generative_train.py

#!/usr/bin/env python
# coding=utf-8

import argparse
import json
import logging
import math
import os
import re
from collections import Counter

import datasets
import numpy as np
import torch
from accelerate import Accelerator
from accelerate.logging import get_logger
from accelerate.utils import set_seed
from datasets import load_dataset
from nlgeval import NLGEval
from nltk import ngrams
from nltk.tokenize import word_tokenize
from torch.utils.data import DataLoader
from torchvision.io import ImageReadMode, read_image
from tqdm.auto import tqdm

import transformers
from transformers import (
    AutoConfig,
    AutoImageProcessor,
    AutoTokenizer,
    DataCollatorForSeq2Seq,
    SchedulerType,
    GenerationConfig,
    get_scheduler,
)

from contrastive_train import Transform
from model.modeling_zrigf import ZRIGFForConditionalGeneration

torch.backends.cuda.matmul.allow_tf32 = True

logger = get_logger(__name__)


def parse_args():
    parser = argparse.ArgumentParser(description="Finetune a transformers model on a summarization task")
    parser.add_argument(
        "--dataset_name",
        type=str,
        default=None,
        help="The name of the dataset to use (via the datasets library).",
    )
    parser.add_argument(
        "--dataset_config_name",
        type=str,
        default=None,
        help="The configuration name of the dataset to use (via the datasets library).",
    )
    parser.add_argument(
        "--train_file", type=str, default=None, help="A csv or a json file containing the training data."
    )
    parser.add_argument(
        "--test_file", type=str, default=None, help="A csv or a json file containing the testing data."
    )
    parser.add_argument(
        "--contrastive_model_path",
        type=str,
        help="Path to contrastive pretrained model",
        required=True,
    )
    parser.add_argument(
        "--ignore_pad_token_for_loss",
        type=bool,
        default=True,
        help="Whether to ignore the tokens corresponding to padded labels in the loss computation or not.",
    )
    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", type=bool, default=None, help="Overwrite the cached training and evaluation sets"
    )
    parser.add_argument(
        "--max_source_length",
        type=int,
        default=1024,
        help=(
            "The maximum total input sequence length after "
            "tokenization.Sequences longer than this will be truncated, sequences shorter will be padded."
        ),
    )
    parser.add_argument(
        "--max_target_length",
        type=int,
        default=128,
        help=(
            "The maximum total sequence length for target text after "
            "tokenization. Sequences longer than this will be truncated, sequences shorter will be padded."
            "during ``evaluate`` and ``predict``."
        ),
    )
    parser.add_argument(
        "--val_max_target_length",
        type=int,
        default=None,
        help=(
            "The maximum total sequence length for validation "
            "target text after tokenization.Sequences longer than this will be truncated, sequences shorter will be "
            "padded. Will default to `max_target_length`.This argument is also used to override the ``max_length`` "
            "param of ``model.generate``, which is used during ``evaluate`` and ``predict``."
        ),
    )
    parser.add_argument(
        "--num_beams",
        type=int,
        default=None,
        help=(
            "Number of beams to use for evaluation. This argument will be "
            "passed to ``model.generate``, which is used during ``evaluate`` and ``predict``."
        ),
    )
    parser.add_argument(
        "--pad_to_max_length",
        action="store_true",
        help="If passed, pad all samples to `max_length`. Otherwise, dynamic padding is used.",
    )
    parser.add_argument(
        "--per_device_train_batch_size",
        type=int,
        default=8,
        help="Batch size (per device) for the training dataloader.",
    )
    parser.add_argument(
        "--per_device_eval_batch_size",
        type=int,
        default=8,
        help="Batch size (per device) for the evaluation dataloader.",
    )
    parser.add_argument(
        "--learning_rate",
        type=float,
        default=5e-5,
        help="Initial learning rate (after the potential warmup period) to use.",
    )
    parser.add_argument("--weight_decay", type=float, default=0.0, help="Weight decay to use.")
    parser.add_argument("--num_train_epochs", type=int, default=3, help="Total number of training epochs to perform.")
    parser.add_argument(
        "--max_train_steps",
        type=int,
        default=None,
        help="Total number of training steps to perform. If provided, overrides num_train_epochs.",
    )
    parser.add_argument(
        "--gradient_accumulation_steps",
        type=int,
        default=1,
        help="Number of updates steps to accumulate before performing a backward/update pass.",
    )
    parser.add_argument(
        "--lr_scheduler_type",
        type=SchedulerType,
        default="linear",
        help="The scheduler type to use.",
        choices=["linear", "cosine", "cosine_with_restarts", "polynomial", "constant", "constant_with_warmup"],
    )
    parser.add_argument(
        "--num_warmup_steps", type=int, default=0, help="Number of steps for the warmup in the lr scheduler."
    )
    parser.add_argument("--output_dir", type=str, default=None, help="Where to store the final model.")
    parser.add_argument("--seed", type=int, default=None, help="A seed for reproducible training.")
    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(
        "--with_tracking",
        action="store_true",
        help="Whether to enable experiment trackers for logging.",
    )
    parser.add_argument(
        "--report_to",
        type=str,
        default="all",
        help=(
            'The integration to report the results and logs to. Supported platforms are `"tensorboard"`,'
            ' `"wandb"`, `"comet_ml"` and `"clearml"`. Use `"all"` (default) to report to all integrations.'
            "Only applicable when `--with_tracking` is passed."
        ),
    )
    parser.add_argument(
        "--label_smoothing_factor",
        type=float,
        default=0,
    )
    parser.add_argument(
        "--wo_mf",
        type=bool,
        default=False,
    )
    parser.add_argument(
        "--wo_it",
        type=bool,
        default=False,
    )
    args = parser.parse_args()

    return args


def main():
    args = parse_args()

    # Initialize the accelerator. We will let the accelerator handle device placement for us in this example.
    # If we're using tracking, we also need to initialize it here and it will by default pick up all supported trackers
    # in the environment
    accelerator_kwargs = {}
    if args.with_tracking:
        accelerator_kwargs["log_with"] = args.report_to
        accelerator_kwargs["project_dir"] = args.output_dir
    accelerator = Accelerator(gradient_accumulation_steps=args.gradient_accumulation_steps, **accelerator_kwargs)

    # 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=True)
    # datasets.utils.logging.set_verbosity_error()
    # transformers.utils.logging.set_verbosity_error()
    if accelerator.is_local_main_process:
        datasets.utils.logging.set_verbosity_warning()
        transformers.utils.logging.set_verbosity_warning()
    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)
        with open(os.path.join(args.output_dir, 'commandline_args.txt'), 'w') as f:
            json.dump(args.__dict__, f, indent=2)
    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 = {}
        if args.train_file is not None:
            data_files["train"] = args.train_file
        if args.test_file is not None:
            data_files["test"] = args.test_file
        extension = args.train_file.split(".")[-1]
        raw_datasets = load_dataset(extension, data_files=data_files)
    # See more about loading any type of standard or custom dataset (from files, python dict, pandas DataFrame, etc) at
    # https://huggingface.co/docs/datasets/loading_datasets.html.

    # Load pretrained model and tokenizer
    config = AutoConfig.from_pretrained(args.contrastive_model_path)
    config.label_smoothing_factor = args.label_smoothing_factor
    config.contrastive_pretrain = False
    config.wo_mf = args.wo_mf
    config.wo_it = args.wo_it
    tokenizer = AutoTokenizer.from_pretrained(args.contrastive_model_path)
    image_processor = AutoImageProcessor.from_pretrained(args.contrastive_model_path)
    model = ZRIGFForConditionalGeneration.from_pretrained(args.contrastive_model_path, config=config)

    generation_config = GenerationConfig.from_pretrained(args.contrastive_model_path)
    generation_config.max_new_tokens = args.val_max_target_length if args.val_max_target_length is not None else args.max_target_length
    generation_config.renormalize_logits = True
    if args.num_beams is not None:
        generation_config.num_beams = args.num_beams

    model.resize_token_embeddings(len(tokenizer))
    if model.config.decoder_start_token_id is None:
        raise ValueError("Make sure that `config.decoder_start_token_id` is correctly defined")

    # Preprocessing the datasets.
    # First we tokenize all the texts.
    column_names = raw_datasets["train"].column_names

    # Temporarily set max_target_length for training.
    max_source_length = args.max_source_length
    max_target_length = args.max_target_length
    padding = "max_length" if args.pad_to_max_length else False

    # Initialize torchvision transforms and jit it for faster processing.
    image_transformations = Transform(
        model.model.vision_encoder.config.image_size, image_processor.image_mean, image_processor.image_std
    )

    def preprocess_function(examples):
        inputs = examples["context"]
        targets = examples["response"]
        model_inputs = tokenizer(inputs, max_length=max_source_length, padding=padding, truncation=True)
        # Tokenize targets with the `text_target` keyword argument
        labels = tokenizer(text_target=targets, max_length=max_target_length, padding=padding, truncation=True)

        # If we are padding here, replace all tokenizer.pad_token_id in the labels by -100 when we want to ignore
        # padding in the loss.
        if padding == "max_length" and args.ignore_pad_token_for_loss:
            labels["input_ids"] = [
                [(l if l != tokenizer.pad_token_id else -100) for l in label] for label in labels["input_ids"]
            ]

        model_inputs["labels"] = labels["input_ids"]
        return model_inputs

    def transform_images(examples):
        images = [[read_image(image_path, mode=ImageReadMode.RGB) for image_path in image_paths[:3]]
                  for image_paths in examples['image_paths']]
        examples["pixel_values"] = [torch.stack([image_transformations(i) for i in image]) for image in images]
        return examples

    with accelerator.main_process_first():
        processed_datasets = raw_datasets.map(
            preprocess_function,
            batched=True,
            num_proc=args.preprocessing_num_workers,
            remove_columns=[col for col in column_names if col != 'image_paths'],
            load_from_cache_file=not args.overwrite_cache,
            desc="Running tokenizer on dataset",
        )
        processed_datasets.set_transform(transform_images)

    train_dataset = processed_datasets["train"]
    eval_dataset = processed_datasets["test"]

    label_pad_token_id = -100 if args.ignore_pad_token_for_loss else tokenizer.pad_token_id
    tokenizer.deprecation_warnings["Asking-to-pad-a-fast-tokenizer"] = True
    data_collator = DataCollatorForSeq2Seq(
        tokenizer,
        model=model,
        label_pad_token_id=label_pad_token_id,
        pad_to_multiple_of=8 if accelerator.use_fp16 else None,
    )

    def normalize_answer(s):
        """
        Lower text and remove punctuation, articles and extra whitespace.
        """

        s = s.lower()
        re_art = re.compile(r'\b(a|an|the)\b')
        re_punc = re.compile(r'[!"#$%&()*+,-./:;<=>?@\[\]\\^`{|}~_\']')
        s = re_punc.sub(' ', s)
        s = re_art.sub(' ', s)
        # TODO: this could almost certainly be faster with a regex \s+ -> ' '
        s = ' '.join(s.split())
        return s

    def postprocess_text(preds, labels):
        preds = [pred.strip() for pred in preds]
        labels = [label.strip() for label in labels]

        preds = [' '.join(word_tokenize(pred)) for pred in preds]
        labels = [' '.join(word_tokenize(label)) for label in labels]

        return preds, labels

    def compute_distinct(preds):
        unigram_counter, bigram_counter = Counter([]), Counter([])
        for pred in preds:
            pred_for_cal = pred.split()
            unigram_counter.update(pred_for_cal)
            bigram_counter.update(ngrams(pred_for_cal, 2))

        try:
            distinct_1 = len(unigram_counter) / (sum(unigram_counter.values()))
        except ZeroDivisionError:
            distinct_1 = 0
        try:
            distinct_2 = len(bigram_counter) / (sum(bigram_counter.values()))
        except ZeroDivisionError:
            distinct_2 = 0

        return distinct_1, distinct_2

    def collate_fn(examples):
        pixel_values = torch.stack([example.pop("pixel_values") for example in examples])
        [example.pop("image_paths") for example in examples]
        features = data_collator(examples)
        features["pixel_values"] = pixel_values
        return features

    train_dataloader = DataLoader(
        train_dataset, shuffle=True, collate_fn=collate_fn, batch_size=args.per_device_train_batch_size,
        num_workers=args.preprocessing_num_workers, pin_memory=True
    )
    eval_dataloader = DataLoader(
        eval_dataset, collate_fn=collate_fn, batch_size=args.per_device_eval_batch_size,
        num_workers=args.preprocessing_num_workers, pin_memory=True
    )

    # Optimizer
    # Split weights in two groups, one with weight decay and the other not.
    no_decay = ["bias", "LayerNorm.weight", "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

    lr_scheduler = get_scheduler(
        name=args.lr_scheduler_type,
        optimizer=optimizer,
        num_warmup_steps=args.num_warmup_steps * args.gradient_accumulation_steps,
        num_training_steps=args.max_train_steps * args.gradient_accumulation_steps,
    )

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

    # 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)

    # We need to initialize the trackers we use, and also store our configuration.
    # The trackers initializes automatically on the main process.
    if args.with_tracking:
        experiment_config = vars(args)
        # TensorBoard cannot log Enums, need the raw value
        experiment_config["lr_scheduler_type"] = experiment_config["lr_scheduler_type"].value
        accelerator.init_trackers("runs", experiment_config)

    # Metric
    nlgeval = NLGEval(no_skipthoughts=True, no_glove=True)

    # 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:
            resume_step = int(training_difference.replace("step_", ""))
            starting_epoch = resume_step // len(train_dataloader)
            resume_step -= starting_epoch * len(train_dataloader)

        completed_steps = starting_epoch * num_update_steps_per_epoch
        progress_bar.update(completed_steps)

    best_score = 0
    for epoch in range(starting_epoch, args.num_train_epochs):
        model.train()
        if args.with_tracking:
            total_loss = 0
        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:
                    progress_bar.update(1)
                    completed_steps += 1
                    continue

            with accelerator.accumulate(model):
                with accelerator.autocast():
                    outputs = model(**batch)
                    loss = outputs.loss
                # We keep track of the loss at each epoch
                if args.with_tracking:
                    total_loss += loss.detach().float()
                accelerator.backward(loss)
                optimizer.step()
                lr_scheduler.step()
                optimizer.zero_grad(set_to_none=True)

            # 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)
                    accelerator.save_state(output_dir)

            if completed_steps >= args.max_train_steps:
                break

        model.eval()

        hyp_list = []
        ref_list = []
        losses = []
        for step, batch in enumerate(eval_dataloader):
            with torch.no_grad():
                outputs = model(**batch)
                labels = batch["labels"]
                loss_fct = torch.nn.CrossEntropyLoss()
                loss = loss_fct(outputs.logits.view(-1, config.vocab_size), labels.view(-1))
                losses.append(accelerator.gather_for_metrics(loss.repeat(args.per_device_eval_batch_size)))

                generated_tokens = accelerator.unwrap_model(model).generate(
                    batch["input_ids"],
                    generation_config=generation_config,
                    attention_mask=batch["attention_mask"],
                    pixel_values=batch["pixel_values"],
                )

                generated_tokens = accelerator.pad_across_processes(
                    generated_tokens, dim=1, pad_index=tokenizer.pad_token_id
                )
                if not args.pad_to_max_length:
                    # If we did not pad to max length, we need to pad the labels too
                    labels = accelerator.pad_across_processes(batch["labels"], dim=1, pad_index=tokenizer.pad_token_id)

                generated_tokens, labels = accelerator.gather_for_metrics((generated_tokens, labels))
                generated_tokens = generated_tokens.cpu().numpy()
                labels = labels.cpu().numpy()

                if args.ignore_pad_token_for_loss:
                    # Replace -100 in the labels as we can't decode them.
                    labels = np.where(labels != -100, labels, tokenizer.pad_token_id)
                if isinstance(generated_tokens, tuple):
                    generated_tokens = generated_tokens[0]
                decoded_preds = tokenizer.batch_decode(generated_tokens, skip_special_tokens=True)
                decoded_labels = tokenizer.batch_decode(labels, skip_special_tokens=True)

                hyp_list.extend(decoded_preds)
                ref_list.extend(decoded_labels)

        losses = torch.cat(losses)
        try:
            eval_loss = torch.mean(losses)
            perplexity = math.exp(eval_loss)
        except OverflowError:
            perplexity = float("inf")

        logger.info(f"epoch {epoch}: perplexity: {perplexity} eval_loss: {eval_loss}")

        distinct_1, distinct_2 = compute_distinct([normalize_answer(s) for s in hyp_list])
        hyp_list, ref_list = postprocess_text(hyp_list, ref_list)
        result = nlgeval.compute_metrics([ref_list], hyp_list)
        result['distinct_1'] = distinct_1
        result['distinct_2'] = distinct_2
        result = {k: round(v * 100, 4) for k, v in result.items()}
        logger.info(result)
        result['perplexity'] = perplexity

        if args.with_tracking:
            result["train_loss"] = total_loss.item() / len(train_dataloader)
            result["eval_loss"] = eval_loss.item()
            result["epoch"] = epoch
            result["step"] = completed_steps
            accelerator.log(result, step=completed_steps)

        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)
            with open(os.path.join(output_dir, "results_hyp.txt"), "w") as f:
                f.writelines([p + '\n' for p in hyp_list])
            with open(os.path.join(output_dir, "results_ref.txt"), "w") as f:
                f.writelines([p + '\n' for p in ref_list])

        if result['Bleu_1'] > best_score:
            best_score = result['Bleu_1']
            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)
                image_processor.save_pretrained(args.output_dir)
                with open(os.path.join(args.output_dir, "results_hyp.txt"), "w") as f:
                    f.writelines([p + '\n' for p in hyp_list])
                with open(os.path.join(args.output_dir, "results_ref.txt"), "w") as f:
                    f.writelines([p + '\n' for p in ref_list])
                with open(os.path.join(args.output_dir, "all_results.json"), "w") as f:
                    json.dump(result, f)


if __name__ == "__main__":
    main()