question_answering.py

import logging
import os
import glob
import random
import sys
from dataclasses import dataclass, field
from typing import Optional

import datasets
from datasets import load_dataset, load_metric

import transformers
from transformers import (
    AutoConfig,
    AutoModelForQuestionAnswering,
    AutoTokenizer,
    DataCollatorWithPadding,
    EvalPrediction,
    HfArgumentParser,
    PreTrainedTokenizerFast,
    Trainer,
    TrainingArguments,
    default_data_collator,
    set_seed,
)
from transformers.trainer_utils import get_last_checkpoint
from transformers.utils import check_min_version
from transformers.utils.versions import require_version
from normalizer import normalize
from utils import (
    QADatasetReader,
    find_all_indices,
    postprocess_qa_predictions,
    QuestionAnsweringTrainer
)

EXT2CONFIG = {
    "json": (QADatasetReader, {})
}

logger = logging.getLogger(__name__)


@dataclass
class DataTrainingArguments:
   
    dataset_dir: Optional[str] = field(
        default=None, metadata={
            "help": "Path to the directory containing the data files. (.json)"
            "File datatypes will be identified with their prefix names as follows: "
            "`train`- Training file(s) e.g. `train.json`/ `train_part1.json` etc. "
            "`validation`- Evaluation file(s) e.g. `validation.json`/ `validation_part1.json` etc. "
            "`test`- Test file(s) e.g. `test.json`/ `test_part1.json` etc. "
            "All files for must have the same extension."
        }
    )

    train_file: Optional[str] = field(default=None, metadata={"help": "The input training data file (a json file)."})
    validation_file: Optional[str] = field(
        default=None,
        metadata={"help": "An optional input evaluation data file to evaluate the perplexity on (a json file)."},
    )
    test_file: Optional[str] = field(
        default=None,
        metadata={"help": "An optional input test data file to evaluate the perplexity on (a json file)."},
    )
    overwrite_cache: bool = field(
        default=False, metadata={"help": "Overwrite the cached training and evaluation sets"}
    )
    preprocessing_num_workers: Optional[int] = field(
        default=None,
        metadata={"help": "The number of processes to use for the preprocessing."},
    )
    max_seq_length: int = field(
        default=384,
        metadata={
            "help": "The maximum total input sequence length after tokenization. Sequences longer "
            "than this will be truncated, sequences shorter will be padded."
        },
    )
    pad_to_max_length: bool = field(
        default=True,
        metadata={
            "help": "Whether to pad all samples to `max_seq_length`. "
            "If False, will pad the samples dynamically when batching to the maximum length in the batch (which can "
            "be faster on GPU but will be slower on TPU)."
        },
    )
    max_train_samples: Optional[int] = field(
        default=None,
        metadata={
            "help": "For debugging purposes or quicker training, truncate the number of training examples to this "
            "value if set."
        },
    )
    max_eval_samples: Optional[int] = field(
        default=None,
        metadata={
            "help": "For debugging purposes or quicker training, truncate the number of evaluation examples to this "
            "value if set."
        },
    )
    max_predict_samples: Optional[int] = field(
        default=None,
        metadata={
            "help": "For debugging purposes or quicker training, truncate the number of prediction examples to this "
            "value if set."
        },
    )
    allow_null_ans: bool = field(
        default=False, metadata={"help": "If true, some of the examples do not have an answer."}
    )
    null_score_diff_threshold: float = field(
        default=0.0,
        metadata={
            "help": "The threshold used to select the null answer: if the best answer has a score that is less than "
            "the score of the null answer minus this threshold, the null answer is selected for this example. "
            "Only useful when `allow_null_ans=True`."
        },
    )
    doc_stride: int = field(
        default=128,
        metadata={"help": "When splitting up a long document into chunks, how much stride to take between chunks."},
    )
    n_best_size: int = field(
        default=20,
        metadata={"help": "The total number of n-best predictions to generate when looking for an answer."},
    )
    max_answer_length: int = field(
        default=30,
        metadata={
            "help": "The maximum length of an answer that can be generated. This is needed because the start "
            "and end predictions are not conditioned on one another."
        },
    )
    do_normalize: Optional[bool] = field(default=True, metadata={"help": "Normalize text before feeding to the model."})
    unicode_norm: Optional[str] = field(default="NFKC", metadata={"help": "Type of unicode normalization"})
   

    def __post_init__(self):
        if self.train_file is not None and self.validation_file is not None:
            train_extension = self.train_file.split(".")[-1]
            assert train_extension in ["csv", "json", "tsv"], "`train_file` should be a csv / tsv / json file."
            validation_extension = self.validation_file.split(".")[-1]
            assert (
                validation_extension == train_extension
            ), "`validation_file` should have the same extension csv / tsv / json as `train_file`."


@dataclass
class ModelArguments:
    
    model_name_or_path: str = field(
        metadata={"help": "Path to pretrained model or model identifier from huggingface.co/models"}
    )
    cache_dir: Optional[str] = field(
        default=None,
        metadata={"help": "Where do you want to store the pretrained models downloaded from huggingface.co"},
    )


def main():
    
    parser = HfArgumentParser((ModelArguments, DataTrainingArguments, TrainingArguments))
    if len(sys.argv) == 2 and sys.argv[1].endswith(".json"):
        model_args, data_args, training_args = parser.parse_json_file(json_file=os.path.abspath(sys.argv[1]))
    else:
        model_args, data_args, training_args = parser.parse_args_into_dataclasses()

    # Setup logging
    logging.basicConfig(
        format="%(asctime)s - %(levelname)s - %(name)s - %(message)s",
        datefmt="%m/%d/%Y %H:%M:%S",
        handlers=[logging.StreamHandler(sys.stdout)],
    )

    log_level = training_args.get_process_log_level()
    logger.setLevel(log_level)
    datasets.utils.logging.set_verbosity(log_level)
    transformers.utils.logging.set_verbosity(log_level)
    transformers.utils.logging.enable_default_handler()
    transformers.utils.logging.enable_explicit_format()

    logger.warning(
        f"Process rank: {training_args.local_rank}, device: {training_args.device}, n_gpu: {training_args.n_gpu}"
        + f"distributed training: {bool(training_args.local_rank != -1)}, 16-bits training: {training_args.fp16}"
    )
    logger.info(f"Training/evaluation parameters {training_args}")

    last_checkpoint = None
    if os.path.isdir(training_args.output_dir) and training_args.do_train and not training_args.overwrite_output_dir:
        last_checkpoint = get_last_checkpoint(training_args.output_dir)
        if last_checkpoint is None and len(os.listdir(training_args.output_dir)) > 0:
            raise ValueError(
                f"Output directory ({training_args.output_dir}) already exists and is not empty. "
                "Use --overwrite_output_dir to overcome."
            )
        elif last_checkpoint is not None and training_args.resume_from_checkpoint is None:
            logger.info(
                f"Checkpoint detected, resuming training at {last_checkpoint}. To avoid this behavior, change "
                "the `--output_dir` or add `--overwrite_output_dir` to train from scratch."
            )

    # Set seed before initializing model.
    set_seed(training_args.seed)

    has_ext = lambda path: len(os.path.basename(path).split(".")) > 1
    get_ext = lambda path: os.path.basename(path).split(".")[-1]

    if data_args.dataset_dir is not None:
        data_files = {}
        all_files = glob.glob(
            os.path.join(
                data_args.dataset_dir,
                "*"
            )
        )
        all_exts = [get_ext(k) for k in all_files if has_ext(k)]
        if not all_exts:
            raise ValueError("The `dataset_dir` doesnt have any valid file.")
            
        selected_ext = max(set(all_exts), key=all_exts.count)
        for search_prefix in ["train", "validation", "test"]:
            found_files = glob.glob(
                os.path.join(
                    data_args.dataset_dir,
                    search_prefix + "*" + selected_ext
                )
            )
            if not found_files:
                continue

            data_files[search_prefix] = found_files
        
    else:
        data_files = {
            "train": data_args.train_file, 
            "validation": data_args.validation_file,
            "test": data_args.test_file
        }

        data_files = {k: v for k, v in data_files.items() if v is not None}
        
        if not data_files:
            raise ValueError("No valid input file found.")

        selected_ext = get_ext(list(data_files.values())[0])


    dataset_configs = EXT2CONFIG[selected_ext]
    raw_datasets = dataset_configs[0](
        data_files, 
        **dataset_configs[1]
    ).read()

    config = AutoConfig.from_pretrained(
        model_args.model_name_or_path,
        cache_dir=model_args.cache_dir
    )

    tokenizer_kwargs = {"add_prefix_space": True} if config.model_type in {"gpt2", "roberta"} else {}   
    tokenizer = AutoTokenizer.from_pretrained(
        model_args.model_name_or_path,
        cache_dir=model_args.cache_dir,
        use_fast=True,
        **tokenizer_kwargs
    )
    model = AutoModelForQuestionAnswering.from_pretrained(
        model_args.model_name_or_path,
        from_tf=bool(".ckpt" in model_args.model_name_or_path),
        config=config,
        cache_dir=model_args.cache_dir
    )

    if not isinstance(tokenizer, PreTrainedTokenizerFast):
        raise ValueError(
            "This script only works for models that have a fast tokenizer. Checkout the big table of models "
            "at https://huggingface.co/transformers/index.html#supported-frameworks to find the model types that meet this "
            "requirement"
        )

    if training_args.do_train:
        column_names = raw_datasets["train"].column_names
    elif training_args.do_eval:
        column_names = raw_datasets["validation"].column_names
    else:
        column_names = raw_datasets["test"].column_names

    question_column_name = "question" if "question" in column_names else column_names[0]
    context_column_name = "context" if "context" in column_names else column_names[1]
    answer_column_name = "answers" if "answers" in column_names else column_names[2]


    if data_args.do_normalize:
        normalization_kwargs = {
            "unicode_norm": data_args.unicode_norm,
        }
        required_column_names = [
            question_column_name,
            context_column_name,
            answer_column_name
        ]

        def normalize_example(example):
            required_row_values = [example[k] for k in required_column_names if k in example]
            question, context = required_row_values[:2]
            example[question_column_name] = normalize(question, **normalization_kwargs)
            example[context_column_name] = normalize(context, **normalization_kwargs)

            if len(required_row_values) == 3:
                answer = required_row_values[2]
                for i, ans in enumerate(answer["text"]):
                    prev_position = answer["answer_start"][i]
                    answer["text"][i] = normalize(ans, **normalization_kwargs)

                    replace_index = -1
                    for j, pos in enumerate(find_all_indices(ans, context)):
                        replace_index = j
                        if pos == prev_position:
                            break

                    if replace_index != -1:
                        index_iterator = find_all_indices(
                            answer["text"][i],
                            example[context_column_name]
                        )
                        for j, pos in enumerate(index_iterator):
                            if j == replace_index:
                                answer["answer_start"][i] = pos
                                assert answer["text"][i] == example[context_column_name][pos: pos + len(answer["text"][i])]
                                break

                example[answer_column_name] = answer

            return example

        raw_datasets = raw_datasets.map(
            normalize_example,
            desc="Running normalization on dataset",
            load_from_cache_file=not data_args.overwrite_cache
        )
    
    pad_on_right = tokenizer.padding_side == "right"

    if data_args.max_seq_length > tokenizer.model_max_length:
        logger.warning(
            f"The max_seq_length passed ({data_args.max_seq_length}) is larger than the maximum length for the"
            f"model ({tokenizer.model_max_length}). Using max_seq_length={tokenizer.model_max_length}."
        )
    max_seq_length = min(data_args.max_seq_length, tokenizer.model_max_length)



    def prepare_train_features(examples):
        tokenized_examples = tokenizer(
            examples[question_column_name if pad_on_right else context_column_name],
            examples[context_column_name if pad_on_right else question_column_name],
            truncation="only_second" if pad_on_right else "only_first",
            max_length=max_seq_length,
            stride=data_args.doc_stride,
            return_overflowing_tokens=True,
            return_offsets_mapping=True,
            padding="max_length" if data_args.pad_to_max_length else False,
        )

        sample_mapping = tokenized_examples.pop("overflow_to_sample_mapping")
        offset_mapping = tokenized_examples.pop("offset_mapping")

        tokenized_examples["start_positions"] = []
        tokenized_examples["end_positions"] = []

        for i, offsets in enumerate(offset_mapping):
            input_ids = tokenized_examples["input_ids"][i]
            cls_index = input_ids.index(tokenizer.cls_token_id)
            sequence_ids = tokenized_examples.sequence_ids(i)

            sample_index = sample_mapping[i]
            answers = examples[answer_column_name][sample_index]
            if len(answers["answer_start"]) == 0:
                tokenized_examples["start_positions"].append(cls_index)
                tokenized_examples["end_positions"].append(cls_index)
            else:
                start_char = answers["answer_start"][0]
                end_char = start_char + len(answers["text"][0])

                token_start_index = 0
                while sequence_ids[token_start_index] != (1 if pad_on_right else 0):
                    token_start_index += 1

                token_end_index = len(input_ids) - 1
                while sequence_ids[token_end_index] != (1 if pad_on_right else 0):
                    token_end_index -= 1

                if not (offsets[token_start_index][0] <= start_char and offsets[token_end_index][1] >= end_char):
                    tokenized_examples["start_positions"].append(cls_index)
                    tokenized_examples["end_positions"].append(cls_index)
                else:
                    while token_start_index < len(offsets) and offsets[token_start_index][0] <= start_char:
                        token_start_index += 1
                    tokenized_examples["start_positions"].append(token_start_index - 1)
                    while offsets[token_end_index][1] >= end_char:
                        token_end_index -= 1
                    tokenized_examples["end_positions"].append(token_end_index + 1)

        return tokenized_examples

    if training_args.do_train:
        if "train" not in raw_datasets:
            raise ValueError("--do_train requires a train dataset")
        train_dataset = raw_datasets["train"]
        if data_args.max_train_samples is not None:
            answerable_indices = [i for i, data in enumerate(train_dataset)
                                     if data['answers']['text']]
            unanswerable_indices = [i for i, data in enumerate(train_dataset)
                                     if not data['answers']['text']]

            if (
                len(answerable_indices) >= data_args.max_train_samples // 2 and
                len(unanswerable_indices) >= data_args.max_train_samples // 2
            ):
                selected_answerable_indices = answerable_indices[:data_args.max_train_samples // 2]
                selected_unanswerable_indices = unanswerable_indices[:data_args.max_train_samples - len(selected_answerable_indices)]
                train_dataset = train_dataset.select(
                    selected_answerable_indices + 
                    selected_unanswerable_indices
                )
            else:
                train_dataset = train_dataset.select(range(data_args.max_train_samples))
        
        with training_args.main_process_first(desc="train dataset map pre-processing"):
            train_dataset = train_dataset.map(
                prepare_train_features,
                batched=True,
                num_proc=data_args.preprocessing_num_workers,
                remove_columns=column_names,
                load_from_cache_file=not data_args.overwrite_cache,
                desc="Running tokenizer on train dataset",
            )
        if data_args.max_train_samples is not None:
            train_dataset = train_dataset.select(range(data_args.max_train_samples))

    def prepare_validation_features(examples):
        
        tokenized_examples = tokenizer(
            examples[question_column_name if pad_on_right else context_column_name],
            examples[context_column_name if pad_on_right else question_column_name],
            truncation="only_second" if pad_on_right else "only_first",
            max_length=max_seq_length,
            stride=data_args.doc_stride,
            return_overflowing_tokens=True,
            return_offsets_mapping=True,
            padding="max_length" if data_args.pad_to_max_length else False,
        )

        sample_mapping = tokenized_examples.pop("overflow_to_sample_mapping")

        tokenized_examples["example_id"] = []

        for i in range(len(tokenized_examples["input_ids"])):
            sequence_ids = tokenized_examples.sequence_ids(i)
            context_index = 1 if pad_on_right else 0

            sample_index = sample_mapping[i]
            tokenized_examples["example_id"].append(examples["id"][sample_index])

            tokenized_examples["offset_mapping"][i] = [
                (o if sequence_ids[k] == context_index else None)
                for k, o in enumerate(tokenized_examples["offset_mapping"][i])
            ]

        return tokenized_examples

    if training_args.do_eval:
        if "validation" not in raw_datasets:
            raise ValueError("--do_eval requires a validation dataset")
        eval_examples = raw_datasets["validation"]
        if data_args.max_eval_samples is not None:
            eval_examples = eval_examples.select(range(data_args.max_eval_samples))
        
        with training_args.main_process_first(desc="validation dataset map pre-processing"):
            eval_dataset = eval_examples.map(
                prepare_validation_features,
                batched=True,
                num_proc=data_args.preprocessing_num_workers,
                remove_columns=column_names,
                load_from_cache_file=not data_args.overwrite_cache,
                desc="Running tokenizer on validation dataset",
            )
        if data_args.max_eval_samples is not None:
            # During Feature creation dataset samples might increase, we will select required samples again
            eval_dataset = eval_dataset.select(range(data_args.max_eval_samples))

    if training_args.do_predict:
        if "test" not in raw_datasets:
            raise ValueError("--do_predict requires a test dataset")
        predict_examples = raw_datasets["test"]
        if data_args.max_predict_samples is not None:
            predict_examples = predict_examples.select(range(data_args.max_predict_samples))
        
        with training_args.main_process_first(desc="prediction dataset map pre-processing"):
            predict_dataset = predict_examples.map(
                prepare_validation_features,
                batched=True,
                num_proc=data_args.preprocessing_num_workers,
                remove_columns=column_names,
                load_from_cache_file=not data_args.overwrite_cache,
                desc="Running tokenizer on prediction dataset",
            )
        if data_args.max_predict_samples is not None:
            predict_dataset = predict_dataset.select(range(data_args.max_predict_samples))

    data_collator = (
        default_data_collator
        if data_args.pad_to_max_length
        else DataCollatorWithPadding(tokenizer, pad_to_multiple_of=8 if training_args.fp16 else None)
    )

    # Log a few random samples from the training set:
    if training_args.do_train:
        for index in random.sample(range(len(train_dataset)), 3):
            logger.info(f"Sample {index} of the training set: {train_dataset[index]}.")

    def post_processing_function(examples, features, predictions, stage="eval"):
        predictions = postprocess_qa_predictions(
            examples=examples,
            features=features,
            predictions=predictions,
            allow_null_ans=data_args.allow_null_ans,
            n_best_size=data_args.n_best_size,
            max_answer_length=data_args.max_answer_length,
            null_score_diff_threshold=data_args.null_score_diff_threshold,
            output_dir=training_args.output_dir,
            log_level=log_level,
            prefix=stage,
        )
        
        if data_args.allow_null_ans:
            formatted_predictions = [
                {"id": k, "prediction_text": v, "no_answer_probability": 0.0} for k, v in predictions.items()
            ]
        else:
            formatted_predictions = [{"id": k, "prediction_text": v} for k, v in predictions.items()]

        references = [{"id": ex["id"], "answers": ex[answer_column_name]} for ex in examples]
        return EvalPrediction(predictions=formatted_predictions, label_ids=references)

    metric = load_metric("squad_v2" if data_args.allow_null_ans else "squad")

    def compute_metrics(p: EvalPrediction):
        return metric.compute(predictions=p.predictions, references=p.label_ids)

    # Initialize our Trainer
    trainer = QuestionAnsweringTrainer(
        model=model,
        args=training_args,
        train_dataset=train_dataset if training_args.do_train else None,
        eval_dataset=eval_dataset if training_args.do_eval else None,
        eval_examples=eval_examples if training_args.do_eval else None,
        tokenizer=tokenizer,
        data_collator=data_collator,
        post_process_function=post_processing_function,
        compute_metrics=compute_metrics,
    )

    # Training
    if training_args.do_train:
        checkpoint = None
        if training_args.resume_from_checkpoint is not None:
            checkpoint = training_args.resume_from_checkpoint
        elif last_checkpoint is not None:
            checkpoint = last_checkpoint
        train_result = trainer.train(resume_from_checkpoint=checkpoint)
        trainer.save_model()  # Saves the tokenizer too for easy upload

        metrics = train_result.metrics
        max_train_samples = (
            data_args.max_train_samples if data_args.max_train_samples is not None else len(train_dataset)
        )
        metrics["train_samples"] = min(max_train_samples, len(train_dataset))

        trainer.log_metrics("train", metrics)
        trainer.save_metrics("train", metrics)
        trainer.save_state()

    # Evaluation
    if training_args.do_eval:
        logger.info("*** Evaluate ***")
        metrics = trainer.evaluate()

        max_eval_samples = data_args.max_eval_samples if data_args.max_eval_samples is not None else len(eval_dataset)
        metrics["eval_samples"] = min(max_eval_samples, len(eval_dataset))

        trainer.log_metrics("eval", metrics)
        trainer.save_metrics("eval", metrics)

    # Prediction
    if training_args.do_predict:
        logger.info("*** Predict ***")
        results = trainer.predict(predict_dataset, predict_examples)
        metrics = results.metrics

        max_predict_samples = (
            data_args.max_predict_samples if data_args.max_predict_samples is not None else len(predict_dataset)
        )
        metrics["predict_samples"] = min(max_predict_samples, len(predict_dataset))

        trainer.log_metrics("predict", metrics)
        trainer.save_metrics("predict", metrics)


def _mp_fn(index):
    # For xla_spawn (TPUs)
    main()


if __name__ == "__main__":
    main()