train.py

#!/usr/bin/env python
# coding=utf-8
# Copyright 2021 The HuggingFace Inc. team. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
"""
Fine-tuning the library models for causal language modeling (GPT, GPT-2, CTRL, ...)
on a text file or a dataset without using HuggingFace Trainer.
Here is the full list of checkpoints on the hub that can be fine-tuned by this script:
https://huggingface.co/models?filter=text-generation
"""
# You can also adapt this script on your own causal language modeling task. Pointers for this are left as comments.

import argparse
import logging
import math
import os
import random
from itertools import chain

import datasets
import torch
import numpy as np
from datasets import load_dataset
from torch.utils.data import DataLoader
from tqdm.auto import tqdm

import transformers
from accelerate import Accelerator, DistributedType
from accelerate.logging import get_logger
from accelerate.utils import set_seed
from transformers import (
    CONFIG_MAPPING,
    MODEL_MAPPING,
    AutoConfig,
    AutoModelForCausalLM,
    AutoTokenizer,
    default_data_collator,
)
from transformers.utils.versions import require_version


logger = get_logger(__name__)

require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/language-modeling/requirements.txt")

MODEL_CONFIG_CLASSES = list(MODEL_MAPPING.keys())
MODEL_TYPES = tuple(conf.model_type for conf in MODEL_CONFIG_CLASSES)


def parse_args():

    parser = argparse.ArgumentParser(description="Finetune large language models on causal language modeling tasks")
    parser.add_argument("--dataset_name", type=str, default=None, help="The name of the dataset to use (via the datasets library).")
    parser.add_argument("--dataset_config_name", type=str, default=None, help="The configuration name of the dataset to use (via the datasets library).")
    parser.add_argument("--train_file", type=str, default=None, help="A csv or a json file containing the training data.")
    parser.add_argument("--model_name_or_path", type=str, help="Path to pretrained model or model identifier from huggingface.co/models.", required=False)
    parser.add_argument("--config_name", type=str, default=None, help="Pretrained config name or path if not the same as model_name")
    parser.add_argument("--tokenizer_name", type=str, default=None, help="Pretrained tokenizer name or path if not the same as model_name")
    parser.add_argument("--use_slow_tokenizer", action="store_true", help="If passed, will use a slow tokenizer (not backed by the 🤗 Tokenizers library).")
    parser.add_argument("--per_device_train_batch_size", type=int, default=8, help="Batch size (per device) for the training dataloader.")
    parser.add_argument("--learning_rate", type=float, default=0.0001, help="Initial learning rate (after the potential warmup period) to use.")
    parser.add_argument("--weight_decay", type=float, default=0.0, help="Weight decay to use.")
    parser.add_argument("--num_train_epochs", type=int, default=1, help="Total number of training epochs to perform.")
    parser.add_argument("--max_train_steps", type=int, default=None, help="Total number of training steps to perform. If provided, overrides num_train_epochs.")
    parser.add_argument("--gradient_accumulation_steps", type=int, default=1, help="Number of updates steps to accumulate before performing a backward/update pass.")
    parser.add_argument("--output_dir", type=str, default=None, help="Where to store the final model.")
    parser.add_argument("--seed", type=int, default=None, help="A seed for reproducible training.")
    parser.add_argument("--model_type", type=str, default=None, help="Model type to use if training from scratch.", choices=MODEL_TYPES)
    parser.add_argument("--block_size", type=int, default=None, help="The training dataset will be truncated to blocks of this size (after tokenization) for training.")
    parser.add_argument("--preprocessing_num_workers", type=int, default=None, help="The number of processes to use for the preprocessing.")
    parser.add_argument("--overwrite_cache", action="store_true", help="Overwrite the cached training and evaluation sets")
    parser.add_argument("--no_keep_linebreaks", action="store_true", help="Do not keep line breaks when using TXT files.")
    parser.add_argument("--checkpointing_steps", type=str, default=None, help="Whether the various states should be saved at the end of every n steps, or 'epoch' for each epoch.")
    parser.add_argument("--resume_from_checkpoint", type=str, default=None, help="If the training should continue from a checkpoint folder.")
    parser.add_argument("--save_prefix", type=str, default='', help="Informative string prefix for saving purposes.")
    parser.add_argument("--use_pretrained_weights", type=bool, default=True, help="Whether to use pretrained weights.")

    args = parser.parse_args()

    # Sanity checks
    if args.dataset_name is None and args.train_file is None:
        raise ValueError("Need either a dataset name or a training file.")
    else:
        if args.train_file is not None:
            extension = args.train_file.split(".")[-1]
            assert extension in ["csv", "json", "txt"], "`train_file` should be a csv, json or txt file."

    return args


def main():
    args = parse_args()

    # Initialize the accelerator
    accelerator = Accelerator(gradient_accumulation_steps=args.gradient_accumulation_steps)
    
    # Make one log on every process with the configuration for debugging.
    logging.basicConfig(
        format="%(asctime)s - %(levelname)s - %(name)s - %(message)s",
        datefmt="%m/%d/%Y %H:%M:%S",
        level=logging.INFO,
    )
    logger.info(accelerator.state, main_process_only=False)
    if accelerator.is_local_main_process:
        datasets.utils.logging.set_verbosity_warning()
        transformers.utils.logging.set_verbosity_info()
    else:
        datasets.utils.logging.set_verbosity_error()
        transformers.utils.logging.set_verbosity_error()

    # If passed along, set the training seed now.
    if args.seed is not None:
        set_seed(args.seed)

    # Handle the repository creation
    if accelerator.is_main_process:
        if args.output_dir is not None:
            os.makedirs(args.output_dir, exist_ok=True)
    accelerator.wait_for_everyone()

    # Get the datasets: you can either provide your own CSV/JSON/TXT training and evaluation files (see below)
    # or just provide the name of one of the public datasets available on the hub at https://huggingface.co/datasets/
    # (the dataset will be downloaded automatically from the datasets Hub).
    #
    # For CSV/JSON files, this script will use the column called 'text' or the first column if no column called
    # 'text' is found. You can easily tweak this behavior (see below).
    #
    # In distributed training, the load_dataset function guarantee that only one local process can concurrently
    # download the dataset.
    if args.dataset_name is not None:
        # Downloading and loading a dataset from the hub.
        raw_datasets = load_dataset(args.dataset_name, args.dataset_config_name)
    else:
        data_files = {"train": args.train_file}
        dataset_args = {}
        extension = args.train_file.split(".")[-1]
        if extension == "txt":
            extension = "text"
            dataset_args["keep_linebreaks"] = not args.no_keep_linebreaks
        raw_datasets = load_dataset(extension, data_files=data_files, **dataset_args)

    # See more about loading any type of standard or custom dataset (from files, python dict, pandas DataFrame, etc) at
    # https://huggingface.co/docs/datasets/loading_datasets.html.

    # Load pretrained model and tokenizer
 
    # In distributed training, the .from_pretrained methods guarantee that only one local process can concurrently
    # download model & vocab.
    if args.config_name:
        config = AutoConfig.from_pretrained(args.config_name)
    elif args.model_name_or_path:
        config = AutoConfig.from_pretrained(args.model_name_or_path)
    else:
        config = CONFIG_MAPPING[args.model_type]()
        logger.warning("You are instantiating a new config instance from scratch.")

    if args.tokenizer_name:
        tokenizer = AutoTokenizer.from_pretrained(args.tokenizer_name, use_fast=not args.use_slow_tokenizer, model_max_length=2048)  # TODO: pass this more beautifully
    elif args.model_name_or_path:
        tokenizer = AutoTokenizer.from_pretrained(args.model_name_or_path, use_fast=not args.use_slow_tokenizer, model_max_length=2048)  # TODO: pass this more beautifully
        if args.model_name_or_path.startswith("gpt2") or args.model_name_or_path.startswith("EleutherAI"):
            tokenizer.pad_token = tokenizer.eos_token
    else:
        raise ValueError(
            "You are instantiating a new tokenizer from scratch. This is not supported by this script."
            "You can do it from another script, save it, and load it from here, using --tokenizer_name."
        )

    if args.model_name_or_path and args.use_pretrained_weights:
        model = AutoModelForCausalLM.from_pretrained(args.model_name_or_path, from_tf=bool(".ckpt" in args.model_name_or_path), config=config)
    else:
        logger.info("Training new model from scratch")
        model = AutoModelForCausalLM.from_config(config)

    model.resize_token_embeddings(len(tokenizer))

    # Preprocessing the datasets. First we tokenize all the texts.
    column_names = raw_datasets["train"].column_names
    text_column_name = "text" if "text" in column_names else column_names[0]

    if args.block_size is None:
        block_size = tokenizer.model_max_length
        if block_size > 1024:
            logger.warning(
                f"The tokenizer picked seems to have a very large `model_max_length` ({tokenizer.model_max_length})."
                "Picking 1024 instead. You can change that default value by passing --block_size xxx."
            )
            block_size = 1024
    else:
        block_size = args.block_size
        if args.block_size > tokenizer.model_max_length:
            logger.warning(
                f"The block_size passed ({args.block_size}) is larger than the maximum length for the model"
                f"({tokenizer.model_max_length}). Using block_size={tokenizer.model_max_length}."
            )
            block_size = tokenizer.model_max_length
    print('Block size:', block_size)

    # TODO: revert this back later on
    def tokenize_function(examples):
        return tokenizer(examples[text_column_name], padding='max_length', truncation=True, max_length=block_size)

    def tokenize_function_original(examples):
        return tokenizer(examples[text_column_name])

    with accelerator.main_process_first():
        tokenized_datasets = raw_datasets.map(
            tokenize_function,
            batched=True,
            num_proc=args.preprocessing_num_workers,
            remove_columns=column_names,
            load_from_cache_file=not args.overwrite_cache,
            desc="Running tokenizer on dataset",
        )

    # TODO: Might have to change this to group_text as in the original example
    # TODO: revert this back later on
    # Main data processing function.
    def preprocess_function(examples):
        examples["labels"] = examples["input_ids"].copy()
        # pad token must be set to -100 in labels to make sure it's ignored when computing the loss
        examples["labels"] = [[(l if l != tokenizer.pad_token_id else -100) for l in label] for label in examples["labels"]]
        return examples

    def preprocess_function_original(examples):
        # Concatenate all texts.
        concatenated_examples = {k: list(chain(*examples[k])) for k in examples.keys()}
        total_length = len(concatenated_examples[list(examples.keys())[0]])
        # We drop the small remainder, we could add padding if the model supported it instead of this drop, you can
        # customize this part to your needs.
        if total_length >= block_size:
            total_length = (total_length // block_size) * block_size
        # Split by chunks of max_len.
        result = {
            k: [t[i : i + block_size] for i in range(0, total_length, block_size)]
            for k, t in concatenated_examples.items()
        }
        result["labels"] = result["input_ids"].copy()
        return result
    
    with accelerator.main_process_first():
        lm_datasets = tokenized_datasets.map(
            preprocess_function,
            batched=True,
            num_proc=args.preprocessing_num_workers,
            load_from_cache_file=not args.overwrite_cache,
            desc=f"Not grouping text.",
        )

    # dataset & dataloader creation
    train_dataset = lm_datasets["train"]
    train_dataloader = DataLoader(train_dataset, shuffle=True, collate_fn=default_data_collator, batch_size=args.per_device_train_batch_size)

    # Log a few random samples from the training set:
    for index in random.sample(range(len(train_dataset)), 3):
        logger.info(f"Sample {index} of the training set: {train_dataset[index]}.")
        logger.info(f"Sample {index} of the training set (decoded): {tokenizer.decode(train_dataset[index]['input_ids'], skip_special_tokens=True)}.")

    model = accelerator.prepare(model)
    
    # Optimizer
    # Split weights in two groups, one with weight decay and the other not.
    no_decay = ["bias", "layer_norm.weight"]
    optimizer_grouped_parameters = [
        {"params": [p for n, p in model.named_parameters() if not any(nd in n for nd in no_decay)], "weight_decay": args.weight_decay},
        {"params": [p for n, p in model.named_parameters() if any(nd in n for nd in no_decay)], "weight_decay": 0.0},
    ]
    optimizer = torch.optim.AdamW(optimizer_grouped_parameters, lr=args.learning_rate)

    # Scheduler and math around the number of training steps.
    overrode_max_train_steps = False
    num_update_steps_per_epoch = math.ceil(len(train_dataloader) / args.gradient_accumulation_steps)
    if args.max_train_steps is None:
        args.max_train_steps = args.num_train_epochs * num_update_steps_per_epoch
        overrode_max_train_steps = True

    # Prepare everything with our `accelerator`.
    optimizer, train_dataloader = accelerator.prepare(optimizer, train_dataloader)

    # On TPU, the tie weights in our model have been disconnected, so we need to restore the ties.
    if accelerator.distributed_type == DistributedType.TPU:
        model.tie_weights()

    # We need to recalculate our total training steps as the size of the training dataloader may have changed.
    num_update_steps_per_epoch = math.ceil(len(train_dataloader) / args.gradient_accumulation_steps)
    if overrode_max_train_steps:
        args.max_train_steps = args.num_train_epochs * num_update_steps_per_epoch

    # Afterwards we recalculate our number of training epochs
    args.num_train_epochs = math.ceil(args.max_train_steps / num_update_steps_per_epoch)

    # Figure out how many steps we should save the Accelerator states
    checkpointing_steps = args.checkpointing_steps
    if checkpointing_steps is not None and checkpointing_steps.isdigit():
        checkpointing_steps = int(checkpointing_steps)

    # Train!
    total_batch_size = args.per_device_train_batch_size * accelerator.num_processes * args.gradient_accumulation_steps

    logger.info("***** Running training *****")
    logger.info(f"  Num examples = {len(train_dataset)}")
    logger.info(f"  Num Epochs = {args.num_train_epochs}")
    logger.info(f"  Instantaneous batch size per device = {args.per_device_train_batch_size}")
    logger.info(f"  Total train batch size (w. parallel, distributed & accumulation) = {total_batch_size}")
    logger.info(f"  Gradient Accumulation steps = {args.gradient_accumulation_steps}")
    logger.info(f"  Total optimization steps = {args.max_train_steps}")
    
    # Only show the progress bar once on each machine.
    progress_bar = tqdm(range(args.max_train_steps), disable=not accelerator.is_local_main_process)
    completed_steps = 0
    starting_epoch = 0

    # Potentially load in the weights and states from a previous save
    if args.resume_from_checkpoint:
        if args.resume_from_checkpoint is not None or args.resume_from_checkpoint != "":
            accelerator.print(f"Resumed from checkpoint: {args.resume_from_checkpoint}")
            accelerator.load_state(args.resume_from_checkpoint)
            path = os.path.basename(args.resume_from_checkpoint)
        else:
            # Get the most recent checkpoint
            dirs = [f.name for f in os.scandir(os.getcwd()) if f.is_dir()]
            dirs.sort(key=os.path.getctime)
            path = dirs[-1]  # Sorts folders by date modified, most recent checkpoint is the last
        # Extract `epoch_{i}` or `step_{i}`
        training_difference = os.path.splitext(path)[0]

        if "epoch" in training_difference:
            starting_epoch = int(training_difference.replace("epoch_", "")) + 1
            resume_step = None
        else:
            # need to multiply `gradient_accumulation_steps` to reflect real steps
            resume_step = int(training_difference.replace("step_", "")) * args.gradient_accumulation_steps
            starting_epoch = resume_step // len(train_dataloader)
            resume_step -= starting_epoch * len(train_dataloader)

    # update the progress_bar if load from checkpoint
    progress_bar.update(starting_epoch * num_update_steps_per_epoch)
    completed_steps = starting_epoch * num_update_steps_per_epoch

    train_losses = []
    train_losses_all = []

    for epoch in range(starting_epoch, args.num_train_epochs):
        model.train()
        for step, batch in enumerate(train_dataloader):
            # We need to skip steps until we reach the resumed step
            if args.resume_from_checkpoint and epoch == starting_epoch:
                if resume_step is not None and step < resume_step:
                    if step % args.gradient_accumulation_steps == 0:
                        progress_bar.update(1)
                        completed_steps += 1
                    continue

            with accelerator.accumulate(model):
                outputs = model(**batch)
                loss = outputs.loss
                # keep track of the loss at each epoch
                train_losses.append(loss.detach().unsqueeze(0))
                train_losses_all.append(loss.detach().unsqueeze(0))
                accelerator.backward(loss)
                optimizer.step()
                optimizer.zero_grad()

            # Checks if the accelerator has performed an optimization step behind the scenes
            if accelerator.sync_gradients:
                progress_bar.update(1)
                completed_steps += 1

            if isinstance(checkpointing_steps, int):
                if completed_steps % checkpointing_steps == 0:
                    output_dir = f"step_{completed_steps}"
                    if args.output_dir is not None:
                        output_dir = os.path.join(args.output_dir, output_dir)

                    # save model and tokenizer
                    accelerator.wait_for_everyone()
                    unwrapped_model = accelerator.unwrap_model(model)
                    unwrapped_model.save_pretrained(output_dir, is_main_process=accelerator.is_main_process, save_function=accelerator.save)
                    if accelerator.is_main_process:
                        tokenizer.save_pretrained(output_dir)

                    # save train_losses
                    train_losses_ckpt = torch.cat(train_losses)
                    train_losses_ckpt = train_losses_ckpt.cpu().numpy()

                    train_losses_all_ckpt = torch.cat(train_losses_all)
                    train_losses_all_ckpt = train_losses_all_ckpt.cpu().numpy()

                    logger.info(f"Mean train loss: {np.mean(train_losses_ckpt)}")

                    save_path = os.path.join(output_dir, 'train_losses.npz')
                    np.savez(save_path, train_losses=train_losses_all_ckpt, completed_steps=completed_steps)

                    # re-initialize losses
                    train_losses = []

            if completed_steps >= args.max_train_steps:
                break

        if args.checkpointing_steps == "epoch":
            output_dir = f"epoch_{epoch}"
            if args.output_dir is not None:
                output_dir = os.path.join(args.output_dir, output_dir)
            accelerator.save_state(output_dir)

            # save train_losses
            train_losses_ckpt = torch.cat(train_losses)
            train_losses_ckpt = train_losses_ckpt.cpu().numpy()
            logger.info(f"Mean train loss: {np.mean(train_losses_ckpt)}")

            save_path = os.path.join(output_dir, args.save_prefix + '_results.npz')
            np.savez(save_path, train_losses_ckpt=train_losses_ckpt, completed_steps=completed_steps)

    if args.checkpointing_steps is None and args.output_dir is not None:
        # save model and tokenizer
        accelerator.wait_for_everyone()
        unwrapped_model = accelerator.unwrap_model(model)
        unwrapped_model.save_pretrained(args.output_dir, is_main_process=accelerator.is_main_process, save_function=accelerator.save)
        if accelerator.is_main_process:
            tokenizer.save_pretrained(args.output_dir)

        # save train_losses
        train_losses_ckpt = torch.cat(train_losses)
        train_losses_ckpt = train_losses_ckpt.cpu().numpy()
        logger.info(f"Final mean train loss: {np.mean(train_losses_ckpt)}")

        # save results
        save_path = os.path.join(args.output_dir, args.save_prefix + '_results.npz')
        np.savez(save_path, train_losses_ckpt=train_losses_ckpt, completed_steps=completed_steps)


if __name__ == "__main__":
    main()