int64)
tmp2 = tmp0 >= tmp1
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tmp2 & tmp4
tmp6 = -1 + x0
tmp7 = tmp6 >= tmp1
tmp8 = tmp6 < tmp3
tmp9 = tmp7 & tmp8
tmp10 = tmp5 & tmp9
tmp11 = tl.load(in_ptr0 + (-5 + x4), tmp10 & xmask, other=float('-inf'))
tmp12 = x0
tmp13 = tmp12 >= tmp1
tmp14 = tmp12 < tmp3
tmp15 = tmp13 & tmp14
tmp16 = tmp5 & tmp15
tmp17 = tl.load(in_ptr0 + (-4 + x4), tmp16 & xmask, other=float('-inf'))
tmp18 = triton_helpers.maximum(tmp17, tmp11)
tmp19 = 1 + x0
tmp20 = tmp19 >= tmp1
tmp21 = tmp19 < tmp3
tmp22 = tmp20 & tmp21
tmp23 = tmp5 & tmp22
tmp24 = tl.load(in_ptr0 + (-3 + x4), tmp23 & xmask, other=float('-inf'))
tmp25 = triton_helpers.maximum(tmp24, tmp18)
tmp26 = x1
tmp27 = tmp26 >= tmp1
tmp28 = tmp26 < tmp3
tmp29 = tmp27 & tmp28
tmp30 = tmp29 & tmp9
tmp31 = tl.load(in_ptr0 + (-1 + x4), tmp30 & xmask, other=float('-inf'))
tmp32 = triton_helpers.maximum(tmp31, tmp25)
tmp33 = tmp29 & tmp15
tmp34 = tl.load(in_ptr0 + x4, tmp33 & xmask, other=float('-inf'))
tmp35 = triton_helpers.maximum(tmp34, tmp32)
tmp36 = tmp29 & tmp22
tmp37 = tl.load(in_ptr0 + (1 + x4), tmp36 & xmask, other=float('-inf'))
tmp38 = triton_helpers.maximum(tmp37, tmp35)
tmp39 = 1 + x1
tmp40 = tmp39 >= tmp1
tmp41 = tmp39 < tmp3
tmp42 = tmp40 & tmp41
tmp43 = tmp42 & tmp9
tmp44 = tl.load(in_ptr0 + (3 + x4), tmp43 & xmask, other=float('-inf'))
tmp45 = triton_helpers.maximum(tmp44, tmp38)
tmp46 = tmp42 & tmp15
tmp47 = tl.load(in_ptr0 + (4 + x4), tmp46 & xmask, other=float('-inf'))
tmp48 = triton_helpers.maximum(tmp47, tmp45)
tmp49 = tmp42 & tmp22
tmp50 = tl.load(in_ptr0 + (5 + x4), tmp49 & xmask, other=float('-inf'))
tmp51 = triton_helpers.maximum(tmp50, tmp48)
tl.store(out_ptr0 + x4, tmp51, xmask)
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_max_pool2d_with_indices_0[grid(256)](arg0_1, buf0,
256, XBLOCK=128, num_warps=4, num_stages=1)
del arg0_1
return buf0,
class MaxPool3x3New(nn.Module):
"""3x3 max pool with no subsampling."""
def __init__(self, in_channels, out_channels, kernel_size=3, stride=1,
padding=1):
super(MaxPool3x3New, self).__init__()
self.maxpool = nn.MaxPool2d(kernel_size, stride, padding)
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
|
KelvinYang0320/nas-without-training
|
MaxPool3x3
| false
| 13,932
|
[
"MIT"
] | 385
|
5ed77a06726a73233a5a93b8f70a7172ce570029
|
https://github.com/KelvinYang0320/nas-without-training/tree/5ed77a06726a73233a5a93b8f70a7172ce570029
|
PseudoCoord
|
import torch
import torch.nn as nn
import torch.utils.data
class PseudoCoord(nn.Module):
def __init__(self):
super(PseudoCoord, self).__init__()
def forward(self, b):
"""
Input:
b: bounding box [batch, num_obj, 4] (x1,y1,x2,y2)
Output:
pseudo_coord [batch, num_obj, num_obj, 2] (rho, theta)
"""
batch_size, num_obj, _ = b.shape
centers = (b[:, :, 2:] + b[:, :, :2]) * 0.5
relative_coord = centers.view(batch_size, num_obj, 1, 2
) - centers.view(batch_size, 1, num_obj, 2)
rho = torch.sqrt(relative_coord[:, :, :, 0] ** 2 + relative_coord[:,
:, :, 1] ** 2)
theta = torch.atan2(relative_coord[:, :, :, 0], relative_coord[:, :,
:, 1])
new_coord = torch.cat((rho.unsqueeze(-1), theta.unsqueeze(-1)), dim=-1)
return new_coord
def get_inputs():
return [torch.rand([4, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice
import torch.nn as nn
import torch.utils.data
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_cat_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 128
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 2
x5 = xindex // 8
x1 = xindex // 2 % 4
x3 = xindex // 32
x6 = xindex
tmp0 = x0
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 1, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (2 + 4 * x5), tmp4 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp6 = tl.load(in_ptr0 + 4 * x5, tmp4 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp7 = tmp5 + tmp6
tmp8 = 0.5
tmp9 = tmp7 * tmp8
tmp10 = tl.load(in_ptr0 + (2 + 4 * x1 + 16 * x3), tmp4 & xmask,
eviction_policy='evict_last', other=0.0)
tmp11 = tl.load(in_ptr0 + (4 * x1 + 16 * x3), tmp4 & xmask,
eviction_policy='evict_last', other=0.0)
tmp12 = tmp10 + tmp11
tmp13 = tmp12 * tmp8
tmp14 = tmp9 - tmp13
tmp15 = tmp14 * tmp14
tmp16 = tl.load(in_ptr0 + (3 + 4 * x5), tmp4 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp17 = tl.load(in_ptr0 + (1 + 4 * x5), tmp4 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp18 = tmp16 + tmp17
tmp19 = tmp18 * tmp8
tmp20 = tl.load(in_ptr0 + (3 + 4 * x1 + 16 * x3), tmp4 & xmask,
eviction_policy='evict_last', other=0.0)
tmp21 = tl.load(in_ptr0 + (1 + 4 * x1 + 16 * x3), tmp4 & xmask,
eviction_policy='evict_last', other=0.0)
tmp22 = tmp20 + tmp21
tmp23 = tmp22 * tmp8
tmp24 = tmp19 - tmp23
tmp25 = tmp24 * tmp24
tmp26 = tmp15 + tmp25
tmp27 = libdevice.sqrt(tmp26)
tmp28 = tl.full(tmp27.shape, 0.0, tmp27.dtype)
tmp29 = tl.where(tmp4, tmp27, tmp28)
tmp30 = tmp0 >= tmp3
tl.full([1], 2, tl.int64)
tmp33 = tl.load(in_ptr0 + (2 + 4 * x5), tmp30 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp34 = tl.load(in_ptr0 + 4 * x5, tmp30 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp35 = tmp33 + tmp34
tmp36 = tmp35 * tmp8
tmp37 = tl.load(in_ptr0 + (2 + 4 * x1 + 16 * x3), tmp30 & xmask,
eviction_policy='evict_last', other=0.0)
tmp38 = tl.load(in_ptr0 + (4 * x1 + 16 * x3), tmp30 & xmask,
eviction_policy='evict_last', other=0.0)
tmp39 = tmp37 + tmp38
tmp40 = tmp39 * tmp8
tmp41 = tmp36 - tmp40
tmp42 = tl.load(in_ptr0 + (3 + 4 * x5), tmp30 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp43 = tl.load(in_ptr0 + (1 + 4 * x5), tmp30 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp44 = tmp42 + tmp43
tmp45 = tmp44 * tmp8
tmp46 = tl.load(in_ptr0 + (3 + 4 * x1 + 16 * x3), tmp30 & xmask,
eviction_policy='evict_last', other=0.0)
tmp47 = tl.load(in_ptr0 + (1 + 4 * x1 + 16 * x3), tmp30 & xmask,
eviction_policy='evict_last', other=0.0)
tmp48 = tmp46 + tmp47
tmp49 = tmp48 * tmp8
tmp50 = tmp45 - tmp49
tmp51 = libdevice.atan2(tmp41, tmp50)
tmp52 = tl.full(tmp51.shape, 0.0, tmp51.dtype)
tmp53 = tl.where(tmp30, tmp51, tmp52)
tmp54 = tl.where(tmp4, tmp29, tmp53)
tl.store(out_ptr0 + x6, tmp54, xmask)
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4), (16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 2), (32, 8, 2, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_cat_0[grid(128)](arg0_1, buf0, 128, XBLOCK=128,
num_warps=4, num_stages=1)
del arg0_1
return buf0,
class PseudoCoordNew(nn.Module):
def __init__(self):
super(PseudoCoordNew, self).__init__()
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
|
KaihuaTang/VQA2.0-Recent-Approachs-2018.pytorch
|
PseudoCoord
| false
| 13,933
|
[
"MIT"
] | 298
|
52e1ba5a7f3b88c617115ccc755e2e7868e8de2b
|
https://github.com/KaihuaTang/VQA2.0-Recent-Approachs-2018.pytorch/tree/52e1ba5a7f3b88c617115ccc755e2e7868e8de2b
|
Conv2d
|
from torch.autograd import Function
import torch
import numpy as np
import torchvision.transforms.functional as F
import torch.nn as nn
import torch.nn.functional as F
def _setup_kernel(k):
k = np.asarray(k, dtype=np.float32)
if k.ndim == 1:
k = np.outer(k, k)
k /= np.sum(k)
assert k.ndim == 2
assert k.shape[0] == k.shape[1]
return k
def upfirdn2d_native(input, kernel, up_x, up_y, down_x, down_y, pad_x0,
pad_x1, pad_y0, pad_y1):
_, channel, in_h, in_w = input.shape
input = input.reshape(-1, in_h, in_w, 1)
_, in_h, in_w, minor = input.shape
kernel_h, kernel_w = kernel.shape
out = input.view(-1, in_h, 1, in_w, 1, minor)
out = F.pad(out, [0, 0, 0, up_x - 1, 0, 0, 0, up_y - 1])
out = out.view(-1, in_h * up_y, in_w * up_x, minor)
out = F.pad(out, [0, 0, max(pad_x0, 0), max(pad_x1, 0), max(pad_y0, 0),
max(pad_y1, 0)])
out = out[:, max(-pad_y0, 0):out.shape[1] - max(-pad_y1, 0), max(-
pad_x0, 0):out.shape[2] - max(-pad_x1, 0), :]
out = out.permute(0, 3, 1, 2)
out = out.reshape([-1, 1, in_h * up_y + pad_y0 + pad_y1, in_w * up_x +
pad_x0 + pad_x1])
w = torch.flip(kernel, [0, 1]).view(1, 1, kernel_h, kernel_w)
out = F.conv2d(out, w)
out = out.reshape(-1, minor, in_h * up_y + pad_y0 + pad_y1 - kernel_h +
1, in_w * up_x + pad_x0 + pad_x1 - kernel_w + 1)
out = out.permute(0, 2, 3, 1)
out = out[:, ::down_y, ::down_x, :]
out_h = (in_h * up_y + pad_y0 + pad_y1 - kernel_h) // down_y + 1
out_w = (in_w * up_x + pad_x0 + pad_x1 - kernel_w) // down_x + 1
return out.view(-1, channel, out_h, out_w)
def upfirdn2d(input, kernel, up=1, down=1, pad=(0, 0)):
if input.device.type == 'cpu':
out = upfirdn2d_native(input, kernel, up, up, down, down, pad[0],
pad[1], pad[0], pad[1])
else:
out = UpFirDn2d.apply(input, kernel, (up, up), (down, down), (pad[0
], pad[1], pad[0], pad[1]))
return out
def conv_downsample_2d(x, w, k=None, factor=2, gain=1):
"""Fused `tf.nn.conv2d()` followed by `downsample_2d()`.
Padding is performed only once at the beginning, not between the operations.
The fused op is considerably more efficient than performing the same
calculation
using standard TensorFlow ops. It supports gradients of arbitrary order.
Args:
x: Input tensor of the shape `[N, C, H, W]` or `[N, H, W,
C]`.
w: Weight tensor of the shape `[filterH, filterW, inChannels,
outChannels]`. Grouped convolution can be performed by `inChannels =
x.shape[0] // numGroups`.
k: FIR filter of the shape `[firH, firW]` or `[firN]`
(separable). The default is `[1] * factor`, which corresponds to
average pooling.
factor: Integer downsampling factor (default: 2).
gain: Scaling factor for signal magnitude (default: 1.0).
Returns:
Tensor of the shape `[N, C, H // factor, W // factor]` or
`[N, H // factor, W // factor, C]`, and same datatype as `x`.
"""
assert isinstance(factor, int) and factor >= 1
_outC, _inC, convH, convW = w.shape
assert convW == convH
if k is None:
k = [1] * factor
k = _setup_kernel(k) * gain
p = k.shape[0] - factor + (convW - 1)
s = [factor, factor]
x = upfirdn2d(x, torch.tensor(k, device=x.device), pad=((p + 1) // 2, p //
2))
return F.conv2d(x, w, stride=s, padding=0)
def _shape(x, dim):
return x.shape[dim]
def upsample_conv_2d(x, w, k=None, factor=2, gain=1):
"""Fused `upsample_2d()` followed by `tf.nn.conv2d()`.
Padding is performed only once at the beginning, not between the
operations.
The fused op is considerably more efficient than performing the same
calculation
using standard TensorFlow ops. It supports gradients of arbitrary order.
Args:
x: Input tensor of the shape `[N, C, H, W]` or `[N, H, W,
C]`.
w: Weight tensor of the shape `[filterH, filterW, inChannels,
outChannels]`. Grouped convolution can be performed by `inChannels =
x.shape[0] // numGroups`.
k: FIR filter of the shape `[firH, firW]` or `[firN]`
(separable). The default is `[1] * factor`, which corresponds to
nearest-neighbor upsampling.
factor: Integer upsampling factor (default: 2).
gain: Scaling factor for signal magnitude (default: 1.0).
Returns:
Tensor of the shape `[N, C, H * factor, W * factor]` or
`[N, H * factor, W * factor, C]`, and same datatype as `x`.
"""
assert isinstance(factor, int) and factor >= 1
assert len(w.shape) == 4
convH = w.shape[2]
convW = w.shape[3]
inC = w.shape[1]
w.shape[0]
assert convW == convH
if k is None:
k = [1] * factor
k = _setup_kernel(k) * (gain * factor ** 2)
p = k.shape[0] - factor - (convW - 1)
stride = factor, factor
stride = [1, 1, factor, factor]
output_shape = (_shape(x, 2) - 1) * factor + convH, (_shape(x, 3) - 1
) * factor + convW
output_padding = output_shape[0] - (_shape(x, 2) - 1) * stride[0
] - convH, output_shape[1] - (_shape(x, 3) - 1) * stride[1] - convW
assert output_padding[0] >= 0 and output_padding[1] >= 0
num_groups = _shape(x, 1) // inC
w = torch.reshape(w, (num_groups, -1, inC, convH, convW))
w = w[..., ::-1, ::-1].permute(0, 2, 1, 3, 4)
w = torch.reshape(w, (num_groups * inC, -1, convH, convW))
x = F.conv_transpose2d(x, w, stride=stride, output_padding=
output_padding, padding=0)
return upfirdn2d(x, torch.tensor(k, device=x.device), pad=((p + 1) // 2 +
factor - 1, p // 2 + 1))
class UpFirDn2dBackward(Function):
@staticmethod
def forward(ctx, grad_output, kernel, grad_kernel, up, down, pad, g_pad,
in_size, out_size):
up_x, up_y = up
down_x, down_y = down
g_pad_x0, g_pad_x1, g_pad_y0, g_pad_y1 = g_pad
grad_output = grad_output.reshape(-1, out_size[0], out_size[1], 1)
grad_input = upfirdn2d_op.upfirdn2d(grad_output, grad_kernel,
down_x, down_y, up_x, up_y, g_pad_x0, g_pad_x1, g_pad_y0, g_pad_y1)
grad_input = grad_input.view(in_size[0], in_size[1], in_size[2],
in_size[3])
ctx.save_for_backward(kernel)
pad_x0, pad_x1, pad_y0, pad_y1 = pad
ctx.up_x = up_x
ctx.up_y = up_y
ctx.down_x = down_x
ctx.down_y = down_y
ctx.pad_x0 = pad_x0
ctx.pad_x1 = pad_x1
ctx.pad_y0 = pad_y0
ctx.pad_y1 = pad_y1
ctx.in_size = in_size
ctx.out_size = out_size
return grad_input
@staticmethod
def backward(ctx, gradgrad_input):
kernel, = ctx.saved_tensors
gradgrad_input = gradgrad_input.reshape(-1, ctx.in_size[2], ctx.
in_size[3], 1)
gradgrad_out = upfirdn2d_op.upfirdn2d(gradgrad_input, kernel, ctx.
up_x, ctx.up_y, ctx.down_x, ctx.down_y, ctx.pad_x0, ctx.pad_x1,
ctx.pad_y0, ctx.pad_y1)
gradgrad_out = gradgrad_out.view(ctx.in_size[0], ctx.in_size[1],
ctx.out_size[0], ctx.out_size[1])
return gradgrad_out, None, None, None, None, None, None, None, None
class UpFirDn2d(Function):
@staticmethod
def forward(ctx, input, kernel, up, down, pad):
up_x, up_y = up
down_x, down_y = down
pad_x0, pad_x1, pad_y0, pad_y1 = pad
kernel_h, kernel_w = kernel.shape
_batch, channel, in_h, in_w = input.shape
ctx.in_size = input.shape
input = input.reshape(-1, in_h, in_w, 1)
ctx.save_for_backward(kernel, torch.flip(kernel, [0, 1]))
out_h = (in_h * up_y + pad_y0 + pad_y1 - kernel_h) // down_y + 1
out_w = (in_w * up_x + pad_x0 + pad_x1 - kernel_w) // down_x + 1
ctx.out_size = out_h, out_w
ctx.up = up_x, up_y
ctx.down = down_x, down_y
ctx.pad = pad_x0, pad_x1, pad_y0, pad_y1
g_pad_x0 = kernel_w - pad_x0 - 1
g_pad_y0 = kernel_h - pad_y0 - 1
g_pad_x1 = in_w * up_x - out_w * down_x + pad_x0 - up_x + 1
g_pad_y1 = in_h * up_y - out_h * down_y + pad_y0 - up_y + 1
ctx.g_pad = g_pad_x0, g_pad_x1, g_pad_y0, g_pad_y1
out = upfirdn2d_op.upfirdn2d(input, kernel, up_x, up_y, down_x,
down_y, pad_x0, pad_x1, pad_y0, pad_y1)
out = out.view(-1, channel, out_h, out_w)
return out
@staticmethod
def backward(ctx, grad_output):
kernel, grad_kernel = ctx.saved_tensors
grad_input = UpFirDn2dBackward.apply(grad_output, kernel,
grad_kernel, ctx.up, ctx.down, ctx.pad, ctx.g_pad, ctx.in_size,
ctx.out_size)
return grad_input, None, None, None, None
class Conv2d(nn.Module):
"""Conv2d layer with optimal upsampling and downsampling (StyleGAN2)."""
def __init__(self, in_ch, out_ch, kernel, up=False, down=False,
resample_kernel=(1, 3, 3, 1), use_bias=True, kernel_init=None):
super().__init__()
assert not (up and down)
assert kernel >= 1 and kernel % 2 == 1
self.weight = nn.Parameter(torch.zeros(out_ch, in_ch, kernel, kernel))
if kernel_init is not None:
self.weight.data = kernel_init(self.weight.data.shape)
if use_bias:
self.bias = nn.Parameter(torch.zeros(out_ch))
self.up = up
self.down = down
self.resample_kernel = resample_kernel
self.kernel = kernel
self.use_bias = use_bias
def forward(self, x):
if self.up:
x = upsample_conv_2d(x, self.weight, k=self.resample_kernel)
elif self.down:
x = conv_downsample_2d(x, self.weight, k=self.resample_kernel)
else:
x = F.conv2d(x, self.weight, stride=1, padding=self.kernel // 2)
if self.use_bias:
x = x + self.bias.reshape(1, -1, 1, 1)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_ch': 4, 'out_ch': 4, 'kernel': 1}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch.autograd import Function
import numpy as np
import torchvision.transforms.functional as F
import torch.nn as nn
import torch.nn.functional as F
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
@triton.jit
def triton_poi_fused_add_0(in_out_ptr0, in_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = xindex // 16 % 4
tmp0 = tl.load(in_out_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + x1, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(in_out_ptr0 + x3, tmp2, xmask)
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 1, 1), (4, 1, 1, 1))
assert_size_stride(primals_2, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_3, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(primals_2, primals_1, stride=(1,
1), padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 4, 4, 4), (64, 16, 4, 1))
buf1 = buf0
del buf0
get_raw_stream(0)
triton_poi_fused_add_0[grid(256)](buf1, primals_3, 256, XBLOCK=128,
num_warps=4, num_stages=1)
del primals_3
return buf1, primals_1, primals_2
def _setup_kernel(k):
k = np.asarray(k, dtype=np.float32)
if k.ndim == 1:
k = np.outer(k, k)
k /= np.sum(k)
assert k.ndim == 2
assert k.shape[0] == k.shape[1]
return k
def upfirdn2d_native(input, kernel, up_x, up_y, down_x, down_y, pad_x0,
pad_x1, pad_y0, pad_y1):
_, channel, in_h, in_w = input.shape
input = input.reshape(-1, in_h, in_w, 1)
_, in_h, in_w, minor = input.shape
kernel_h, kernel_w = kernel.shape
out = input.view(-1, in_h, 1, in_w, 1, minor)
out = F.pad(out, [0, 0, 0, up_x - 1, 0, 0, 0, up_y - 1])
out = out.view(-1, in_h * up_y, in_w * up_x, minor)
out = F.pad(out, [0, 0, max(pad_x0, 0), max(pad_x1, 0), max(pad_y0, 0),
max(pad_y1, 0)])
out = out[:, max(-pad_y0, 0):out.shape[1] - max(-pad_y1, 0), max(-
pad_x0, 0):out.shape[2] - max(-pad_x1, 0), :]
out = out.permute(0, 3, 1, 2)
out = out.reshape([-1, 1, in_h * up_y + pad_y0 + pad_y1, in_w * up_x +
pad_x0 + pad_x1])
w = torch.flip(kernel, [0, 1]).view(1, 1, kernel_h, kernel_w)
out = F.conv2d(out, w)
out = out.reshape(-1, minor, in_h * up_y + pad_y0 + pad_y1 - kernel_h +
1, in_w * up_x + pad_x0 + pad_x1 - kernel_w + 1)
out = out.permute(0, 2, 3, 1)
out = out[:, ::down_y, ::down_x, :]
out_h = (in_h * up_y + pad_y0 + pad_y1 - kernel_h) // down_y + 1
out_w = (in_w * up_x + pad_x0 + pad_x1 - kernel_w) // down_x + 1
return out.view(-1, channel, out_h, out_w)
def upfirdn2d(input, kernel, up=1, down=1, pad=(0, 0)):
if input.device.type == 'cpu':
out = upfirdn2d_native(input, kernel, up, up, down, down, pad[0],
pad[1], pad[0], pad[1])
else:
out = UpFirDn2d.apply(input, kernel, (up, up), (down, down), (pad[0
], pad[1], pad[0], pad[1]))
return out
def conv_downsample_2d(x, w, k=None, factor=2, gain=1):
"""Fused `tf.nn.conv2d()` followed by `downsample_2d()`.
Padding is performed only once at the beginning, not between the operations.
The fused op is considerably more efficient than performing the same
calculation
using standard TensorFlow ops. It supports gradients of arbitrary order.
Args:
x: Input tensor of the shape `[N, C, H, W]` or `[N, H, W,
C]`.
w: Weight tensor of the shape `[filterH, filterW, inChannels,
outChannels]`. Grouped convolution can be performed by `inChannels =
x.shape[0] // numGroups`.
k: FIR filter of the shape `[firH, firW]` or `[firN]`
(separable). The default is `[1] * factor`, which corresponds to
average pooling.
factor: Integer downsampling factor (default: 2).
gain: Scaling factor for signal magnitude (default: 1.0).
Returns:
Tensor of the shape `[N, C, H // factor, W // factor]` or
`[N, H // factor, W // factor, C]`, and same datatype as `x`.
"""
assert isinstance(factor, int) and factor >= 1
_outC, _inC, convH, convW = w.shape
assert convW == convH
if k is None:
k = [1] * factor
k = _setup_kernel(k) * gain
p = k.shape[0] - factor + (convW - 1)
s = [factor, factor]
x = upfirdn2d(x, torch.tensor(k, device=x.device), pad=((p + 1) // 2, p //
2))
return F.conv2d(x, w, stride=s, padding=0)
def _shape(x, dim):
return x.shape[dim]
def upsample_conv_2d(x, w, k=None, factor=2, gain=1):
"""Fused `upsample_2d()` followed by `tf.nn.conv2d()`.
Padding is performed only once at the beginning, not between the
operations.
The fused op is considerably more efficient than performing the same
calculation
using standard TensorFlow ops. It supports gradients of arbitrary order.
Args:
x: Input tensor of the shape `[N, C, H, W]` or `[N, H, W,
C]`.
w: Weight tensor of the shape `[filterH, filterW, inChannels,
outChannels]`. Grouped convolution can be performed by `inChannels =
x.shape[0] // numGroups`.
k: FIR filter of the shape `[firH, firW]` or `[firN]`
(separable). The default is `[1] * factor`, which corresponds to
nearest-neighbor upsampling.
factor: Integer upsampling factor (default: 2).
gain: Scaling factor for signal magnitude (default: 1.0).
Returns:
Tensor of the shape `[N, C, H * factor, W * factor]` or
`[N, H * factor, W * factor, C]`, and same datatype as `x`.
"""
assert isinstance(factor, int) and factor >= 1
assert len(w.shape) == 4
convH = w.shape[2]
convW = w.shape[3]
inC = w.shape[1]
w.shape[0]
assert convW == convH
if k is None:
k = [1] * factor
k = _setup_kernel(k) * (gain * factor ** 2)
p = k.shape[0] - factor - (convW - 1)
stride = factor, factor
stride = [1, 1, factor, factor]
output_shape = (_shape(x, 2) - 1) * factor + convH, (_shape(x, 3) - 1
) * factor + convW
output_padding = output_shape[0] - (_shape(x, 2) - 1) * stride[0
] - convH, output_shape[1] - (_shape(x, 3) - 1) * stride[1] - convW
assert output_padding[0] >= 0 and output_padding[1] >= 0
num_groups = _shape(x, 1) // inC
w = torch.reshape(w, (num_groups, -1, inC, convH, convW))
w = w[..., ::-1, ::-1].permute(0, 2, 1, 3, 4)
w = torch.reshape(w, (num_groups * inC, -1, convH, convW))
x = F.conv_transpose2d(x, w, stride=stride, output_padding=
output_padding, padding=0)
return upfirdn2d(x, torch.tensor(k, device=x.device), pad=((p + 1) // 2 +
factor - 1, p // 2 + 1))
class UpFirDn2dBackward(Function):
@staticmethod
def forward(ctx, grad_output, kernel, grad_kernel, up, down, pad, g_pad,
in_size, out_size):
up_x, up_y = up
down_x, down_y = down
g_pad_x0, g_pad_x1, g_pad_y0, g_pad_y1 = g_pad
grad_output = grad_output.reshape(-1, out_size[0], out_size[1], 1)
grad_input = upfirdn2d_op.upfirdn2d(grad_output, grad_kernel,
down_x, down_y, up_x, up_y, g_pad_x0, g_pad_x1, g_pad_y0, g_pad_y1)
grad_input = grad_input.view(in_size[0], in_size[1], in_size[2],
in_size[3])
ctx.save_for_backward(kernel)
pad_x0, pad_x1, pad_y0, pad_y1 = pad
ctx.up_x = up_x
ctx.up_y = up_y
ctx.down_x = down_x
ctx.down_y = down_y
ctx.pad_x0 = pad_x0
ctx.pad_x1 = pad_x1
ctx.pad_y0 = pad_y0
ctx.pad_y1 = pad_y1
ctx.in_size = in_size
ctx.out_size = out_size
return grad_input
@staticmethod
def backward(ctx, gradgrad_input):
kernel, = ctx.saved_tensors
gradgrad_input = gradgrad_input.reshape(-1, ctx.in_size[2], ctx.
in_size[3], 1)
gradgrad_out = upfirdn2d_op.upfirdn2d(gradgrad_input, kernel, ctx.
up_x, ctx.up_y, ctx.down_x, ctx.down_y, ctx.pad_x0, ctx.pad_x1,
ctx.pad_y0, ctx.pad_y1)
gradgrad_out = gradgrad_out.view(ctx.in_size[0], ctx.in_size[1],
ctx.out_size[0], ctx.out_size[1])
return gradgrad_out, None, None, None, None, None, None, None, None
class UpFirDn2d(Function):
@staticmethod
def forward(ctx, input, kernel, up, down, pad):
up_x, up_y = up
down_x, down_y = down
pad_x0, pad_x1, pad_y0, pad_y1 = pad
kernel_h, kernel_w = kernel.shape
_batch, channel, in_h, in_w = input.shape
ctx.in_size = input.shape
input = input.reshape(-1, in_h, in_w, 1)
ctx.save_for_backward(kernel, torch.flip(kernel, [0, 1]))
out_h = (in_h * up_y + pad_y0 + pad_y1 - kernel_h) // down_y + 1
out_w = (in_w * up_x + pad_x0 + pad_x1 - kernel_w) // down_x + 1
ctx.out_size = out_h, out_w
ctx.up = up_x, up_y
ctx.down = down_x, down_y
ctx.pad = pad_x0, pad_x1, pad_y0, pad_y1
g_pad_x0 = kernel_w - pad_x0 - 1
g_pad_y0 = kernel_h - pad_y0 - 1
g_pad_x1 = in_w * up_x - out_w * down_x + pad_x0 - up_x + 1
g_pad_y1 = in_h * up_y - out_h * down_y + pad_y0 - up_y + 1
ctx.g_pad = g_pad_x0, g_pad_x1, g_pad_y0, g_pad_y1
out = upfirdn2d_op.upfirdn2d(input, kernel, up_x, up_y, down_x,
down_y, pad_x0, pad_x1, pad_y0, pad_y1)
out = out.view(-1, channel, out_h, out_w)
return out
@staticmethod
def backward(ctx, grad_output):
kernel, grad_kernel = ctx.saved_tensors
grad_input = UpFirDn2dBackward.apply(grad_output, kernel,
grad_kernel, ctx.up, ctx.down, ctx.pad, ctx.g_pad, ctx.in_size,
ctx.out_size)
return grad_input, None, None, None, None
class Conv2dNew(nn.Module):
"""Conv2d layer with optimal upsampling and downsampling (StyleGAN2)."""
def __init__(self, in_ch, out_ch, kernel, up=False, down=False,
resample_kernel=(1, 3, 3, 1), use_bias=True, kernel_init=None):
super().__init__()
assert not (up and down)
assert kernel >= 1 and kernel % 2 == 1
self.weight = nn.Parameter(torch.zeros(out_ch, in_ch, kernel, kernel))
if kernel_init is not None:
self.weight.data = kernel_init(self.weight.data.shape)
if use_bias:
self.bias = nn.Parameter(torch.zeros(out_ch))
self.up = up
self.down = down
self.resample_kernel = resample_kernel
self.kernel = kernel
self.use_bias = use_bias
def forward(self, input_0):
primals_1 = self.weight
primals_3 = self.bias
primals_2 = input_0
output = call([primals_1, primals_2, primals_3])
return output[0]
|
DeepTitan/PNDM
|
Conv2d
| false
| 13,934
|
[
"Apache-2.0"
] | 61
|
4037a4f40011c9a0d47b92303e64d47fcc7ed56a
|
https://github.com/DeepTitan/PNDM/tree/4037a4f40011c9a0d47b92303e64d47fcc7ed56a
|
SymmSoftplus
|
import torch
from torch.utils.data import Dataset as Dataset
import torch.utils.data
def symm_softplus(x, softplus_=torch.nn.functional.softplus):
return softplus_(x) - 0.5 * x
class SymmSoftplus(torch.nn.Module):
def forward(self, x):
return symm_softplus(x)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch.utils.data import Dataset as Dataset
import torch.utils.data
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_mul_softplus_sub_0(in_ptr0, out_ptr0, xnumel, XBLOCK:
tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = 20.0
tmp2 = tmp0 > tmp1
tmp3 = tl_math.exp(tmp0)
tmp4 = libdevice.log1p(tmp3)
tmp5 = tl.where(tmp2, tmp0, tmp4)
tmp6 = 0.5
tmp7 = tmp0 * tmp6
tmp8 = tmp5 - tmp7
tl.store(out_ptr0 + x0, tmp8, xmask)
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_mul_softplus_sub_0[grid(256)](arg0_1, buf0, 256,
XBLOCK=128, num_warps=4, num_stages=1)
del arg0_1
return buf0,
def symm_softplus(x, softplus_=torch.nn.functional.softplus):
return softplus_(x) - 0.5 * x
class SymmSoftplusNew(torch.nn.Module):
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
|
KelvinKan/CP-Flow
|
SymmSoftplus
| false
| 13,935
|
[
"MIT"
] | 64
|
d01303cb4ebeb5a0bbfca638ffaf5b7a8ec22fb1
|
https://github.com/KelvinKan/CP-Flow/tree/d01303cb4ebeb5a0bbfca638ffaf5b7a8ec22fb1
|
UpBlock
|
import torch
import torch.nn as nn
class UpBlock(nn.Module):
def __init__(self, in_f, out_f, stride=2, add_blur=False):
super(UpBlock, self).__init__()
self.shuffle = nn.ConvTranspose2d(in_f, out_f, kernel_size=3,
stride=stride, padding=0)
self.has_blur = add_blur
if self.has_blur:
self.blur = nn.AvgPool2d(2, 1)
def forward(self, x):
x = self.shuffle(x)
if self.has_blur:
x = self.blur(x)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_f': 4, 'out_f': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
@triton.jit
def triton_poi_fused_convolution_0(in_out_ptr0, in_ptr0, xnumel, XBLOCK: tl
.constexpr):
xnumel = 1296
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = xindex // 81 % 4
tmp0 = tl.load(in_out_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + x1, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(in_out_ptr0 + x3, tmp2, xmask)
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 3, 3), (36, 9, 3, 1))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(primals_3, primals_1, stride=(2,
2), padding=(0, 0), dilation=(1, 1), transposed=True,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 4, 9, 9), (324, 81, 9, 1))
buf1 = buf0
del buf0
get_raw_stream(0)
triton_poi_fused_convolution_0[grid(1296)](buf1, primals_2, 1296,
XBLOCK=256, num_warps=4, num_stages=1)
del primals_2
return buf1, primals_1, primals_3
class UpBlockNew(nn.Module):
def __init__(self, in_f, out_f, stride=2, add_blur=False):
super(UpBlockNew, self).__init__()
self.shuffle = nn.ConvTranspose2d(in_f, out_f, kernel_size=3,
stride=stride, padding=0)
self.has_blur = add_blur
if self.has_blur:
self.blur = nn.AvgPool2d(2, 1)
def forward(self, input_0):
primals_1 = self.shuffle.weight
primals_2 = self.shuffle.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3])
return output[0]
|
Kash6/AnimeBot
|
UpBlock
| false
| 13,936
|
[
"MIT"
] | 177
|
99c68bdb03501d6919669c4aabbb9fe5ea92ec8e
|
https://github.com/Kash6/AnimeBot/tree/99c68bdb03501d6919669c4aabbb9fe5ea92ec8e
|
FCNet
|
import torch
import torch.nn as nn
from torch.nn.utils import weight_norm
import torch.utils.data
class FCNet(nn.Module):
def __init__(self, in_size, out_size, activate=None, drop=0.0):
super(FCNet, self).__init__()
self.lin = weight_norm(nn.Linear(in_size, out_size), dim=None)
self.drop_value = drop
self.drop = nn.Dropout(drop)
self.activate = activate.lower() if activate is not None else None
if activate == 'relu':
self.ac_fn = nn.ReLU()
elif activate == 'sigmoid':
self.ac_fn = nn.Sigmoid()
elif activate == 'tanh':
self.ac_fn = nn.Tanh()
def forward(self, x):
if self.drop_value > 0:
x = self.drop(x)
x = self.lin(x)
if self.activate is not None:
x = self.ac_fn(x)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_size': 4, 'out_size': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice
import torch.nn as nn
from torch.nn.utils import weight_norm
import torch.utils.data
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_per_fused_div_mul_norm_0(in_out_ptr0, in_ptr0, in_ptr1, out_ptr0,
xnumel, rnumel, XBLOCK: tl.constexpr):
RBLOCK: tl.constexpr = 16
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp6 = tl.load(in_ptr1 + 0)
tmp7 = tl.broadcast_to(tmp6, [XBLOCK, RBLOCK])
tmp1 = tmp0 * tmp0
tmp2 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp4 = tl.sum(tmp2, 1)[:, None]
tmp5 = libdevice.sqrt(tmp4)
tmp8 = tmp7 / tmp5
tmp9 = tmp0 * tmp8
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([XBLOCK, 1], 0, tl.int32), tmp5, None)
tl.store(out_ptr0 + tl.broadcast_to(r0, [XBLOCK, RBLOCK]), tmp9, None)
def call(args):
primals_1, primals_2, primals_3, primals_4 = args
args.clear()
assert_size_stride(primals_1, (), ())
assert_size_stride(primals_2, (4, 4), (4, 1))
assert_size_stride(primals_3, (4,), (1,))
assert_size_stride(primals_4, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((), (), torch.float32)
buf1 = buf0
del buf0
buf2 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
get_raw_stream(0)
triton_per_fused_div_mul_norm_0[grid(1)](buf1, primals_2, primals_1,
buf2, 1, 16, XBLOCK=1, num_warps=2, num_stages=1)
buf3 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_3, reinterpret_tensor(primals_4, (64,
4), (4, 1), 0), reinterpret_tensor(buf2, (4, 4), (1, 4), 0),
alpha=1, beta=1, out=buf3)
del primals_3
return reinterpret_tensor(buf3, (4, 4, 4, 4), (64, 16, 4, 1), 0
), buf2, primals_1, primals_2, buf1, reinterpret_tensor(primals_4,
(64, 4), (4, 1), 0)
class FCNetNew(nn.Module):
def __init__(self, in_size, out_size, activate=None, drop=0.0):
super(FCNetNew, self).__init__()
self.lin = weight_norm(nn.Linear(in_size, out_size), dim=None)
self.drop_value = drop
self.drop = nn.Dropout(drop)
self.activate = activate.lower() if activate is not None else None
if activate == 'relu':
self.ac_fn = nn.ReLU()
elif activate == 'sigmoid':
self.ac_fn = nn.Sigmoid()
elif activate == 'tanh':
self.ac_fn = nn.Tanh()
def forward(self, input_0):
primals_3 = self.lin.bias
primals_1 = self.lin.weight_g
primals_2 = self.lin.weight_v
primals_4 = input_0
output = call([primals_1, primals_2, primals_3, primals_4])
return output[0]
|
KaihuaTang/VQA2.0-Recent-Approachs-2018.pytorch
|
FCNet
| false
| 13,937
|
[
"MIT"
] | 298
|
52e1ba5a7f3b88c617115ccc755e2e7868e8de2b
|
https://github.com/KaihuaTang/VQA2.0-Recent-Approachs-2018.pytorch/tree/52e1ba5a7f3b88c617115ccc755e2e7868e8de2b
|
ModulatedToRGB
|
import torch
import torch.nn as nn
from copy import deepcopy
from functools import partial
from torch.nn import functional as F
from torch.nn.init import _calculate_correct_fan
def equalized_lr(module, name='weight', gain=2 ** 0.5, mode='fan_in',
lr_mul=1.0):
"""Equalized Learning Rate.
This trick is proposed in:
Progressive Growing of GANs for Improved Quality, Stability, and Variation
The general idea is to dynamically rescale the weight in training instead
of in initializing so that the variance of the responses in each layer is
guaranteed with some statistical properties.
Note that this function is always combined with a convolution module which
is initialized with :math:`\\mathcal{N}(0, 1)`.
Args:
module (nn.Module): Module to be wrapped.
name (str | optional): The name of weights. Defaults to 'weight'.
mode (str, optional): The mode of computing ``fan`` which is the
same as ``kaiming_init`` in pytorch. You can choose one from
['fan_in', 'fan_out']. Defaults to 'fan_in'.
Returns:
nn.Module: Module that is registered with equalized lr hook.
"""
EqualizedLR.apply(module, name, gain=gain, mode=mode, lr_mul=lr_mul)
return module
def _make_kernel(k):
k = torch.tensor(k, dtype=torch.float32)
if k.ndim == 1:
k = k[None, :] * k[:, None]
k /= k.sum()
return k
class EqualizedLR:
"""Equalized Learning Rate.
This trick is proposed in:
Progressive Growing of GANs for Improved Quality, Stability, and Variation
The general idea is to dynamically rescale the weight in training instead
of in initializing so that the variance of the responses in each layer is
guaranteed with some statistical properties.
Note that this function is always combined with a convolution module which
is initialized with :math:`\\mathcal{N}(0, 1)`.
Args:
name (str | optional): The name of weights. Defaults to 'weight'.
mode (str, optional): The mode of computing ``fan`` which is the
same as ``kaiming_init`` in pytorch. You can choose one from
['fan_in', 'fan_out']. Defaults to 'fan_in'.
"""
def __init__(self, name='weight', gain=2 ** 0.5, mode='fan_in', lr_mul=1.0
):
self.name = name
self.mode = mode
self.gain = gain
self.lr_mul = lr_mul
def compute_weight(self, module):
"""Compute weight with equalized learning rate.
Args:
module (nn.Module): A module that is wrapped with equalized lr.
Returns:
torch.Tensor: Updated weight.
"""
weight = getattr(module, self.name + '_orig')
if weight.ndim == 5:
fan = _calculate_correct_fan(weight[0], self.mode)
else:
assert weight.ndim <= 4
fan = _calculate_correct_fan(weight, self.mode)
weight = weight * torch.tensor(self.gain, device=weight.device
) * torch.sqrt(torch.tensor(1.0 / fan, device=weight.device)
) * self.lr_mul
return weight
def __call__(self, module, inputs):
"""Standard interface for forward pre hooks."""
setattr(module, self.name, self.compute_weight(module))
@staticmethod
def apply(module, name, gain=2 ** 0.5, mode='fan_in', lr_mul=1.0):
"""Apply function.
This function is to register an equalized learning rate hook in an
``nn.Module``.
Args:
module (nn.Module): Module to be wrapped.
name (str | optional): The name of weights. Defaults to 'weight'.
mode (str, optional): The mode of computing ``fan`` which is the
same as ``kaiming_init`` in pytorch. You can choose one from
['fan_in', 'fan_out']. Defaults to 'fan_in'.
Returns:
nn.Module: Module that is registered with equalized lr hook.
"""
for _, hook in module._forward_pre_hooks.items():
if isinstance(hook, EqualizedLR):
raise RuntimeError(
f'Cannot register two equalized_lr hooks on the same parameter {name} in {module} module.'
)
fn = EqualizedLR(name, gain=gain, mode=mode, lr_mul=lr_mul)
weight = module._parameters[name]
delattr(module, name)
module.register_parameter(name + '_orig', weight)
setattr(module, name, weight.data)
module.register_forward_pre_hook(fn)
return fn
class EqualizedLRLinearModule(nn.Linear):
"""Equalized LR LinearModule.
In this module, we adopt equalized lr in ``nn.Linear``. The equalized
learning rate is proposed in:
Progressive Growing of GANs for Improved Quality, Stability, and Variation
Note that, the initialization of ``self.weight`` will be overwritten as
:math:`\\mathcal{N}(0, 1)`.
Args:
equalized_lr_cfg (dict | None, optional): Config for ``EqualizedLR``.
If ``None``, equalized learning rate is ignored. Defaults to
dict(mode='fan_in').
"""
def __init__(self, *args, equalized_lr_cfg=dict(mode='fan_in'), **kwargs):
super(EqualizedLRLinearModule, self).__init__(*args, **kwargs)
self.with_equlized_lr = equalized_lr_cfg is not None
if self.with_equlized_lr:
self.lr_mul = equalized_lr_cfg.get('lr_mul', 1.0)
else:
self.lr_mul = 1.0
if self.with_equlized_lr:
equalized_lr(self, **equalized_lr_cfg)
self._init_linear_weights()
def _init_linear_weights(self):
"""Initialize linear weights as described in PGGAN."""
nn.init.normal_(self.weight, 0, 1.0 / self.lr_mul)
if self.bias is not None:
nn.init.constant_(self.bias, 0.0)
class EqualLinearActModule(nn.Module):
"""Equalized LR Linear Module with Activation Layer.
Args:
nn ([type]): [description]
"""
def __init__(self, *args, equalized_lr_cfg=dict(gain=1.0, lr_mul=1.0),
bias=True, bias_init=0.0, act_cfg=None, **kwargs):
super(EqualLinearActModule, self).__init__()
self.with_activation = act_cfg is not None
self.linear = EqualizedLRLinearModule(*args, bias=False,
equalized_lr_cfg=equalized_lr_cfg, **kwargs)
if equalized_lr_cfg is not None:
self.lr_mul = equalized_lr_cfg.get('lr_mul', 1.0)
else:
self.lr_mul = 1.0
if bias:
self.bias = nn.Parameter(torch.zeros(self.linear.out_features).
fill_(bias_init))
else:
self.bias = None
if self.with_activation:
act_cfg = deepcopy(act_cfg)
if act_cfg['type'] == 'fused_bias':
self.act_type = act_cfg.pop('type')
assert self.bias is not None
self.activate = partial(fused_bias_leakyrelu, **act_cfg)
else:
self.act_type = 'normal'
self.activate = build_activation_layer(act_cfg)
else:
self.act_type = None
def forward(self, x):
if x.ndim >= 3:
x = x.reshape(x.size(0), -1)
x = self.linear(x)
if self.with_activation and self.act_type == 'fused_bias':
x = self.activate(x, self.bias * self.lr_mul)
elif self.bias is not None and self.with_activation:
x = self.activate(x + self.bias * self.lr_mul)
elif self.bias is not None:
x = x + self.bias * self.lr_mul
elif self.with_activation:
x = self.activate(x)
return x
class Blur(nn.Module):
def __init__(self, kernel, pad, upsample_factor=1):
super(Blur, self).__init__()
kernel = _make_kernel(kernel)
if upsample_factor > 1:
kernel = kernel * upsample_factor ** 2
self.register_buffer('kernel', kernel)
self.pad = pad
def forward(self, x):
return upfirdn2d(x, self.kernel, pad=self.pad)
class ModulatedConv2d(nn.Module):
"""Modulated Conv2d in StyleGANv2.
Attention:
#. ``style_bias`` is provided to check the difference between official TF
implementation and other PyTorch implementation.
In TF, Tero explicitly add the ``1.`` after style code, while unofficial
implementation adopts bias initialization with ``1.``.
Details can be found in:
https://github.com/rosinality/stylegan2-pytorch/blob/master/model.py#L214
https://github.com/NVlabs/stylegan2/blob/master/training/networks_stylegan2.py#L99
"""
def __init__(self, in_channels, out_channels, kernel_size,
style_channels, demodulate=True, upsample=False, downsample=False,
blur_kernel=[1, 3, 3, 1], equalized_lr_cfg=dict(mode='fan_in',
lr_mul=1.0, gain=1.0), style_mod_cfg=dict(bias_init=1.0),
style_bias=0.0, eps=1e-08):
super(ModulatedConv2d, self).__init__()
self.in_channels = in_channels
self.out_channels = out_channels
self.kernel_size = kernel_size
self.style_channels = style_channels
self.demodulate = demodulate
assert isinstance(self.kernel_size, int) and (self.kernel_size >= 1 and
self.kernel_size % 2 == 1)
self.upsample = upsample
self.downsample = downsample
self.style_bias = style_bias
self.eps = eps
style_mod_cfg = dict() if style_mod_cfg is None else style_mod_cfg
self.style_modulation = EqualLinearActModule(style_channels,
in_channels, **style_mod_cfg)
lr_mul_ = 1.0
if equalized_lr_cfg is not None:
lr_mul_ = equalized_lr_cfg.get('lr_mul', 1.0)
self.weight = nn.Parameter(torch.randn(1, out_channels, in_channels,
kernel_size, kernel_size).div_(lr_mul_))
if upsample:
factor = 2
p = len(blur_kernel) - factor - (kernel_size - 1)
pad0 = (p + 1) // 2 + factor - 1
pad1 = p // 2 + 1
self.blur = Blur(blur_kernel, (pad0, pad1), upsample_factor=factor)
if downsample:
factor = 2
p = len(blur_kernel) - factor + (kernel_size - 1)
pad0 = (p + 1) // 2
pad1 = p // 2
self.blur = Blur(blur_kernel, pad=(pad0, pad1))
if equalized_lr_cfg is not None:
equalized_lr(self, **equalized_lr_cfg)
self.padding = kernel_size // 2
def forward(self, x, style):
n, c, h, w = x.shape
style = self.style_modulation(style).view(n, 1, c, 1, 1
) + self.style_bias
weight = self.weight * style
if self.demodulate:
demod = torch.rsqrt(weight.pow(2).sum([2, 3, 4]) + self.eps)
weight = weight * demod.view(n, self.out_channels, 1, 1, 1)
weight = weight.view(n * self.out_channels, c, self.kernel_size,
self.kernel_size)
if self.upsample:
x = x.reshape(1, n * c, h, w)
weight = weight.view(n, self.out_channels, c, self.kernel_size,
self.kernel_size)
weight = weight.transpose(1, 2).reshape(n * c, self.
out_channels, self.kernel_size, self.kernel_size)
x = F.conv_transpose2d(x, weight, padding=0, stride=2, groups=n)
x = x.reshape(n, self.out_channels, *x.shape[-2:])
x = self.blur(x)
elif self.downsample:
x = self.blur(x)
x = x.view(1, n * self.in_channels, *x.shape[-2:])
x = F.conv2d(x, weight, stride=2, padding=0, groups=n)
x = x.view(n, self.out_channels, *x.shape[-2:])
else:
x = x.view(1, n * c, h, w)
x = F.conv2d(x, weight, stride=1, padding=self.padding, groups=n)
x = x.view(n, self.out_channels, *x.shape[-2:])
return x
class UpsampleUpFIRDn(nn.Module):
def __init__(self, kernel, factor=2):
super(UpsampleUpFIRDn, self).__init__()
self.factor = factor
kernel = _make_kernel(kernel) * factor ** 2
self.register_buffer('kernel', kernel)
p = kernel.shape[0] - factor
pad0 = (p + 1) // 2 + factor - 1
pad1 = p // 2
self.pad = pad0, pad1
def forward(self, x):
out = upfirdn2d(x, self.kernel, up=self.factor, down=1, pad=self.pad)
return out
class ModulatedToRGB(nn.Module):
def __init__(self, in_channels, style_channels, out_channels=3,
upsample=True, blur_kernel=[1, 3, 3, 1], style_mod_cfg=dict(
bias_init=1.0), style_bias=0.0):
super(ModulatedToRGB, self).__init__()
if upsample:
self.upsample = UpsampleUpFIRDn(blur_kernel)
self.conv = ModulatedConv2d(in_channels, out_channels=out_channels,
kernel_size=1, style_channels=style_channels, demodulate=False,
style_mod_cfg=style_mod_cfg, style_bias=style_bias)
self.bias = nn.Parameter(torch.zeros(1, 3, 1, 1))
def forward(self, x, style, skip=None):
out = self.conv(x, style)
out = out + self.bias
if skip is not None:
skip = self.upsample(skip)
out = out + skip
return out
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4])]
def get_init_inputs():
return [[], {'in_channels': 4, 'style_channels': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
import torch.nn as nn
from copy import deepcopy
from functools import partial
from torch.nn import functional as F
from torch.nn.init import _calculate_correct_fan
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_mul_sqrt_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 12
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = 0.5
tmp2 = tmp0 * tmp1
tmp3 = 1.0
tmp4 = tmp2 * tmp3
tl.store(out_ptr0 + x0, tmp4, xmask)
@triton.jit
def triton_poi_fused_mul_sqrt_1(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = 0.5
tmp2 = tmp0 * tmp1
tmp3 = 1.0
tmp4 = tmp2 * tmp3
tl.store(out_ptr0 + x0, tmp4, xmask)
@triton.jit
def triton_poi_fused_add_mul_2(in_ptr0, in_ptr1, in_ptr2, out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 48
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex % 12
x0 = xindex % 4
x2 = xindex // 12
x4 = xindex
tmp0 = tl.load(in_ptr0 + x3, xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + (x0 + 4 * x2), xmask, eviction_policy='evict_last'
)
tmp2 = tl.load(in_ptr2 + x0, xmask, eviction_policy='evict_last')
tmp3 = 1.0
tmp4 = tmp2 * tmp3
tmp5 = tmp1 + tmp4
tmp6 = 0.0
tmp7 = tmp5 + tmp6
tmp8 = tmp0 * tmp7
tl.store(out_ptr0 + x4, tmp8, xmask)
@triton.jit
def triton_poi_fused_add_3(in_out_ptr0, in_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 192
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = xindex // 16 % 3
tmp0 = tl.load(in_out_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + x1, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(in_out_ptr0 + x3, tmp2, xmask)
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6 = args
args.clear()
assert_size_stride(primals_1, (1, 3, 4, 1, 1), (12, 4, 1, 1, 1))
assert_size_stride(primals_2, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_3, (4, 4), (4, 1))
assert_size_stride(primals_4, (4, 4), (4, 1))
assert_size_stride(primals_5, (4,), (1,))
assert_size_stride(primals_6, (1, 3, 1, 1), (3, 1, 1, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((1, 3, 4, 1, 1), (12, 4, 1, 1, 1), torch.
float32)
get_raw_stream(0)
triton_poi_fused_mul_sqrt_0[grid(12)](primals_1, buf0, 12, XBLOCK=
16, num_warps=1, num_stages=1)
del primals_1
buf1 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
triton_poi_fused_mul_sqrt_1[grid(16)](primals_4, buf1, 16, XBLOCK=
16, num_warps=1, num_stages=1)
del primals_4
buf2 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.mm(primals_3, reinterpret_tensor(buf1, (4, 4), (1, 4
), 0), out=buf2)
buf3 = empty_strided_cuda((4, 3, 4, 1, 1), (12, 4, 1, 1, 1), torch.
float32)
triton_poi_fused_add_mul_2[grid(48)](buf0, buf2, primals_5, buf3,
48, XBLOCK=64, num_warps=1, num_stages=1)
buf4 = extern_kernels.convolution(reinterpret_tensor(primals_2, (1,
16, 4, 4), (256, 16, 4, 1), 0), reinterpret_tensor(buf3, (12, 4,
1, 1), (4, 1, 0, 0), 0), stride=(1, 1), padding=(0, 0),
dilation=(1, 1), transposed=False, output_padding=(0, 0),
groups=4, bias=None)
assert_size_stride(buf4, (1, 12, 4, 4), (192, 16, 4, 1))
buf5 = reinterpret_tensor(buf4, (4, 3, 4, 4), (48, 16, 4, 1), 0)
del buf4
triton_poi_fused_add_3[grid(192)](buf5, primals_6, 192, XBLOCK=256,
num_warps=4, num_stages=1)
del primals_6
return (buf5, buf0, buf1, primals_3, primals_5, buf0, buf2,
reinterpret_tensor(buf3, (12, 4, 1, 1), (4, 1, 1, 1), 0),
reinterpret_tensor(primals_2, (1, 16, 4, 4), (256, 16, 4, 1), 0))
def equalized_lr(module, name='weight', gain=2 ** 0.5, mode='fan_in',
lr_mul=1.0):
"""Equalized Learning Rate.
This trick is proposed in:
Progressive Growing of GANs for Improved Quality, Stability, and Variation
The general idea is to dynamically rescale the weight in training instead
of in initializing so that the variance of the responses in each layer is
guaranteed with some statistical properties.
Note that this function is always combined with a convolution module which
is initialized with :math:`\\mathcal{N}(0, 1)`.
Args:
module (nn.Module): Module to be wrapped.
name (str | optional): The name of weights. Defaults to 'weight'.
mode (str, optional): The mode of computing ``fan`` which is the
same as ``kaiming_init`` in pytorch. You can choose one from
['fan_in', 'fan_out']. Defaults to 'fan_in'.
Returns:
nn.Module: Module that is registered with equalized lr hook.
"""
EqualizedLR.apply(module, name, gain=gain, mode=mode, lr_mul=lr_mul)
return module
def _make_kernel(k):
k = torch.tensor(k, dtype=torch.float32)
if k.ndim == 1:
k = k[None, :] * k[:, None]
k /= k.sum()
return k
class EqualizedLR:
"""Equalized Learning Rate.
This trick is proposed in:
Progressive Growing of GANs for Improved Quality, Stability, and Variation
The general idea is to dynamically rescale the weight in training instead
of in initializing so that the variance of the responses in each layer is
guaranteed with some statistical properties.
Note that this function is always combined with a convolution module which
is initialized with :math:`\\mathcal{N}(0, 1)`.
Args:
name (str | optional): The name of weights. Defaults to 'weight'.
mode (str, optional): The mode of computing ``fan`` which is the
same as ``kaiming_init`` in pytorch. You can choose one from
['fan_in', 'fan_out']. Defaults to 'fan_in'.
"""
def __init__(self, name='weight', gain=2 ** 0.5, mode='fan_in', lr_mul=1.0
):
self.name = name
self.mode = mode
self.gain = gain
self.lr_mul = lr_mul
def compute_weight(self, module):
"""Compute weight with equalized learning rate.
Args:
module (nn.Module): A module that is wrapped with equalized lr.
Returns:
torch.Tensor: Updated weight.
"""
weight = getattr(module, self.name + '_orig')
if weight.ndim == 5:
fan = _calculate_correct_fan(weight[0], self.mode)
else:
assert weight.ndim <= 4
fan = _calculate_correct_fan(weight, self.mode)
weight = weight * torch.tensor(self.gain, device=weight.device
) * torch.sqrt(torch.tensor(1.0 / fan, device=weight.device)
) * self.lr_mul
return weight
def __call__(self, module, inputs):
"""Standard interface for forward pre hooks."""
setattr(module, self.name, self.compute_weight(module))
@staticmethod
def apply(module, name, gain=2 ** 0.5, mode='fan_in', lr_mul=1.0):
"""Apply function.
This function is to register an equalized learning rate hook in an
``nn.Module``.
Args:
module (nn.Module): Module to be wrapped.
name (str | optional): The name of weights. Defaults to 'weight'.
mode (str, optional): The mode of computing ``fan`` which is the
same as ``kaiming_init`` in pytorch. You can choose one from
['fan_in', 'fan_out']. Defaults to 'fan_in'.
Returns:
nn.Module: Module that is registered with equalized lr hook.
"""
for _, hook in module._forward_pre_hooks.items():
if isinstance(hook, EqualizedLR):
raise RuntimeError(
f'Cannot register two equalized_lr hooks on the same parameter {name} in {module} module.'
)
fn = EqualizedLR(name, gain=gain, mode=mode, lr_mul=lr_mul)
weight = module._parameters[name]
delattr(module, name)
module.register_parameter(name + '_orig', weight)
setattr(module, name, weight.data)
module.register_forward_pre_hook(fn)
return fn
class EqualizedLRLinearModule(nn.Linear):
"""Equalized LR LinearModule.
In this module, we adopt equalized lr in ``nn.Linear``. The equalized
learning rate is proposed in:
Progressive Growing of GANs for Improved Quality, Stability, and Variation
Note that, the initialization of ``self.weight`` will be overwritten as
:math:`\\mathcal{N}(0, 1)`.
Args:
equalized_lr_cfg (dict | None, optional): Config for ``EqualizedLR``.
If ``None``, equalized learning rate is ignored. Defaults to
dict(mode='fan_in').
"""
def __init__(self, *args, equalized_lr_cfg=dict(mode='fan_in'), **kwargs):
super(EqualizedLRLinearModule, self).__init__(*args, **kwargs)
self.with_equlized_lr = equalized_lr_cfg is not None
if self.with_equlized_lr:
self.lr_mul = equalized_lr_cfg.get('lr_mul', 1.0)
else:
self.lr_mul = 1.0
if self.with_equlized_lr:
equalized_lr(self, **equalized_lr_cfg)
self._init_linear_weights()
def _init_linear_weights(self):
"""Initialize linear weights as described in PGGAN."""
nn.init.normal_(self.weight, 0, 1.0 / self.lr_mul)
if self.bias is not None:
nn.init.constant_(self.bias, 0.0)
class EqualLinearActModule(nn.Module):
"""Equalized LR Linear Module with Activation Layer.
Args:
nn ([type]): [description]
"""
def __init__(self, *args, equalized_lr_cfg=dict(gain=1.0, lr_mul=1.0),
bias=True, bias_init=0.0, act_cfg=None, **kwargs):
super(EqualLinearActModule, self).__init__()
self.with_activation = act_cfg is not None
self.linear = EqualizedLRLinearModule(*args, bias=False,
equalized_lr_cfg=equalized_lr_cfg, **kwargs)
if equalized_lr_cfg is not None:
self.lr_mul = equalized_lr_cfg.get('lr_mul', 1.0)
else:
self.lr_mul = 1.0
if bias:
self.bias = nn.Parameter(torch.zeros(self.linear.out_features).
fill_(bias_init))
else:
self.bias = None
if self.with_activation:
act_cfg = deepcopy(act_cfg)
if act_cfg['type'] == 'fused_bias':
self.act_type = act_cfg.pop('type')
assert self.bias is not None
self.activate = partial(fused_bias_leakyrelu, **act_cfg)
else:
self.act_type = 'normal'
self.activate = build_activation_layer(act_cfg)
else:
self.act_type = None
def forward(self, x):
if x.ndim >= 3:
x = x.reshape(x.size(0), -1)
x = self.linear(x)
if self.with_activation and self.act_type == 'fused_bias':
x = self.activate(x, self.bias * self.lr_mul)
elif self.bias is not None and self.with_activation:
x = self.activate(x + self.bias * self.lr_mul)
elif self.bias is not None:
x = x + self.bias * self.lr_mul
elif self.with_activation:
x = self.activate(x)
return x
class Blur(nn.Module):
def __init__(self, kernel, pad, upsample_factor=1):
super(Blur, self).__init__()
kernel = _make_kernel(kernel)
if upsample_factor > 1:
kernel = kernel * upsample_factor ** 2
self.register_buffer('kernel', kernel)
self.pad = pad
def forward(self, x):
return upfirdn2d(x, self.kernel, pad=self.pad)
class ModulatedConv2d(nn.Module):
"""Modulated Conv2d in StyleGANv2.
Attention:
#. ``style_bias`` is provided to check the difference between official TF
implementation and other PyTorch implementation.
In TF, Tero explicitly add the ``1.`` after style code, while unofficial
implementation adopts bias initialization with ``1.``.
Details can be found in:
https://github.com/rosinality/stylegan2-pytorch/blob/master/model.py#L214
https://github.com/NVlabs/stylegan2/blob/master/training/networks_stylegan2.py#L99
"""
def __init__(self, in_channels, out_channels, kernel_size,
style_channels, demodulate=True, upsample=False, downsample=False,
blur_kernel=[1, 3, 3, 1], equalized_lr_cfg=dict(mode='fan_in',
lr_mul=1.0, gain=1.0), style_mod_cfg=dict(bias_init=1.0),
style_bias=0.0, eps=1e-08):
super(ModulatedConv2d, self).__init__()
self.in_channels = in_channels
self.out_channels = out_channels
self.kernel_size = kernel_size
self.style_channels = style_channels
self.demodulate = demodulate
assert isinstance(self.kernel_size, int) and (self.kernel_size >= 1 and
self.kernel_size % 2 == 1)
self.upsample = upsample
self.downsample = downsample
self.style_bias = style_bias
self.eps = eps
style_mod_cfg = dict() if style_mod_cfg is None else style_mod_cfg
self.style_modulation = EqualLinearActModule(style_channels,
in_channels, **style_mod_cfg)
lr_mul_ = 1.0
if equalized_lr_cfg is not None:
lr_mul_ = equalized_lr_cfg.get('lr_mul', 1.0)
self.weight = nn.Parameter(torch.randn(1, out_channels, in_channels,
kernel_size, kernel_size).div_(lr_mul_))
if upsample:
factor = 2
p = len(blur_kernel) - factor - (kernel_size - 1)
pad0 = (p + 1) // 2 + factor - 1
pad1 = p // 2 + 1
self.blur = Blur(blur_kernel, (pad0, pad1), upsample_factor=factor)
if downsample:
factor = 2
p = len(blur_kernel) - factor + (kernel_size - 1)
pad0 = (p + 1) // 2
pad1 = p // 2
self.blur = Blur(blur_kernel, pad=(pad0, pad1))
if equalized_lr_cfg is not None:
equalized_lr(self, **equalized_lr_cfg)
self.padding = kernel_size // 2
def forward(self, x, style):
n, c, h, w = x.shape
style = self.style_modulation(style).view(n, 1, c, 1, 1
) + self.style_bias
weight = self.weight * style
if self.demodulate:
demod = torch.rsqrt(weight.pow(2).sum([2, 3, 4]) + self.eps)
weight = weight * demod.view(n, self.out_channels, 1, 1, 1)
weight = weight.view(n * self.out_channels, c, self.kernel_size,
self.kernel_size)
if self.upsample:
x = x.reshape(1, n * c, h, w)
weight = weight.view(n, self.out_channels, c, self.kernel_size,
self.kernel_size)
weight = weight.transpose(1, 2).reshape(n * c, self.
out_channels, self.kernel_size, self.kernel_size)
x = F.conv_transpose2d(x, weight, padding=0, stride=2, groups=n)
x = x.reshape(n, self.out_channels, *x.shape[-2:])
x = self.blur(x)
elif self.downsample:
x = self.blur(x)
x = x.view(1, n * self.in_channels, *x.shape[-2:])
x = F.conv2d(x, weight, stride=2, padding=0, groups=n)
x = x.view(n, self.out_channels, *x.shape[-2:])
else:
x = x.view(1, n * c, h, w)
x = F.conv2d(x, weight, stride=1, padding=self.padding, groups=n)
x = x.view(n, self.out_channels, *x.shape[-2:])
return x
class UpsampleUpFIRDn(nn.Module):
def __init__(self, kernel, factor=2):
super(UpsampleUpFIRDn, self).__init__()
self.factor = factor
kernel = _make_kernel(kernel) * factor ** 2
self.register_buffer('kernel', kernel)
p = kernel.shape[0] - factor
pad0 = (p + 1) // 2 + factor - 1
pad1 = p // 2
self.pad = pad0, pad1
def forward(self, x):
out = upfirdn2d(x, self.kernel, up=self.factor, down=1, pad=self.pad)
return out
class ModulatedToRGBNew(nn.Module):
def __init__(self, in_channels, style_channels, out_channels=3,
upsample=True, blur_kernel=[1, 3, 3, 1], style_mod_cfg=dict(
bias_init=1.0), style_bias=0.0):
super(ModulatedToRGBNew, self).__init__()
if upsample:
self.upsample = UpsampleUpFIRDn(blur_kernel)
self.conv = ModulatedConv2d(in_channels, out_channels=out_channels,
kernel_size=1, style_channels=style_channels, demodulate=False,
style_mod_cfg=style_mod_cfg, style_bias=style_bias)
self.bias = nn.Parameter(torch.zeros(1, 3, 1, 1))
def forward(self, input_0, input_1):
primals_6 = self.bias
primals_1 = self.conv.weight_orig
primals_5 = self.conv.style_modulation.bias
primals_3 = self.conv.style_modulation.linear.weight_orig
primals_2 = input_0
primals_4 = input_1
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6])
return output[0]
|
Juggernaut93/mmediting
|
ModulatedToRGB
| false
| 13,938
|
[
"Apache-2.0"
] | 1,884
|
8ef46ace29756dd2df1d92f2f73a33646e33e007
|
https://github.com/Juggernaut93/mmediting/tree/8ef46ace29756dd2df1d92f2f73a33646e33e007
|
PosLinear
|
import torch
from torch import Tensor
from torch.utils.data import Dataset as Dataset
import torch.nn as nn
import torch.utils.data
class PosLinear(torch.nn.Linear):
def forward(self, x: 'Tensor') ->Tensor:
gain = 1 / x.size(1)
return nn.functional.linear(x, torch.nn.functional.softplus(self.
weight), self.bias) * gain
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_features': 4, 'out_features': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch.utils.data import Dataset as Dataset
import torch.utils.data
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_softplus_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = 20.0
tmp2 = tmp0 > tmp1
tmp3 = tl_math.exp(tmp0)
tmp4 = libdevice.log1p(tmp3)
tmp5 = tl.where(tmp2, tmp0, tmp4)
tl.store(out_ptr0 + x0, tmp5, xmask)
@triton.jit
def triton_poi_fused_mul_1(in_out_ptr0, in_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_out_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = 0.25
tmp4 = tmp2 * tmp3
tl.store(in_out_ptr0 + x2, tmp4, xmask)
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4, 4), (4, 1))
assert_size_stride(primals_3, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_softplus_0[grid(16)](primals_2, buf0, 16, XBLOCK=
16, num_warps=1, num_stages=1)
buf1 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_1, (64, 4), (4, 1), 0),
reinterpret_tensor(buf0, (4, 4), (1, 4), 0), out=buf1)
del buf0
buf2 = reinterpret_tensor(buf1, (4, 4, 4, 4), (64, 16, 4, 1), 0)
del buf1
triton_poi_fused_mul_1[grid(256)](buf2, primals_3, 256, XBLOCK=256,
num_warps=4, num_stages=1)
del primals_3
return buf2, primals_2, reinterpret_tensor(primals_1, (64, 4), (4, 1), 0)
class PosLinearNew(torch.nn.Linear):
def forward(self, input_0):
primals_2 = self.weight
primals_3 = self.bias
primals_1 = input_0
output = call([primals_1, primals_2, primals_3])
return output[0]
|
KelvinKan/CP-Flow
|
PosLinear
| false
| 13,939
|
[
"MIT"
] | 64
|
d01303cb4ebeb5a0bbfca638ffaf5b7a8ec22fb1
|
https://github.com/KelvinKan/CP-Flow/tree/d01303cb4ebeb5a0bbfca638ffaf5b7a8ec22fb1
|
MeanDistLoss
|
import torch
class MeanDistLoss(torch.nn.Module):
def __init__(self, p=2):
super().__init__()
self.p = p
def forward(self, x, y):
return torch.mean(torch.cdist(x, y, p=self.p))
def extra_repr(self):
return c_f.extra_repr(self, ['p'])
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_per_fused_mean_0(in_out_ptr0, in_ptr0, xnumel, rnumel):
XBLOCK: tl.constexpr = 1
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
tl.full([1], xoffset, tl.int32)
tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
tl.full([RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp1 = tl.broadcast_to(tmp0, [RBLOCK])
tmp3 = triton_helpers.promote_to_tensor(tl.sum(tmp1, 0))
tmp4 = 256.0
tmp5 = tmp3 / tmp4
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([1], 0, tl.int32), tmp5, None)
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = torch.ops.aten._cdist_forward.default(arg1_1, arg0_1, 2.0, None)
del arg0_1
del arg1_1
buf1 = buf0
del buf0
buf2 = empty_strided_cuda((), (), torch.float32)
buf3 = buf2
del buf2
get_raw_stream(0)
triton_per_fused_mean_0[grid(1)](buf3, buf1, 1, 256, num_warps=2,
num_stages=1)
del buf1
return buf3,
class MeanDistLossNew(torch.nn.Module):
def __init__(self, p=2):
super().__init__()
self.p = p
def extra_repr(self):
return c_f.extra_repr(self, ['p'])
def forward(self, input_0, input_1):
arg0_1 = input_0
arg1_1 = input_1
output = call([arg0_1, arg1_1])
return output[0]
|
KevinMusgrave/pytorch-adapt
|
MeanDistLoss
| false
| 13,940
|
[
"MIT"
] | 131
|
ff1491e1bfcc586afb8ee619712c8816ddf10358
|
https://github.com/KevinMusgrave/pytorch-adapt/tree/ff1491e1bfcc586afb8ee619712c8816ddf10358
|
M2
|
import torch
import torch.nn as nn
import torch.nn.functional as F
class Conv2D(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, same_padding
=False, stride=1, relu=True, bn=False):
super(Conv2D, self).__init__()
padding = int((kernel_size - 1) / 2) if same_padding else 0
self.conv = nn.Conv2d(in_channels, out_channels, kernel_size,
stride, padding=padding)
self.bn = nn.BatchNorm2d(out_channels, eps=0.001, momentum=0,
affine=True) if bn else None
self.relu = nn.ReLU(inplace=True) if relu else None
def forward(self, x):
x = self.conv(x)
if self.bn is not None:
x = self.bn(x)
if self.relu is not None:
x = self.relu(x)
return x
class M2(nn.Module):
def __init__(self, in_channels):
super(M2, self).__init__()
self.m2_ssh_3x3 = Conv2D(in_channels, 256, 3, True, 1, False)
self.m2_ssh_dimred = Conv2D(in_channels, 128, 3, True, 1, True)
self.m2_ssh_5x5 = Conv2D(128, 128, 3, True, 1, False)
self.m2_ssh_7x7_1 = Conv2D(128, 128, 3, True, 1, True)
self.m2_ssh_7x7 = Conv2D(128, 128, 3, True, 1, False)
self.m2_ssh_cls_score = Conv2D(128 * 2 + 256, 4, 1, False, 1, False)
self.m2_ssh_bbox_pred = Conv2D(128 * 2 + 256, 8, 1, False, 1, False)
def forward(self, conv5_3):
m2_ssh_dimred = self.m2_ssh_dimred(conv5_3)
m2_ssh_3x3 = self.m2_ssh_3x3(conv5_3)
m2_ssh_5x5 = self.m2_ssh_5x5(m2_ssh_dimred)
m2_ssh_7x7_1 = self.m2_ssh_7x7_1(m2_ssh_dimred)
m2_ssh_7x7 = self.m2_ssh_7x7(m2_ssh_7x7_1)
m2_ssh_output = F.relu(torch.cat((m2_ssh_3x3, m2_ssh_5x5,
m2_ssh_7x7), dim=1))
m2_ssh_cls_score = self.m2_ssh_cls_score(m2_ssh_output)
m2_ssh_bbox_pred = self.m2_ssh_bbox_pred(m2_ssh_output)
return m2_ssh_cls_score, m2_ssh_bbox_pred
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_channels': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_0(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK: tl.
constexpr, XBLOCK: tl.constexpr):
ynumel = 512
xnumel = 9
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y3 = yindex
y0 = yindex % 4
y1 = yindex // 4
tmp0 = tl.load(in_ptr0 + (x2 + 9 * y3), xmask & ymask, eviction_policy=
'evict_last')
tl.store(out_ptr0 + (y0 + 4 * x2 + 36 * y1), tmp0, xmask & ymask)
@triton.jit
def triton_poi_fused_1(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK: tl.
constexpr, XBLOCK: tl.constexpr):
ynumel = 16
xnumel = 16
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y3 = yindex
y0 = yindex % 4
y1 = yindex // 4
tmp0 = tl.load(in_ptr0 + (x2 + 16 * y3), xmask & ymask)
tl.store(out_ptr0 + (y0 + 4 * x2 + 64 * y1), tmp0, xmask & ymask)
@triton.jit
def triton_poi_fused_2(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK: tl.
constexpr, XBLOCK: tl.constexpr):
xnumel = 9
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
tl.full([XBLOCK, YBLOCK], True, tl.int1)
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y3 = yindex
y0 = yindex % 4
y1 = yindex // 4
tmp0 = tl.load(in_ptr0 + (x2 + 9 * y3), xmask, eviction_policy='evict_last'
)
tl.store(out_ptr0 + (y0 + 4 * x2 + 36 * y1), tmp0, xmask)
@triton.jit
def triton_poi_fused_3(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK: tl.
constexpr, XBLOCK: tl.constexpr):
xnumel = 9
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
tl.full([XBLOCK, YBLOCK], True, tl.int1)
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y3 = yindex
y0 = yindex % 128
y1 = yindex // 128
tmp0 = tl.load(in_ptr0 + (x2 + 9 * y3), xmask, eviction_policy='evict_last'
)
tl.store(out_ptr0 + (y0 + 128 * x2 + 1152 * y1), tmp0, xmask)
@triton.jit
def triton_poi_fused_convolution_relu_4(in_out_ptr0, in_ptr0, xnumel,
XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x2 = xindex
x0 = xindex % 128
tmp0 = tl.load(in_out_ptr0 + x2, None)
tmp1 = tl.load(in_ptr0 + x0, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + x2, tmp4, None)
@triton.jit
def triton_poi_fused_cat_relu_5(in_ptr0, in_ptr1, in_ptr2, in_ptr3, in_ptr4,
in_ptr5, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x0 = xindex % 512
x1 = xindex // 512
x2 = xindex
tmp0 = x0
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 256, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (256 * x1 + x0), tmp4, eviction_policy=
'evict_last', other=0.0)
tmp6 = tl.load(in_ptr1 + x0, tmp4, eviction_policy='evict_last', other=0.0)
tmp7 = tmp5 + tmp6
tmp8 = tl.full(tmp7.shape, 0.0, tmp7.dtype)
tmp9 = tl.where(tmp4, tmp7, tmp8)
tmp10 = tmp0 >= tmp3
tmp11 = tl.full([1], 384, tl.int64)
tmp12 = tmp0 < tmp11
tmp13 = tmp10 & tmp12
tmp14 = tl.load(in_ptr2 + (128 * x1 + (-256 + x0)), tmp13,
eviction_policy='evict_last', other=0.0)
tmp15 = tl.load(in_ptr3 + (-256 + x0), tmp13, eviction_policy=
'evict_last', other=0.0)
tmp16 = tmp14 + tmp15
tmp17 = tl.full(tmp16.shape, 0.0, tmp16.dtype)
tmp18 = tl.where(tmp13, tmp16, tmp17)
tmp19 = tmp0 >= tmp11
tl.full([1], 512, tl.int64)
tmp22 = tl.load(in_ptr4 + (128 * x1 + (-384 + x0)), tmp19,
eviction_policy='evict_last', other=0.0)
tmp23 = tl.load(in_ptr5 + (-384 + x0), tmp19, eviction_policy=
'evict_last', other=0.0)
tmp24 = tmp22 + tmp23
tmp25 = tl.full(tmp24.shape, 0.0, tmp24.dtype)
tmp26 = tl.where(tmp19, tmp24, tmp25)
tmp27 = tl.where(tmp13, tmp18, tmp26)
tmp28 = tl.where(tmp4, tmp9, tmp27)
tmp29 = tl.full([1], 0, tl.int32)
tmp30 = triton_helpers.maximum(tmp29, tmp28)
tl.store(out_ptr0 + x2, tmp30, None)
@triton.jit
def triton_poi_fused_convolution_6(in_ptr0, in_ptr1, out_ptr0, ynumel,
xnumel, YBLOCK: tl.constexpr, XBLOCK: tl.constexpr):
ynumel = 16
xnumel = 16
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 4
y1 = yindex // 4
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + 4 * x2 + 64 * y1), xmask & ymask)
tmp1 = tl.load(in_ptr1 + y0, ymask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(out_ptr0 + (x2 + 16 * y3), tmp2, xmask & ymask)
@triton.jit
def triton_poi_fused_convolution_7(in_ptr0, in_ptr1, out_ptr0, ynumel,
xnumel, YBLOCK: tl.constexpr, XBLOCK: tl.constexpr):
ynumel = 32
xnumel = 16
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 8
y1 = yindex // 8
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + 8 * x2 + 128 * y1), xmask & ymask,
eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + y0, ymask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(out_ptr0 + (x2 + 16 * y3), tmp2, xmask & ymask)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7, primals_8, primals_9, primals_10, primals_11, primals_12,
primals_13, primals_14, primals_15) = args
args.clear()
assert_size_stride(primals_1, (128, 4, 3, 3), (36, 9, 3, 1))
assert_size_stride(primals_2, (128,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (256, 4, 3, 3), (36, 9, 3, 1))
assert_size_stride(primals_5, (256,), (1,))
assert_size_stride(primals_6, (128, 128, 3, 3), (1152, 9, 3, 1))
assert_size_stride(primals_7, (128,), (1,))
assert_size_stride(primals_8, (128, 128, 3, 3), (1152, 9, 3, 1))
assert_size_stride(primals_9, (128,), (1,))
assert_size_stride(primals_10, (128, 128, 3, 3), (1152, 9, 3, 1))
assert_size_stride(primals_11, (128,), (1,))
assert_size_stride(primals_12, (4, 512, 1, 1), (512, 1, 1, 1))
assert_size_stride(primals_13, (4,), (1,))
assert_size_stride(primals_14, (8, 512, 1, 1), (512, 1, 1, 1))
assert_size_stride(primals_15, (8,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((128, 4, 3, 3), (36, 1, 12, 4), torch.float32
)
get_raw_stream(0)
triton_poi_fused_0[grid(512, 9)](primals_1, buf0, 512, 9, XBLOCK=16,
YBLOCK=64, num_warps=4, num_stages=1)
del primals_1
buf1 = empty_strided_cuda((4, 4, 4, 4), (64, 1, 16, 4), torch.float32)
triton_poi_fused_1[grid(16, 16)](primals_3, buf1, 16, 16, XBLOCK=16,
YBLOCK=16, num_warps=4, num_stages=1)
del primals_3
buf2 = empty_strided_cuda((256, 4, 3, 3), (36, 1, 12, 4), torch.float32
)
triton_poi_fused_2[grid(1024, 9)](primals_4, buf2, 1024, 9, XBLOCK=
16, YBLOCK=64, num_warps=4, num_stages=1)
del primals_4
buf3 = empty_strided_cuda((128, 128, 3, 3), (1152, 1, 384, 128),
torch.float32)
triton_poi_fused_3[grid(16384, 9)](primals_6, buf3, 16384, 9,
XBLOCK=16, YBLOCK=64, num_warps=4, num_stages=1)
del primals_6
buf4 = empty_strided_cuda((128, 128, 3, 3), (1152, 1, 384, 128),
torch.float32)
triton_poi_fused_3[grid(16384, 9)](primals_8, buf4, 16384, 9,
XBLOCK=16, YBLOCK=64, num_warps=4, num_stages=1)
del primals_8
buf5 = empty_strided_cuda((128, 128, 3, 3), (1152, 1, 384, 128),
torch.float32)
triton_poi_fused_3[grid(16384, 9)](primals_10, buf5, 16384, 9,
XBLOCK=16, YBLOCK=64, num_warps=4, num_stages=1)
del primals_10
buf6 = extern_kernels.convolution(buf1, buf0, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf6, (4, 128, 4, 4), (2048, 1, 512, 128))
buf7 = buf6
del buf6
triton_poi_fused_convolution_relu_4[grid(8192)](buf7, primals_2,
8192, XBLOCK=128, num_warps=4, num_stages=1)
del primals_2
buf8 = extern_kernels.convolution(buf1, buf2, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf8, (4, 256, 4, 4), (4096, 1, 1024, 256))
buf9 = extern_kernels.convolution(buf7, buf3, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf9, (4, 128, 4, 4), (2048, 1, 512, 128))
buf10 = extern_kernels.convolution(buf7, buf4, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf10, (4, 128, 4, 4), (2048, 1, 512, 128))
buf11 = buf10
del buf10
triton_poi_fused_convolution_relu_4[grid(8192)](buf11, primals_9,
8192, XBLOCK=128, num_warps=4, num_stages=1)
del primals_9
buf12 = extern_kernels.convolution(buf11, buf5, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf12, (4, 128, 4, 4), (2048, 1, 512, 128))
buf13 = empty_strided_cuda((4, 512, 4, 4), (8192, 1, 2048, 512),
torch.float32)
triton_poi_fused_cat_relu_5[grid(32768)](buf8, primals_5, buf9,
primals_7, buf12, primals_11, buf13, 32768, XBLOCK=256,
num_warps=4, num_stages=1)
del buf12
del buf8
del buf9
del primals_11
del primals_5
del primals_7
buf14 = extern_kernels.convolution(buf13, primals_12, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf14, (4, 4, 4, 4), (64, 1, 16, 4))
buf15 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused_convolution_6[grid(16, 16)](buf14, primals_13,
buf15, 16, 16, XBLOCK=16, YBLOCK=16, num_warps=4, num_stages=1)
del buf14
del primals_13
buf16 = extern_kernels.convolution(buf13, primals_14, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf16, (4, 8, 4, 4), (128, 1, 32, 8))
buf17 = empty_strided_cuda((4, 8, 4, 4), (128, 16, 4, 1), torch.float32
)
triton_poi_fused_convolution_7[grid(32, 16)](buf16, primals_15,
buf17, 32, 16, XBLOCK=16, YBLOCK=16, num_warps=4, num_stages=1)
del buf16
del primals_15
return (buf15, buf17, buf0, buf1, buf2, buf3, buf4, buf5, primals_12,
primals_14, buf7, buf11, buf13)
class Conv2D(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, same_padding
=False, stride=1, relu=True, bn=False):
super(Conv2D, self).__init__()
padding = int((kernel_size - 1) / 2) if same_padding else 0
self.conv = nn.Conv2d(in_channels, out_channels, kernel_size,
stride, padding=padding)
self.bn = nn.BatchNorm2d(out_channels, eps=0.001, momentum=0,
affine=True) if bn else None
self.relu = nn.ReLU(inplace=True) if relu else None
def forward(self, x):
x = self.conv(x)
if self.bn is not None:
x = self.bn(x)
if self.relu is not None:
x = self.relu(x)
return x
class M2New(nn.Module):
def __init__(self, in_channels):
super(M2New, self).__init__()
self.m2_ssh_3x3 = Conv2D(in_channels, 256, 3, True, 1, False)
self.m2_ssh_dimred = Conv2D(in_channels, 128, 3, True, 1, True)
self.m2_ssh_5x5 = Conv2D(128, 128, 3, True, 1, False)
self.m2_ssh_7x7_1 = Conv2D(128, 128, 3, True, 1, True)
self.m2_ssh_7x7 = Conv2D(128, 128, 3, True, 1, False)
self.m2_ssh_cls_score = Conv2D(128 * 2 + 256, 4, 1, False, 1, False)
self.m2_ssh_bbox_pred = Conv2D(128 * 2 + 256, 8, 1, False, 1, False)
def forward(self, input_0):
primals_4 = self.m2_ssh_3x3.conv.weight
primals_5 = self.m2_ssh_3x3.conv.bias
primals_1 = self.m2_ssh_dimred.conv.weight
primals_2 = self.m2_ssh_dimred.conv.bias
primals_6 = self.m2_ssh_5x5.conv.weight
primals_7 = self.m2_ssh_5x5.conv.bias
primals_8 = self.m2_ssh_7x7_1.conv.weight
primals_9 = self.m2_ssh_7x7_1.conv.bias
primals_10 = self.m2_ssh_7x7.conv.weight
primals_11 = self.m2_ssh_7x7.conv.bias
primals_12 = self.m2_ssh_cls_score.conv.weight
primals_13 = self.m2_ssh_cls_score.conv.bias
primals_14 = self.m2_ssh_bbox_pred.conv.weight
primals_15 = self.m2_ssh_bbox_pred.conv.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7, primals_8, primals_9,
primals_10, primals_11, primals_12, primals_13, primals_14,
primals_15])
return output[0], output[1]
|
Juggernaut93/SSH-pytorch
|
M2
| false
| 13,941
|
[
"MIT"
] | 63
|
8ea205fb1a3adfc32b5a4e35f68ed4d385ddbc31
|
https://github.com/Juggernaut93/SSH-pytorch/tree/8ea205fb1a3adfc32b5a4e35f68ed4d385ddbc31
|
AbsLoss
|
import torch
class AbsLoss(torch.nn.Module):
"""
The mean absolute value.
"""
def forward(self, x):
""""""
return torch.mean(torch.abs(x))
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import math as tl_math
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_per_fused_abs_mean_0(in_out_ptr0, in_ptr0, xnumel, rnumel):
XBLOCK: tl.constexpr = 1
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
tl.full([1], xoffset, tl.int32)
tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
tl.full([RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp1 = tl_math.abs(tmp0)
tmp2 = tl.broadcast_to(tmp1, [RBLOCK])
tmp4 = triton_helpers.promote_to_tensor(tl.sum(tmp2, 0))
tmp5 = 256.0
tmp6 = tmp4 / tmp5
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([1], 0, tl.int32), tmp6, None)
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((), (), torch.float32)
buf1 = buf0
del buf0
get_raw_stream(0)
triton_per_fused_abs_mean_0[grid(1)](buf1, arg0_1, 1, 256,
num_warps=2, num_stages=1)
del arg0_1
return buf1,
class AbsLossNew(torch.nn.Module):
"""
The mean absolute value.
"""
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
|
KevinMusgrave/pytorch-adapt
|
AbsLoss
| false
| 13,942
|
[
"MIT"
] | 131
|
ff1491e1bfcc586afb8ee619712c8816ddf10358
|
https://github.com/KevinMusgrave/pytorch-adapt/tree/ff1491e1bfcc586afb8ee619712c8816ddf10358
|
M3
|
import torch
import torch.nn as nn
import torch.nn.functional as F
class Conv2D(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, same_padding
=False, stride=1, relu=True, bn=False):
super(Conv2D, self).__init__()
padding = int((kernel_size - 1) / 2) if same_padding else 0
self.conv = nn.Conv2d(in_channels, out_channels, kernel_size,
stride, padding=padding)
self.bn = nn.BatchNorm2d(out_channels, eps=0.001, momentum=0,
affine=True) if bn else None
self.relu = nn.ReLU(inplace=True) if relu else None
def forward(self, x):
x = self.conv(x)
if self.bn is not None:
x = self.bn(x)
if self.relu is not None:
x = self.relu(x)
return x
class M3(nn.Module):
def __init__(self, in_channels):
super(M3, self).__init__()
self.m3_ssh_3x3 = Conv2D(in_channels, 256, 3, True, 1, False)
self.m3_ssh_dimred = Conv2D(in_channels, 128, 3, True, 1, True)
self.m3_ssh_5x5 = Conv2D(128, 128, 3, True, 1, False)
self.m3_ssh_7x7_1 = Conv2D(128, 128, 3, True, 1, True)
self.m3_ssh_7x7 = Conv2D(128, 128, 3, True, 1, False)
self.m3_ssh_cls_score = Conv2D(128 * 2 + 256, 4, 1, False, 1, False)
self.m3_ssh_bbox_pred = Conv2D(128 * 2 + 256, 8, 1, False, 1, False)
def forward(self, pool6):
m3_ssh_3x3 = self.m3_ssh_3x3(pool6)
m3_ssh_dimred = self.m3_ssh_dimred(pool6)
m3_ssh_5x5 = self.m3_ssh_5x5(m3_ssh_dimred)
m3_ssh_7x7_1 = self.m3_ssh_7x7_1(m3_ssh_dimred)
m3_ssh_7x7 = self.m3_ssh_7x7(m3_ssh_7x7_1)
m3_ssh_output = F.relu(torch.cat((m3_ssh_3x3, m3_ssh_5x5,
m3_ssh_7x7), dim=1))
m3_ssh_cls_score = self.m3_ssh_cls_score(m3_ssh_output)
m3_ssh_bbox_pred = self.m3_ssh_bbox_pred(m3_ssh_output)
return m3_ssh_cls_score, m3_ssh_bbox_pred
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_channels': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_0(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK: tl.
constexpr, XBLOCK: tl.constexpr):
xnumel = 9
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
tl.full([XBLOCK, YBLOCK], True, tl.int1)
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y3 = yindex
y0 = yindex % 4
y1 = yindex // 4
tmp0 = tl.load(in_ptr0 + (x2 + 9 * y3), xmask, eviction_policy='evict_last'
)
tl.store(out_ptr0 + (y0 + 4 * x2 + 36 * y1), tmp0, xmask)
@triton.jit
def triton_poi_fused_1(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK: tl.
constexpr, XBLOCK: tl.constexpr):
ynumel = 16
xnumel = 16
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y3 = yindex
y0 = yindex % 4
y1 = yindex // 4
tmp0 = tl.load(in_ptr0 + (x2 + 16 * y3), xmask & ymask)
tl.store(out_ptr0 + (y0 + 4 * x2 + 64 * y1), tmp0, xmask & ymask)
@triton.jit
def triton_poi_fused_2(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK: tl.
constexpr, XBLOCK: tl.constexpr):
ynumel = 512
xnumel = 9
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y3 = yindex
y0 = yindex % 4
y1 = yindex // 4
tmp0 = tl.load(in_ptr0 + (x2 + 9 * y3), xmask & ymask, eviction_policy=
'evict_last')
tl.store(out_ptr0 + (y0 + 4 * x2 + 36 * y1), tmp0, xmask & ymask)
@triton.jit
def triton_poi_fused_3(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK: tl.
constexpr, XBLOCK: tl.constexpr):
xnumel = 9
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
tl.full([XBLOCK, YBLOCK], True, tl.int1)
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y3 = yindex
y0 = yindex % 128
y1 = yindex // 128
tmp0 = tl.load(in_ptr0 + (x2 + 9 * y3), xmask, eviction_policy='evict_last'
)
tl.store(out_ptr0 + (y0 + 128 * x2 + 1152 * y1), tmp0, xmask)
@triton.jit
def triton_poi_fused_convolution_relu_4(in_out_ptr0, in_ptr0, xnumel,
XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x2 = xindex
x0 = xindex % 128
tmp0 = tl.load(in_out_ptr0 + x2, None)
tmp1 = tl.load(in_ptr0 + x0, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + x2, tmp4, None)
@triton.jit
def triton_poi_fused_cat_relu_5(in_ptr0, in_ptr1, in_ptr2, in_ptr3, in_ptr4,
in_ptr5, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x0 = xindex % 512
x1 = xindex // 512
x2 = xindex
tmp0 = x0
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 256, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (256 * x1 + x0), tmp4, eviction_policy=
'evict_last', other=0.0)
tmp6 = tl.load(in_ptr1 + x0, tmp4, eviction_policy='evict_last', other=0.0)
tmp7 = tmp5 + tmp6
tmp8 = tl.full(tmp7.shape, 0.0, tmp7.dtype)
tmp9 = tl.where(tmp4, tmp7, tmp8)
tmp10 = tmp0 >= tmp3
tmp11 = tl.full([1], 384, tl.int64)
tmp12 = tmp0 < tmp11
tmp13 = tmp10 & tmp12
tmp14 = tl.load(in_ptr2 + (128 * x1 + (-256 + x0)), tmp13,
eviction_policy='evict_last', other=0.0)
tmp15 = tl.load(in_ptr3 + (-256 + x0), tmp13, eviction_policy=
'evict_last', other=0.0)
tmp16 = tmp14 + tmp15
tmp17 = tl.full(tmp16.shape, 0.0, tmp16.dtype)
tmp18 = tl.where(tmp13, tmp16, tmp17)
tmp19 = tmp0 >= tmp11
tl.full([1], 512, tl.int64)
tmp22 = tl.load(in_ptr4 + (128 * x1 + (-384 + x0)), tmp19,
eviction_policy='evict_last', other=0.0)
tmp23 = tl.load(in_ptr5 + (-384 + x0), tmp19, eviction_policy=
'evict_last', other=0.0)
tmp24 = tmp22 + tmp23
tmp25 = tl.full(tmp24.shape, 0.0, tmp24.dtype)
tmp26 = tl.where(tmp19, tmp24, tmp25)
tmp27 = tl.where(tmp13, tmp18, tmp26)
tmp28 = tl.where(tmp4, tmp9, tmp27)
tmp29 = tl.full([1], 0, tl.int32)
tmp30 = triton_helpers.maximum(tmp29, tmp28)
tl.store(out_ptr0 + x2, tmp30, None)
@triton.jit
def triton_poi_fused_convolution_6(in_ptr0, in_ptr1, out_ptr0, ynumel,
xnumel, YBLOCK: tl.constexpr, XBLOCK: tl.constexpr):
ynumel = 16
xnumel = 16
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 4
y1 = yindex // 4
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + 4 * x2 + 64 * y1), xmask & ymask)
tmp1 = tl.load(in_ptr1 + y0, ymask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(out_ptr0 + (x2 + 16 * y3), tmp2, xmask & ymask)
@triton.jit
def triton_poi_fused_convolution_7(in_ptr0, in_ptr1, out_ptr0, ynumel,
xnumel, YBLOCK: tl.constexpr, XBLOCK: tl.constexpr):
ynumel = 32
xnumel = 16
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 8
y1 = yindex // 8
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + 8 * x2 + 128 * y1), xmask & ymask,
eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + y0, ymask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(out_ptr0 + (x2 + 16 * y3), tmp2, xmask & ymask)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7, primals_8, primals_9, primals_10, primals_11, primals_12,
primals_13, primals_14, primals_15) = args
args.clear()
assert_size_stride(primals_1, (256, 4, 3, 3), (36, 9, 3, 1))
assert_size_stride(primals_2, (256,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (128, 4, 3, 3), (36, 9, 3, 1))
assert_size_stride(primals_5, (128,), (1,))
assert_size_stride(primals_6, (128, 128, 3, 3), (1152, 9, 3, 1))
assert_size_stride(primals_7, (128,), (1,))
assert_size_stride(primals_8, (128, 128, 3, 3), (1152, 9, 3, 1))
assert_size_stride(primals_9, (128,), (1,))
assert_size_stride(primals_10, (128, 128, 3, 3), (1152, 9, 3, 1))
assert_size_stride(primals_11, (128,), (1,))
assert_size_stride(primals_12, (4, 512, 1, 1), (512, 1, 1, 1))
assert_size_stride(primals_13, (4,), (1,))
assert_size_stride(primals_14, (8, 512, 1, 1), (512, 1, 1, 1))
assert_size_stride(primals_15, (8,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((256, 4, 3, 3), (36, 1, 12, 4), torch.float32
)
get_raw_stream(0)
triton_poi_fused_0[grid(1024, 9)](primals_1, buf0, 1024, 9, XBLOCK=
16, YBLOCK=64, num_warps=4, num_stages=1)
del primals_1
buf1 = empty_strided_cuda((4, 4, 4, 4), (64, 1, 16, 4), torch.float32)
triton_poi_fused_1[grid(16, 16)](primals_3, buf1, 16, 16, XBLOCK=16,
YBLOCK=16, num_warps=4, num_stages=1)
del primals_3
buf2 = empty_strided_cuda((128, 4, 3, 3), (36, 1, 12, 4), torch.float32
)
triton_poi_fused_2[grid(512, 9)](primals_4, buf2, 512, 9, XBLOCK=16,
YBLOCK=64, num_warps=4, num_stages=1)
del primals_4
buf3 = empty_strided_cuda((128, 128, 3, 3), (1152, 1, 384, 128),
torch.float32)
triton_poi_fused_3[grid(16384, 9)](primals_6, buf3, 16384, 9,
XBLOCK=16, YBLOCK=64, num_warps=4, num_stages=1)
del primals_6
buf4 = empty_strided_cuda((128, 128, 3, 3), (1152, 1, 384, 128),
torch.float32)
triton_poi_fused_3[grid(16384, 9)](primals_8, buf4, 16384, 9,
XBLOCK=16, YBLOCK=64, num_warps=4, num_stages=1)
del primals_8
buf5 = empty_strided_cuda((128, 128, 3, 3), (1152, 1, 384, 128),
torch.float32)
triton_poi_fused_3[grid(16384, 9)](primals_10, buf5, 16384, 9,
XBLOCK=16, YBLOCK=64, num_warps=4, num_stages=1)
del primals_10
buf6 = extern_kernels.convolution(buf1, buf0, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf6, (4, 256, 4, 4), (4096, 1, 1024, 256))
buf7 = extern_kernels.convolution(buf1, buf2, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf7, (4, 128, 4, 4), (2048, 1, 512, 128))
buf8 = buf7
del buf7
triton_poi_fused_convolution_relu_4[grid(8192)](buf8, primals_5,
8192, XBLOCK=128, num_warps=4, num_stages=1)
del primals_5
buf9 = extern_kernels.convolution(buf8, buf3, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf9, (4, 128, 4, 4), (2048, 1, 512, 128))
buf10 = extern_kernels.convolution(buf8, buf4, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf10, (4, 128, 4, 4), (2048, 1, 512, 128))
buf11 = buf10
del buf10
triton_poi_fused_convolution_relu_4[grid(8192)](buf11, primals_9,
8192, XBLOCK=128, num_warps=4, num_stages=1)
del primals_9
buf12 = extern_kernels.convolution(buf11, buf5, stride=(1, 1),
padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf12, (4, 128, 4, 4), (2048, 1, 512, 128))
buf13 = empty_strided_cuda((4, 512, 4, 4), (8192, 1, 2048, 512),
torch.float32)
triton_poi_fused_cat_relu_5[grid(32768)](buf6, primals_2, buf9,
primals_7, buf12, primals_11, buf13, 32768, XBLOCK=256,
num_warps=4, num_stages=1)
del buf12
del buf6
del buf9
del primals_11
del primals_2
del primals_7
buf14 = extern_kernels.convolution(buf13, primals_12, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf14, (4, 4, 4, 4), (64, 1, 16, 4))
buf15 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused_convolution_6[grid(16, 16)](buf14, primals_13,
buf15, 16, 16, XBLOCK=16, YBLOCK=16, num_warps=4, num_stages=1)
del buf14
del primals_13
buf16 = extern_kernels.convolution(buf13, primals_14, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf16, (4, 8, 4, 4), (128, 1, 32, 8))
buf17 = empty_strided_cuda((4, 8, 4, 4), (128, 16, 4, 1), torch.float32
)
triton_poi_fused_convolution_7[grid(32, 16)](buf16, primals_15,
buf17, 32, 16, XBLOCK=16, YBLOCK=16, num_warps=4, num_stages=1)
del buf16
del primals_15
return (buf15, buf17, buf0, buf1, buf2, buf3, buf4, buf5, primals_12,
primals_14, buf8, buf11, buf13)
class Conv2D(nn.Module):
def __init__(self, in_channels, out_channels, kernel_size, same_padding
=False, stride=1, relu=True, bn=False):
super(Conv2D, self).__init__()
padding = int((kernel_size - 1) / 2) if same_padding else 0
self.conv = nn.Conv2d(in_channels, out_channels, kernel_size,
stride, padding=padding)
self.bn = nn.BatchNorm2d(out_channels, eps=0.001, momentum=0,
affine=True) if bn else None
self.relu = nn.ReLU(inplace=True) if relu else None
def forward(self, x):
x = self.conv(x)
if self.bn is not None:
x = self.bn(x)
if self.relu is not None:
x = self.relu(x)
return x
class M3New(nn.Module):
def __init__(self, in_channels):
super(M3New, self).__init__()
self.m3_ssh_3x3 = Conv2D(in_channels, 256, 3, True, 1, False)
self.m3_ssh_dimred = Conv2D(in_channels, 128, 3, True, 1, True)
self.m3_ssh_5x5 = Conv2D(128, 128, 3, True, 1, False)
self.m3_ssh_7x7_1 = Conv2D(128, 128, 3, True, 1, True)
self.m3_ssh_7x7 = Conv2D(128, 128, 3, True, 1, False)
self.m3_ssh_cls_score = Conv2D(128 * 2 + 256, 4, 1, False, 1, False)
self.m3_ssh_bbox_pred = Conv2D(128 * 2 + 256, 8, 1, False, 1, False)
def forward(self, input_0):
primals_1 = self.m3_ssh_3x3.conv.weight
primals_2 = self.m3_ssh_3x3.conv.bias
primals_4 = self.m3_ssh_dimred.conv.weight
primals_5 = self.m3_ssh_dimred.conv.bias
primals_6 = self.m3_ssh_5x5.conv.weight
primals_7 = self.m3_ssh_5x5.conv.bias
primals_8 = self.m3_ssh_7x7_1.conv.weight
primals_9 = self.m3_ssh_7x7_1.conv.bias
primals_10 = self.m3_ssh_7x7.conv.weight
primals_11 = self.m3_ssh_7x7.conv.bias
primals_12 = self.m3_ssh_cls_score.conv.weight
primals_13 = self.m3_ssh_cls_score.conv.bias
primals_14 = self.m3_ssh_bbox_pred.conv.weight
primals_15 = self.m3_ssh_bbox_pred.conv.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7, primals_8, primals_9,
primals_10, primals_11, primals_12, primals_13, primals_14,
primals_15])
return output[0], output[1]
|
Juggernaut93/SSH-pytorch
|
M3
| false
| 13,943
|
[
"MIT"
] | 63
|
8ea205fb1a3adfc32b5a4e35f68ed4d385ddbc31
|
https://github.com/Juggernaut93/SSH-pytorch/tree/8ea205fb1a3adfc32b5a4e35f68ed4d385ddbc31
|
AdaptiveFeatureNorm
|
import torch
class AdaptiveFeatureNorm(torch.nn.Module):
"""
Implementation of the loss in
[Larger Norm More Transferable:
An Adaptive Feature Norm Approach for
Unsupervised Domain Adaptation](https://arxiv.org/abs/1811.07456).
Encourages features to gradually have larger and larger L2 norms.
"""
def __init__(self, step_size: 'float'=1):
"""
Arguments:
step_size: The desired increase in L2 norm at each iteration.
Note that the loss will always be equal to ```step_size```
because the goal is always to make the L2 norm ```step_size```
larger than whatever the current L2 norm is.
"""
super().__init__()
self.step_size = step_size
def forward(self, x):
""""""
l2_norm = x.norm(p=2, dim=1)
radius = l2_norm.detach() + self.step_size
return torch.mean((l2_norm - radius) ** 2)
def extra_repr(self):
""""""
return c_f.extra_repr(self, ['step_size'])
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_per_fused_add_linalg_vector_norm_mean_pow_sub_0(in_out_ptr0,
in_ptr0, xnumel, rnumel, XBLOCK: tl.constexpr):
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex % 16
r1 = rindex // 16
tmp0 = tl.load(in_ptr0 + (r0 + 64 * r1), None)
tmp2 = tl.load(in_ptr0 + (16 + r0 + 64 * r1), None)
tmp5 = tl.load(in_ptr0 + (32 + r0 + 64 * r1), None)
tmp8 = tl.load(in_ptr0 + (48 + r0 + 64 * r1), None)
tmp1 = tmp0 * tmp0
tmp3 = tmp2 * tmp2
tmp4 = tmp1 + tmp3
tmp6 = tmp5 * tmp5
tmp7 = tmp4 + tmp6
tmp9 = tmp8 * tmp8
tmp10 = tmp7 + tmp9
tmp11 = libdevice.sqrt(tmp10)
tmp12 = 1.0
tmp13 = tmp11 + tmp12
tmp14 = tmp11 - tmp13
tmp15 = tmp14 * tmp14
tmp16 = tl.broadcast_to(tmp15, [XBLOCK, RBLOCK])
tmp18 = tl.sum(tmp16, 1)[:, None]
tmp19 = 64.0
tmp20 = tmp18 / tmp19
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([XBLOCK, 1], 0, tl.int32), tmp20, None)
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((), (), torch.float32)
buf1 = buf0
del buf0
get_raw_stream(0)
triton_per_fused_add_linalg_vector_norm_mean_pow_sub_0[grid(1)](buf1,
arg0_1, 1, 64, XBLOCK=1, num_warps=2, num_stages=1)
del arg0_1
return buf1,
class AdaptiveFeatureNormNew(torch.nn.Module):
"""
Implementation of the loss in
[Larger Norm More Transferable:
An Adaptive Feature Norm Approach for
Unsupervised Domain Adaptation](https://arxiv.org/abs/1811.07456).
Encourages features to gradually have larger and larger L2 norms.
"""
def __init__(self, step_size: 'float'=1):
"""
Arguments:
step_size: The desired increase in L2 norm at each iteration.
Note that the loss will always be equal to ```step_size```
because the goal is always to make the L2 norm ```step_size```
larger than whatever the current L2 norm is.
"""
super().__init__()
self.step_size = step_size
def extra_repr(self):
""""""
return c_f.extra_repr(self, ['step_size'])
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
|
KevinMusgrave/pytorch-adapt
|
AdaptiveFeatureNorm
| false
| 13,944
|
[
"MIT"
] | 131
|
ff1491e1bfcc586afb8ee619712c8816ddf10358
|
https://github.com/KevinMusgrave/pytorch-adapt/tree/ff1491e1bfcc586afb8ee619712c8816ddf10358
|
UniformDistributionLoss
|
import torch
import torch.nn.functional as F
class UniformDistributionLoss(torch.nn.Module):
"""
Implementation of the confusion loss from
[Simultaneous Deep Transfer Across Domains and Tasks](https://arxiv.org/abs/1510.02192).
"""
def forward(self, x, *args):
""""""
probs = F.log_softmax(x, dim=1)
avg_probs = torch.mean(probs, dim=1)
return -torch.mean(avg_probs)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import math as tl_math
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused__log_softmax_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.
constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x0 = xindex % 16
x2 = xindex // 64
tmp0 = tl.load(in_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + (x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp2 = tl.load(in_ptr0 + (16 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp4 = tl.load(in_ptr0 + (32 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp6 = tl.load(in_ptr0 + (48 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp3 = triton_helpers.maximum(tmp1, tmp2)
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp0 - tmp7
tl.store(out_ptr0 + x3, tmp8, xmask)
@triton.jit
def triton_per_fused__log_softmax_mean_neg_1(in_out_ptr0, in_ptr0, xnumel,
rnumel, XBLOCK: tl.constexpr):
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex % 16
r1 = rindex // 16
tmp0 = tl.load(in_ptr0 + (r0 + 64 * r1), None)
tmp2 = tl.load(in_ptr0 + (16 + r0 + 64 * r1), None)
tmp5 = tl.load(in_ptr0 + (32 + r0 + 64 * r1), None)
tmp8 = tl.load(in_ptr0 + (48 + r0 + 64 * r1), None)
tmp1 = tl_math.exp(tmp0)
tmp3 = tl_math.exp(tmp2)
tmp4 = tmp1 + tmp3
tmp6 = tl_math.exp(tmp5)
tmp7 = tmp4 + tmp6
tmp9 = tl_math.exp(tmp8)
tmp10 = tmp7 + tmp9
tmp11 = tl_math.log(tmp10)
tmp12 = tmp0 - tmp11
tmp13 = tmp2 - tmp11
tmp14 = tmp12 + tmp13
tmp15 = tmp5 - tmp11
tmp16 = tmp14 + tmp15
tmp17 = tmp8 - tmp11
tmp18 = tmp16 + tmp17
tmp19 = 4.0
tmp20 = tmp18 / tmp19
tmp21 = tl.broadcast_to(tmp20, [XBLOCK, RBLOCK])
tmp23 = tl.sum(tmp21, 1)[:, None]
tmp24 = 64.0
tmp25 = tmp23 / tmp24
tmp26 = -tmp25
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([XBLOCK, 1], 0, tl.int32), tmp26, None)
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused__log_softmax_0[grid(256)](arg0_1, buf0, 256,
XBLOCK=128, num_warps=4, num_stages=1)
del arg0_1
buf1 = empty_strided_cuda((), (), torch.float32)
buf2 = buf1
del buf1
triton_per_fused__log_softmax_mean_neg_1[grid(1)](buf2, buf0, 1, 64,
XBLOCK=1, num_warps=2, num_stages=1)
del buf0
return buf2,
class UniformDistributionLossNew(torch.nn.Module):
"""
Implementation of the confusion loss from
[Simultaneous Deep Transfer Across Domains and Tasks](https://arxiv.org/abs/1510.02192).
"""
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
|
KevinMusgrave/pytorch-adapt
|
UniformDistributionLoss
| false
| 13,945
|
[
"MIT"
] | 131
|
ff1491e1bfcc586afb8ee619712c8816ddf10358
|
https://github.com/KevinMusgrave/pytorch-adapt/tree/ff1491e1bfcc586afb8ee619712c8816ddf10358
|
BatchSpectralLoss
|
import torch
def batch_spectral_loss(x, k):
singular_values = torch.linalg.svdvals(x)
return torch.sum(singular_values[:k] ** 2)
class BatchSpectralLoss(torch.nn.Module):
"""
Implementation of the loss in
[Transferability vs. Discriminability: Batch Spectral
Penalization for Adversarial Domain Adaptation](http://proceedings.mlr.press/v97/chen19i.html).
The loss is the sum of the squares of the first k singular values.
"""
def __init__(self, k: 'int'=1):
"""
Arguments:
k: the number of singular values to include in the loss
"""
super().__init__()
self.k = k
def forward(self, x):
""""""
return batch_spectral_loss(x, self.k)
def extra_repr(self):
""""""
return c_f.extra_repr(self, ['k'])
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_per_fused_pow_sum_0(in_ptr0, out_ptr0, xnumel, rnumel, XBLOCK:
tl.constexpr):
RBLOCK: tl.constexpr = 16
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp1 = tmp0 * tmp0
tmp2 = tl.broadcast_to(tmp1, [XBLOCK, RBLOCK])
tmp4 = tl.sum(tmp2, 1)[:, None]
tl.store(out_ptr0 + tl.full([XBLOCK, 1], 0, tl.int32), tmp4, None)
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = torch.ops.aten._linalg_svd.default(arg0_1)
del arg0_1
buf2 = buf0[1]
del buf0
buf4 = empty_strided_cuda((), (), torch.float32)
get_raw_stream(0)
triton_per_fused_pow_sum_0[grid(1)](buf2, buf4, 1, 16, XBLOCK=1,
num_warps=2, num_stages=1)
del buf2
return buf4,
def batch_spectral_loss(x, k):
singular_values = torch.linalg.svdvals(x)
return torch.sum(singular_values[:k] ** 2)
class BatchSpectralLossNew(torch.nn.Module):
"""
Implementation of the loss in
[Transferability vs. Discriminability: Batch Spectral
Penalization for Adversarial Domain Adaptation](http://proceedings.mlr.press/v97/chen19i.html).
The loss is the sum of the squares of the first k singular values.
"""
def __init__(self, k: 'int'=1):
"""
Arguments:
k: the number of singular values to include in the loss
"""
super().__init__()
self.k = k
def extra_repr(self):
""""""
return c_f.extra_repr(self, ['k'])
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
|
KevinMusgrave/pytorch-adapt
|
BatchSpectralLoss
| false
| 13,946
|
[
"MIT"
] | 131
|
ff1491e1bfcc586afb8ee619712c8816ddf10358
|
https://github.com/KevinMusgrave/pytorch-adapt/tree/ff1491e1bfcc586afb8ee619712c8816ddf10358
|
CORALLoss
|
import torch
def covariance(x):
batch_size = x.shape[0]
mm1 = torch.mm(x.t(), x)
cols_summed = torch.sum(x, dim=0)
mm2 = torch.mm(cols_summed.unsqueeze(1), cols_summed.unsqueeze(0))
return 1.0 / (batch_size - 1) * (mm1 - 1.0 / batch_size * mm2)
class CORALLoss(torch.nn.Module):
"""
Implementation of [Deep CORAL:
Correlation Alignment for
Deep Domain Adaptation](https://arxiv.org/abs/1607.01719)
"""
def forward(self, x: 'torch.Tensor', y: 'torch.Tensor'):
"""
Arguments:
x: features from one domain
y: features from the other domain
"""
embedding_size = x.shape[1]
cx = covariance(x)
cy = covariance(y)
squared_fro_norm = torch.linalg.norm(cx - cy, ord='fro') ** 2
return squared_fro_norm / (4 * embedding_size ** 2)
def get_inputs():
return [torch.rand([4, 4]), torch.rand([4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_sum_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = tl.load(in_ptr0 + (4 + x0), xmask)
tmp3 = tl.load(in_ptr0 + (8 + x0), xmask)
tmp5 = tl.load(in_ptr0 + (12 + x0), xmask)
tmp2 = tmp0 + tmp1
tmp4 = tmp2 + tmp3
tmp6 = tmp4 + tmp5
tl.store(out_ptr0 + x0, tmp6, xmask)
@triton.jit
def triton_per_fused_div_linalg_vector_norm_mul_pow_sub_1(in_out_ptr0,
in_ptr0, in_ptr1, in_ptr2, in_ptr3, xnumel, rnumel, XBLOCK: tl.constexpr):
RBLOCK: tl.constexpr = 16
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp1 = tl.load(in_ptr1 + r0, None)
tmp7 = tl.load(in_ptr2 + r0, None)
tmp8 = tl.load(in_ptr3 + r0, None)
tmp2 = 0.25
tmp3 = tmp1 * tmp2
tmp4 = tmp0 - tmp3
tmp5 = 0.3333333333333333
tmp6 = tmp4 * tmp5
tmp9 = tmp8 * tmp2
tmp10 = tmp7 - tmp9
tmp11 = tmp10 * tmp5
tmp12 = tmp6 - tmp11
tmp13 = tmp12 * tmp12
tmp14 = tl.broadcast_to(tmp13, [XBLOCK, RBLOCK])
tmp16 = tl.sum(tmp14, 1)[:, None]
tmp17 = libdevice.sqrt(tmp16)
tmp18 = tmp17 * tmp17
tmp19 = 0.015625
tmp20 = tmp18 * tmp19
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([XBLOCK, 1], 0, tl.int32), tmp20, None)
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4), (4, 1))
assert_size_stride(arg1_1, (4, 4), (4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(arg0_1, (4, 4), (1, 4), 0),
arg0_1, out=buf0)
buf1 = empty_strided_cuda((4,), (1,), torch.float32)
get_raw_stream(0)
triton_poi_fused_sum_0[grid(4)](arg0_1, buf1, 4, XBLOCK=4,
num_warps=1, num_stages=1)
del arg0_1
buf2 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf1, (4, 1), (1, 0), 0),
reinterpret_tensor(buf1, (1, 4), (0, 1), 0), out=buf2)
buf3 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(arg1_1, (4, 4), (1, 4), 0),
arg1_1, out=buf3)
buf4 = buf1
del buf1
triton_poi_fused_sum_0[grid(4)](arg1_1, buf4, 4, XBLOCK=4,
num_warps=1, num_stages=1)
del arg1_1
buf5 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf4, (4, 1), (1, 0), 0),
reinterpret_tensor(buf4, (1, 4), (0, 1), 0), out=buf5)
del buf4
buf6 = empty_strided_cuda((), (), torch.float32)
buf7 = buf6
del buf6
triton_per_fused_div_linalg_vector_norm_mul_pow_sub_1[grid(1)](buf7,
buf0, buf2, buf3, buf5, 1, 16, XBLOCK=1, num_warps=2, num_stages=1)
del buf0
del buf2
del buf3
del buf5
return buf7,
def covariance(x):
batch_size = x.shape[0]
mm1 = torch.mm(x.t(), x)
cols_summed = torch.sum(x, dim=0)
mm2 = torch.mm(cols_summed.unsqueeze(1), cols_summed.unsqueeze(0))
return 1.0 / (batch_size - 1) * (mm1 - 1.0 / batch_size * mm2)
class CORALLossNew(torch.nn.Module):
"""
Implementation of [Deep CORAL:
Correlation Alignment for
Deep Domain Adaptation](https://arxiv.org/abs/1607.01719)
"""
def forward(self, input_0, input_1):
arg0_1 = input_0
arg1_1 = input_1
output = call([arg0_1, arg1_1])
return output[0]
|
KevinMusgrave/pytorch-adapt
|
CORALLoss
| false
| 13,947
|
[
"MIT"
] | 131
|
ff1491e1bfcc586afb8ee619712c8816ddf10358
|
https://github.com/KevinMusgrave/pytorch-adapt/tree/ff1491e1bfcc586afb8ee619712c8816ddf10358
|
SumNormalizer
|
import torch
def sum_normalizer(x, detach=False, scale_by_batch_size=False):
y = torch.sum(x)
if detach:
y = y.detach()
if scale_by_batch_size:
x = x * x.shape[0]
return x / y
class SumNormalizer(torch.nn.Module):
def __init__(self, detach=False, scale_by_batch_size=False):
super().__init__()
self.detach = detach
self.scale_by_batch_size = scale_by_batch_size
def forward(self, x):
return sum_normalizer(x, detach=self.detach, scale_by_batch_size=
self.scale_by_batch_size)
def extra_repr(self):
return c_f.extra_repr(self, ['detach', 'scale_by_batch_size'])
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_per_fused_div_sum_0(in_ptr0, out_ptr1, xnumel, rnumel):
XBLOCK: tl.constexpr = 1
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
tl.full([1], xoffset, tl.int32)
tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
tl.full([RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp1 = tl.broadcast_to(tmp0, [RBLOCK])
tmp3 = triton_helpers.promote_to_tensor(tl.sum(tmp1, 0))
tmp4 = tmp0 / tmp3
tl.store(out_ptr1 + tl.broadcast_to(r0, [RBLOCK]), tmp4, None)
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf1 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_per_fused_div_sum_0[grid(1)](arg0_1, buf1, 1, 256, num_warps
=2, num_stages=1)
del arg0_1
return buf1,
def sum_normalizer(x, detach=False, scale_by_batch_size=False):
y = torch.sum(x)
if detach:
y = y.detach()
if scale_by_batch_size:
x = x * x.shape[0]
return x / y
class SumNormalizerNew(torch.nn.Module):
def __init__(self, detach=False, scale_by_batch_size=False):
super().__init__()
self.detach = detach
self.scale_by_batch_size = scale_by_batch_size
def extra_repr(self):
return c_f.extra_repr(self, ['detach', 'scale_by_batch_size'])
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
|
KevinMusgrave/pytorch-adapt
|
SumNormalizer
| false
| 13,948
|
[
"MIT"
] | 131
|
ff1491e1bfcc586afb8ee619712c8816ddf10358
|
https://github.com/KevinMusgrave/pytorch-adapt/tree/ff1491e1bfcc586afb8ee619712c8816ddf10358
|
Encoder
|
import torch
from torch import nn
import torch.hub
import torch.nn.functional as F
class Encoder(nn.Module):
"""Estimation of the nonnegative mixture weight by a 1-D conv layer.
"""
def __init__(self, L, N, audio_channels):
super(Encoder, self).__init__()
self.L, self.N = L, N
self.conv1d_U = nn.Conv1d(audio_channels, N, kernel_size=L, stride=
L // 2, bias=False)
def forward(self, mixture):
"""
Args:
mixture: [M, T], M is batch size, T is #samples
Returns:
mixture_w: [M, N, K], where K = (T-L)/(L/2)+1 = 2T/L-1
"""
mixture_w = F.relu(self.conv1d_U(mixture))
return mixture_w
def get_inputs():
return [torch.rand([4, 4])]
def get_init_inputs():
return [[], {'L': 4, 'N': 4, 'audio_channels': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch import nn
import torch.hub
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_relu_threshold_backward_0(in_out_ptr0, out_ptr0,
xnumel, XBLOCK: tl.constexpr):
xnumel = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_out_ptr0 + x0, xmask)
tmp1 = tl.full([1], 0, tl.int32)
tmp2 = triton_helpers.maximum(tmp1, tmp0)
tmp3 = 0.0
tmp4 = tmp2 <= tmp3
tl.store(in_out_ptr0 + x0, tmp2, xmask)
tl.store(out_ptr0 + x0, tmp4, xmask)
def call(args):
primals_1, primals_2 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_2, (4, 4), (4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(reinterpret_tensor(primals_2, (1,
4, 4), (16, 4, 1), 0), primals_1, stride=(2,), padding=(0,),
dilation=(1,), transposed=False, output_padding=(0,), groups=1,
bias=None)
assert_size_stride(buf0, (1, 4, 1), (4, 1, 1))
buf1 = reinterpret_tensor(buf0, (4, 1), (1, 1), 0)
del buf0
buf2 = empty_strided_cuda((4, 1), (1, 1), torch.bool)
get_raw_stream(0)
triton_poi_fused_relu_threshold_backward_0[grid(4)](buf1, buf2, 4,
XBLOCK=4, num_warps=1, num_stages=1)
return buf1, primals_1, reinterpret_tensor(primals_2, (1, 4, 4), (16, 4,
1), 0), buf2
class EncoderNew(nn.Module):
"""Estimation of the nonnegative mixture weight by a 1-D conv layer.
"""
def __init__(self, L, N, audio_channels):
super(EncoderNew, self).__init__()
self.L, self.N = L, N
self.conv1d_U = nn.Conv1d(audio_channels, N, kernel_size=L, stride=
L // 2, bias=False)
def forward(self, input_0):
primals_1 = self.conv1d_U.weight
primals_2 = input_0
output = call([primals_1, primals_2])
return output[0]
|
KilianRuiz2B/demucs
|
Encoder
| false
| 13,949
|
[
"MIT"
] | 3,013
|
a6fbf3806b018634f68563887feaee64c5e36600
|
https://github.com/KilianRuiz2B/demucs/tree/a6fbf3806b018634f68563887feaee64c5e36600
|
BNMLoss
|
import torch
class BNMLoss(torch.nn.Module):
"""
Implementation of the loss in
[Towards Discriminability and Diversity:
Batch Nuclear-norm Maximization
under Label Insufficient Situations](https://arxiv.org/abs/2003.12237).
"""
def forward(self, x):
""""""
x = torch.nn.functional.softmax(x, dim=1)
return -torch.linalg.norm(x, 'nuc') / x.shape[0]
def get_inputs():
return [torch.rand([4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import math as tl_math
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused__softmax_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + 4 * x1, xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + 4 * x1), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (2 + 4 * x1), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (3 + 4 * x1), xmask, eviction_policy='evict_last')
tmp3 = triton_helpers.maximum(tmp1, tmp2)
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp0 - tmp7
tmp9 = tl_math.exp(tmp8)
tl.store(out_ptr0 + x2, tmp9, xmask)
@triton.jit
def triton_poi_fused__softmax_1(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + 4 * x1, xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + 4 * x1), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (2 + 4 * x1), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (3 + 4 * x1), xmask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp5 = tmp3 + tmp4
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tl.store(out_ptr0 + x2, tmp8, xmask)
@triton.jit
def triton_per_fused_div_neg_sum_2(in_out_ptr0, in_ptr0, xnumel, rnumel,
XBLOCK: tl.constexpr):
RBLOCK: tl.constexpr = 4
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp1 = tl.broadcast_to(tmp0, [XBLOCK, RBLOCK])
tmp3 = tl.sum(tmp1, 1)[:, None]
tmp4 = -tmp3
tmp5 = 0.25
tmp6 = tmp4 * tmp5
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([XBLOCK, 1], 0, tl.int32), tmp6, None)
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4), (4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused__softmax_0[grid(16)](arg0_1, buf0, 16, XBLOCK=16,
num_warps=1, num_stages=1)
del arg0_1
buf1 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
triton_poi_fused__softmax_1[grid(16)](buf0, buf1, 16, XBLOCK=16,
num_warps=1, num_stages=1)
del buf0
buf2 = torch.ops.aten._linalg_svd.default(buf1)
del buf1
buf4 = buf2[1]
del buf2
buf6 = empty_strided_cuda((), (), torch.float32)
buf7 = buf6
del buf6
triton_per_fused_div_neg_sum_2[grid(1)](buf7, buf4, 1, 4, XBLOCK=1,
num_warps=2, num_stages=1)
del buf4
return buf7,
class BNMLossNew(torch.nn.Module):
"""
Implementation of the loss in
[Towards Discriminability and Diversity:
Batch Nuclear-norm Maximization
under Label Insufficient Situations](https://arxiv.org/abs/2003.12237).
"""
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
|
KevinMusgrave/pytorch-adapt
|
BNMLoss
| false
| 13,950
|
[
"MIT"
] | 131
|
ff1491e1bfcc586afb8ee619712c8816ddf10358
|
https://github.com/KevinMusgrave/pytorch-adapt/tree/ff1491e1bfcc586afb8ee619712c8816ddf10358
|
MinMaxNormalizer
|
import torch
def min_max_normalizer(x, detach=False):
x_min = torch.min(x)
x_max = torch.max(x)
if detach:
x_min = x_min.detach()
x_max = x_max.detach()
return (x - x_min) / (x_max - x_min)
class MinMaxNormalizer(torch.nn.Module):
def __init__(self, detach=False):
super().__init__()
self.detach = detach
def forward(self, x):
return min_max_normalizer(x, detach=self.detach)
def extra_repr(self):
return c_f.extra_repr(self, ['detach'])
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_per_fused_div_max_min_sub_0(in_ptr0, out_ptr2, xnumel, rnumel):
XBLOCK: tl.constexpr = 1
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
tl.full([1], xoffset, tl.int32)
tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
tl.full([RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp1 = tl.broadcast_to(tmp0, [RBLOCK])
tmp3 = triton_helpers.promote_to_tensor(triton_helpers.min2(tmp1, 0))
tmp5 = triton_helpers.promote_to_tensor(triton_helpers.max2(tmp1, 0))
tmp6 = tmp0 - tmp3
tmp7 = tmp5 - tmp3
tmp8 = tmp6 / tmp7
tl.store(out_ptr2 + tl.broadcast_to(r0, [RBLOCK]), tmp8, None)
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf2 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_per_fused_div_max_min_sub_0[grid(1)](arg0_1, buf2, 1, 256,
num_warps=2, num_stages=1)
del arg0_1
return buf2,
def min_max_normalizer(x, detach=False):
x_min = torch.min(x)
x_max = torch.max(x)
if detach:
x_min = x_min.detach()
x_max = x_max.detach()
return (x - x_min) / (x_max - x_min)
class MinMaxNormalizerNew(torch.nn.Module):
def __init__(self, detach=False):
super().__init__()
self.detach = detach
def extra_repr(self):
return c_f.extra_repr(self, ['detach'])
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
|
KevinMusgrave/pytorch-adapt
|
MinMaxNormalizer
| false
| 13,951
|
[
"MIT"
] | 131
|
ff1491e1bfcc586afb8ee619712c8816ddf10358
|
https://github.com/KevinMusgrave/pytorch-adapt/tree/ff1491e1bfcc586afb8ee619712c8816ddf10358
|
SineLayer
|
import torch
import numpy as np
import torch.nn as nn
class SineLayer(nn.Module):
def __init__(self, in_features, out_features, bias=True, is_first=False,
omega_0=30):
super().__init__()
self.omega_0 = omega_0
self.is_first = is_first
self.in_features = in_features
self.linear = nn.Linear(in_features, out_features, bias=bias)
self.init_weights()
def init_weights(self):
with torch.no_grad():
if self.is_first:
self.linear.weight.uniform_(-1 / self.in_features, 1 / self
.in_features)
else:
self.linear.weight.uniform_(-np.sqrt(6 / self.in_features) /
self.omega_0, np.sqrt(6 / self.in_features) / self.omega_0)
def forward(self, input):
return torch.sin(self.omega_0 * self.linear(input))
def forward_with_intermediate(self, input):
intermediate = self.omega_0 * self.linear(input)
return torch.sin(intermediate), intermediate
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_features': 4, 'out_features': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import math as tl_math
import numpy as np
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_mul_sin_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = 30.0
tmp2 = tmp0 * tmp1
tmp3 = tl_math.sin(tmp2)
tl.store(out_ptr0 + x0, tmp3, xmask)
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_2, reinterpret_tensor(primals_3, (64,
4), (4, 1), 0), reinterpret_tensor(primals_1, (4, 4), (1, 4), 0
), alpha=1, beta=1, out=buf0)
del primals_1
del primals_2
buf1 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_mul_sin_0[grid(256)](buf0, buf1, 256, XBLOCK=256,
num_warps=4, num_stages=1)
return buf1, reinterpret_tensor(primals_3, (64, 4), (4, 1), 0), buf0
class SineLayerNew(nn.Module):
def __init__(self, in_features, out_features, bias=True, is_first=False,
omega_0=30):
super().__init__()
self.omega_0 = omega_0
self.is_first = is_first
self.in_features = in_features
self.linear = nn.Linear(in_features, out_features, bias=bias)
self.init_weights()
def init_weights(self):
with torch.no_grad():
if self.is_first:
self.linear.weight.uniform_(-1 / self.in_features, 1 / self
.in_features)
else:
self.linear.weight.uniform_(-np.sqrt(6 / self.in_features) /
self.omega_0, np.sqrt(6 / self.in_features) / self.omega_0)
def forward_with_intermediate(self, input):
intermediate = self.omega_0 * self.linear(input)
return torch.sin(intermediate), intermediate
def forward(self, input_0):
primals_1 = self.linear.weight
primals_2 = self.linear.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3])
return output[0]
|
Juju-botu/diffeqml-research
|
SineLayer
| false
| 13,952
|
[
"Apache-2.0"
] | 49
|
aa796c87447e5299ec4f25a07fc4d032afb1f63e
|
https://github.com/Juju-botu/diffeqml-research/tree/aa796c87447e5299ec4f25a07fc4d032afb1f63e
|
LRN
|
import torch
import torch.nn as nn
class LRN(nn.Module):
def __init__(self, local_size=1, alpha=0.0001, beta=0.75,
ACROSS_CHANNELS=False):
super(LRN, self).__init__()
self.ACROSS_CHANNELS = ACROSS_CHANNELS
if self.ACROSS_CHANNELS:
self.average = nn.AvgPool3d(kernel_size=(local_size, 1, 1),
stride=1, padding=(int((local_size - 1.0) / 2), 0, 0))
else:
self.average = nn.AvgPool2d(kernel_size=local_size, stride=1,
padding=int((local_size - 1.0) / 2))
self.alpha = alpha
self.beta = beta
def forward(self, x):
if self.ACROSS_CHANNELS:
div = x.pow(2).unsqueeze(1)
div = self.average(div).squeeze(1)
div = div.mul(self.alpha).add(2.0).pow(self.beta)
else:
div = x.pow(2)
div = self.average(div)
div = div.mul(self.alpha).add(2.0).pow(self.beta)
x = x.div(div)
return x
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_add_avg_pool2d_div_mul_pow_0(in_ptr0, out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = tmp0 * tmp0
tmp2 = 1.0
tmp3 = tmp1 * tmp2
tmp4 = 0.0001
tmp5 = tmp3 * tmp4
tmp6 = 2.0
tmp7 = tmp5 + tmp6
tmp8 = 0.75
tmp9 = libdevice.pow(tmp7, tmp8)
tmp10 = tmp0 / tmp9
tl.store(out_ptr0 + x0, tmp10, xmask)
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_add_avg_pool2d_div_mul_pow_0[grid(256)](arg0_1,
buf0, 256, XBLOCK=256, num_warps=4, num_stages=1)
del arg0_1
return buf0,
class LRNNew(nn.Module):
def __init__(self, local_size=1, alpha=0.0001, beta=0.75,
ACROSS_CHANNELS=False):
super(LRNNew, self).__init__()
self.ACROSS_CHANNELS = ACROSS_CHANNELS
if self.ACROSS_CHANNELS:
self.average = nn.AvgPool3d(kernel_size=(local_size, 1, 1),
stride=1, padding=(int((local_size - 1.0) / 2), 0, 0))
else:
self.average = nn.AvgPool2d(kernel_size=local_size, stride=1,
padding=int((local_size - 1.0) / 2))
self.alpha = alpha
self.beta = beta
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
|
Kitware/VAIME
|
LRN
| false
| 13,953
|
[
"BSD-3-Clause"
] | 127
|
47b24b9d8a208cf8c621e5bb1088c61fcf507af6
|
https://github.com/Kitware/VAIME/tree/47b24b9d8a208cf8c621e5bb1088c61fcf507af6
|
SDFNetwork
|
import torch
import numpy as np
import torch.nn as nn
def get_embedder(multires, input_dims=3):
embed_kwargs = {'include_input': True, 'input_dims': input_dims,
'max_freq_log2': multires - 1, 'num_freqs': multires,
'log_sampling': True, 'periodic_fns': [torch.sin, torch.cos]}
embedder_obj = Embedder(**embed_kwargs)
def embed(x, eo=embedder_obj):
return eo.embed(x)
return embed, embedder_obj.out_dim
class Embedder:
def __init__(self, **kwargs):
self.kwargs = kwargs
self.create_embedding_fn()
def create_embedding_fn(self):
embed_fns = []
d = self.kwargs['input_dims']
out_dim = 0
if self.kwargs['include_input']:
embed_fns.append(lambda x: x)
out_dim += d
max_freq = self.kwargs['max_freq_log2']
N_freqs = self.kwargs['num_freqs']
if self.kwargs['log_sampling']:
freq_bands = 2.0 ** torch.linspace(0.0, max_freq, N_freqs)
else:
freq_bands = torch.linspace(2.0 ** 0.0, 2.0 ** max_freq, N_freqs)
for freq in freq_bands:
for p_fn in self.kwargs['periodic_fns']:
embed_fns.append(lambda x, p_fn=p_fn, freq=freq: p_fn(x * freq)
)
out_dim += d
self.embed_fns = embed_fns
self.out_dim = out_dim
def embed(self, inputs):
return torch.cat([fn(inputs) for fn in self.embed_fns], -1)
class SDFNetwork(nn.Module):
def __init__(self, d_in, d_out, d_hidden, n_layers, skip_in=(4,),
multires=0, bias=0.5, scale=1, geometric_init=True, weight_norm=
True, inside_outside=False):
super(SDFNetwork, self).__init__()
dims = [d_in] + [d_hidden for _ in range(n_layers)] + [d_out]
self.embed_fn_fine = None
if multires > 0:
embed_fn, input_ch = get_embedder(multires, input_dims=d_in)
self.embed_fn_fine = embed_fn
dims[0] = input_ch
self.num_layers = len(dims)
self.skip_in = skip_in
self.scale = scale
for l in range(0, self.num_layers - 1):
if l + 1 in self.skip_in:
out_dim = dims[l + 1] - dims[0]
else:
out_dim = dims[l + 1]
lin = nn.Linear(dims[l], out_dim)
if geometric_init:
if l == self.num_layers - 2:
if not inside_outside:
torch.nn.init.normal_(lin.weight, mean=np.sqrt(np.
pi) / np.sqrt(dims[l]), std=0.0001)
torch.nn.init.constant_(lin.bias, -bias)
else:
torch.nn.init.normal_(lin.weight, mean=-np.sqrt(np.
pi) / np.sqrt(dims[l]), std=0.0001)
torch.nn.init.constant_(lin.bias, bias)
elif multires > 0 and l == 0:
torch.nn.init.constant_(lin.bias, 0.0)
torch.nn.init.constant_(lin.weight[:, 3:], 0.0)
torch.nn.init.normal_(lin.weight[:, :3], 0.0, np.sqrt(2
) / np.sqrt(out_dim))
elif multires > 0 and l in self.skip_in:
torch.nn.init.constant_(lin.bias, 0.0)
torch.nn.init.normal_(lin.weight, 0.0, np.sqrt(2) / np.
sqrt(out_dim))
torch.nn.init.constant_(lin.weight[:, -(dims[0] - 3):], 0.0
)
else:
torch.nn.init.constant_(lin.bias, 0.0)
torch.nn.init.normal_(lin.weight, 0.0, np.sqrt(2) / np.
sqrt(out_dim))
if weight_norm:
lin = nn.utils.weight_norm(lin)
setattr(self, 'lin' + str(l), lin)
self.activation = nn.Softplus(beta=100)
def forward(self, inputs):
inputs = inputs * self.scale
if self.embed_fn_fine is not None:
inputs = self.embed_fn_fine(inputs)
x = inputs
for l in range(0, self.num_layers - 1):
lin = getattr(self, 'lin' + str(l))
if l in self.skip_in:
x = torch.cat([x, inputs], 1) / np.sqrt(2)
x = lin(x)
if l < self.num_layers - 2:
x = self.activation(x)
return torch.cat([x[:, :1] / self.scale, x[:, 1:]], dim=-1)
def sdf(self, x):
return self.forward(x)[:, :1]
def sdf_hidden_appearance(self, x):
return self.forward(x)
def gradient(self, x):
x.requires_grad_(True)
y = self.sdf(x)
d_output = torch.ones_like(y, requires_grad=False, device=y.device)
gradients = torch.autograd.grad(outputs=y, inputs=x, grad_outputs=
d_output, create_graph=True, retain_graph=True, only_inputs=True)[0
]
return gradients.unsqueeze(1)
def get_inputs():
return [torch.rand([4, 4])]
def get_init_inputs():
return [[], {'d_in': 4, 'd_out': 4, 'd_hidden': 4, 'n_layers': 1}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
import numpy as np
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_mul_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = 1.0
tmp2 = tmp0 * tmp1
tl.store(out_ptr0 + x0, tmp2, xmask)
@triton.jit
def triton_poi_fused__weight_norm_interface_1(in_ptr0, out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + 4 * x0, xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + 4 * x0), xmask, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (2 + 4 * x0), xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr0 + (3 + 4 * x0), xmask, eviction_policy='evict_last')
tmp1 = tmp0 * tmp0
tmp3 = tmp2 * tmp2
tmp4 = tmp1 + tmp3
tmp6 = tmp5 * tmp5
tmp7 = tmp4 + tmp6
tmp9 = tmp8 * tmp8
tmp10 = tmp7 + tmp9
tmp11 = libdevice.sqrt(tmp10)
tl.store(out_ptr0 + x0, tmp11, xmask)
@triton.jit
def triton_poi_fused__weight_norm_interface_2(in_ptr0, in_ptr1, in_ptr2,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr1 + x1, xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr2 + x1, xmask, eviction_policy='evict_last')
tmp3 = tmp1 / tmp2
tmp4 = tmp0 * tmp3
tl.store(out_ptr0 + x2, tmp4, xmask)
@triton.jit
def triton_poi_fused_softplus_3(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = 100.0
tmp2 = tmp0 * tmp1
tmp3 = 20.0
tmp4 = tmp2 > tmp3
tmp5 = tl_math.exp(tmp2)
tmp6 = libdevice.log1p(tmp5)
tmp7 = 0.01
tmp8 = tmp6 * tmp7
tmp9 = tl.where(tmp4, tmp0, tmp8)
tl.store(out_ptr0 + x0, tmp9, xmask)
@triton.jit
def triton_poi_fused_cat_4(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 4
x1 = xindex // 4
x2 = xindex
tmp0 = x0
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 1, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + 4 * x1, tmp4 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp6 = 1.0
tmp7 = tmp5 * tmp6
tmp8 = tl.full(tmp7.shape, 0.0, tmp7.dtype)
tmp9 = tl.where(tmp4, tmp7, tmp8)
tmp10 = tmp0 >= tmp3
tl.full([1], 4, tl.int64)
tmp13 = tl.load(in_ptr0 + (1 + 4 * x1 + (-1 + x0)), tmp10 & xmask,
eviction_policy='evict_last', other=0.0)
tmp14 = tl.where(tmp4, tmp9, tmp13)
tl.store(out_ptr0 + x2, tmp14, xmask)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7) = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4, 1), (1, 1))
assert_size_stride(primals_3, (4, 4), (4, 1))
assert_size_stride(primals_4, (4,), (1,))
assert_size_stride(primals_5, (4, 1), (1, 1))
assert_size_stride(primals_6, (4, 4), (4, 1))
assert_size_stride(primals_7, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_mul_0[grid(16)](primals_1, buf0, 16, XBLOCK=16,
num_warps=1, num_stages=1)
del primals_1
buf1 = empty_strided_cuda((4, 1), (1, 1), torch.float32)
triton_poi_fused__weight_norm_interface_1[grid(4)](primals_3, buf1,
4, XBLOCK=4, num_warps=1, num_stages=1)
buf2 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
triton_poi_fused__weight_norm_interface_2[grid(16)](primals_3,
primals_2, buf1, buf2, 16, XBLOCK=16, num_warps=1, num_stages=1)
buf3 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_4, buf0, reinterpret_tensor(buf2, (4,
4), (1, 4), 0), alpha=1, beta=1, out=buf3)
del primals_4
buf4 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
triton_poi_fused_softplus_3[grid(16)](buf3, buf4, 16, XBLOCK=16,
num_warps=1, num_stages=1)
buf5 = empty_strided_cuda((4, 1), (1, 1), torch.float32)
triton_poi_fused__weight_norm_interface_1[grid(4)](primals_6, buf5,
4, XBLOCK=4, num_warps=1, num_stages=1)
buf6 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
triton_poi_fused__weight_norm_interface_2[grid(16)](primals_6,
primals_5, buf5, buf6, 16, XBLOCK=16, num_warps=1, num_stages=1)
buf7 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_7, buf4, reinterpret_tensor(buf6, (4,
4), (1, 4), 0), alpha=1, beta=1, out=buf7)
del primals_7
buf8 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
triton_poi_fused_cat_4[grid(16)](buf7, buf8, 16, XBLOCK=16,
num_warps=1, num_stages=1)
del buf7
return (buf8, buf2, buf6, primals_2, primals_3, primals_5, primals_6,
buf0, buf1, buf3, buf4, buf5, buf6)
def get_embedder(multires, input_dims=3):
embed_kwargs = {'include_input': True, 'input_dims': input_dims,
'max_freq_log2': multires - 1, 'num_freqs': multires,
'log_sampling': True, 'periodic_fns': [torch.sin, torch.cos]}
embedder_obj = Embedder(**embed_kwargs)
def embed(x, eo=embedder_obj):
return eo.embed(x)
return embed, embedder_obj.out_dim
class Embedder:
def __init__(self, **kwargs):
self.kwargs = kwargs
self.create_embedding_fn()
def create_embedding_fn(self):
embed_fns = []
d = self.kwargs['input_dims']
out_dim = 0
if self.kwargs['include_input']:
embed_fns.append(lambda x: x)
out_dim += d
max_freq = self.kwargs['max_freq_log2']
N_freqs = self.kwargs['num_freqs']
if self.kwargs['log_sampling']:
freq_bands = 2.0 ** torch.linspace(0.0, max_freq, N_freqs)
else:
freq_bands = torch.linspace(2.0 ** 0.0, 2.0 ** max_freq, N_freqs)
for freq in freq_bands:
for p_fn in self.kwargs['periodic_fns']:
embed_fns.append(lambda x, p_fn=p_fn, freq=freq: p_fn(x * freq)
)
out_dim += d
self.embed_fns = embed_fns
self.out_dim = out_dim
def embed(self, inputs):
return torch.cat([fn(inputs) for fn in self.embed_fns], -1)
class SDFNetworkNew(nn.Module):
def __init__(self, d_in, d_out, d_hidden, n_layers, skip_in=(4,),
multires=0, bias=0.5, scale=1, geometric_init=True, weight_norm=
True, inside_outside=False):
super(SDFNetworkNew, self).__init__()
dims = [d_in] + [d_hidden for _ in range(n_layers)] + [d_out]
self.embed_fn_fine = None
if multires > 0:
embed_fn, input_ch = get_embedder(multires, input_dims=d_in)
self.embed_fn_fine = embed_fn
dims[0] = input_ch
self.num_layers = len(dims)
self.skip_in = skip_in
self.scale = scale
for l in range(0, self.num_layers - 1):
if l + 1 in self.skip_in:
out_dim = dims[l + 1] - dims[0]
else:
out_dim = dims[l + 1]
lin = nn.Linear(dims[l], out_dim)
if geometric_init:
if l == self.num_layers - 2:
if not inside_outside:
torch.nn.init.normal_(lin.weight, mean=np.sqrt(np.
pi) / np.sqrt(dims[l]), std=0.0001)
torch.nn.init.constant_(lin.bias, -bias)
else:
torch.nn.init.normal_(lin.weight, mean=-np.sqrt(np.
pi) / np.sqrt(dims[l]), std=0.0001)
torch.nn.init.constant_(lin.bias, bias)
elif multires > 0 and l == 0:
torch.nn.init.constant_(lin.bias, 0.0)
torch.nn.init.constant_(lin.weight[:, 3:], 0.0)
torch.nn.init.normal_(lin.weight[:, :3], 0.0, np.sqrt(2
) / np.sqrt(out_dim))
elif multires > 0 and l in self.skip_in:
torch.nn.init.constant_(lin.bias, 0.0)
torch.nn.init.normal_(lin.weight, 0.0, np.sqrt(2) / np.
sqrt(out_dim))
torch.nn.init.constant_(lin.weight[:, -(dims[0] - 3):], 0.0
)
else:
torch.nn.init.constant_(lin.bias, 0.0)
torch.nn.init.normal_(lin.weight, 0.0, np.sqrt(2) / np.
sqrt(out_dim))
if weight_norm:
lin = nn.utils.weight_norm(lin)
setattr(self, 'lin' + str(l), lin)
self.activation = nn.Softplus(beta=100)
def sdf(self, x):
return self.forward(x)[:, :1]
def sdf_hidden_appearance(self, x):
return self.forward(x)
def gradient(self, x):
x.requires_grad_(True)
y = self.sdf(x)
d_output = torch.ones_like(y, requires_grad=False, device=y.device)
gradients = torch.autograd.grad(outputs=y, inputs=x, grad_outputs=
d_output, create_graph=True, retain_graph=True, only_inputs=True)[0
]
return gradients.unsqueeze(1)
def forward(self, input_0):
primals_4 = self.lin0.bias
primals_2 = self.lin0.weight_g
primals_1 = self.lin0.weight_v
primals_7 = self.lin1.bias
primals_5 = self.lin1.weight_g
primals_3 = self.lin1.weight_v
primals_6 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7])
return output[0]
|
Junlin-Yin/NeuS
|
SDFNetwork
| false
| 13,954
|
[
"MIT"
] | 345
|
b13dba90ba1c65d0ccaaca6b9d65225d5dfa8fe2
|
https://github.com/Junlin-Yin/NeuS/tree/b13dba90ba1c65d0ccaaca6b9d65225d5dfa8fe2
|
PosLinear2
|
import torch
from torch import Tensor
from torch.utils.data import Dataset as Dataset
import torch.nn as nn
import torch.utils.data
class PosLinear2(torch.nn.Linear):
def forward(self, x: 'Tensor') ->Tensor:
return nn.functional.linear(x, torch.nn.functional.softmax(self.
weight, 1), self.bias)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_features': 4, 'out_features': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import math as tl_math
from torch.utils.data import Dataset as Dataset
import torch.utils.data
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused__softmax_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + 4 * x1, xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + 4 * x1), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (2 + 4 * x1), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (3 + 4 * x1), xmask, eviction_policy='evict_last')
tmp3 = triton_helpers.maximum(tmp1, tmp2)
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp0 - tmp7
tmp9 = tl_math.exp(tmp8)
tl.store(out_ptr0 + x2, tmp9, xmask)
@triton.jit
def triton_poi_fused__softmax_1(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + 4 * x1, xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + 4 * x1), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (2 + 4 * x1), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (3 + 4 * x1), xmask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp5 = tmp3 + tmp4
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tl.store(out_ptr0 + x2, tmp8, xmask)
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused__softmax_0[grid(16)](primals_1, buf0, 16, XBLOCK=
16, num_warps=1, num_stages=1)
buf1 = empty_strided_cuda((4, 4), (4, 1), torch.float32)
triton_poi_fused__softmax_1[grid(16)](buf0, buf1, 16, XBLOCK=16,
num_warps=1, num_stages=1)
del buf0
buf2 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_2, reinterpret_tensor(primals_3, (64,
4), (4, 1), 0), reinterpret_tensor(buf1, (4, 4), (1, 4), 0),
alpha=1, beta=1, out=buf2)
del buf1
del primals_2
return reinterpret_tensor(buf2, (4, 4, 4, 4), (64, 16, 4, 1), 0
), primals_1, reinterpret_tensor(primals_3, (64, 4), (4, 1), 0)
class PosLinear2New(torch.nn.Linear):
def forward(self, input_0):
primals_1 = self.weight
primals_2 = self.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3])
return output[0]
|
KelvinKan/CP-Flow
|
PosLinear2
| false
| 13,955
|
[
"MIT"
] | 64
|
d01303cb4ebeb5a0bbfca638ffaf5b7a8ec22fb1
|
https://github.com/KelvinKan/CP-Flow/tree/d01303cb4ebeb5a0bbfca638ffaf5b7a8ec22fb1
|
UpsamplingBilinear2d
|
import torch
import torch.nn as nn
import torch.nn.functional as F
class UpsamplingBilinear2d(nn.Module):
def __init__(self, scale_factor=2.0):
super().__init__()
self.scale_factor = scale_factor
def forward(self, x):
return F.interpolate(x, scale_factor=self.scale_factor, mode=
'bilinear', align_corners=True)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused__to_copy__unsafe_index_add_arange_clamp_mul_sub_0(
in_out_ptr0, in_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 1024
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 8 % 8
x0 = xindex % 8
x2 = xindex // 64
x4 = xindex
tmp0 = x1
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.42857142857142855
tmp3 = tmp1 * tmp2
tmp4 = 0.0
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp6 = tmp5.to(tl.int32)
tmp7 = tl.full([1], 1, tl.int64)
tmp8 = tmp6 + tmp7
tmp9 = tl.full([1], 3, tl.int64)
tmp10 = triton_helpers.minimum(tmp8, tmp9)
tmp11 = x0
tmp12 = tmp11.to(tl.float32)
tmp13 = tmp12 * tmp2
tmp14 = triton_helpers.maximum(tmp13, tmp4)
tmp15 = tmp14.to(tl.int32)
tmp16 = tl.load(in_ptr0 + (tmp15 + 4 * tmp10 + 16 * x2), xmask,
eviction_policy='evict_last')
tmp17 = tmp15 + tmp7
tmp18 = triton_helpers.minimum(tmp17, tmp9)
tmp19 = tl.load(in_ptr0 + (tmp18 + 4 * tmp10 + 16 * x2), xmask,
eviction_policy='evict_last')
tmp20 = tmp19 - tmp16
tmp21 = tmp15.to(tl.float32)
tmp22 = tmp14 - tmp21
tmp23 = triton_helpers.maximum(tmp22, tmp4)
tmp24 = 1.0
tmp25 = triton_helpers.minimum(tmp23, tmp24)
tmp26 = tmp20 * tmp25
tmp27 = tmp16 + tmp26
tmp28 = tl.load(in_ptr0 + (tmp15 + 4 * tmp6 + 16 * x2), xmask,
eviction_policy='evict_last')
tmp29 = tl.load(in_ptr0 + (tmp18 + 4 * tmp6 + 16 * x2), xmask,
eviction_policy='evict_last')
tmp30 = tmp29 - tmp28
tmp31 = tmp30 * tmp25
tmp32 = tmp28 + tmp31
tmp33 = tmp27 - tmp32
tmp34 = tmp6.to(tl.float32)
tmp35 = tmp5 - tmp34
tmp36 = triton_helpers.maximum(tmp35, tmp4)
tmp37 = triton_helpers.minimum(tmp36, tmp24)
tmp38 = tmp33 * tmp37
tmp39 = tmp32 + tmp38
tl.store(in_out_ptr0 + x4, tmp39, xmask)
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 8, 8), (256, 64, 8, 1), torch.float32)
buf1 = buf0
del buf0
get_raw_stream(0)
triton_poi_fused__to_copy__unsafe_index_add_arange_clamp_mul_sub_0[grid
(1024)](buf1, arg0_1, 1024, XBLOCK=128, num_warps=4, num_stages=1)
del arg0_1
return buf1,
class UpsamplingBilinear2dNew(nn.Module):
def __init__(self, scale_factor=2.0):
super().__init__()
self.scale_factor = scale_factor
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
|
KyleDavisSA/pde-surrogate
|
UpsamplingBilinear2d
| false
| 13,956
|
[
"MIT"
] | 62
|
41ad2c9eb73c323e389174080f4b3df6cbd3c900
|
https://github.com/KyleDavisSA/pde-surrogate/tree/41ad2c9eb73c323e389174080f4b3df6cbd3c900
|
RewardCriterion
|
import torch
from torch import nn
import torch.nn.init
class RewardCriterion(nn.Module):
def __init__(self):
super(RewardCriterion, self).__init__()
def forward(self, input, seq, reward):
input = input.contiguous().view(-1)
reward = reward.contiguous().view(-1)
mask = (seq > 0).float()
mask = torch.cat([mask.new(mask.size(0), 1).fill_(1), mask[:, :-1]], 1
).contiguous().view(-1)
output = -input * reward * mask
output = torch.sum(output) / torch.sum(mask)
return output
def get_inputs():
return [torch.rand([4, 4]), torch.rand([4, 4]), torch.rand([4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch import nn
import torch.nn.init
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_per_fused_div_mul_neg_sum_0(in_out_ptr0, in_ptr0, in_ptr1,
in_ptr2, xnumel, rnumel, XBLOCK: tl.constexpr):
RBLOCK: tl.constexpr = 16
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp2 = tl.load(in_ptr1 + r0, None)
tmp1 = -tmp0
tmp3 = tmp1 * tmp2
tmp4 = r0 % 4
tl.full([1, 1], 0, tl.int64)
tmp7 = tl.full([1, 1], 1, tl.int64)
tmp8 = tmp4 < tmp7
tmp9 = 1.0
tmp10 = tl.full(tmp9.shape, 0.0, tmp9.dtype)
tmp11 = tl.where(tmp8, tmp9, tmp10)
tmp12 = tmp4 >= tmp7
tl.full([1, 1], 4, tl.int64)
tmp15 = tl.load(in_ptr2 + tl.broadcast_to(4 * (r0 // 4) + (-1 + r0 % 4),
[XBLOCK, RBLOCK]), tmp12, eviction_policy='evict_last', other=0.0)
tmp16 = 0.0
tmp17 = tmp15 > tmp16
tmp18 = tmp17.to(tl.float32)
tmp19 = tl.full(tmp18.shape, 0.0, tmp18.dtype)
tmp20 = tl.where(tmp12, tmp18, tmp19)
tmp21 = tl.where(tmp8, tmp11, tmp20)
tmp22 = tmp3 * tmp21
tmp23 = tl.broadcast_to(tmp22, [XBLOCK, RBLOCK])
tmp25 = tl.sum(tmp23, 1)[:, None]
tmp26 = tl.broadcast_to(tmp21, [XBLOCK, RBLOCK])
tmp28 = tl.sum(tmp26, 1)[:, None]
tmp29 = tmp25 / tmp28
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([XBLOCK, 1], 0, tl.int32), tmp29, None)
def call(args):
arg0_1, arg1_1, arg2_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4), (4, 1))
assert_size_stride(arg1_1, (4, 4), (4, 1))
assert_size_stride(arg2_1, (4, 4), (4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((), (), torch.float32)
buf2 = buf0
del buf0
get_raw_stream(0)
triton_per_fused_div_mul_neg_sum_0[grid(1)](buf2, arg0_1, arg1_1,
arg2_1, 1, 16, XBLOCK=1, num_warps=2, num_stages=1)
del arg0_1
del arg1_1
del arg2_1
return buf2,
class RewardCriterionNew(nn.Module):
def __init__(self):
super(RewardCriterionNew, self).__init__()
def forward(self, input_0, input_1, input_2):
arg0_1 = input_0
arg1_1 = input_1
arg2_1 = input_2
output = call([arg0_1, arg1_1, arg2_1])
return output[0]
|
KunpengLi1994/VSRN
|
RewardCriterion
| false
| 13,957
|
[
"Apache-2.0"
] | 238
|
777ae74326fdb6abe69dbd3911d0e545322520d1
|
https://github.com/KunpengLi1994/VSRN/tree/777ae74326fdb6abe69dbd3911d0e545322520d1
|
MVCRegularizer
|
import torch
import torch.nn.parallel
import torch.utils.data
class MVCRegularizer(torch.nn.Module):
"""
penalize MVC with large absolute value and negative values
alpha * large_weight^2 + beta * (negative_weight)^2
"""
def __init__(self, alpha=1.0, beta=1.0, threshold=5.0):
super().__init__()
self.alpha = alpha
self.beta = beta
self.threshold = threshold
def forward(self, weights):
loss = 0
if self.alpha > 0:
large_loss = torch.log(torch.nn.functional.relu(weights.abs() -
self.threshold) + 1)
loss += torch.mean(large_loss) * self.alpha
if self.beta > 0:
neg_loss = torch.nn.functional.relu(-weights)
neg_loss = neg_loss ** 2
loss += torch.mean(neg_loss) * self.beta
return loss
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import math as tl_math
import torch.nn.parallel
import torch.utils.data
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_per_fused_abs_add_log_mean_mul_neg_pow_relu_sub_0(in_out_ptr0,
in_ptr0, xnumel, rnumel):
XBLOCK: tl.constexpr = 1
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
tl.full([1], xoffset, tl.int32)
tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
tl.full([RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp1 = tl_math.abs(tmp0)
tmp2 = 5.0
tmp3 = tmp1 - tmp2
tmp4 = tl.full([1], 0, tl.int32)
tmp5 = triton_helpers.maximum(tmp4, tmp3)
tmp6 = 1.0
tmp7 = tmp5 + tmp6
tmp8 = tl_math.log(tmp7)
tmp9 = tl.broadcast_to(tmp8, [RBLOCK])
tmp11 = triton_helpers.promote_to_tensor(tl.sum(tmp9, 0))
tmp12 = -tmp0
tmp13 = triton_helpers.maximum(tmp4, tmp12)
tmp14 = tmp13 * tmp13
tmp15 = tl.broadcast_to(tmp14, [RBLOCK])
tmp17 = triton_helpers.promote_to_tensor(tl.sum(tmp15, 0))
tmp18 = 256.0
tmp19 = tmp11 / tmp18
tmp20 = tmp19 * tmp6
tmp21 = 0.0
tmp22 = tmp20 + tmp21
tmp23 = tmp17 / tmp18
tmp24 = tmp23 * tmp6
tmp25 = tmp22 + tmp24
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([1], 0, tl.int32), tmp25, None)
def call(args):
arg0_1, = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((), (), torch.float32)
buf2 = buf0
del buf0
get_raw_stream(0)
triton_per_fused_abs_add_log_mean_mul_neg_pow_relu_sub_0[grid(1)](buf2,
arg0_1, 1, 256, num_warps=2, num_stages=1)
del arg0_1
return buf2,
class MVCRegularizerNew(torch.nn.Module):
"""
penalize MVC with large absolute value and negative values
alpha * large_weight^2 + beta * (negative_weight)^2
"""
def __init__(self, alpha=1.0, beta=1.0, threshold=5.0):
super().__init__()
self.alpha = alpha
self.beta = beta
self.threshold = threshold
def forward(self, input_0):
arg0_1 = input_0
output = call([arg0_1])
return output[0]
|
KunalMGupta/deep_cage
|
MVCRegularizer
| false
| 13,958
|
[
"MIT"
] | 123
|
d8454c40d650911341b7f594af2fcefcf26f3d1d
|
https://github.com/KunalMGupta/deep_cage/tree/d8454c40d650911341b7f594af2fcefcf26f3d1d
|
MultiplicativeIntegration
|
import torch
import torch.nn as nn
from typing import List
class MultiplicativeIntegration(nn.Module):
def __init__(self, inputs_sizes: 'List[int]', output_sizes: 'List[int]',
bias: 'bool', bias_start: 'float'=0.0, alpha_start: 'float'=1.0,
beta_start: 'float'=1.0):
super().__init__()
self.inputs_sizes = inputs_sizes
self.output_sizes = output_sizes
total_output_size = sum(output_sizes)
total_input_size = sum(inputs_sizes)
self.bias_start = bias_start
self.alpha_start = alpha_start
self.beta_start = beta_start
self.weights = nn.Parameter(torch.empty(total_input_size,
total_output_size))
self.alphas = nn.Parameter(torch.empty([total_output_size]))
self.betas = nn.Parameter(torch.empty([2 * total_output_size]))
self.biases = nn.Parameter(torch.empty([total_output_size])
) if bias else None
self.reset_parameters()
def forward(self, input0, input1):
w1, w2 = torch.split(self.weights, self.inputs_sizes, dim=0)
b1, b2 = torch.split(self.betas, sum(self.output_sizes), dim=0)
wx1, wx2 = input0 @ w1, input1 @ w2
res = self.alphas * wx1 * wx2 + b1 * wx1 + b2 * wx2
if self.biases is not None:
res += self.biases
return res
def reset_parameters(self):
nn.init.xavier_uniform_(self.weights, gain=1.0)
nn.init.constant_(self.alphas, self.alpha_start)
nn.init.constant_(self.betas, self.beta_start)
if self.biases is not None:
nn.init.constant_(self.biases, self.bias_start)
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'inputs_sizes': [4, 4], 'output_sizes': [4, 4], 'bias': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
import torch.nn as nn
from typing import List
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_add_mul_0(in_ptr0, in_ptr1, in_ptr2, in_ptr3, in_ptr4,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 512
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 8
x2 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp1 = tl.load(in_ptr1 + x2, xmask)
tmp3 = tl.load(in_ptr2 + x2, xmask)
tmp5 = tl.load(in_ptr3 + x0, xmask, eviction_policy='evict_last')
tmp8 = tl.load(in_ptr3 + (8 + x0), xmask, eviction_policy='evict_last')
tmp11 = tl.load(in_ptr4 + x0, xmask, eviction_policy='evict_last')
tmp2 = tmp0 * tmp1
tmp4 = tmp2 * tmp3
tmp6 = tmp5 * tmp1
tmp7 = tmp4 + tmp6
tmp9 = tmp8 * tmp3
tmp10 = tmp7 + tmp9
tmp12 = tmp10 + tmp11
tl.store(out_ptr0 + x2, tmp12, xmask)
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5, primals_6 = args
args.clear()
assert_size_stride(primals_1, (8, 8), (8, 1))
assert_size_stride(primals_2, (16,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_5, (8,), (1,))
assert_size_stride(primals_6, (8,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 8), (8, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_3, (64, 4), (4, 1), 0),
reinterpret_tensor(primals_1, (4, 8), (8, 1), 0), out=buf0)
buf1 = empty_strided_cuda((64, 8), (8, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_4, (64, 4), (4, 1), 0),
reinterpret_tensor(primals_1, (4, 8), (8, 1), 32), out=buf1)
del primals_1
buf2 = empty_strided_cuda((4, 4, 4, 8), (128, 32, 8, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_add_mul_0[grid(512)](primals_5, buf0, buf1,
primals_2, primals_6, buf2, 512, XBLOCK=256, num_warps=4,
num_stages=1)
del primals_6
return buf2, primals_5, reinterpret_tensor(primals_2, (8,), (1,), 0
), reinterpret_tensor(primals_2, (8,), (1,), 8
), buf0, buf1, reinterpret_tensor(primals_4, (4, 64), (1, 4), 0
), reinterpret_tensor(primals_3, (4, 64), (1, 4), 0)
class MultiplicativeIntegrationNew(nn.Module):
def __init__(self, inputs_sizes: 'List[int]', output_sizes: 'List[int]',
bias: 'bool', bias_start: 'float'=0.0, alpha_start: 'float'=1.0,
beta_start: 'float'=1.0):
super().__init__()
self.inputs_sizes = inputs_sizes
self.output_sizes = output_sizes
total_output_size = sum(output_sizes)
total_input_size = sum(inputs_sizes)
self.bias_start = bias_start
self.alpha_start = alpha_start
self.beta_start = beta_start
self.weights = nn.Parameter(torch.empty(total_input_size,
total_output_size))
self.alphas = nn.Parameter(torch.empty([total_output_size]))
self.betas = nn.Parameter(torch.empty([2 * total_output_size]))
self.biases = nn.Parameter(torch.empty([total_output_size])
) if bias else None
self.reset_parameters()
def reset_parameters(self):
nn.init.xavier_uniform_(self.weights, gain=1.0)
nn.init.constant_(self.alphas, self.alpha_start)
nn.init.constant_(self.betas, self.beta_start)
if self.biases is not None:
nn.init.constant_(self.biases, self.bias_start)
def forward(self, input_0, input_1):
primals_1 = self.weights
primals_5 = self.alphas
primals_2 = self.betas
primals_6 = self.biases
primals_3 = input_0
primals_4 = input_1
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6])
return output[0]
|
KnowingNothing/FlexTensor
|
MultiplicativeIntegration
| false
| 13,959
|
[
"MIT"
] | 135
|
00f6cd7e038af7714b833fde7034d465fe2dc4a7
|
https://github.com/KnowingNothing/FlexTensor/tree/00f6cd7e038af7714b833fde7034d465fe2dc4a7
|
QuanConv
|
from torch.autograd import Function
import torch
import torch.utils.data.distributed
import torch.nn as nn
import torch.nn.functional as F
import torch.utils.data
def quantize(input, nbit):
return Quantizer.apply(input, nbit)
def dorefa_a(input, nbit_a):
return quantize(torch.clamp(0.1 * input, 0, 1), nbit_a)
def scale_sign(input):
return ScaleSigner.apply(input)
def dorefa_w(w, nbit_w):
if nbit_w == 1:
w = scale_sign(w)
else:
w = torch.tanh(w)
w = w / (2 * torch.max(torch.abs(w))) + 0.5
w = 2 * quantize(w, nbit_w) - 1
return w
class Quantizer(Function):
@staticmethod
def forward(ctx, input, nbit):
scale = 2 ** nbit - 1
return torch.round(input * scale) / scale
@staticmethod
def backward(ctx, grad_output):
return grad_output, None
class ScaleSigner(Function):
"""take a real value x, output sign(x)*E(|x|)"""
@staticmethod
def forward(ctx, input):
return torch.sign(input) * torch.mean(torch.abs(input))
@staticmethod
def backward(ctx, grad_output):
return grad_output
class QuanConv(nn.Conv2d):
"""docstring for QuanConv"""
def __init__(self, in_channels, out_channels, kernel_size, quan_name_w=
'dorefa', quan_name_a='dorefa', nbit_w=1, nbit_a=1, stride=1,
padding=0, dilation=1, groups=1, bias=True):
super(QuanConv, self).__init__(in_channels, out_channels,
kernel_size, stride, padding, dilation, groups, bias)
self.nbit_w = nbit_w
self.nbit_a = nbit_a
name_w_dict = {'dorefa': dorefa_w}
name_a_dict = {'dorefa': dorefa_a}
self.quan_w = name_w_dict[quan_name_w]
self.quan_a = name_a_dict[quan_name_a]
def forward(self, input):
if self.nbit_w < 32:
w = self.quan_w(self.weight, self.nbit_w)
else:
w = self.weight
if self.nbit_a < 32:
x = self.quan_a(input, self.nbit_a)
else:
x = input
output = F.conv2d(x, w, self.bias, self.stride, self.padding, self.
dilation, self.groups)
return output
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_channels': 4, 'out_channels': 4, 'kernel_size': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import libdevice, math as tl_math
from torch.autograd import Function
import torch.utils.data.distributed
import torch.nn as nn
import torch.utils.data
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_per_fused_abs_mean_mul_sign_0(in_ptr0, out_ptr1, xnumel, rnumel):
XBLOCK: tl.constexpr = 1
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
tl.full([1], xoffset, tl.int32)
tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
tl.full([RBLOCK], True, tl.int1)
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, None)
tmp1 = tl_math.abs(tmp0)
tmp2 = tl.broadcast_to(tmp1, [RBLOCK])
tmp4 = triton_helpers.promote_to_tensor(tl.sum(tmp2, 0))
tmp5 = tl.full([1], 0, tl.int32)
tmp6 = tmp5 < tmp0
tmp7 = tmp6.to(tl.int8)
tmp8 = tmp0 < tmp5
tmp9 = tmp8.to(tl.int8)
tmp10 = tmp7 - tmp9
tmp11 = tmp10.to(tmp0.dtype)
tmp12 = 256.0
tmp13 = tmp4 / tmp12
tmp14 = tmp11 * tmp13
tl.store(out_ptr1 + tl.broadcast_to(r0, [RBLOCK]), tmp14, None)
@triton.jit
def triton_poi_fused_clamp_div_mul_round_1(in_ptr0, out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_ptr0 + x0, xmask)
tmp1 = 0.1
tmp2 = tmp0 * tmp1
tmp3 = 0.0
tmp4 = triton_helpers.maximum(tmp2, tmp3)
tmp5 = 1.0
tmp6 = triton_helpers.minimum(tmp4, tmp5)
tmp7 = tmp6 * tmp5
tmp8 = libdevice.nearbyint(tmp7)
tmp9 = tmp8 * tmp5
tl.store(out_ptr0 + x0, tmp9, xmask)
@triton.jit
def triton_poi_fused_convolution_2(in_out_ptr0, in_ptr0, xnumel, XBLOCK: tl
.constexpr):
xnumel = 16
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_out_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tl.store(in_out_ptr0 + x2, tmp2, xmask)
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_3, (4,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf1 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_per_fused_abs_mean_mul_sign_0[grid(1)](primals_1, buf1, 1,
256, num_warps=2, num_stages=1)
del primals_1
buf2 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused_clamp_div_mul_round_1[grid(256)](primals_2, buf2,
256, XBLOCK=128, num_warps=4, num_stages=1)
del primals_2
buf3 = extern_kernels.convolution(buf2, buf1, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf3, (4, 4, 1, 1), (4, 1, 1, 1))
buf4 = buf3
del buf3
triton_poi_fused_convolution_2[grid(16)](buf4, primals_3, 16,
XBLOCK=16, num_warps=1, num_stages=1)
del primals_3
return buf4, buf1, buf2
def quantize(input, nbit):
return Quantizer.apply(input, nbit)
def dorefa_a(input, nbit_a):
return quantize(torch.clamp(0.1 * input, 0, 1), nbit_a)
def scale_sign(input):
return ScaleSigner.apply(input)
def dorefa_w(w, nbit_w):
if nbit_w == 1:
w = scale_sign(w)
else:
w = torch.tanh(w)
w = w / (2 * torch.max(torch.abs(w))) + 0.5
w = 2 * quantize(w, nbit_w) - 1
return w
class Quantizer(Function):
@staticmethod
def forward(ctx, input, nbit):
scale = 2 ** nbit - 1
return torch.round(input * scale) / scale
@staticmethod
def backward(ctx, grad_output):
return grad_output, None
class ScaleSigner(Function):
"""take a real value x, output sign(x)*E(|x|)"""
@staticmethod
def forward(ctx, input):
return torch.sign(input) * torch.mean(torch.abs(input))
@staticmethod
def backward(ctx, grad_output):
return grad_output
class QuanConvNew(nn.Conv2d):
"""docstring for QuanConv"""
def __init__(self, in_channels, out_channels, kernel_size, quan_name_w=
'dorefa', quan_name_a='dorefa', nbit_w=1, nbit_a=1, stride=1,
padding=0, dilation=1, groups=1, bias=True):
super(QuanConvNew, self).__init__(in_channels, out_channels,
kernel_size, stride, padding, dilation, groups, bias)
self.nbit_w = nbit_w
self.nbit_a = nbit_a
name_w_dict = {'dorefa': dorefa_w}
name_a_dict = {'dorefa': dorefa_a}
self.quan_w = name_w_dict[quan_name_w]
self.quan_a = name_a_dict[quan_name_a]
def forward(self, input_0):
primals_1 = self.weight
primals_3 = self.bias
primals_2 = input_0
output = call([primals_1, primals_2, primals_3])
return output[0]
|
Jzz24/pytorch_quantization
|
QuanConv
| false
| 13,960
|
[
"MIT"
] | 71
|
0c2d93c8ce4f85dd2c34ea6f36c58d14db21bf8e
|
https://github.com/Jzz24/pytorch_quantization/tree/0c2d93c8ce4f85dd2c34ea6f36c58d14db21bf8e
|
SlicedWasserstein
|
import torch
class SlicedWasserstein(torch.nn.Module):
"""
Implementation of the loss used in
[Sliced Wasserstein Discrepancy for Unsupervised Domain Adaptation](https://arxiv.org/abs/1903.04064)
"""
def __init__(self, m: 'int'=128):
"""
Arguments:
m: The dimensionality to project to.
"""
super().__init__()
self.m = 128
def forward(self, x: 'torch.Tensor', y: 'torch.Tensor') ->torch.Tensor:
"""
Arguments:
x: a batch of class predictions
y: the other batch of class predictions
Returns:
The discrepancy between the two batches of class predictions.
"""
d = x.shape[1]
proj = torch.randn(d, self.m, device=x.device)
proj = torch.nn.functional.normalize(proj, dim=0)
x = torch.matmul(x, proj)
y = torch.matmul(y, proj)
x, _ = torch.sort(x, dim=0)
y, _ = torch.sort(y, dim=0)
return torch.mean((x - y) ** 2)
def extra_repr(self):
""""""
return c_f.extra_repr(self, ['m'])
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
from torch import device
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import libdevice
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_div_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 512
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 128
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (128 + x0), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (256 + x0), xmask, eviction_policy='evict_last')
tmp9 = tl.load(in_ptr0 + (384 + x0), xmask, eviction_policy='evict_last')
tmp2 = tmp1 * tmp1
tmp4 = tmp3 * tmp3
tmp5 = tmp2 + tmp4
tmp7 = tmp6 * tmp6
tmp8 = tmp5 + tmp7
tmp10 = tmp9 * tmp9
tmp11 = tmp8 + tmp10
tmp12 = libdevice.sqrt(tmp11)
tmp13 = 1e-12
tmp14 = triton_helpers.maximum(tmp12, tmp13)
tmp15 = tmp0 / tmp14
tl.store(out_ptr0 + x2, tmp15, xmask)
@triton.jit
def triton_per_fused_sort_1(in_ptr0, out_ptr0, xnumel, rnumel, XBLOCK: tl.
constexpr):
RBLOCK: tl.constexpr = 4
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r1 = rindex
x0 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + 2048 * r1), None)
tmp1 = r1
tmp2 = tmp1.to(tl.int16)
tmp3 = tl.broadcast_to(tmp0, [XBLOCK, RBLOCK])
tmp4 = tl.broadcast_to(tmp2, [XBLOCK, RBLOCK])
tmp5, _tmp6 = triton_helpers.sort_with_index(tmp3, tmp4, None, 1,
stable=False, descending=False)
tl.store(out_ptr0 + (x0 + 2048 * r1), tmp5, None)
@triton.jit
def triton_red_fused_mean_pow_sub_2(in_out_ptr0, in_ptr0, in_ptr1, xnumel,
rnumel, XBLOCK: tl.constexpr, RBLOCK: tl.constexpr):
rnumel = 8192
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rbase = tl.arange(0, RBLOCK)[None, :]
_tmp5 = tl.full([XBLOCK, RBLOCK], 0, tl.float32)
for roffset in range(0, rnumel, RBLOCK):
rindex = roffset + rbase
rmask = rindex < rnumel
r0 = rindex
tmp0 = tl.load(in_ptr0 + r0, rmask, eviction_policy='evict_first',
other=0.0)
tmp1 = tl.load(in_ptr1 + r0, rmask, eviction_policy='evict_first',
other=0.0)
tmp2 = tmp0 - tmp1
tmp3 = tmp2 * tmp2
tmp4 = tl.broadcast_to(tmp3, [XBLOCK, RBLOCK])
tmp6 = _tmp5 + tmp4
_tmp5 = tl.where(rmask, tmp6, _tmp5)
tmp5 = tl.sum(_tmp5, 1)[:, None]
tmp7 = 8192.0
tmp8 = tmp5 / tmp7
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([XBLOCK, 1], 0, tl.int32), tmp8, None)
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = torch.ops.aten.randn.default([4, 128], device=device(type=
'cuda', index=0), pin_memory=False)
buf1 = buf0
del buf0
buf2 = empty_strided_cuda((4, 128), (128, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_div_0[grid(512)](buf1, buf2, 512, XBLOCK=256,
num_warps=4, num_stages=1)
del buf1
buf3 = empty_strided_cuda((64, 128), (128, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(arg0_1, (64, 4), (4, 1), 0),
buf2, out=buf3)
del arg0_1
buf4 = empty_strided_cuda((4, 4, 4, 128), (2048, 512, 128, 1),
torch.float32)
triton_per_fused_sort_1[grid(2048)](buf3, buf4, 2048, 4, XBLOCK=32,
num_warps=2, num_stages=1)
buf6 = buf3
del buf3
extern_kernels.mm(reinterpret_tensor(arg1_1, (64, 4), (4, 1), 0),
buf2, out=buf6)
del arg1_1
del buf2
buf7 = empty_strided_cuda((4, 4, 4, 128), (2048, 512, 128, 1),
torch.float32)
triton_per_fused_sort_1[grid(2048)](buf6, buf7, 2048, 4, XBLOCK=32,
num_warps=2, num_stages=1)
del buf6
buf9 = empty_strided_cuda((), (), torch.float32)
buf10 = buf9
del buf9
triton_red_fused_mean_pow_sub_2[grid(1)](buf10, buf4, buf7, 1, 8192,
XBLOCK=1, RBLOCK=2048, num_warps=16, num_stages=1)
del buf4
del buf7
return buf10,
class SlicedWassersteinNew(torch.nn.Module):
"""
Implementation of the loss used in
[Sliced Wasserstein Discrepancy for Unsupervised Domain Adaptation](https://arxiv.org/abs/1903.04064)
"""
def __init__(self, m: 'int'=128):
"""
Arguments:
m: The dimensionality to project to.
"""
super().__init__()
self.m = 128
def extra_repr(self):
""""""
return c_f.extra_repr(self, ['m'])
def forward(self, input_0, input_1):
arg0_1 = input_0
arg1_1 = input_1
output = call([arg0_1, arg1_1])
return output[0]
|
KevinMusgrave/pytorch-adapt
|
SlicedWasserstein
| false
| 13,961
|
[
"MIT"
] | 131
|
ff1491e1bfcc586afb8ee619712c8816ddf10358
|
https://github.com/KevinMusgrave/pytorch-adapt/tree/ff1491e1bfcc586afb8ee619712c8816ddf10358
|
EncoderImagePrecomp
|
import torch
import numpy as np
from torch import nn
from collections import OrderedDict
import torch.nn.init
def l2norm(X):
"""L2-normalize columns of X
"""
norm = torch.pow(X, 2).sum(dim=1, keepdim=True).sqrt()
X = torch.div(X, norm)
return X
class EncoderImagePrecomp(nn.Module):
def __init__(self, img_dim, embed_size, use_abs=False, no_imgnorm=False):
super(EncoderImagePrecomp, self).__init__()
self.embed_size = embed_size
self.no_imgnorm = no_imgnorm
self.use_abs = use_abs
self.fc = nn.Linear(img_dim, embed_size)
self.init_weights()
def init_weights(self):
"""Xavier initialization for the fully connected layer
"""
r = np.sqrt(6.0) / np.sqrt(self.fc.in_features + self.fc.out_features)
self.fc.weight.data.uniform_(-r, r)
self.fc.bias.data.fill_(0)
def forward(self, images):
"""Extract image feature vectors."""
features = self.fc(images)
if not self.no_imgnorm:
features = l2norm(features)
if self.use_abs:
features = torch.abs(features)
return features
def load_state_dict(self, state_dict):
"""Copies parameters. overwritting the default one to
accept state_dict from Full model
"""
own_state = self.state_dict()
new_state = OrderedDict()
for name, param in state_dict.items():
if name in own_state:
new_state[name] = param
super(EncoderImagePrecomp, self).load_state_dict(new_state)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'img_dim': 4, 'embed_size': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice
import numpy as np
from torch import nn
from collections import OrderedDict
import torch.nn.init
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_div_pow_sqrt_sum_0(in_ptr0, out_ptr0, xnumel, XBLOCK:
tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x0 = xindex % 16
x2 = xindex // 64
tmp0 = tl.load(in_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + (x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp3 = tl.load(in_ptr0 + (16 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp6 = tl.load(in_ptr0 + (32 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp9 = tl.load(in_ptr0 + (48 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp2 = tmp1 * tmp1
tmp4 = tmp3 * tmp3
tmp5 = tmp2 + tmp4
tmp7 = tmp6 * tmp6
tmp8 = tmp5 + tmp7
tmp10 = tmp9 * tmp9
tmp11 = tmp8 + tmp10
tmp12 = libdevice.sqrt(tmp11)
tmp13 = tmp0 / tmp12
tl.store(out_ptr0 + x3, tmp13, xmask)
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4), (4, 1))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 4), (4, 1), torch.float32)
extern_kernels.addmm(primals_2, reinterpret_tensor(primals_3, (64,
4), (4, 1), 0), reinterpret_tensor(primals_1, (4, 4), (1, 4), 0
), alpha=1, beta=1, out=buf0)
del primals_1
del primals_2
buf1 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_div_pow_sqrt_sum_0[grid(256)](buf0, buf1, 256,
XBLOCK=256, num_warps=4, num_stages=1)
return buf1, reinterpret_tensor(primals_3, (64, 4), (4, 1), 0), buf0
def l2norm(X):
"""L2-normalize columns of X
"""
norm = torch.pow(X, 2).sum(dim=1, keepdim=True).sqrt()
X = torch.div(X, norm)
return X
class EncoderImagePrecompNew(nn.Module):
def __init__(self, img_dim, embed_size, use_abs=False, no_imgnorm=False):
super(EncoderImagePrecompNew, self).__init__()
self.embed_size = embed_size
self.no_imgnorm = no_imgnorm
self.use_abs = use_abs
self.fc = nn.Linear(img_dim, embed_size)
self.init_weights()
def init_weights(self):
"""Xavier initialization for the fully connected layer
"""
r = np.sqrt(6.0) / np.sqrt(self.fc.in_features + self.fc.out_features)
self.fc.weight.data.uniform_(-r, r)
self.fc.bias.data.fill_(0)
def load_state_dict(self, state_dict):
"""Copies parameters. overwritting the default one to
accept state_dict from Full model
"""
own_state = self.state_dict()
new_state = OrderedDict()
for name, param in state_dict.items():
if name in own_state:
new_state[name] = param
super(EncoderImagePrecompNew, self).load_state_dict(new_state)
def forward(self, input_0):
primals_1 = self.fc.weight
primals_2 = self.fc.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3])
return output[0]
|
KunpengLi1994/VSRN
|
EncoderImagePrecomp
| false
| 13,962
|
[
"Apache-2.0"
] | 238
|
777ae74326fdb6abe69dbd3911d0e545322520d1
|
https://github.com/KunpengLi1994/VSRN/tree/777ae74326fdb6abe69dbd3911d0e545322520d1
|
SppBlock
|
import torch
import torch.nn.functional as F
from torch import nn
class SppBlock(nn.Module):
def __init__(self, in_channels):
super(SppBlock, self).__init__()
self.pool1 = nn.MaxPool2d(kernel_size=[2, 2], stride=2)
self.pool2 = nn.MaxPool2d(kernel_size=[3, 3], stride=3)
self.pool3 = nn.MaxPool2d(kernel_size=[5, 5], stride=5)
self.pool4 = nn.MaxPool2d(kernel_size=[6, 6], stride=6)
self.conv = nn.Conv2d(in_channels=in_channels, out_channels=1,
kernel_size=1, padding=0)
def forward(self, x):
self.in_channels, h, w = x.size(1), x.size(2), x.size(3)
self.layer1 = F.interpolate(self.conv(self.pool1(x)), size=(h, w),
mode='bilinear', align_corners=True)
self.layer2 = F.interpolate(self.conv(self.pool2(x)), size=(h, w),
mode='bilinear', align_corners=True)
self.layer3 = F.interpolate(self.conv(self.pool3(x)), size=(h, w),
mode='bilinear', align_corners=True)
self.layer4 = F.interpolate(self.conv(self.pool4(x)), size=(h, w),
mode='bilinear', align_corners=True)
out = torch.cat([self.layer1, self.layer2, self.layer3, self.layer4,
x], 1)
return out
def get_inputs():
return [torch.rand([4, 4, 64, 64])]
def get_init_inputs():
return [[], {'in_channels': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch import nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_max_pool2d_with_indices_0(in_ptr0, out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x0 = xindex % 32
x1 = xindex // 32
x2 = xindex
tmp0 = tl.load(in_ptr0 + (2 * x0 + 128 * x1), None, eviction_policy=
'evict_last')
tmp1 = tl.load(in_ptr0 + (1 + 2 * x0 + 128 * x1), None, eviction_policy
='evict_last')
tmp3 = tl.load(in_ptr0 + (64 + 2 * x0 + 128 * x1), None,
eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (65 + 2 * x0 + 128 * x1), None,
eviction_policy='evict_last')
tmp2 = triton_helpers.maximum(tmp1, tmp0)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp6 = triton_helpers.maximum(tmp5, tmp4)
tl.store(out_ptr0 + x2, tmp6, None)
@triton.jit
def triton_poi_fused__to_copy_1(out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = x0
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.49206349206349204
tmp3 = tmp1 * tmp2
tmp4 = 0.0
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp6 = tmp5.to(tl.int32)
tl.store(out_ptr0 + x0, tmp6, xmask)
@triton.jit
def triton_poi_fused_add_clamp_2(out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = x0
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.49206349206349204
tmp3 = tmp1 * tmp2
tmp4 = 0.0
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp6 = tmp5.to(tl.int32)
tmp7 = tl.full([1], 1, tl.int64)
tmp8 = tmp6 + tmp7
tmp9 = tl.full([1], 31, tl.int64)
tmp10 = triton_helpers.minimum(tmp8, tmp9)
tl.store(out_ptr0 + x0, tmp10, xmask)
@triton.jit
def triton_poi_fused__to_copy_arange_clamp_mul_sub_3(out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = x0
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.49206349206349204
tmp3 = tmp1 * tmp2
tmp4 = 0.0
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp6 = tmp5.to(tl.int32)
tmp7 = tmp6.to(tl.float32)
tmp8 = tmp5 - tmp7
tmp9 = triton_helpers.maximum(tmp8, tmp4)
tmp10 = 1.0
tmp11 = triton_helpers.minimum(tmp9, tmp10)
tl.store(out_ptr0 + x0, tmp11, xmask)
@triton.jit
def triton_poi_fused_max_pool2d_with_indices_4(in_ptr0, out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 7056
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 21
x1 = xindex // 21 % 21
x4 = xindex // 441
x3 = xindex // 1764
x5 = xindex % 1764
tmp0 = tl.load(in_ptr0 + (3 * x0 + 192 * x1 + 4096 * x4), xmask,
eviction_policy='evict_last')
tmp1 = tl.load(in_ptr0 + (1 + 3 * x0 + 192 * x1 + 4096 * x4), xmask,
eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (2 + 3 * x0 + 192 * x1 + 4096 * x4), xmask,
eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (64 + 3 * x0 + 192 * x1 + 4096 * x4), xmask,
eviction_policy='evict_last')
tmp7 = tl.load(in_ptr0 + (65 + 3 * x0 + 192 * x1 + 4096 * x4), xmask,
eviction_policy='evict_last')
tmp9 = tl.load(in_ptr0 + (66 + 3 * x0 + 192 * x1 + 4096 * x4), xmask,
eviction_policy='evict_last')
tmp11 = tl.load(in_ptr0 + (128 + 3 * x0 + 192 * x1 + 4096 * x4), xmask,
eviction_policy='evict_last')
tmp13 = tl.load(in_ptr0 + (129 + 3 * x0 + 192 * x1 + 4096 * x4), xmask,
eviction_policy='evict_last')
tmp15 = tl.load(in_ptr0 + (130 + 3 * x0 + 192 * x1 + 4096 * x4), xmask,
eviction_policy='evict_last')
tmp2 = triton_helpers.maximum(tmp1, tmp0)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp6 = triton_helpers.maximum(tmp5, tmp4)
tmp8 = triton_helpers.maximum(tmp7, tmp6)
tmp10 = triton_helpers.maximum(tmp9, tmp8)
tmp12 = triton_helpers.maximum(tmp11, tmp10)
tmp14 = triton_helpers.maximum(tmp13, tmp12)
tmp16 = triton_helpers.maximum(tmp15, tmp14)
tl.store(out_ptr0 + (x5 + 1792 * x3), tmp16, xmask)
@triton.jit
def triton_poi_fused__to_copy_5(out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = x0
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.31746031746031744
tmp3 = tmp1 * tmp2
tmp4 = 0.0
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp6 = tmp5.to(tl.int32)
tl.store(out_ptr0 + x0, tmp6, xmask)
@triton.jit
def triton_poi_fused_add_clamp_6(out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = x0
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.31746031746031744
tmp3 = tmp1 * tmp2
tmp4 = 0.0
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp6 = tmp5.to(tl.int32)
tmp7 = tl.full([1], 1, tl.int64)
tmp8 = tmp6 + tmp7
tmp9 = tl.full([1], 20, tl.int64)
tmp10 = triton_helpers.minimum(tmp8, tmp9)
tl.store(out_ptr0 + x0, tmp10, xmask)
@triton.jit
def triton_poi_fused__to_copy_arange_clamp_mul_sub_7(out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = x0
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.31746031746031744
tmp3 = tmp1 * tmp2
tmp4 = 0.0
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp6 = tmp5.to(tl.int32)
tmp7 = tmp6.to(tl.float32)
tmp8 = tmp5 - tmp7
tmp9 = triton_helpers.maximum(tmp8, tmp4)
tmp10 = 1.0
tmp11 = triton_helpers.minimum(tmp9, tmp10)
tl.store(out_ptr0 + x0, tmp11, xmask)
@triton.jit
def triton_poi_fused_max_pool2d_with_indices_8(in_ptr0, out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 2304
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 12
x1 = xindex // 12 % 12
x2 = xindex // 144
x3 = xindex
tmp0 = tl.load(in_ptr0 + (5 * x0 + 320 * x1 + 4096 * x2), xmask,
eviction_policy='evict_last')
tmp1 = tl.load(in_ptr0 + (1 + 5 * x0 + 320 * x1 + 4096 * x2), xmask,
eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (2 + 5 * x0 + 320 * x1 + 4096 * x2), xmask,
eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (3 + 5 * x0 + 320 * x1 + 4096 * x2), xmask,
eviction_policy='evict_last')
tmp7 = tl.load(in_ptr0 + (4 + 5 * x0 + 320 * x1 + 4096 * x2), xmask,
eviction_policy='evict_last')
tmp9 = tl.load(in_ptr0 + (64 + 5 * x0 + 320 * x1 + 4096 * x2), xmask,
eviction_policy='evict_last')
tmp11 = tl.load(in_ptr0 + (65 + 5 * x0 + 320 * x1 + 4096 * x2), xmask,
eviction_policy='evict_last')
tmp13 = tl.load(in_ptr0 + (66 + 5 * x0 + 320 * x1 + 4096 * x2), xmask,
eviction_policy='evict_last')
tmp15 = tl.load(in_ptr0 + (67 + 5 * x0 + 320 * x1 + 4096 * x2), xmask,
eviction_policy='evict_last')
tmp17 = tl.load(in_ptr0 + (68 + 5 * x0 + 320 * x1 + 4096 * x2), xmask,
eviction_policy='evict_last')
tmp19 = tl.load(in_ptr0 + (128 + 5 * x0 + 320 * x1 + 4096 * x2), xmask,
eviction_policy='evict_last')
tmp21 = tl.load(in_ptr0 + (129 + 5 * x0 + 320 * x1 + 4096 * x2), xmask,
eviction_policy='evict_last')
tmp23 = tl.load(in_ptr0 + (130 + 5 * x0 + 320 * x1 + 4096 * x2), xmask,
eviction_policy='evict_last')
tmp25 = tl.load(in_ptr0 + (131 + 5 * x0 + 320 * x1 + 4096 * x2), xmask,
eviction_policy='evict_last')
tmp27 = tl.load(in_ptr0 + (132 + 5 * x0 + 320 * x1 + 4096 * x2), xmask,
eviction_policy='evict_last')
tmp29 = tl.load(in_ptr0 + (192 + 5 * x0 + 320 * x1 + 4096 * x2), xmask,
eviction_policy='evict_last')
tmp31 = tl.load(in_ptr0 + (193 + 5 * x0 + 320 * x1 + 4096 * x2), xmask,
eviction_policy='evict_last')
tmp33 = tl.load(in_ptr0 + (194 + 5 * x0 + 320 * x1 + 4096 * x2), xmask,
eviction_policy='evict_last')
tmp35 = tl.load(in_ptr0 + (195 + 5 * x0 + 320 * x1 + 4096 * x2), xmask,
eviction_policy='evict_last')
tmp37 = tl.load(in_ptr0 + (196 + 5 * x0 + 320 * x1 + 4096 * x2), xmask,
eviction_policy='evict_last')
tmp39 = tl.load(in_ptr0 + (256 + 5 * x0 + 320 * x1 + 4096 * x2), xmask,
eviction_policy='evict_last')
tmp41 = tl.load(in_ptr0 + (257 + 5 * x0 + 320 * x1 + 4096 * x2), xmask,
eviction_policy='evict_last')
tmp43 = tl.load(in_ptr0 + (258 + 5 * x0 + 320 * x1 + 4096 * x2), xmask,
eviction_policy='evict_last')
tmp45 = tl.load(in_ptr0 + (259 + 5 * x0 + 320 * x1 + 4096 * x2), xmask,
eviction_policy='evict_last')
tmp47 = tl.load(in_ptr0 + (260 + 5 * x0 + 320 * x1 + 4096 * x2), xmask,
eviction_policy='evict_last')
tmp2 = triton_helpers.maximum(tmp1, tmp0)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp6 = triton_helpers.maximum(tmp5, tmp4)
tmp8 = triton_helpers.maximum(tmp7, tmp6)
tmp10 = triton_helpers.maximum(tmp9, tmp8)
tmp12 = triton_helpers.maximum(tmp11, tmp10)
tmp14 = triton_helpers.maximum(tmp13, tmp12)
tmp16 = triton_helpers.maximum(tmp15, tmp14)
tmp18 = triton_helpers.maximum(tmp17, tmp16)
tmp20 = triton_helpers.maximum(tmp19, tmp18)
tmp22 = triton_helpers.maximum(tmp21, tmp20)
tmp24 = triton_helpers.maximum(tmp23, tmp22)
tmp26 = triton_helpers.maximum(tmp25, tmp24)
tmp28 = triton_helpers.maximum(tmp27, tmp26)
tmp30 = triton_helpers.maximum(tmp29, tmp28)
tmp32 = triton_helpers.maximum(tmp31, tmp30)
tmp34 = triton_helpers.maximum(tmp33, tmp32)
tmp36 = triton_helpers.maximum(tmp35, tmp34)
tmp38 = triton_helpers.maximum(tmp37, tmp36)
tmp40 = triton_helpers.maximum(tmp39, tmp38)
tmp42 = triton_helpers.maximum(tmp41, tmp40)
tmp44 = triton_helpers.maximum(tmp43, tmp42)
tmp46 = triton_helpers.maximum(tmp45, tmp44)
tmp48 = triton_helpers.maximum(tmp47, tmp46)
tl.store(out_ptr0 + x3, tmp48, xmask)
@triton.jit
def triton_poi_fused__to_copy_9(out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = x0
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.1746031746031746
tmp3 = tmp1 * tmp2
tmp4 = 0.0
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp6 = tmp5.to(tl.int32)
tl.store(out_ptr0 + x0, tmp6, xmask)
@triton.jit
def triton_poi_fused_add_clamp_10(out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = x0
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.1746031746031746
tmp3 = tmp1 * tmp2
tmp4 = 0.0
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp6 = tmp5.to(tl.int32)
tmp7 = tl.full([1], 1, tl.int64)
tmp8 = tmp6 + tmp7
tmp9 = tl.full([1], 11, tl.int64)
tmp10 = triton_helpers.minimum(tmp8, tmp9)
tl.store(out_ptr0 + x0, tmp10, xmask)
@triton.jit
def triton_poi_fused__to_copy_arange_clamp_mul_sub_11(out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = x0
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.1746031746031746
tmp3 = tmp1 * tmp2
tmp4 = 0.0
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp6 = tmp5.to(tl.int32)
tmp7 = tmp6.to(tl.float32)
tmp8 = tmp5 - tmp7
tmp9 = triton_helpers.maximum(tmp8, tmp4)
tmp10 = 1.0
tmp11 = triton_helpers.minimum(tmp9, tmp10)
tl.store(out_ptr0 + x0, tmp11, xmask)
@triton.jit
def triton_poi_fused__to_copy_12(out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = x0
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.14285714285714285
tmp3 = tmp1 * tmp2
tmp4 = 0.0
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp6 = tmp5.to(tl.int32)
tl.store(out_ptr0 + x0, tmp6, xmask)
@triton.jit
def triton_poi_fused_add_clamp_13(out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = x0
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.14285714285714285
tmp3 = tmp1 * tmp2
tmp4 = 0.0
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp6 = tmp5.to(tl.int32)
tmp7 = tl.full([1], 1, tl.int64)
tmp8 = tmp6 + tmp7
tmp9 = tl.full([1], 9, tl.int64)
tmp10 = triton_helpers.minimum(tmp8, tmp9)
tl.store(out_ptr0 + x0, tmp10, xmask)
@triton.jit
def triton_poi_fused__to_copy_arange_clamp_mul_sub_14(out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = x0
tmp1 = tmp0.to(tl.float32)
tmp2 = 0.14285714285714285
tmp3 = tmp1 * tmp2
tmp4 = 0.0
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp6 = tmp5.to(tl.int32)
tmp7 = tmp6.to(tl.float32)
tmp8 = tmp5 - tmp7
tmp9 = triton_helpers.maximum(tmp8, tmp4)
tmp10 = 1.0
tmp11 = triton_helpers.minimum(tmp9, tmp10)
tl.store(out_ptr0 + x0, tmp11, xmask)
@triton.jit
def triton_poi_fused__unsafe_index_add_convolution_mul_sub_15(in_out_ptr0,
in_out_ptr1, in_out_ptr2, in_out_ptr3, in_ptr0, in_ptr1, in_ptr2,
in_ptr3, in_ptr4, in_ptr5, in_ptr6, in_ptr7, in_ptr8, in_ptr9, in_ptr10,
in_ptr11, in_ptr12, in_ptr13, in_ptr14, in_ptr15, in_ptr16, in_ptr17,
in_ptr18, in_ptr19, in_ptr20, in_ptr21, in_ptr22, in_ptr23, in_ptr24,
in_ptr25, in_ptr26, in_ptr27, in_ptr28, xnumel, XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x1 = xindex // 64 % 64
x0 = xindex % 64
x2 = xindex // 4096
x3 = xindex
tmp0 = tl.load(in_ptr0 + x1, None, eviction_policy='evict_last')
tmp5 = tl.load(in_ptr1 + x0, None, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr3 + 0)
tmp11 = tl.broadcast_to(tmp10, [XBLOCK])
tmp13 = tl.load(in_ptr4 + x0, None, eviction_policy='evict_last')
tmp20 = tl.load(in_ptr5 + x0, None, eviction_policy='evict_last')
tmp23 = tl.load(in_ptr6 + x1, None, eviction_policy='evict_last')
tmp35 = tl.load(in_ptr7 + x1, None, eviction_policy='evict_last')
tmp38 = tl.load(in_ptr8 + x1, None, eviction_policy='evict_last')
tmp43 = tl.load(in_ptr9 + x0, None, eviction_policy='evict_last')
tmp49 = tl.load(in_ptr11 + x0, None, eviction_policy='evict_last')
tmp56 = tl.load(in_ptr12 + x0, None, eviction_policy='evict_last')
tmp59 = tl.load(in_ptr13 + x1, None, eviction_policy='evict_last')
tmp71 = tl.load(in_ptr14 + x1, None, eviction_policy='evict_last')
tmp74 = tl.load(in_ptr15 + x1, None, eviction_policy='evict_last')
tmp79 = tl.load(in_ptr16 + x0, None, eviction_policy='evict_last')
tmp85 = tl.load(in_ptr18 + x0, None, eviction_policy='evict_last')
tmp92 = tl.load(in_ptr19 + x0, None, eviction_policy='evict_last')
tmp95 = tl.load(in_ptr20 + x1, None, eviction_policy='evict_last')
tmp107 = tl.load(in_ptr21 + x1, None, eviction_policy='evict_last')
tmp110 = tl.load(in_ptr22 + x1, None, eviction_policy='evict_last')
tmp115 = tl.load(in_ptr23 + x0, None, eviction_policy='evict_last')
tmp121 = tl.load(in_ptr25 + x0, None, eviction_policy='evict_last')
tmp128 = tl.load(in_ptr26 + x0, None, eviction_policy='evict_last')
tmp131 = tl.load(in_ptr27 + x1, None, eviction_policy='evict_last')
tmp143 = tl.load(in_ptr28 + x1, None, eviction_policy='evict_last')
tmp1 = tl.full([XBLOCK], 32, tl.int32)
tmp2 = tmp0 + tmp1
tmp3 = tmp0 < 0
tmp4 = tl.where(tmp3, tmp2, tmp0)
tmp6 = tmp5 + tmp1
tmp7 = tmp5 < 0
tmp8 = tl.where(tmp7, tmp6, tmp5)
tmp9 = tl.load(in_ptr2 + (tmp8 + 32 * tmp4 + 1024 * x2), None,
eviction_policy='evict_last')
tmp12 = tmp9 + tmp11
tmp14 = tmp13 + tmp1
tmp15 = tmp13 < 0
tmp16 = tl.where(tmp15, tmp14, tmp13)
tmp17 = tl.load(in_ptr2 + (tmp16 + 32 * tmp4 + 1024 * x2), None,
eviction_policy='evict_last')
tmp18 = tmp17 + tmp11
tmp19 = tmp18 - tmp12
tmp21 = tmp19 * tmp20
tmp22 = tmp12 + tmp21
tmp24 = tmp23 + tmp1
tmp25 = tmp23 < 0
tmp26 = tl.where(tmp25, tmp24, tmp23)
tmp27 = tl.load(in_ptr2 + (tmp8 + 32 * tmp26 + 1024 * x2), None,
eviction_policy='evict_last')
tmp28 = tmp27 + tmp11
tmp29 = tl.load(in_ptr2 + (tmp16 + 32 * tmp26 + 1024 * x2), None,
eviction_policy='evict_last')
tmp30 = tmp29 + tmp11
tmp31 = tmp30 - tmp28
tmp32 = tmp31 * tmp20
tmp33 = tmp28 + tmp32
tmp34 = tmp33 - tmp22
tmp36 = tmp34 * tmp35
tmp37 = tmp22 + tmp36
tmp39 = tl.full([XBLOCK], 21, tl.int32)
tmp40 = tmp38 + tmp39
tmp41 = tmp38 < 0
tmp42 = tl.where(tmp41, tmp40, tmp38)
tmp44 = tmp43 + tmp39
tmp45 = tmp43 < 0
tmp46 = tl.where(tmp45, tmp44, tmp43)
tmp47 = tl.load(in_ptr10 + (tmp46 + 21 * tmp42 + 441 * x2), None,
eviction_policy='evict_last')
tmp48 = tmp47 + tmp11
tmp50 = tmp49 + tmp39
tmp51 = tmp49 < 0
tmp52 = tl.where(tmp51, tmp50, tmp49)
tmp53 = tl.load(in_ptr10 + (tmp52 + 21 * tmp42 + 441 * x2), None,
eviction_policy='evict_last')
tmp54 = tmp53 + tmp11
tmp55 = tmp54 - tmp48
tmp57 = tmp55 * tmp56
tmp58 = tmp48 + tmp57
tmp60 = tmp59 + tmp39
tmp61 = tmp59 < 0
tmp62 = tl.where(tmp61, tmp60, tmp59)
tmp63 = tl.load(in_ptr10 + (tmp46 + 21 * tmp62 + 441 * x2), None,
eviction_policy='evict_last')
tmp64 = tmp63 + tmp11
tmp65 = tl.load(in_ptr10 + (tmp52 + 21 * tmp62 + 441 * x2), None,
eviction_policy='evict_last')
tmp66 = tmp65 + tmp11
tmp67 = tmp66 - tmp64
tmp68 = tmp67 * tmp56
tmp69 = tmp64 + tmp68
tmp70 = tmp69 - tmp58
tmp72 = tmp70 * tmp71
tmp73 = tmp58 + tmp72
tmp75 = tl.full([XBLOCK], 12, tl.int32)
tmp76 = tmp74 + tmp75
tmp77 = tmp74 < 0
tmp78 = tl.where(tmp77, tmp76, tmp74)
tmp80 = tmp79 + tmp75
tmp81 = tmp79 < 0
tmp82 = tl.where(tmp81, tmp80, tmp79)
tmp83 = tl.load(in_ptr17 + (tmp82 + 12 * tmp78 + 144 * x2), None,
eviction_policy='evict_last')
tmp84 = tmp83 + tmp11
tmp86 = tmp85 + tmp75
tmp87 = tmp85 < 0
tmp88 = tl.where(tmp87, tmp86, tmp85)
tmp89 = tl.load(in_ptr17 + (tmp88 + 12 * tmp78 + 144 * x2), None,
eviction_policy='evict_last')
tmp90 = tmp89 + tmp11
tmp91 = tmp90 - tmp84
tmp93 = tmp91 * tmp92
tmp94 = tmp84 + tmp93
tmp96 = tmp95 + tmp75
tmp97 = tmp95 < 0
tmp98 = tl.where(tmp97, tmp96, tmp95)
tmp99 = tl.load(in_ptr17 + (tmp82 + 12 * tmp98 + 144 * x2), None,
eviction_policy='evict_last')
tmp100 = tmp99 + tmp11
tmp101 = tl.load(in_ptr17 + (tmp88 + 12 * tmp98 + 144 * x2), None,
eviction_policy='evict_last')
tmp102 = tmp101 + tmp11
tmp103 = tmp102 - tmp100
tmp104 = tmp103 * tmp92
tmp105 = tmp100 + tmp104
tmp106 = tmp105 - tmp94
tmp108 = tmp106 * tmp107
tmp109 = tmp94 + tmp108
tmp111 = tl.full([XBLOCK], 10, tl.int32)
tmp112 = tmp110 + tmp111
tmp113 = tmp110 < 0
tmp114 = tl.where(tmp113, tmp112, tmp110)
tmp116 = tmp115 + tmp111
tmp117 = tmp115 < 0
tmp118 = tl.where(tmp117, tmp116, tmp115)
tmp119 = tl.load(in_ptr24 + (tmp118 + 10 * tmp114 + 100 * x2), None,
eviction_policy='evict_last')
tmp120 = tmp119 + tmp11
tmp122 = tmp121 + tmp111
tmp123 = tmp121 < 0
tmp124 = tl.where(tmp123, tmp122, tmp121)
tmp125 = tl.load(in_ptr24 + (tmp124 + 10 * tmp114 + 100 * x2), None,
eviction_policy='evict_last')
tmp126 = tmp125 + tmp11
tmp127 = tmp126 - tmp120
tmp129 = tmp127 * tmp128
tmp130 = tmp120 + tmp129
tmp132 = tmp131 + tmp111
tmp133 = tmp131 < 0
tmp134 = tl.where(tmp133, tmp132, tmp131)
tmp135 = tl.load(in_ptr24 + (tmp118 + 10 * tmp134 + 100 * x2), None,
eviction_policy='evict_last')
tmp136 = tmp135 + tmp11
tmp137 = tl.load(in_ptr24 + (tmp124 + 10 * tmp134 + 100 * x2), None,
eviction_policy='evict_last')
tmp138 = tmp137 + tmp11
tmp139 = tmp138 - tmp136
tmp140 = tmp139 * tmp128
tmp141 = tmp136 + tmp140
tmp142 = tmp141 - tmp130
tmp144 = tmp142 * tmp143
tmp145 = tmp130 + tmp144
tl.store(in_out_ptr0 + x3, tmp37, None)
tl.store(in_out_ptr1 + x3, tmp73, None)
tl.store(in_out_ptr2 + x3, tmp109, None)
tl.store(in_out_ptr3 + x3, tmp145, None)
@triton.jit
def triton_poi_fused_cat_16(in_ptr0, in_ptr1, in_ptr2, in_ptr3, in_ptr4,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x1 = xindex // 4096 % 8
x0 = xindex % 4096
x2 = xindex // 32768
x3 = xindex
tmp0 = x1
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 1, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (x0 + 4096 * x2), tmp4, eviction_policy=
'evict_last', other=0.0)
tmp6 = tmp0 >= tmp3
tmp7 = tl.full([1], 2, tl.int64)
tmp8 = tmp0 < tmp7
tmp9 = tmp6 & tmp8
tmp10 = tl.load(in_ptr1 + (x0 + 4096 * x2), tmp9, eviction_policy=
'evict_last', other=0.0)
tmp11 = tmp0 >= tmp7
tmp12 = tl.full([1], 3, tl.int64)
tmp13 = tmp0 < tmp12
tmp14 = tmp11 & tmp13
tmp15 = tl.load(in_ptr2 + (x0 + 4096 * x2), tmp14, eviction_policy=
'evict_last', other=0.0)
tmp16 = tmp0 >= tmp12
tmp17 = tl.full([1], 4, tl.int64)
tmp18 = tmp0 < tmp17
tmp19 = tmp16 & tmp18
tmp20 = tl.load(in_ptr3 + (x0 + 4096 * x2), tmp19, eviction_policy=
'evict_last', other=0.0)
tmp21 = tmp0 >= tmp17
tl.full([1], 8, tl.int64)
tmp24 = tl.load(in_ptr4 + (x0 + 4096 * (-4 + x1) + 16384 * x2), tmp21,
other=0.0)
tmp25 = tl.where(tmp19, tmp20, tmp24)
tmp26 = tl.where(tmp14, tmp15, tmp25)
tmp27 = tl.where(tmp9, tmp10, tmp26)
tmp28 = tl.where(tmp4, tmp5, tmp27)
tl.store(out_ptr0 + x3, tmp28, None)
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 64, 64), (16384, 4096, 64, 1))
assert_size_stride(primals_2, (1, 4, 1, 1), (4, 1, 1, 1))
assert_size_stride(primals_3, (1,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 32, 32), (4096, 1024, 32, 1),
torch.float32)
get_raw_stream(0)
triton_poi_fused_max_pool2d_with_indices_0[grid(16384)](primals_1,
buf0, 16384, XBLOCK=128, num_warps=4, num_stages=1)
buf1 = extern_kernels.convolution(buf0, primals_2, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf1, (4, 1, 32, 32), (1024, 1024, 32, 1))
buf2 = empty_strided_cuda((64, 1), (1, 1), torch.int64)
triton_poi_fused__to_copy_1[grid(64)](buf2, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf3 = empty_strided_cuda((64, 1), (1, 1), torch.int64)
triton_poi_fused_add_clamp_2[grid(64)](buf3, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf4 = empty_strided_cuda((64,), (1,), torch.int64)
triton_poi_fused__to_copy_1[grid(64)](buf4, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf5 = empty_strided_cuda((64,), (1,), torch.int64)
triton_poi_fused_add_clamp_2[grid(64)](buf5, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf6 = empty_strided_cuda((64,), (1,), torch.float32)
triton_poi_fused__to_copy_arange_clamp_mul_sub_3[grid(64)](buf6, 64,
XBLOCK=64, num_warps=1, num_stages=1)
buf11 = empty_strided_cuda((4, 4, 21, 21), (1792, 441, 21, 1),
torch.float32)
triton_poi_fused_max_pool2d_with_indices_4[grid(7056)](primals_1,
buf11, 7056, XBLOCK=256, num_warps=4, num_stages=1)
buf12 = extern_kernels.convolution(buf11, primals_2, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf12, (4, 1, 21, 21), (441, 441, 21, 1))
buf13 = empty_strided_cuda((64, 1), (1, 1), torch.int64)
triton_poi_fused__to_copy_5[grid(64)](buf13, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf14 = empty_strided_cuda((64, 1), (1, 1), torch.int64)
triton_poi_fused_add_clamp_6[grid(64)](buf14, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf15 = empty_strided_cuda((64,), (1,), torch.int64)
triton_poi_fused__to_copy_5[grid(64)](buf15, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf16 = empty_strided_cuda((64,), (1,), torch.int64)
triton_poi_fused_add_clamp_6[grid(64)](buf16, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf17 = empty_strided_cuda((64,), (1,), torch.float32)
triton_poi_fused__to_copy_arange_clamp_mul_sub_7[grid(64)](buf17,
64, XBLOCK=64, num_warps=1, num_stages=1)
buf19 = empty_strided_cuda((64, 1), (1, 1), torch.float32)
triton_poi_fused__to_copy_arange_clamp_mul_sub_7[grid(64)](buf19,
64, XBLOCK=64, num_warps=1, num_stages=1)
buf22 = empty_strided_cuda((4, 4, 12, 12), (576, 144, 12, 1), torch
.float32)
triton_poi_fused_max_pool2d_with_indices_8[grid(2304)](primals_1,
buf22, 2304, XBLOCK=128, num_warps=4, num_stages=1)
buf23 = extern_kernels.convolution(buf22, primals_2, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf23, (4, 1, 12, 12), (144, 144, 12, 1))
buf24 = empty_strided_cuda((64, 1), (1, 1), torch.int64)
triton_poi_fused__to_copy_9[grid(64)](buf24, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf25 = empty_strided_cuda((64, 1), (1, 1), torch.int64)
triton_poi_fused_add_clamp_10[grid(64)](buf25, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf26 = empty_strided_cuda((64,), (1,), torch.int64)
triton_poi_fused__to_copy_9[grid(64)](buf26, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf27 = empty_strided_cuda((64,), (1,), torch.int64)
triton_poi_fused_add_clamp_10[grid(64)](buf27, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf28 = empty_strided_cuda((64,), (1,), torch.float32)
triton_poi_fused__to_copy_arange_clamp_mul_sub_11[grid(64)](buf28,
64, XBLOCK=64, num_warps=1, num_stages=1)
buf30 = empty_strided_cuda((64, 1), (1, 1), torch.float32)
triton_poi_fused__to_copy_arange_clamp_mul_sub_11[grid(64)](buf30,
64, XBLOCK=64, num_warps=1, num_stages=1)
buf33 = torch.ops.aten.max_pool2d_with_indices.default(primals_1, [
6, 6], [6, 6])
buf34 = buf33[0]
del buf33
buf36 = extern_kernels.convolution(buf34, primals_2, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf36, (4, 1, 10, 10), (100, 100, 10, 1))
buf37 = empty_strided_cuda((64, 1), (1, 1), torch.int64)
triton_poi_fused__to_copy_12[grid(64)](buf37, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf38 = empty_strided_cuda((64, 1), (1, 1), torch.int64)
triton_poi_fused_add_clamp_13[grid(64)](buf38, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf39 = empty_strided_cuda((64,), (1,), torch.int64)
triton_poi_fused__to_copy_12[grid(64)](buf39, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf40 = empty_strided_cuda((64,), (1,), torch.int64)
triton_poi_fused_add_clamp_13[grid(64)](buf40, 64, XBLOCK=64,
num_warps=1, num_stages=1)
buf41 = empty_strided_cuda((64,), (1,), torch.float32)
triton_poi_fused__to_copy_arange_clamp_mul_sub_14[grid(64)](buf41,
64, XBLOCK=64, num_warps=1, num_stages=1)
buf43 = empty_strided_cuda((64, 1), (1, 1), torch.float32)
triton_poi_fused__to_copy_arange_clamp_mul_sub_14[grid(64)](buf43,
64, XBLOCK=64, num_warps=1, num_stages=1)
buf8 = empty_strided_cuda((64, 1), (1, 1), torch.float32)
triton_poi_fused__to_copy_arange_clamp_mul_sub_3[grid(64)](buf8, 64,
XBLOCK=64, num_warps=1, num_stages=1)
buf9 = empty_strided_cuda((4, 1, 64, 64), (4096, 16384, 64, 1),
torch.float32)
buf10 = reinterpret_tensor(buf9, (4, 1, 64, 64), (4096, 4096, 64, 1), 0
)
del buf9
buf20 = empty_strided_cuda((4, 1, 64, 64), (4096, 16384, 64, 1),
torch.float32)
buf21 = reinterpret_tensor(buf20, (4, 1, 64, 64), (4096, 4096, 64,
1), 0)
del buf20
buf31 = empty_strided_cuda((4, 1, 64, 64), (4096, 16384, 64, 1),
torch.float32)
buf32 = reinterpret_tensor(buf31, (4, 1, 64, 64), (4096, 4096, 64,
1), 0)
del buf31
buf44 = empty_strided_cuda((4, 1, 64, 64), (4096, 16384, 64, 1),
torch.float32)
buf45 = reinterpret_tensor(buf44, (4, 1, 64, 64), (4096, 4096, 64,
1), 0)
del buf44
triton_poi_fused__unsafe_index_add_convolution_mul_sub_15[grid(16384)](
buf10, buf21, buf32, buf45, buf2, buf4, buf1, primals_3, buf5,
buf6, buf3, buf8, buf13, buf15, buf12, buf16, buf17, buf14,
buf19, buf24, buf26, buf23, buf27, buf28, buf25, buf30, buf37,
buf39, buf36, buf40, buf41, buf38, buf43, 16384, XBLOCK=256,
num_warps=4, num_stages=1)
del buf1
del buf12
del buf23
del buf36
del primals_3
buf46 = empty_strided_cuda((4, 8, 64, 64), (32768, 4096, 64, 1),
torch.float32)
triton_poi_fused_cat_16[grid(131072)](buf10, buf21, buf32, buf45,
primals_1, buf46, 131072, XBLOCK=512, num_warps=8, num_stages=1)
del primals_1
return (buf46, buf45, buf32, buf21, buf10, primals_2, buf0, buf2, buf3,
buf4, buf5, buf6, buf8, buf11, buf13, buf14, buf15, buf16, buf17,
buf19, buf22, buf24, buf25, buf26, buf27, buf28, buf30, buf34,
buf37, buf38, buf39, buf40, buf41, buf43)
class SppBlockNew(nn.Module):
def __init__(self, in_channels):
super(SppBlockNew, self).__init__()
self.pool1 = nn.MaxPool2d(kernel_size=[2, 2], stride=2)
self.pool2 = nn.MaxPool2d(kernel_size=[3, 3], stride=3)
self.pool3 = nn.MaxPool2d(kernel_size=[5, 5], stride=5)
self.pool4 = nn.MaxPool2d(kernel_size=[6, 6], stride=6)
self.conv = nn.Conv2d(in_channels=in_channels, out_channels=1,
kernel_size=1, padding=0)
def forward(self, input_0):
primals_2 = self.conv.weight
primals_3 = self.conv.bias
primals_1 = input_0
output = call([primals_1, primals_2, primals_3])
return output[0]
|
JiYuanFeng/MCTrans
|
SppBlock
| false
| 13,963
|
[
"Apache-2.0"
] | 84
|
9b8b5677eef584b423d5e1630680a4b667cbe823
|
https://github.com/JiYuanFeng/MCTrans/tree/9b8b5677eef584b423d5e1630680a4b667cbe823
|
SymNetsCategoryLoss
|
import torch
import torch.nn.functional as F
def split_half(x, dim):
d = x.shape[dim] // 2
return torch.split(x, d, dim=dim)
class ConcatSoftmax(torch.nn.Module):
"""
Applies softmax to the concatenation of a list of tensors.
"""
def __init__(self, dim: 'int'=1):
"""
Arguments:
dim: a dimension along which softmax will be computed
"""
super().__init__()
self.dim = dim
def forward(self, *x: torch.Tensor):
"""
Arguments:
*x: A sequence of tensors to be concatenated
"""
all_logits = torch.cat(x, dim=self.dim)
return torch.nn.functional.softmax(all_logits, dim=self.dim)
def extra_repr(self):
""""""
return c_f.extra_repr(self, ['dim'])
class SymNetsCategoryLoss(torch.nn.Module):
def __init__(self):
super().__init__()
self.softmax_fn = ConcatSoftmax()
def forward(self, x, y, src_labels):
x = self.softmax_fn(x, y)
x, y = split_half(x, dim=1)
x_loss = F.cross_entropy(x, src_labels)
y_loss = F.cross_entropy(y, src_labels)
return x_loss + y_loss
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4]), torch.rand(
[4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import math as tl_math
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_per_fused__log_softmax__softmax_cat_0(in_ptr0, in_ptr1, out_ptr0,
out_ptr1, out_ptr2, out_ptr3, xnumel, rnumel, XBLOCK: tl.constexpr):
xnumel = 64
RBLOCK: tl.constexpr = 8
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r2 = rindex
x0 = xindex % 16
x1 = xindex // 16
x3 = xindex
tmp0 = r2
tmp1 = tl.full([1, 1], 0, tl.int64)
tmp3 = tl.full([1, 1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (x0 + 16 * r2 + 64 * x1), tmp4 & xmask, other=0.0)
tmp6 = tmp0 >= tmp3
tl.full([1, 1], 8, tl.int64)
tmp9 = tl.load(in_ptr1 + (x0 + 16 * (-4 + r2) + 64 * x1), tmp6 & xmask,
other=0.0)
tmp10 = tl.where(tmp4, tmp5, tmp9)
tmp11 = tl.broadcast_to(tmp10, [XBLOCK, RBLOCK])
tmp13 = tl.where(xmask, tmp11, float('-inf'))
tmp14 = triton_helpers.max2(tmp13, 1)[:, None]
tmp15 = tmp10 - tmp14
tmp16 = tl_math.exp(tmp15)
tmp17 = tl.broadcast_to(tmp16, [XBLOCK, RBLOCK])
tmp19 = tl.where(xmask, tmp17, 0)
tmp20 = tl.sum(tmp19, 1)[:, None]
tmp22 = tmp1 < tmp3
tmp23 = tl.load(in_ptr0 + (x0 + 16 * 0 + 64 * x1), tmp22 & xmask,
eviction_policy='evict_last', other=0.0)
tmp24 = tmp1 >= tmp3
tmp26 = tl.load(in_ptr1 + (x0 + 16 * -4 + 64 * x1), tmp24 & xmask,
eviction_policy='evict_last', other=0.0)
tmp27 = tl.where(tmp22, tmp23, tmp26)
tmp28 = tmp27 - tmp14
tmp29 = tl_math.exp(tmp28)
tmp30 = tmp29 / tmp20
tmp31 = tl.full([1, 1], 1, tl.int64)
tmp33 = tmp31 < tmp3
tmp34 = tl.load(in_ptr0 + (x0 + 16 * 1 + 64 * x1), tmp33 & xmask,
eviction_policy='evict_last', other=0.0)
tmp35 = tmp31 >= tmp3
tmp37 = tl.load(in_ptr1 + (x0 + 16 * -3 + 64 * x1), tmp35 & xmask,
eviction_policy='evict_last', other=0.0)
tmp38 = tl.where(tmp33, tmp34, tmp37)
tmp39 = tmp38 - tmp14
tmp40 = tl_math.exp(tmp39)
tmp41 = tmp40 / tmp20
tmp42 = triton_helpers.maximum(tmp30, tmp41)
tmp43 = tl.full([1, 1], 2, tl.int64)
tmp45 = tmp43 < tmp3
tmp46 = tl.load(in_ptr0 + (x0 + 16 * 2 + 64 * x1), tmp45 & xmask,
eviction_policy='evict_last', other=0.0)
tmp47 = tmp43 >= tmp3
tmp49 = tl.load(in_ptr1 + (x0 + 16 * -2 + 64 * x1), tmp47 & xmask,
eviction_policy='evict_last', other=0.0)
tmp50 = tl.where(tmp45, tmp46, tmp49)
tmp51 = tmp50 - tmp14
tmp52 = tl_math.exp(tmp51)
tmp53 = tmp52 / tmp20
tmp54 = triton_helpers.maximum(tmp42, tmp53)
tmp55 = tl.full([1, 1], 3, tl.int64)
tmp57 = tmp55 < tmp3
tmp58 = tl.load(in_ptr0 + (x0 + 16 * 3 + 64 * x1), tmp57 & xmask,
eviction_policy='evict_last', other=0.0)
tmp59 = tmp55 >= tmp3
tmp61 = tl.load(in_ptr1 + (x0 + 16 * -1 + 64 * x1), tmp59 & xmask,
eviction_policy='evict_last', other=0.0)
tmp62 = tl.where(tmp57, tmp58, tmp61)
tmp63 = tmp62 - tmp14
tmp64 = tl_math.exp(tmp63)
tmp65 = tmp64 / tmp20
tmp66 = triton_helpers.maximum(tmp54, tmp65)
tmp68 = tmp3 < tmp3
tmp69 = tl.load(in_ptr0 + (x0 + 16 * 4 + 64 * x1), tmp68 & xmask,
eviction_policy='evict_last', other=0.0)
tmp70 = tmp3 >= tmp3
tmp72 = tl.load(in_ptr1 + (x0 + 16 * 0 + 64 * x1), tmp70 & xmask,
eviction_policy='evict_last', other=0.0)
tmp73 = tl.where(tmp68, tmp69, tmp72)
tmp74 = tmp73 - tmp14
tmp75 = tl_math.exp(tmp74)
tmp76 = tmp75 / tmp20
tmp77 = tl.full([1, 1], 5, tl.int64)
tmp79 = tmp77 < tmp3
tmp80 = tl.load(in_ptr0 + (x0 + 16 * 5 + 64 * x1), tmp79 & xmask,
eviction_policy='evict_last', other=0.0)
tmp81 = tmp77 >= tmp3
tmp83 = tl.load(in_ptr1 + (x0 + 16 * 1 + 64 * x1), tmp81 & xmask,
eviction_policy='evict_last', other=0.0)
tmp84 = tl.where(tmp79, tmp80, tmp83)
tmp85 = tmp84 - tmp14
tmp86 = tl_math.exp(tmp85)
tmp87 = tmp86 / tmp20
tmp88 = triton_helpers.maximum(tmp76, tmp87)
tmp89 = tl.full([1, 1], 6, tl.int64)
tmp91 = tmp89 < tmp3
tmp92 = tl.load(in_ptr0 + (x0 + 16 * 6 + 64 * x1), tmp91 & xmask,
eviction_policy='evict_last', other=0.0)
tmp93 = tmp89 >= tmp3
tmp95 = tl.load(in_ptr1 + (x0 + 16 * 2 + 64 * x1), tmp93 & xmask,
eviction_policy='evict_last', other=0.0)
tmp96 = tl.where(tmp91, tmp92, tmp95)
tmp97 = tmp96 - tmp14
tmp98 = tl_math.exp(tmp97)
tmp99 = tmp98 / tmp20
tmp100 = triton_helpers.maximum(tmp88, tmp99)
tmp101 = tl.full([1, 1], 7, tl.int64)
tmp103 = tmp101 < tmp3
tmp104 = tl.load(in_ptr0 + (x0 + 16 * 7 + 64 * x1), tmp103 & xmask,
eviction_policy='evict_last', other=0.0)
tmp105 = tmp101 >= tmp3
tmp107 = tl.load(in_ptr1 + (x0 + 16 * 3 + 64 * x1), tmp105 & xmask,
eviction_policy='evict_last', other=0.0)
tmp108 = tl.where(tmp103, tmp104, tmp107)
tmp109 = tmp108 - tmp14
tmp110 = tl_math.exp(tmp109)
tmp111 = tmp110 / tmp20
tmp112 = triton_helpers.maximum(tmp100, tmp111)
tl.store(out_ptr2 + x3, tmp66, xmask)
tl.store(out_ptr3 + x3, tmp112, xmask)
tl.store(out_ptr0 + x3, tmp14, xmask)
tl.store(out_ptr1 + x3, tmp20, xmask)
@triton.jit
def triton_poi_fused__log_softmax_1(in_ptr0, in_ptr1, in_ptr2, in_ptr3,
in_ptr4, in_ptr5, out_ptr0, out_ptr1, xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 16 % 4
x0 = xindex % 16
x2 = xindex // 64
x3 = xindex
tmp11 = tl.load(in_ptr2 + (x0 + 16 * x2), xmask, eviction_policy=
'evict_last')
tmp14 = tl.load(in_ptr3 + (x0 + 16 * x2), xmask, eviction_policy=
'evict_last')
tmp16 = tl.load(in_ptr4 + (x0 + 16 * x2), xmask, eviction_policy=
'evict_last')
tmp29 = tl.load(in_ptr5 + (x0 + 16 * x2), xmask, eviction_policy=
'evict_last')
tmp0 = x1
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (x0 + 16 * x1 + 64 * x2), tmp4 & xmask, other=0.0)
tmp6 = tmp0 >= tmp3
tl.full([1], 8, tl.int64)
tmp9 = tl.load(in_ptr1 + (x0 + 16 * (-4 + x1) + 64 * x2), tmp6 & xmask,
other=0.0)
tmp10 = tl.where(tmp4, tmp5, tmp9)
tmp12 = tmp10 - tmp11
tmp13 = tl_math.exp(tmp12)
tmp15 = tmp13 / tmp14
tmp17 = tmp15 - tmp16
tmp18 = 4 + x1
tmp20 = tmp18 < tmp3
tmp21 = tl.load(in_ptr0 + (x0 + 16 * (4 + x1) + 64 * x2), tmp20 & xmask,
other=0.0)
tmp22 = tmp18 >= tmp3
tmp24 = tl.load(in_ptr1 + (x0 + 16 * x1 + 64 * x2), tmp22 & xmask,
other=0.0)
tmp25 = tl.where(tmp20, tmp21, tmp24)
tmp26 = tmp25 - tmp11
tmp27 = tl_math.exp(tmp26)
tmp28 = tmp27 / tmp14
tmp30 = tmp28 - tmp29
tl.store(out_ptr0 + x3, tmp17, xmask)
tl.store(out_ptr1 + x3, tmp30, xmask)
@triton.jit
def triton_per_fused__log_softmax_add_div_mul_neg_sum_2(in_out_ptr0,
in_ptr0, in_ptr1, in_ptr2, xnumel, rnumel):
XBLOCK: tl.constexpr = 1
RBLOCK: tl.constexpr = 256
xoffset = tl.program_id(0) * XBLOCK
tl.full([1], xoffset, tl.int32)
tl.full([RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[:]
tl.full([RBLOCK], True, tl.int1)
r3 = rindex
r0 = rindex % 16
r2 = rindex // 64
tmp0 = tl.load(in_ptr0 + r3, None)
tmp1 = tl.load(in_ptr0 + (r0 + 64 * r2), None, eviction_policy='evict_last'
)
tmp3 = tl.load(in_ptr0 + (16 + r0 + 64 * r2), None, eviction_policy=
'evict_last')
tmp6 = tl.load(in_ptr0 + (32 + r0 + 64 * r2), None, eviction_policy=
'evict_last')
tmp9 = tl.load(in_ptr0 + (48 + r0 + 64 * r2), None, eviction_policy=
'evict_last')
tmp14 = tl.load(in_ptr1 + r3, None)
tmp19 = tl.load(in_ptr2 + r3, None)
tmp20 = tl.load(in_ptr2 + (r0 + 64 * r2), None, eviction_policy=
'evict_last')
tmp22 = tl.load(in_ptr2 + (16 + r0 + 64 * r2), None, eviction_policy=
'evict_last')
tmp25 = tl.load(in_ptr2 + (32 + r0 + 64 * r2), None, eviction_policy=
'evict_last')
tmp28 = tl.load(in_ptr2 + (48 + r0 + 64 * r2), None, eviction_policy=
'evict_last')
tmp2 = tl_math.exp(tmp1)
tmp4 = tl_math.exp(tmp3)
tmp5 = tmp2 + tmp4
tmp7 = tl_math.exp(tmp6)
tmp8 = tmp5 + tmp7
tmp10 = tl_math.exp(tmp9)
tmp11 = tmp8 + tmp10
tmp12 = tl_math.log(tmp11)
tmp13 = tmp0 - tmp12
tmp15 = tmp13 * tmp14
tmp16 = tl.broadcast_to(tmp15, [RBLOCK])
tmp18 = triton_helpers.promote_to_tensor(tl.sum(tmp16, 0))
tmp21 = tl_math.exp(tmp20)
tmp23 = tl_math.exp(tmp22)
tmp24 = tmp21 + tmp23
tmp26 = tl_math.exp(tmp25)
tmp27 = tmp24 + tmp26
tmp29 = tl_math.exp(tmp28)
tmp30 = tmp27 + tmp29
tmp31 = tl_math.log(tmp30)
tmp32 = tmp19 - tmp31
tmp33 = tmp32 * tmp14
tmp34 = tl.broadcast_to(tmp33, [RBLOCK])
tmp36 = triton_helpers.promote_to_tensor(tl.sum(tmp34, 0))
tmp37 = -tmp18
tmp38 = 0.015625
tmp39 = tmp37 * tmp38
tmp40 = -tmp36
tmp41 = tmp40 * tmp38
tmp42 = tmp39 + tmp41
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([1], 0, tl.int32), tmp42, None)
def call(args):
arg0_1, arg1_1, arg2_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg2_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 1, 4, 4), (16, 64, 4, 1), torch.float32)
buf1 = empty_strided_cuda((4, 1, 4, 4), (16, 64, 4, 1), torch.float32)
buf2 = empty_strided_cuda((4, 1, 4, 4), (16, 64, 4, 1), torch.float32)
buf5 = empty_strided_cuda((4, 1, 4, 4), (16, 64, 4, 1), torch.float32)
get_raw_stream(0)
triton_per_fused__log_softmax__softmax_cat_0[grid(64)](arg0_1,
arg1_1, buf0, buf1, buf2, buf5, 64, 8, XBLOCK=1, num_warps=2,
num_stages=1)
buf3 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
buf6 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused__log_softmax_1[grid(256)](arg0_1, arg1_1, buf0,
buf1, buf2, buf5, buf3, buf6, 256, XBLOCK=128, num_warps=4,
num_stages=1)
del arg0_1
del arg1_1
del buf0
del buf1
del buf2
del buf5
buf4 = empty_strided_cuda((), (), torch.float32)
buf8 = buf4
del buf4
triton_per_fused__log_softmax_add_div_mul_neg_sum_2[grid(1)](buf8,
buf3, arg2_1, buf6, 1, 256, num_warps=2, num_stages=1)
del arg2_1
del buf3
del buf6
return buf8,
def split_half(x, dim):
d = x.shape[dim] // 2
return torch.split(x, d, dim=dim)
class ConcatSoftmax(torch.nn.Module):
"""
Applies softmax to the concatenation of a list of tensors.
"""
def __init__(self, dim: 'int'=1):
"""
Arguments:
dim: a dimension along which softmax will be computed
"""
super().__init__()
self.dim = dim
def forward(self, *x: torch.Tensor):
"""
Arguments:
*x: A sequence of tensors to be concatenated
"""
all_logits = torch.cat(x, dim=self.dim)
return torch.nn.functional.softmax(all_logits, dim=self.dim)
def extra_repr(self):
""""""
return c_f.extra_repr(self, ['dim'])
class SymNetsCategoryLossNew(torch.nn.Module):
def __init__(self):
super().__init__()
self.softmax_fn = ConcatSoftmax()
def forward(self, input_0, input_1, input_2):
arg0_1 = input_0
arg1_1 = input_1
arg2_1 = input_2
output = call([arg0_1, arg1_1, arg2_1])
return output[0]
|
KevinMusgrave/pytorch-adapt
|
SymNetsCategoryLoss
| false
| 13,964
|
[
"MIT"
] | 131
|
ff1491e1bfcc586afb8ee619712c8816ddf10358
|
https://github.com/KevinMusgrave/pytorch-adapt/tree/ff1491e1bfcc586afb8ee619712c8816ddf10358
|
Snake
|
import torch
import torch.nn as nn
from torch import sin
from torch import pow
from torch.nn import Parameter
from torch.distributions.exponential import Exponential
class Snake(nn.Module):
"""
Implementation of the serpentine-like sine-based periodic activation function
.. math::
Snake_a := x + rac{1}{a} sin^2(ax) = x - rac{1}{2a}cos{2ax} + rac{1}{2a}
Shape:
- Input: (N, *) where * means, any number of additional
dimensions
- Output: (N, *), same shape as the input
Parameters:
- a - trainable parameter
References:
- This activation function is from this paper by Liu Ziyin, Tilman Hartwig, Masahito Ueda:
https://arxiv.org/abs/2006.08195
Examples:
>>> a1 = snake(256)
>>> x = torch.randn(256)
>>> x = a1(x)
"""
def __init__(self, in_features, a=None, trainable=True):
"""
Initialization.
Args:
in_features: shape of the input
a: trainable parameter
trainable: sets `a` as a trainable parameter
`a` is initialized to 1 by default, higher values = higher-frequency,
5-50 is a good starting point if you already think your data is periodic,
consider starting lower e.g. 0.5 if you think not, but don't worry,
`a` will be trained along with the rest of your model.
"""
super(Snake, self).__init__()
self.in_features = in_features if isinstance(in_features, list) else [
in_features]
if a is not None:
self.a = Parameter(torch.ones(self.in_features) * a)
else:
m = Exponential(torch.tensor([0.1]))
self.a = Parameter(m.rsample(self.in_features).squeeze())
self.a.requiresGrad = trainable
def forward(self, x):
"""
Forward pass of the function.
Applies the function to the input elementwise.
Snake ∶= x + 1/a* sin^2 (xa)
"""
return x + 1.0 / self.a * pow(sin(x * self.a), 2)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_features': 4}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import math as tl_math
import torch.nn as nn
from torch.nn import Parameter
from torch.distributions.exponential import Exponential
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_add_mul_pow_reciprocal_sin_0(in_ptr0, in_ptr1,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr1 + x0, xmask, eviction_policy='evict_last')
tmp2 = tl.full([1], 1, tl.int32)
tmp3 = tmp2 / tmp1
tmp4 = 1.0
tmp5 = tmp3 * tmp4
tmp6 = tmp0 * tmp1
tmp7 = tl_math.sin(tmp6)
tmp8 = tmp7 * tmp7
tmp9 = tmp5 * tmp8
tmp10 = tmp0 + tmp9
tl.store(out_ptr0 + x2, tmp10, xmask)
def call(args):
primals_1, primals_2 = args
args.clear()
assert_size_stride(primals_1, (4,), (1,))
assert_size_stride(primals_2, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_add_mul_pow_reciprocal_sin_0[grid(256)](primals_2,
primals_1, buf0, 256, XBLOCK=256, num_warps=4, num_stages=1)
return buf0, primals_1, primals_2
class SnakeNew(nn.Module):
"""
Implementation of the serpentine-like sine-based periodic activation function
.. math::
Snake_a := x + rac{1}{a} sin^2(ax) = x - rac{1}{2a}cos{2ax} + rac{1}{2a}
Shape:
- Input: (N, *) where * means, any number of additional
dimensions
- Output: (N, *), same shape as the input
Parameters:
- a - trainable parameter
References:
- This activation function is from this paper by Liu Ziyin, Tilman Hartwig, Masahito Ueda:
https://arxiv.org/abs/2006.08195
Examples:
>>> a1 = snake(256)
>>> x = torch.randn(256)
>>> x = a1(x)
"""
def __init__(self, in_features, a=None, trainable=True):
"""
Initialization.
Args:
in_features: shape of the input
a: trainable parameter
trainable: sets `a` as a trainable parameter
`a` is initialized to 1 by default, higher values = higher-frequency,
5-50 is a good starting point if you already think your data is periodic,
consider starting lower e.g. 0.5 if you think not, but don't worry,
`a` will be trained along with the rest of your model.
"""
super(SnakeNew, self).__init__()
self.in_features = in_features if isinstance(in_features, list) else [
in_features]
if a is not None:
self.a = Parameter(torch.ones(self.in_features) * a)
else:
m = Exponential(torch.tensor([0.1]))
self.a = Parameter(m.rsample(self.in_features).squeeze())
self.a.requiresGrad = trainable
def forward(self, input_0):
primals_1 = self.a
primals_2 = input_0
output = call([primals_1, primals_2])
return output[0]
|
Juju-botu/diffeqml-research
|
Snake
| false
| 13,965
|
[
"Apache-2.0"
] | 49
|
aa796c87447e5299ec4f25a07fc4d032afb1f63e
|
https://github.com/Juju-botu/diffeqml-research/tree/aa796c87447e5299ec4f25a07fc4d032afb1f63e
|
MNISTFeatures
|
import torch
import torch.nn.functional as F
import torch.nn as nn
class MNISTFeatures(nn.Module):
"""
A small convnet for extracting features
from MNIST.
"""
def __init__(self):
""" """
super().__init__()
self.conv1 = nn.Conv2d(3, 32, 5, 1)
self.conv2 = nn.Conv2d(32, 48, 5, 1)
self.fc = nn.Identity()
def forward(self, x):
""" """
x = self.conv1(x)
x = F.relu(x)
x = F.max_pool2d(x, kernel_size=2, stride=2)
x = self.conv2(x)
x = F.relu(x)
x = F.max_pool2d(x, kernel_size=2, stride=2)
x = torch.flatten(x, start_dim=1)
return self.fc(x)
def get_inputs():
return [torch.rand([4, 3, 64, 64])]
def get_init_inputs():
return [[], {}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_convolution_relu_0(in_ptr0, in_ptr1, out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x3 = xindex
x1 = xindex // 3600 % 32
x0 = xindex % 3600
x4 = xindex // 3600
tmp0 = tl.load(in_ptr0 + x3, None)
tmp1 = tl.load(in_ptr1 + x1, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(out_ptr0 + (x0 + 3616 * x4), tmp4, None)
@triton.jit
def triton_poi_fused_max_pool2d_with_indices_1(in_ptr0, out_ptr0, out_ptr1,
xnumel, XBLOCK: tl.constexpr):
xnumel = 115200
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 30
x1 = xindex // 30 % 30
x2 = xindex // 900
x3 = xindex
tmp0 = tl.load(in_ptr0 + (2 * x0 + 120 * x1 + 3616 * x2), xmask,
eviction_policy='evict_last')
tmp1 = tl.load(in_ptr0 + (1 + 2 * x0 + 120 * x1 + 3616 * x2), xmask,
eviction_policy='evict_last')
tmp3 = tl.load(in_ptr0 + (60 + 2 * x0 + 120 * x1 + 3616 * x2), xmask,
eviction_policy='evict_last')
tmp5 = tl.load(in_ptr0 + (61 + 2 * x0 + 120 * x1 + 3616 * x2), xmask,
eviction_policy='evict_last')
tmp2 = triton_helpers.maximum(tmp1, tmp0)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp6 = triton_helpers.maximum(tmp5, tmp4)
tmp7 = tmp1 > tmp0
tmp8 = tl.full([1], 1, tl.int8)
tmp9 = tl.full([1], 0, tl.int8)
tmp10 = tl.where(tmp7, tmp8, tmp9)
tmp11 = tmp3 > tmp2
tmp12 = tl.full([1], 2, tl.int8)
tmp13 = tl.where(tmp11, tmp12, tmp10)
tmp14 = tmp5 > tmp4
tmp15 = tl.full([1], 3, tl.int8)
tmp16 = tl.where(tmp14, tmp15, tmp13)
tl.store(out_ptr0 + x3, tmp6, xmask)
tl.store(out_ptr1 + x3, tmp16, xmask)
@triton.jit
def triton_poi_fused_convolution_relu_2(in_out_ptr0, in_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 129792
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = xindex // 676 % 48
tmp0 = tl.load(in_out_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + x1, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tl.store(in_out_ptr0 + x3, tmp4, xmask)
@triton.jit
def triton_poi_fused_max_pool2d_with_indices_3(in_ptr0, out_ptr0, out_ptr1,
xnumel, XBLOCK: tl.constexpr):
xnumel = 32448
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 13
x3 = xindex // 13
x2 = xindex // 8112
x4 = xindex % 8112
x5 = xindex
tmp0 = tl.load(in_ptr0 + (2 * x0 + 52 * x3), xmask, eviction_policy=
'evict_last')
tmp1 = tl.load(in_ptr0 + (1 + 2 * x0 + 52 * x3), xmask, eviction_policy
='evict_last')
tmp7 = tl.load(in_ptr0 + (26 + 2 * x0 + 52 * x3), xmask,
eviction_policy='evict_last')
tmp12 = tl.load(in_ptr0 + (27 + 2 * x0 + 52 * x3), xmask,
eviction_policy='evict_last')
tmp2 = tmp1 > tmp0
tmp3 = tl.full([1], 1, tl.int8)
tmp4 = tl.full([1], 0, tl.int8)
tmp5 = tl.where(tmp2, tmp3, tmp4)
tmp6 = triton_helpers.maximum(tmp1, tmp0)
tmp8 = tmp7 > tmp6
tmp9 = tl.full([1], 2, tl.int8)
tmp10 = tl.where(tmp8, tmp9, tmp5)
tmp11 = triton_helpers.maximum(tmp7, tmp6)
tmp13 = tmp12 > tmp11
tmp14 = tl.full([1], 3, tl.int8)
tmp15 = tl.where(tmp13, tmp14, tmp10)
tmp16 = triton_helpers.maximum(tmp12, tmp11)
tl.store(out_ptr0 + (x4 + 8192 * x2), tmp15, xmask)
tl.store(out_ptr1 + x5, tmp16, xmask)
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5 = args
args.clear()
assert_size_stride(primals_1, (32, 3, 5, 5), (75, 25, 5, 1))
assert_size_stride(primals_2, (32,), (1,))
assert_size_stride(primals_3, (4, 3, 64, 64), (12288, 4096, 64, 1))
assert_size_stride(primals_4, (48, 32, 5, 5), (800, 25, 5, 1))
assert_size_stride(primals_5, (48,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(primals_3, primals_1, stride=(1,
1), padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 32, 60, 60), (115200, 3600, 60, 1))
buf1 = empty_strided_cuda((4, 32, 60, 60), (115712, 3616, 60, 1),
torch.float32)
get_raw_stream(0)
triton_poi_fused_convolution_relu_0[grid(460800)](buf0, primals_2,
buf1, 460800, XBLOCK=1024, num_warps=4, num_stages=1)
del buf0
del primals_2
buf2 = empty_strided_cuda((4, 32, 30, 30), (28800, 900, 30, 1),
torch.float32)
buf3 = empty_strided_cuda((4, 32, 30, 30), (28800, 900, 30, 1),
torch.int8)
triton_poi_fused_max_pool2d_with_indices_1[grid(115200)](buf1, buf2,
buf3, 115200, XBLOCK=512, num_warps=8, num_stages=1)
buf4 = extern_kernels.convolution(buf2, primals_4, stride=(1, 1),
padding=(0, 0), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf4, (4, 48, 26, 26), (32448, 676, 26, 1))
buf5 = buf4
del buf4
triton_poi_fused_convolution_relu_2[grid(129792)](buf5, primals_5,
129792, XBLOCK=1024, num_warps=4, num_stages=1)
del primals_5
buf6 = empty_strided_cuda((4, 48, 13, 13), (8192, 169, 13, 1),
torch.int8)
buf7 = empty_strided_cuda((4, 48, 13, 13), (8112, 169, 13, 1),
torch.float32)
triton_poi_fused_max_pool2d_with_indices_3[grid(32448)](buf5, buf6,
buf7, 32448, XBLOCK=128, num_warps=4, num_stages=1)
return reinterpret_tensor(buf7, (4, 8112), (8112, 1), 0
), primals_1, primals_3, primals_4, buf1, buf2, buf3, buf5, buf6
class MNISTFeaturesNew(nn.Module):
"""
A small convnet for extracting features
from MNIST.
"""
def __init__(self):
""" """
super().__init__()
self.conv1 = nn.Conv2d(3, 32, 5, 1)
self.conv2 = nn.Conv2d(32, 48, 5, 1)
self.fc = nn.Identity()
def forward(self, input_0):
primals_1 = self.conv1.weight
primals_2 = self.conv1.bias
primals_4 = self.conv2.weight
primals_5 = self.conv2.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4, primals_5])
return output[0]
|
KevinMusgrave/pytorch-adapt
|
MNISTFeatures
| false
| 13,966
|
[
"MIT"
] | 131
|
ff1491e1bfcc586afb8ee619712c8816ddf10358
|
https://github.com/KevinMusgrave/pytorch-adapt/tree/ff1491e1bfcc586afb8ee619712c8816ddf10358
|
MultiHeadAttention
|
import torch
import numpy as np
import torch.nn as nn
def dot_product_attention(queries, keys, values, normalise=True):
"""
:param queries:[batch_size, N_target, key_size]
:param keys:[batch_size, N_context, key_size]
:param values: []
:param normalise:
:return:
"""
key_size = keys.shape[-1]
scale = np.sqrt(key_size)
unnorm_weights = torch.matmul(queries, keys.transpose(-2, -1)) / scale
if normalise:
attention = torch.softmax(unnorm_weights, dim=-1)
else:
attention = torch.sigmoid(unnorm_weights)
output = torch.matmul(attention, values)
return output
class MultiHeadAttention(nn.Module):
"""
Multi-head attention class
"""
def __init__(self, key_size, value_size, num_heads, key_hidden_size,
normalise=True):
"""
:param num_heads:
:param normalise:
"""
super().__init__()
self._key_size = key_size
self._value_size = value_size
self._num_heads = num_heads
self._key_hidden_size = key_hidden_size
self._head_size = int(self._value_size / self._num_heads)
self._normalise = normalise
self._query_transform = nn.Linear(self._key_size, self._num_heads *
self._key_hidden_size, bias=False)
self._key_transform = nn.Linear(self._key_size, self._num_heads *
self._key_hidden_size, bias=False)
self._value_transform = nn.Linear(self._value_size, self._num_heads *
self._head_size, bias=False)
self._head_transform = nn.Linear(self._num_heads * self._head_size,
self._value_size, bias=False)
def forward(self, queries, keys=None, values=None):
"""
:param queries: [batch_size, N_target, key_size]
:param keys: [batch_size, N_context, key_size]
:param values: [batch_size, N_context, value_size]
:return:
"""
if keys is None:
keys = queries
if values is None:
values = queries
self._batch_size = queries.shape[0]
self._n_target = queries.shape[1]
self._n_context = keys.shape[1]
queries = self._query_transform(queries).view(self._batch_size,
self._n_target, self._num_heads, self._key_hidden_size)
keys = self._key_transform(keys).view(self._batch_size, self.
_n_context, self._num_heads, self._key_hidden_size)
values = self._value_transform(values).view(self._batch_size, self.
_n_context, self._num_heads, self._head_size)
queries = queries.transpose(1, 2)
keys = keys.transpose(1, 2)
values = values.transpose(1, 2)
attention = dot_product_attention(queries, keys, values, normalise=
self._normalise)
attention = attention.transpose(1, 2)
attention = attention.reshape(self._batch_size, self._n_target, -1)
output = self._head_transform(attention)
return output
def get_inputs():
return [torch.rand([4, 4, 4])]
def get_init_inputs():
return [[], {'key_size': 4, 'value_size': 4, 'num_heads': 4,
'key_hidden_size': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import math as tl_math
import numpy as np
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_clone_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex % 4
x1 = xindex // 4 % 4
x2 = xindex // 16 % 4
x3 = xindex // 64
x4 = xindex
tmp0 = tl.load(in_ptr0 + (x0 + 4 * x2 + 16 * x1 + 64 * x3), xmask)
tl.store(out_ptr0 + x4, tmp0, xmask)
@triton.jit
def triton_poi_fused_clone_1(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK: tl.
constexpr, XBLOCK: tl.constexpr):
ynumel = 64
xnumel = 4
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 16
y1 = yindex // 16
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + 16 * x2 + 64 * y1), xmask & ymask,
eviction_policy='evict_last')
tl.store(out_ptr0 + (x2 + 4 * y3), tmp0, xmask & ymask)
@triton.jit
def triton_poi_fused__softmax_sqrt_2(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.
constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp8 = tl.load(in_ptr0 + 4 * x1, xmask, eviction_policy='evict_last')
tmp10 = tl.load(in_ptr0 + (1 + 4 * x1), xmask, eviction_policy='evict_last'
)
tmp13 = tl.load(in_ptr0 + (2 + 4 * x1), xmask, eviction_policy='evict_last'
)
tmp16 = tl.load(in_ptr0 + (3 + 4 * x1), xmask, eviction_policy='evict_last'
)
tmp1 = tl.full([1], 2.0, tl.float64)
tmp2 = tl.full([1], 0.0, tl.float64)
tmp3 = tmp1 >= tmp2
tmp4 = 1.0
tmp5 = -1.0
tmp6 = tl.where(tmp3, tmp4, tmp5)
tmp7 = tmp0 * tmp6
tmp9 = tmp8 * tmp6
tmp11 = tmp10 * tmp6
tmp12 = triton_helpers.maximum(tmp9, tmp11)
tmp14 = tmp13 * tmp6
tmp15 = triton_helpers.maximum(tmp12, tmp14)
tmp17 = tmp16 * tmp6
tmp18 = triton_helpers.maximum(tmp15, tmp17)
tmp19 = tmp7 - tmp18
tmp20 = tmp6.to(tl.float64)
tmp21 = tmp20 * tmp1
tmp22 = tmp21.to(tl.float32)
tmp23 = tmp19 / tmp22
tmp24 = tl_math.exp(tmp23)
tl.store(out_ptr0 + x2, tmp24, xmask)
@triton.jit
def triton_poi_fused__softmax_3(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.constexpr
):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x1 = xindex // 4
tmp0 = tl.load(in_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + 4 * x1, xmask, eviction_policy='evict_last')
tmp2 = tl.load(in_ptr0 + (1 + 4 * x1), xmask, eviction_policy='evict_last')
tmp4 = tl.load(in_ptr0 + (2 + 4 * x1), xmask, eviction_policy='evict_last')
tmp6 = tl.load(in_ptr0 + (3 + 4 * x1), xmask, eviction_policy='evict_last')
tmp3 = tmp1 + tmp2
tmp5 = tmp3 + tmp4
tmp7 = tmp5 + tmp6
tmp8 = tmp0 / tmp7
tl.store(out_ptr0 + x2, tmp8, xmask)
@triton.jit
def triton_poi_fused_clone_4(in_ptr0, out_ptr0, ynumel, xnumel, YBLOCK: tl.
constexpr, XBLOCK: tl.constexpr):
ynumel = 16
xnumel = 4
yoffset = tl.program_id(1) * YBLOCK
yindex = yoffset + tl.arange(0, YBLOCK)[None, :]
ymask = yindex < ynumel
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:, None]
xmask = xindex < xnumel
x2 = xindex
y0 = yindex % 4
y1 = yindex // 4
y3 = yindex
tmp0 = tl.load(in_ptr0 + (y0 + 4 * x2 + 16 * y1), xmask & ymask,
eviction_policy='evict_last')
tl.store(out_ptr0 + (x2 + 4 * y3), tmp0, xmask & ymask)
def call(args):
primals_1, primals_2, primals_3, primals_4, primals_5 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4), (16, 4, 1))
assert_size_stride(primals_2, (16, 4), (4, 1))
assert_size_stride(primals_3, (16, 4), (4, 1))
assert_size_stride(primals_4, (4, 4), (4, 1))
assert_size_stride(primals_5, (4, 4), (4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((16, 16), (16, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_1, (16, 4), (4, 1), 0),
reinterpret_tensor(primals_2, (4, 16), (1, 4), 0), out=buf0)
del primals_2
buf1 = empty_strided_cuda((16, 16), (16, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_1, (16, 4), (4, 1), 0),
reinterpret_tensor(primals_3, (4, 16), (1, 4), 0), out=buf1)
del primals_3
buf2 = empty_strided_cuda((16, 4), (4, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_1, (16, 4), (4, 1), 0),
reinterpret_tensor(primals_4, (4, 4), (1, 4), 0), out=buf2)
del primals_4
buf3 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_clone_0[grid(256)](buf0, buf3, 256, XBLOCK=256,
num_warps=4, num_stages=1)
buf4 = reinterpret_tensor(buf0, (4, 4, 4, 4), (64, 16, 4, 1), 0)
del buf0
triton_poi_fused_clone_1[grid(64, 4)](buf1, buf4, 64, 4, XBLOCK=4,
YBLOCK=32, num_warps=4, num_stages=1)
buf5 = reinterpret_tensor(buf1, (16, 4, 4), (16, 4, 1), 0)
del buf1
extern_kernels.bmm(reinterpret_tensor(buf3, (16, 4, 4), (16, 4, 1),
0), reinterpret_tensor(buf4, (16, 4, 4), (16, 4, 1), 0), out=buf5)
buf6 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused__softmax_sqrt_2[grid(256)](buf5, buf6, 256, XBLOCK
=128, num_warps=4, num_stages=1)
buf7 = reinterpret_tensor(buf5, (4, 4, 4, 4), (64, 16, 4, 1), 0)
del buf5
triton_poi_fused__softmax_3[grid(256)](buf6, buf7, 256, XBLOCK=256,
num_warps=4, num_stages=1)
del buf6
buf8 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 1), torch.float32)
triton_poi_fused_clone_4[grid(16, 4)](buf2, buf8, 16, 4, XBLOCK=4,
YBLOCK=16, num_warps=1, num_stages=1)
buf9 = reinterpret_tensor(buf2, (16, 4, 1), (4, 1, 1), 0)
del buf2
extern_kernels.bmm(reinterpret_tensor(buf7, (16, 4, 4), (16, 4, 1),
0), reinterpret_tensor(buf8, (16, 4, 1), (4, 1, 0), 0), out=buf9)
buf10 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
triton_poi_fused_clone_4[grid(16, 4)](buf9, buf10, 16, 4, XBLOCK=4,
YBLOCK=16, num_warps=1, num_stages=1)
buf11 = reinterpret_tensor(buf9, (16, 4), (4, 1), 0)
del buf9
extern_kernels.mm(reinterpret_tensor(buf10, (16, 4), (4, 1), 0),
reinterpret_tensor(primals_5, (4, 4), (1, 4), 0), out=buf11)
return reinterpret_tensor(buf11, (4, 4, 4), (16, 4, 1), 0
), reinterpret_tensor(primals_1, (16, 4), (4, 1), 0
), buf7, reinterpret_tensor(buf10, (16, 4), (4, 1), 0
), primals_5, reinterpret_tensor(buf8, (16, 1, 4), (4, 1, 1), 0
), reinterpret_tensor(buf3, (16, 4, 4), (16, 1, 4), 0
), reinterpret_tensor(buf4, (16, 4, 4), (16, 1, 4), 0)
def dot_product_attention(queries, keys, values, normalise=True):
"""
:param queries:[batch_size, N_target, key_size]
:param keys:[batch_size, N_context, key_size]
:param values: []
:param normalise:
:return:
"""
key_size = keys.shape[-1]
scale = np.sqrt(key_size)
unnorm_weights = torch.matmul(queries, keys.transpose(-2, -1)) / scale
if normalise:
attention = torch.softmax(unnorm_weights, dim=-1)
else:
attention = torch.sigmoid(unnorm_weights)
output = torch.matmul(attention, values)
return output
class MultiHeadAttentionNew(nn.Module):
"""
Multi-head attention class
"""
def __init__(self, key_size, value_size, num_heads, key_hidden_size,
normalise=True):
"""
:param num_heads:
:param normalise:
"""
super().__init__()
self._key_size = key_size
self._value_size = value_size
self._num_heads = num_heads
self._key_hidden_size = key_hidden_size
self._head_size = int(self._value_size / self._num_heads)
self._normalise = normalise
self._query_transform = nn.Linear(self._key_size, self._num_heads *
self._key_hidden_size, bias=False)
self._key_transform = nn.Linear(self._key_size, self._num_heads *
self._key_hidden_size, bias=False)
self._value_transform = nn.Linear(self._value_size, self._num_heads *
self._head_size, bias=False)
self._head_transform = nn.Linear(self._num_heads * self._head_size,
self._value_size, bias=False)
def forward(self, input_0):
primals_2 = self._query_transform.weight
primals_3 = self._key_transform.weight
primals_4 = self._value_transform.weight
primals_5 = self._head_transform.weight
primals_1 = input_0
output = call([primals_1, primals_2, primals_3, primals_4, primals_5])
return output[0]
|
JuliusSchwartz/FlowMO
|
MultiHeadAttention
| false
| 13,967
|
[
"MIT"
] | 53
|
e221d989914f906501e1ad19cd3629d88eac1785
|
https://github.com/JuliusSchwartz/FlowMO/tree/e221d989914f906501e1ad19cd3629d88eac1785
|
PerformanceModel
|
import torch
import torch.nn as nn
class PerformanceModel(nn.Module):
def __init__(self, input_len):
super(PerformanceModel, self).__init__()
self.input_len = input_len
self.linear1 = nn.Linear(self.input_len, 32, bias=True)
self.dropout1 = nn.Dropout(p=0.01)
self.activate1 = torch.relu
self.linear2 = nn.Linear(32, 64, bias=True)
self.dropout2 = nn.Dropout(p=0.01)
self.activate2 = torch.relu
self.linear3 = nn.Linear(64, 128, bias=True)
self.dropout3 = nn.Dropout(p=0.01)
self.activate3 = torch.relu
self.linear4 = nn.Linear(128, 64, bias=True)
self.dropout4 = nn.Dropout(p=0.01)
self.activate4 = torch.relu
self.linear5 = nn.Linear(64, 16, bias=True)
self.activate5 = torch.relu
self.linear6 = nn.Linear(16, 1, bias=True)
self.activate6 = torch.relu
def forward(self, inputs):
output1 = self.activate1(self.dropout1(self.linear1(inputs)))
output2 = self.activate2(self.dropout2(self.linear2(output1)))
output3 = self.activate3(self.dropout3(self.linear3(output2)))
output4 = self.activate4(self.dropout4(self.linear4(output3)))
output5 = self.activate5(self.linear5(output4))
output6 = self.activate6(self.linear6(output5))
return output6
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'input_len': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
reinterpret_tensor = torch._C._dynamo.guards._reinterpret_tensor
@triton.jit
def triton_poi_fused_relu_threshold_backward_0(in_out_ptr0, in_ptr0,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x2 = xindex
x0 = xindex % 32
tmp0 = tl.load(in_out_ptr0 + x2, None)
tmp1 = tl.load(in_ptr0 + x0, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(in_out_ptr0 + x2, tmp4, None)
tl.store(out_ptr0 + x2, tmp6, None)
@triton.jit
def triton_poi_fused_relu_threshold_backward_1(in_out_ptr0, in_ptr0,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x2 = xindex
x0 = xindex % 64
tmp0 = tl.load(in_out_ptr0 + x2, None)
tmp1 = tl.load(in_ptr0 + x0, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(in_out_ptr0 + x2, tmp4, None)
tl.store(out_ptr0 + x2, tmp6, None)
@triton.jit
def triton_poi_fused_relu_threshold_backward_2(in_out_ptr0, in_ptr0,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
tl.full([XBLOCK], True, tl.int1)
x2 = xindex
x0 = xindex % 128
tmp0 = tl.load(in_out_ptr0 + x2, None)
tmp1 = tl.load(in_ptr0 + x0, None, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(in_out_ptr0 + x2, tmp4, None)
tl.store(out_ptr0 + x2, tmp6, None)
@triton.jit
def triton_poi_fused_relu_threshold_backward_3(in_out_ptr0, in_ptr0,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 1024
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x2 = xindex
x0 = xindex % 16
tmp0 = tl.load(in_out_ptr0 + x2, xmask)
tmp1 = tl.load(in_ptr0 + x0, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp3 = tl.full([1], 0, tl.int32)
tmp4 = triton_helpers.maximum(tmp3, tmp2)
tmp5 = 0.0
tmp6 = tmp4 <= tmp5
tl.store(in_out_ptr0 + x2, tmp4, xmask)
tl.store(out_ptr0 + x2, tmp6, xmask)
@triton.jit
def triton_poi_fused_relu_threshold_backward_4(in_out_ptr0, in_ptr0,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 64
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x0 = xindex
tmp0 = tl.load(in_out_ptr0 + x0, xmask)
tmp1 = tl.load(in_ptr0 + 0)
tmp2 = tl.broadcast_to(tmp1, [XBLOCK])
tmp3 = tmp0 + tmp2
tmp4 = tl.full([1], 0, tl.int32)
tmp5 = triton_helpers.maximum(tmp4, tmp3)
tmp6 = 0.0
tmp7 = tmp5 <= tmp6
tl.store(in_out_ptr0 + x0, tmp5, xmask)
tl.store(out_ptr0 + x0, tmp7, xmask)
def call(args):
(primals_1, primals_2, primals_3, primals_4, primals_5, primals_6,
primals_7, primals_8, primals_9, primals_10, primals_11, primals_12,
primals_13) = args
args.clear()
assert_size_stride(primals_1, (32, 4), (4, 1))
assert_size_stride(primals_2, (32,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (64, 32), (32, 1))
assert_size_stride(primals_5, (64,), (1,))
assert_size_stride(primals_6, (128, 64), (64, 1))
assert_size_stride(primals_7, (128,), (1,))
assert_size_stride(primals_8, (64, 128), (128, 1))
assert_size_stride(primals_9, (64,), (1,))
assert_size_stride(primals_10, (16, 64), (64, 1))
assert_size_stride(primals_11, (16,), (1,))
assert_size_stride(primals_12, (1, 16), (16, 1))
assert_size_stride(primals_13, (1,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((64, 32), (32, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(primals_3, (64, 4), (4, 1), 0),
reinterpret_tensor(primals_1, (4, 32), (1, 4), 0), out=buf0)
del primals_1
buf1 = reinterpret_tensor(buf0, (4, 4, 4, 32), (512, 128, 32, 1), 0)
del buf0
buf17 = empty_strided_cuda((4, 4, 4, 32), (512, 128, 32, 1), torch.bool
)
get_raw_stream(0)
triton_poi_fused_relu_threshold_backward_0[grid(2048)](buf1,
primals_2, buf17, 2048, XBLOCK=256, num_warps=4, num_stages=1)
del primals_2
buf2 = empty_strided_cuda((64, 64), (64, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf1, (64, 32), (32, 1), 0),
reinterpret_tensor(primals_4, (32, 64), (1, 32), 0), out=buf2)
buf3 = reinterpret_tensor(buf2, (4, 4, 4, 64), (1024, 256, 64, 1), 0)
del buf2
buf16 = empty_strided_cuda((4, 4, 4, 64), (1024, 256, 64, 1), torch
.bool)
triton_poi_fused_relu_threshold_backward_1[grid(4096)](buf3,
primals_5, buf16, 4096, XBLOCK=256, num_warps=4, num_stages=1)
del primals_5
buf4 = empty_strided_cuda((64, 128), (128, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf3, (64, 64), (64, 1), 0),
reinterpret_tensor(primals_6, (64, 128), (1, 64), 0), out=buf4)
buf5 = reinterpret_tensor(buf4, (4, 4, 4, 128), (2048, 512, 128, 1), 0)
del buf4
buf15 = empty_strided_cuda((4, 4, 4, 128), (2048, 512, 128, 1),
torch.bool)
triton_poi_fused_relu_threshold_backward_2[grid(8192)](buf5,
primals_7, buf15, 8192, XBLOCK=256, num_warps=4, num_stages=1)
del primals_7
buf6 = empty_strided_cuda((64, 64), (64, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf5, (64, 128), (128, 1), 0),
reinterpret_tensor(primals_8, (128, 64), (1, 128), 0), out=buf6)
buf7 = reinterpret_tensor(buf6, (4, 4, 4, 64), (1024, 256, 64, 1), 0)
del buf6
buf14 = empty_strided_cuda((4, 4, 4, 64), (1024, 256, 64, 1), torch
.bool)
triton_poi_fused_relu_threshold_backward_1[grid(4096)](buf7,
primals_9, buf14, 4096, XBLOCK=256, num_warps=4, num_stages=1)
del primals_9
buf8 = empty_strided_cuda((64, 16), (16, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf7, (64, 64), (64, 1), 0),
reinterpret_tensor(primals_10, (64, 16), (1, 64), 0), out=buf8)
buf9 = reinterpret_tensor(buf8, (4, 4, 4, 16), (256, 64, 16, 1), 0)
del buf8
buf13 = empty_strided_cuda((4, 4, 4, 16), (256, 64, 16, 1), torch.bool)
triton_poi_fused_relu_threshold_backward_3[grid(1024)](buf9,
primals_11, buf13, 1024, XBLOCK=128, num_warps=4, num_stages=1)
del primals_11
buf10 = empty_strided_cuda((64, 1), (1, 1), torch.float32)
extern_kernels.mm(reinterpret_tensor(buf9, (64, 16), (16, 1), 0),
reinterpret_tensor(primals_12, (16, 1), (1, 16), 0), out=buf10)
buf11 = reinterpret_tensor(buf10, (4, 4, 4, 1), (16, 4, 1, 1), 0)
del buf10
buf12 = empty_strided_cuda((4, 4, 4, 1), (16, 4, 1, 1), torch.bool)
triton_poi_fused_relu_threshold_backward_4[grid(64)](buf11,
primals_13, buf12, 64, XBLOCK=64, num_warps=1, num_stages=1)
del primals_13
return (buf11, reinterpret_tensor(primals_3, (64, 4), (4, 1), 0),
reinterpret_tensor(buf1, (64, 32), (32, 1), 0), reinterpret_tensor(
buf3, (64, 64), (64, 1), 0), reinterpret_tensor(buf5, (64, 128), (
128, 1), 0), reinterpret_tensor(buf7, (64, 64), (64, 1), 0),
reinterpret_tensor(buf9, (64, 16), (16, 1), 0), buf12, primals_12,
buf13, primals_10, buf14, primals_8, buf15, primals_6, buf16,
primals_4, buf17)
class PerformanceModelNew(nn.Module):
def __init__(self, input_len):
super(PerformanceModelNew, self).__init__()
self.input_len = input_len
self.linear1 = nn.Linear(self.input_len, 32, bias=True)
self.dropout1 = nn.Dropout(p=0.01)
self.activate1 = torch.relu
self.linear2 = nn.Linear(32, 64, bias=True)
self.dropout2 = nn.Dropout(p=0.01)
self.activate2 = torch.relu
self.linear3 = nn.Linear(64, 128, bias=True)
self.dropout3 = nn.Dropout(p=0.01)
self.activate3 = torch.relu
self.linear4 = nn.Linear(128, 64, bias=True)
self.dropout4 = nn.Dropout(p=0.01)
self.activate4 = torch.relu
self.linear5 = nn.Linear(64, 16, bias=True)
self.activate5 = torch.relu
self.linear6 = nn.Linear(16, 1, bias=True)
self.activate6 = torch.relu
def forward(self, input_0):
primals_1 = self.linear1.weight
primals_2 = self.linear1.bias
primals_4 = self.linear2.weight
primals_5 = self.linear2.bias
primals_6 = self.linear3.weight
primals_7 = self.linear3.bias
primals_8 = self.linear4.weight
primals_9 = self.linear4.bias
primals_10 = self.linear5.weight
primals_11 = self.linear5.bias
primals_12 = self.linear6.weight
primals_13 = self.linear6.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4,
primals_5, primals_6, primals_7, primals_8, primals_9,
primals_10, primals_11, primals_12, primals_13])
return output[0]
|
KnowingNothing/FlexTensor
|
PerformanceModel
| false
| 13,968
|
[
"MIT"
] | 135
|
00f6cd7e038af7714b833fde7034d465fe2dc4a7
|
https://github.com/KnowingNothing/FlexTensor/tree/00f6cd7e038af7714b833fde7034d465fe2dc4a7
|
BinaryLoss
|
import torch
import torch.nn as nn
import torch.nn.functional as F
class BinaryLoss(nn.Module):
def __init__(self):
super(BinaryLoss, self).__init__()
def forward(self, pos_score, neg_score):
pos_loss = -F.log_softmax(pos_score)[:, 1]
neg_loss = -F.log_softmax(neg_score)[:, 0]
loss = (pos_loss.sum() + neg_loss.sum()) / (pos_loss.size(0) +
neg_loss.size(0))
return loss
def get_inputs():
return [torch.rand([4, 4, 4, 4]), torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime import triton_helpers
from torch._inductor.runtime.triton_helpers import math as tl_math
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused__log_softmax_0(in_ptr0, out_ptr0, xnumel, XBLOCK: tl.
constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x0 = xindex % 16
x2 = xindex // 64
tmp0 = tl.load(in_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + (x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp2 = tl.load(in_ptr0 + (16 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp4 = tl.load(in_ptr0 + (32 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp6 = tl.load(in_ptr0 + (48 + x0 + 64 * x2), xmask, eviction_policy=
'evict_last')
tmp3 = triton_helpers.maximum(tmp1, tmp2)
tmp5 = triton_helpers.maximum(tmp3, tmp4)
tmp7 = triton_helpers.maximum(tmp5, tmp6)
tmp8 = tmp0 - tmp7
tl.store(out_ptr0 + x3, tmp8, xmask)
@triton.jit
def triton_per_fused_add_div_neg_sum_1(in_out_ptr0, in_ptr0, in_ptr1,
xnumel, rnumel, XBLOCK: tl.constexpr):
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex % 16
r1 = rindex // 16
tmp0 = tl.load(in_ptr0 + (16 + r0 + 64 * r1), None)
tmp1 = tl.load(in_ptr0 + (r0 + 64 * r1), None)
tmp5 = tl.load(in_ptr0 + (32 + r0 + 64 * r1), None)
tmp8 = tl.load(in_ptr0 + (48 + r0 + 64 * r1), None)
tmp17 = tl.load(in_ptr1 + (r0 + 64 * r1), None)
tmp19 = tl.load(in_ptr1 + (16 + r0 + 64 * r1), None)
tmp22 = tl.load(in_ptr1 + (32 + r0 + 64 * r1), None)
tmp25 = tl.load(in_ptr1 + (48 + r0 + 64 * r1), None)
tmp2 = tl_math.exp(tmp1)
tmp3 = tl_math.exp(tmp0)
tmp4 = tmp2 + tmp3
tmp6 = tl_math.exp(tmp5)
tmp7 = tmp4 + tmp6
tmp9 = tl_math.exp(tmp8)
tmp10 = tmp7 + tmp9
tmp11 = tl_math.log(tmp10)
tmp12 = tmp0 - tmp11
tmp13 = -tmp12
tmp14 = tl.broadcast_to(tmp13, [XBLOCK, RBLOCK])
tmp16 = tl.sum(tmp14, 1)[:, None]
tmp18 = tl_math.exp(tmp17)
tmp20 = tl_math.exp(tmp19)
tmp21 = tmp18 + tmp20
tmp23 = tl_math.exp(tmp22)
tmp24 = tmp21 + tmp23
tmp26 = tl_math.exp(tmp25)
tmp27 = tmp24 + tmp26
tmp28 = tl_math.log(tmp27)
tmp29 = tmp17 - tmp28
tmp30 = -tmp29
tmp31 = tl.broadcast_to(tmp30, [XBLOCK, RBLOCK])
tmp33 = tl.sum(tmp31, 1)[:, None]
tmp34 = tmp16 + tmp33
tmp35 = 0.125
tmp36 = tmp34 * tmp35
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([XBLOCK, 1], 0, tl.int32), tmp36, None)
def call(args):
arg0_1, arg1_1 = args
args.clear()
assert_size_stride(arg0_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(arg1_1, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused__log_softmax_0[grid(256)](arg0_1, buf0, 256,
XBLOCK=128, num_warps=4, num_stages=1)
del arg0_1
buf2 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
triton_poi_fused__log_softmax_0[grid(256)](arg1_1, buf2, 256,
XBLOCK=128, num_warps=4, num_stages=1)
del arg1_1
buf1 = empty_strided_cuda((), (), torch.float32)
buf4 = buf1
del buf1
triton_per_fused_add_div_neg_sum_1[grid(1)](buf4, buf0, buf2, 1, 64,
XBLOCK=1, num_warps=2, num_stages=1)
del buf0
del buf2
return buf4,
class BinaryLossNew(nn.Module):
def __init__(self):
super(BinaryLossNew, self).__init__()
def forward(self, input_0, input_1):
arg0_1 = input_0
arg1_1 = input_1
output = call([arg0_1, arg1_1])
return output[0]
|
Kitware/VAIME
|
BinaryLoss
| false
| 13,969
|
[
"BSD-3-Clause"
] | 127
|
47b24b9d8a208cf8c621e5bb1088c61fcf507af6
|
https://github.com/Kitware/VAIME/tree/47b24b9d8a208cf8c621e5bb1088c61fcf507af6
|
Conv2dZeros
|
import torch
import torch.nn as nn
class Conv2dZeros(nn.Module):
"""Normal conv2d for reparameterize the latent variable.
- weight and bias initialized to zero
- scale channel-wise after conv2d
"""
def __init__(self, in_channels, out_channels):
super(Conv2dZeros, self).__init__()
self.conv = nn.Conv2d(in_channels, out_channels, kernel_size=3,
stride=1, padding=1, bias=True)
self.conv.weight.data.zero_()
self.conv.bias.data.zero_()
self.scale = nn.Parameter(torch.zeros(1, out_channels, 1, 1))
def forward(self, x):
x = self.conv(x)
return x * torch.exp(self.scale * 3)
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'in_channels': 4, 'out_channels': 4}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import math as tl_math
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_convolution_exp_mul_0(in_out_ptr0, in_ptr0, in_ptr1,
out_ptr0, xnumel, XBLOCK: tl.constexpr):
xnumel = 256
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x3 = xindex
x1 = xindex // 16 % 4
tmp0 = tl.load(in_out_ptr0 + x3, xmask)
tmp1 = tl.load(in_ptr0 + x1, xmask, eviction_policy='evict_last')
tmp3 = tl.load(in_ptr1 + x1, xmask, eviction_policy='evict_last')
tmp2 = tmp0 + tmp1
tmp4 = 3.0
tmp5 = tmp3 * tmp4
tmp6 = tl_math.exp(tmp5)
tmp7 = tmp2 * tmp6
tl.store(in_out_ptr0 + x3, tmp2, xmask)
tl.store(out_ptr0 + x3, tmp7, xmask)
def call(args):
primals_1, primals_2, primals_3, primals_4 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 3, 3), (36, 9, 3, 1))
assert_size_stride(primals_2, (4,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_4, (1, 4, 1, 1), (4, 1, 1, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(primals_3, primals_1, stride=(1,
1), padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 4, 4, 4), (64, 16, 4, 1))
buf1 = buf0
del buf0
buf2 = empty_strided_cuda((4, 4, 4, 4), (64, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_convolution_exp_mul_0[grid(256)](buf1, primals_2,
primals_4, buf2, 256, XBLOCK=256, num_warps=4, num_stages=1)
del primals_2
return buf2, primals_1, primals_3, primals_4, buf1
class Conv2dZerosNew(nn.Module):
"""Normal conv2d for reparameterize the latent variable.
- weight and bias initialized to zero
- scale channel-wise after conv2d
"""
def __init__(self, in_channels, out_channels):
super(Conv2dZerosNew, self).__init__()
self.conv = nn.Conv2d(in_channels, out_channels, kernel_size=3,
stride=1, padding=1, bias=True)
self.conv.weight.data.zero_()
self.conv.bias.data.zero_()
self.scale = nn.Parameter(torch.zeros(1, out_channels, 1, 1))
def forward(self, input_0):
primals_4 = self.scale
primals_1 = self.conv.weight
primals_2 = self.conv.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3, primals_4])
return output[0]
|
KyleDavisSA/pde-surrogate
|
Conv2dZeros
| false
| 13,970
|
[
"MIT"
] | 62
|
41ad2c9eb73c323e389174080f4b3df6cbd3c900
|
https://github.com/KyleDavisSA/pde-surrogate/tree/41ad2c9eb73c323e389174080f4b3df6cbd3c900
|
RingLoss
|
import torch
import torch.nn as nn
class RingLoss(nn.Module):
"""Ring loss.
Reference:
Zheng et al. Ring loss: Convex Feature Normalization for Face Recognition. CVPR 2018.
"""
def __init__(self, weight_ring=1.0):
super(RingLoss, self).__init__()
self.radius = nn.Parameter(torch.ones(1, dtype=torch.float))
self.weight_ring = weight_ring
def forward(self, x):
l = ((x.norm(p=2, dim=1) - self.radius) ** 2).mean()
return l * self.weight_ring
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {}]
|
import torch
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
from torch._inductor.runtime.triton_helpers import libdevice
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_per_fused_linalg_vector_norm_mean_mul_pow_sub_0(in_out_ptr0,
in_ptr0, in_ptr1, out_ptr1, xnumel, rnumel, XBLOCK: tl.constexpr):
RBLOCK: tl.constexpr = 64
xoffset = tl.program_id(0) * XBLOCK
xoffset + tl.arange(0, XBLOCK)[:, None]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
rindex = tl.arange(0, RBLOCK)[None, :]
tl.full([XBLOCK, RBLOCK], True, tl.int1)
r0 = rindex % 16
r1 = rindex // 16
r2 = rindex
tmp0 = tl.load(in_ptr0 + (r0 + 64 * r1), None)
tmp2 = tl.load(in_ptr0 + (16 + r0 + 64 * r1), None)
tmp5 = tl.load(in_ptr0 + (32 + r0 + 64 * r1), None)
tmp8 = tl.load(in_ptr0 + (48 + r0 + 64 * r1), None)
tmp12 = tl.load(in_ptr1 + 0)
tmp13 = tl.broadcast_to(tmp12, [XBLOCK, RBLOCK])
tmp1 = tmp0 * tmp0
tmp3 = tmp2 * tmp2
tmp4 = tmp1 + tmp3
tmp6 = tmp5 * tmp5
tmp7 = tmp4 + tmp6
tmp9 = tmp8 * tmp8
tmp10 = tmp7 + tmp9
tmp11 = libdevice.sqrt(tmp10)
tmp14 = tmp11 - tmp13
tmp15 = 2.0
tmp16 = tmp14 * tmp15
tmp17 = tmp14 * tmp14
tmp18 = tl.broadcast_to(tmp17, [XBLOCK, RBLOCK])
tmp20 = tl.sum(tmp18, 1)[:, None]
tmp21 = 64.0
tmp22 = tmp20 / tmp21
tmp23 = 1.0
tmp24 = tmp22 * tmp23
tl.store(out_ptr1 + tl.broadcast_to(r2, [XBLOCK, RBLOCK]), tmp16, None)
tl.debug_barrier()
tl.store(in_out_ptr0 + tl.full([XBLOCK, 1], 0, tl.int32), tmp24, None)
def call(args):
primals_1, primals_2 = args
args.clear()
assert_size_stride(primals_1, (4, 4, 4, 4), (64, 16, 4, 1))
assert_size_stride(primals_2, (1,), (1,))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf2 = empty_strided_cuda((4, 4, 4), (16, 4, 1), torch.float32)
buf1 = empty_strided_cuda((), (), torch.float32)
buf3 = buf1
del buf1
get_raw_stream(0)
triton_per_fused_linalg_vector_norm_mean_mul_pow_sub_0[grid(1)](buf3,
primals_1, primals_2, buf2, 1, 64, XBLOCK=1, num_warps=2,
num_stages=1)
del primals_1
del primals_2
return buf3, buf2
class RingLossNew(nn.Module):
"""Ring loss.
Reference:
Zheng et al. Ring loss: Convex Feature Normalization for Face Recognition. CVPR 2018.
"""
def __init__(self, weight_ring=1.0):
super(RingLossNew, self).__init__()
self.radius = nn.Parameter(torch.ones(1, dtype=torch.float))
self.weight_ring = weight_ring
def forward(self, input_0):
primals_2 = self.radius
primals_1 = input_0
output = call([primals_1, primals_2])
return output[0]
|
LT1st/ReID_Alined_beginer
|
RingLoss
| false
| 13,971
|
[
"MIT"
] | 370
|
1a12403a32d99900451ac05cd3623a9b770f6d24
|
https://github.com/LT1st/ReID_Alined_beginer/tree/1a12403a32d99900451ac05cd3623a9b770f6d24
|
_DenseBlockInput
|
import torch
import torch.nn as nn
class _DenseLayer(nn.Sequential):
"""One dense layer within dense block, with bottleneck design.
Args:
in_features (int):
growth_rate (int): # out feature maps of every dense layer
drop_rate (float):
bn_size (int): Specifies maximum # features is `bn_size` *
`growth_rate`
bottleneck (bool, False): If True, enable bottleneck design
"""
def __init__(self, in_features, growth_rate, drop_rate=0.0, bn_size=8,
bottleneck=False):
super(_DenseLayer, self).__init__()
if bottleneck and in_features > bn_size * growth_rate:
self.add_module('norm1', nn.BatchNorm2d(in_features))
self.add_module('relu1', nn.ReLU(inplace=True))
self.add_module('conv1', nn.Conv2d(in_features, bn_size *
growth_rate, kernel_size=1, stride=1, bias=False))
self.add_module('norm2', nn.BatchNorm2d(bn_size * growth_rate))
self.add_module('relu2', nn.ReLU(inplace=True))
self.add_module('conv2', nn.Conv2d(bn_size * growth_rate,
growth_rate, kernel_size=3, stride=1, padding=1, bias=False))
else:
self.add_module('norm1', nn.BatchNorm2d(in_features))
self.add_module('relu1', nn.ReLU(inplace=True))
self.add_module('conv1', nn.Conv2d(in_features, growth_rate,
kernel_size=3, stride=1, padding=1, bias=False))
if drop_rate > 0:
self.add_module('dropout', nn.Dropout2d(p=drop_rate))
def forward(self, x):
y = super(_DenseLayer, self).forward(x)
return torch.cat([x, y], 1)
class _DenseBlockInput(nn.Sequential):
"""For input dense block, feature map size the same as input"""
def __init__(self, num_layers, in_features, init_features, growth_rate,
drop_rate, bn_size=4, bottleneck=False):
super(_DenseBlockInput, self).__init__()
self.num_layers = num_layers
self.add_module('in_conv', nn.Conv2d(in_features, init_features - 1,
kernel_size=3, stride=1, padding=1))
for i in range(num_layers - 1):
layer = _DenseLayer(init_features + i * growth_rate,
growth_rate, drop_rate=drop_rate, bn_size=bn_size,
bottleneck=bottleneck)
self.add_module(f'denselayer{i + 1}', layer)
def forward(self, x):
out = self.in_conv(x)
out = torch.cat((x, out), 1)
for i in range(self.num_layers - 1):
out = self[i + 1](out)
return out
def get_inputs():
return [torch.rand([4, 4, 4, 4])]
def get_init_inputs():
return [[], {'num_layers': 1, 'in_features': 4, 'init_features': 4,
'growth_rate': 4, 'drop_rate': 0.5}]
|
import torch
from torch._inductor.select_algorithm import extern_kernels
import triton
import triton.language as tl
from torch._inductor.runtime.triton_heuristics import grid
from torch._C import _cuda_getCurrentRawStream as get_raw_stream
import torch.nn as nn
assert_size_stride = torch._C._dynamo.guards.assert_size_stride
empty_strided_cuda = torch._C._dynamo.guards._empty_strided_cuda
@triton.jit
def triton_poi_fused_cat_0(in_ptr0, in_ptr1, in_ptr2, out_ptr0, xnumel,
XBLOCK: tl.constexpr):
xnumel = 448
xoffset = tl.program_id(0) * XBLOCK
xindex = xoffset + tl.arange(0, XBLOCK)[:]
xmask = xindex < xnumel
x1 = xindex // 16 % 7
x0 = xindex % 16
x2 = xindex // 112
x3 = xindex
tmp0 = x1
tl.full([1], 0, tl.int64)
tmp3 = tl.full([1], 4, tl.int64)
tmp4 = tmp0 < tmp3
tmp5 = tl.load(in_ptr0 + (x0 + 16 * x1 + 64 * x2), tmp4 & xmask, other=0.0)
tmp6 = tmp0 >= tmp3
tl.full([1], 7, tl.int64)
tmp9 = tl.load(in_ptr1 + (x0 + 16 * (-4 + x1) + 48 * x2), tmp6 & xmask,
other=0.0)
tmp10 = tl.load(in_ptr2 + (-4 + x1), tmp6 & xmask, eviction_policy=
'evict_last', other=0.0)
tmp11 = tmp9 + tmp10
tmp12 = tl.full(tmp11.shape, 0.0, tmp11.dtype)
tmp13 = tl.where(tmp6, tmp11, tmp12)
tmp14 = tl.where(tmp4, tmp5, tmp13)
tl.store(out_ptr0 + x3, tmp14, xmask)
def call(args):
primals_1, primals_2, primals_3 = args
args.clear()
assert_size_stride(primals_1, (3, 4, 3, 3), (36, 9, 3, 1))
assert_size_stride(primals_2, (3,), (1,))
assert_size_stride(primals_3, (4, 4, 4, 4), (64, 16, 4, 1))
with torch.cuda._DeviceGuard(0):
torch.cuda.set_device(0)
buf0 = extern_kernels.convolution(primals_3, primals_1, stride=(1,
1), padding=(1, 1), dilation=(1, 1), transposed=False,
output_padding=(0, 0), groups=1, bias=None)
assert_size_stride(buf0, (4, 3, 4, 4), (48, 16, 4, 1))
buf1 = empty_strided_cuda((4, 7, 4, 4), (112, 16, 4, 1), torch.float32)
get_raw_stream(0)
triton_poi_fused_cat_0[grid(448)](primals_3, buf0, primals_2, buf1,
448, XBLOCK=256, num_warps=4, num_stages=1)
del buf0
del primals_2
return buf1, primals_1, primals_3
class _DenseLayer(nn.Sequential):
"""One dense layer within dense block, with bottleneck design.
Args:
in_features (int):
growth_rate (int): # out feature maps of every dense layer
drop_rate (float):
bn_size (int): Specifies maximum # features is `bn_size` *
`growth_rate`
bottleneck (bool, False): If True, enable bottleneck design
"""
def __init__(self, in_features, growth_rate, drop_rate=0.0, bn_size=8,
bottleneck=False):
super(_DenseLayer, self).__init__()
if bottleneck and in_features > bn_size * growth_rate:
self.add_module('norm1', nn.BatchNorm2d(in_features))
self.add_module('relu1', nn.ReLU(inplace=True))
self.add_module('conv1', nn.Conv2d(in_features, bn_size *
growth_rate, kernel_size=1, stride=1, bias=False))
self.add_module('norm2', nn.BatchNorm2d(bn_size * growth_rate))
self.add_module('relu2', nn.ReLU(inplace=True))
self.add_module('conv2', nn.Conv2d(bn_size * growth_rate,
growth_rate, kernel_size=3, stride=1, padding=1, bias=False))
else:
self.add_module('norm1', nn.BatchNorm2d(in_features))
self.add_module('relu1', nn.ReLU(inplace=True))
self.add_module('conv1', nn.Conv2d(in_features, growth_rate,
kernel_size=3, stride=1, padding=1, bias=False))
if drop_rate > 0:
self.add_module('dropout', nn.Dropout2d(p=drop_rate))
def forward(self, x):
y = super(_DenseLayer, self).forward(x)
return torch.cat([x, y], 1)
class _DenseBlockInputNew(nn.Sequential):
"""For input dense block, feature map size the same as input"""
def __init__(self, num_layers, in_features, init_features, growth_rate,
drop_rate, bn_size=4, bottleneck=False):
super(_DenseBlockInputNew, self).__init__()
self.num_layers = num_layers
self.add_module('in_conv', nn.Conv2d(in_features, init_features - 1,
kernel_size=3, stride=1, padding=1))
for i in range(num_layers - 1):
layer = _DenseLayer(init_features + i * growth_rate,
growth_rate, drop_rate=drop_rate, bn_size=bn_size,
bottleneck=bottleneck)
self.add_module(f'denselayer{i + 1}', layer)
def forward(self, input_0):
primals_1 = self.in_conv.weight
primals_2 = self.in_conv.bias
primals_3 = input_0
output = call([primals_1, primals_2, primals_3])
return output[0]
|
KyleDavisSA/pde-surrogate
|
_DenseBlockInput
| false
| 13,972
|
[
"MIT"
] | 62
|
41ad2c9eb73c323e389174080f4b3df6cbd3c900
|
https://github.com/KyleDavisSA/pde-surrogate/tree/41ad2c9eb73c323e389174080f4b3df6cbd3c900
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.