pipeline.py

import os
from contextlib import nullcontext
import torch
import tiktoken
from model import GPTConfig, GPT
from transformers import  GPT2Tokenizer, GPT2Model

start = "\n"  # or "<|endoftext|>" or etc. Can also specify a file, use as: "FILE:prompt.txt"
num_samples = 1  # number of samples to draw
max_new_tokens = 100  # number of tokens generated in each sample
temperature = 0.6  # 1.0 = no change, < 1.0 = less random, > 1.0 = more random, in predictions
top_k = 200  # retain only the top_k most likely tokens, clamp others to have 0 probability
seed = 1337
device = 'cpu'  # examples: 'cpu', 'cuda', 'cuda:0', 'cuda:1', etc.
dtype = 'float16'

output_dir = ""
model = GPT.from_pretrained(output_dir)
tokenizer = GPT2Tokenizer.from_pretrained(output_dir)

def infer():
    torch.manual_seed(seed)
    torch.cuda.manual_seed(seed)
    torch.backends.cuda.matmul.allow_tf32 = True  # allow tf32 on matmul
    torch.backends.cudnn.allow_tf32 = True  # allow tf32 on cudnn
    device_type = 'cuda' if 'cuda' in device else 'cpu'  # for later use in torch.autocast
    ptdtype = {
        'float32': torch.float32,
        'bfloat16': torch.bfloat16,
        'float16': torch.float16
    }[dtype]
    ctx = nullcontext() if device_type == 'cpu' else torch.amp.autocast(
        device_type=device_type, dtype=ptdtype)

    # ckpt_path = os.path.join(os.getcwd(), out_dir, 'ckpt.pt')
    # checkpoint = torch.load(ckpt_path, map_location=device)
    # gptconf = GPTConfig(**checkpoint['model_args'])
    # model = GPT(gptconf)
    # state_dict = checkpoint['model']
    # unwanted_prefix = '_orig_mod.'
    # for k, v in list(state_dict.items()):
    #     if k.startswith(unwanted_prefix):
    #         state_dict[k[len(unwanted_prefix):]] = state_dict.pop(k)
    # model.load_state_dict(state_dict)
    
    model.eval()
    model.to(device)

    # enc = tiktoken.get_encoding("gpt2")
    # encode = lambda s: enc.encode(s, allowed_special={"<|endoftext|>"})
    # decode = lambda l: enc.decode(l)
    encode = lambda s: tokenizer.encode(s)
    decode = lambda l: tokenizer.decode(l)

    start_ids = encode(start)
    x = (torch.tensor(start_ids, dtype=torch.long, device=device)[None, ...])

    with torch.no_grad():
        with ctx:
            for k in range(num_samples):
                y = model.generate(x,
                                max_new_tokens,
                                temperature=temperature,
                                top_k=top_k)
                return decode(y[0].tolist())