utils.py

import os
import json
import datasets
from typing import Optional
from datasets.io.abc import AbstractDatasetReader
from datasets.utils.typing import NestedDataStructureLike, PathLike
from datasets import Features, NamedSplit
from datasets.tasks import QuestionAnsweringExtractive
import collections
import logging
from typing import Optional, Tuple
import numpy as np
from tqdm.auto import tqdm
from transformers import Trainer, is_torch_tpu_available
from transformers.trainer_utils import PredictionOutput

if is_torch_tpu_available():
    import torch_xla.core.xla_model as xm
    import torch_xla.debug.metrics as met


logger = logging.getLogger(__name__)



class QADatasetBuilder(datasets.GeneratorBasedBuilder):

    def _info(self):
        return datasets.DatasetInfo(
            features=datasets.Features(
                {
                    "id": datasets.Value("string"),
                    "title": datasets.Value("string"),
                    "context": datasets.Value("string"),
                    "question": datasets.Value("string"),
                    "answers": datasets.features.Sequence(
                        {
                            "text": datasets.Value("string"),
                            "answer_start": datasets.Value("int32"),
                        }
                    ),
                }
            ),
            supervised_keys=None,
            task_templates=[
                QuestionAnsweringExtractive(
                    question_column="question", context_column="context", answers_column="answers"
                )
            ],
        )
    
    def _split_generators(self, dl_manager):
        if not self.config.data_files:
            raise ValueError(f"At least one data file must be specified, but got data_files={self.config.data_files}")
        data_files = dl_manager.download_and_extract(self.config.data_files)
        if isinstance(data_files, (str, list, tuple)):
            files = data_files
            if isinstance(files, str):
                files = [files]
            return [datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"files": files})]
        splits = []
        for split_name, files in data_files.items():
            if isinstance(files, str):
                files = [files]
            splits.append(datasets.SplitGenerator(name=split_name, gen_kwargs={"files": files}))
        return splits

    def _generate_examples(self, files):
        for filepath in files:
            with open(filepath, encoding="utf-8") as f:
                squad = json.load(f)
                for example in squad["data"]:
                    title = example.get("title", "")
                    for paragraph in example["paragraphs"]:
                        context = paragraph["context"]
                        for qa in paragraph["qas"]:
                            question = qa["question"]
                            id_ = qa["id"]

                            answer_starts = [answer["answer_start"] for answer in qa["answers"]]
                            answers = [answer["text"] for answer in qa["answers"]]

                            yield id_, {
                                "title": title,
                                "context": context,
                                "question": question,
                                "id": id_,
                                "answers": {
                                    "answer_start": answer_starts,
                                    "text": answers,
                                },
                            }

class QADatasetReader(AbstractDatasetReader):

    def __init__(
        self,
        path_or_paths: NestedDataStructureLike[PathLike],
        split: Optional[NamedSplit] = None,
        features: Optional[Features] = None,
        cache_dir: str = None,
        keep_in_memory: bool = False,
        **kwargs,
    ):
        super().__init__(
            path_or_paths, split=split, features=features, cache_dir=cache_dir, keep_in_memory=keep_in_memory, **kwargs
        )
        path_or_paths = path_or_paths if isinstance(path_or_paths, dict) else {self.split: path_or_paths}
        self.builder = QADatasetBuilder(
            cache_dir=cache_dir,
            data_files=path_or_paths,
            **kwargs,
        )

    def read(self):
        download_config = None
        download_mode = None
        ignore_verifications = True
        try_from_hf_gcs = False
        use_auth_token = None
        base_path = None

        self.builder.download_and_prepare(
            download_config=download_config,
            download_mode=download_mode,
            ignore_verifications=ignore_verifications,
            try_from_hf_gcs=try_from_hf_gcs,
            base_path=base_path,
            use_auth_token=use_auth_token,
        )

        dataset = self.builder.as_dataset(
            split=self.split, ignore_verifications=ignore_verifications, in_memory=self.keep_in_memory
        )
        return dataset

def find_all_indices(pattern_str, source_str, overlapping=True):
    index = source_str.find(pattern_str)
    while index != -1:
        yield index
        index = source_str.find(
            pattern_str,
            index + (1 if overlapping else len(pattern_str))
        )

def postprocess_qa_predictions(
    examples,
    features,
    predictions: Tuple[np.ndarray, np.ndarray],
    allow_null_ans: bool = False,
    n_best_size: int = 20,
    max_answer_length: int = 30,
    null_score_diff_threshold: float = 0.0,
    output_dir: Optional[str] = None,
    prefix: Optional[str] = None,
    log_level: Optional[int] = logging.WARNING,
):
    """
    Post-processes the predictions of a question-answering model to convert them to answers that are substrings of the
    original contexts. This is the base postprocessing functions for models that only return start and end logits.

    Args:
        examples: The non-preprocessed dataset (see the main script for more information).
        features: The processed dataset (see the main script for more information).
        predictions (:obj:`Tuple[np.ndarray, np.ndarray]`):
            The predictions of the model: two arrays containing the start logits and the end logits respectively. Its
            first dimension must match the number of elements of :obj:`features`.
        allow_null_ans (:obj:`bool`, `optional`, defaults to :obj:`False`):
            Whether or not the underlying dataset contains examples with no answers.
        n_best_size (:obj:`int`, `optional`, defaults to 20):
            The total number of n-best predictions to generate when looking for an answer.
        max_answer_length (:obj:`int`, `optional`, defaults to 30):
            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.
        null_score_diff_threshold (:obj:`float`, `optional`, defaults to 0):
            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 (note that the score of
            the null answer for an example giving several features is the minimum of the scores for the null answer on
            each feature: all features must be aligned on the fact they `want` to predict a null answer).

            Only useful when :obj:`allow_null_ans` is :obj:`True`.
        output_dir (:obj:`str`, `optional`):
            If provided, the dictionaries of predictions, n_best predictions (with their scores and logits) and, if
            :obj:`allow_null_ans=True`, the dictionary of the scores differences between best and null
            answers, are saved in `output_dir`.
        prefix (:obj:`str`, `optional`):
            If provided, the dictionaries mentioned above are saved with `prefix` added to their names.
        log_level (:obj:`int`, `optional`, defaults to ``logging.WARNING``):
            ``logging`` log level (e.g., ``logging.WARNING``)
    """
    assert len(predictions) == 2, "`predictions` should be a tuple with two elements (start_logits, end_logits)."
    all_start_logits, all_end_logits = predictions

    assert len(predictions[0]) == len(features), f"Got {len(predictions[0])} predictions and {len(features)} features."

    # Build a map example to its corresponding features.
    example_id_to_index = {k: i for i, k in enumerate(examples["id"])}
    features_per_example = collections.defaultdict(list)
    for i, feature in enumerate(features):
        features_per_example[example_id_to_index[feature["example_id"]]].append(i)

    # The dictionaries we have to fill.
    all_predictions = collections.OrderedDict()
    all_nbest_json = collections.OrderedDict()
    if allow_null_ans:
        scores_diff_json = collections.OrderedDict()

    # Logging.
    logger.setLevel(log_level)
    logger.info(f"Post-processing {len(examples)} example predictions split into {len(features)} features.")

    # Let's loop over all the examples!
    for example_index, example in enumerate(tqdm(examples)):
        # Those are the indices of the features associated to the current example.
        feature_indices = features_per_example[example_index]

        min_null_prediction = None
        prelim_predictions = []

        # Looping through all the features associated to the current example.
        for feature_index in feature_indices:
            # We grab the predictions of the model for this feature.
            start_logits = all_start_logits[feature_index]
            end_logits = all_end_logits[feature_index]
            # This is what will allow us to map some the positions in our logits to span of texts in the original
            # context.
            offset_mapping = features[feature_index]["offset_mapping"]
            # Optional `token_is_max_context`, if provided we will remove answers that do not have the maximum context
            # available in the current feature.
            token_is_max_context = features[feature_index].get("token_is_max_context", None)

            # Update minimum null prediction.
            feature_null_score = start_logits[0] + end_logits[0]
            if min_null_prediction is None or min_null_prediction["score"] > feature_null_score:
                min_null_prediction = {
                    "offsets": (0, 0),
                    "score": feature_null_score,
                    "start_logit": start_logits[0],
                    "end_logit": end_logits[0],
                }

            # Go through all possibilities for the `n_best_size` greater start and end logits.
            start_indexes = np.argsort(start_logits)[-1 : -n_best_size - 1 : -1].tolist()
            end_indexes = np.argsort(end_logits)[-1 : -n_best_size - 1 : -1].tolist()
            for start_index in start_indexes:
                for end_index in end_indexes:
                    # Don't consider out-of-scope answers, either because the indices are out of bounds or correspond
                    # to part of the input_ids that are not in the context.
                    if (
                        start_index >= len(offset_mapping)
                        or end_index >= len(offset_mapping)
                        or offset_mapping[start_index] is None
                        or offset_mapping[end_index] is None
                    ):
                        continue
                    # Don't consider answers with a length that is either < 0 or > max_answer_length.
                    if end_index < start_index or end_index - start_index + 1 > max_answer_length:
                        continue
                    # Don't consider answer that don't have the maximum context available (if such information is
                    # provided).
                    if token_is_max_context is not None and not token_is_max_context.get(str(start_index), False):
                        continue
                    prelim_predictions.append(
                        {
                            "offsets": (offset_mapping[start_index][0], offset_mapping[end_index][1]),
                            "score": start_logits[start_index] + end_logits[end_index],
                            "start_logit": start_logits[start_index],
                            "end_logit": end_logits[end_index],
                        }
                    )
        if allow_null_ans:
            # Add the minimum null prediction
            prelim_predictions.append(min_null_prediction)
            null_score = min_null_prediction["score"]

        # Only keep the best `n_best_size` predictions.
        predictions = sorted(prelim_predictions, key=lambda x: x["score"], reverse=True)[:n_best_size]

        # Add back the minimum null prediction if it was removed because of its low score.
        if allow_null_ans and not any(p["offsets"] == (0, 0) for p in predictions):
            predictions.append(min_null_prediction)

        # Use the offsets to gather the answer text in the original context.
        context = example["context"]
        for pred in predictions:
            offsets = pred.pop("offsets")
            pred["text"] = context[offsets[0] : offsets[1]]

        # In the very rare edge case we have not a single non-null prediction, we create a fake prediction to avoid
        # failure.
        if len(predictions) == 0 or (len(predictions) == 1 and predictions[0]["text"] == ""):
            predictions.insert(0, {"text": "empty", "start_logit": 0.0, "end_logit": 0.0, "score": 0.0})

        # Compute the softmax of all scores (we do it with numpy to stay independent from torch/tf in this file, using
        # the LogSumExp trick).
        scores = np.array([pred.pop("score") for pred in predictions])
        exp_scores = np.exp(scores - np.max(scores))
        probs = exp_scores / exp_scores.sum()

        # Include the probabilities in our predictions.
        for prob, pred in zip(probs, predictions):
            pred["probability"] = prob

        # Pick the best prediction. If the null answer is not possible, this is easy.
        if not allow_null_ans:
            all_predictions[example["id"]] = predictions[0]["text"]
        else:
            # Otherwise we first need to find the best non-empty prediction.
            i = 0
            while predictions[i]["text"] == "":
                i += 1
            best_non_null_pred = predictions[i]

            # Then we compare to the null prediction using the threshold.
            score_diff = null_score - best_non_null_pred["start_logit"] - best_non_null_pred["end_logit"]
            scores_diff_json[example["id"]] = float(score_diff)  # To be JSON-serializable.
            if score_diff > null_score_diff_threshold:
                all_predictions[example["id"]] = ""
            else:
                all_predictions[example["id"]] = best_non_null_pred["text"]

        # Make `predictions` JSON-serializable by casting np.float back to float.
        all_nbest_json[example["id"]] = [
            {k: (float(v) if isinstance(v, (np.float16, np.float32, np.float64)) else v) for k, v in pred.items()}
            for pred in predictions
        ]

    # If we have an output_dir, let's save all those dicts.
    if output_dir is not None:
        assert os.path.isdir(output_dir), f"{output_dir} is not a directory."

        prediction_file = os.path.join(
            output_dir, "predictions.json" if prefix is None else f"{prefix}_predictions.json"
        )
        nbest_file = os.path.join(
            output_dir, "nbest_predictions.json" if prefix is None else f"{prefix}_nbest_predictions.json"
        )
        if allow_null_ans:
            null_odds_file = os.path.join(
                output_dir, "null_odds.json" if prefix is None else f"{prefix}_null_odds.json"
            )

        logger.info(f"Saving predictions to {prediction_file}.")
        with open(prediction_file, "w") as writer:
            writer.write(json.dumps(all_predictions, ensure_ascii=False, indent=4) + "\n")
        logger.info(f"Saving nbest_preds to {nbest_file}.")
        with open(nbest_file, "w") as writer:
            writer.write(json.dumps(all_nbest_json, ensure_ascii=False, indent=4) + "\n")
        if allow_null_ans:
            logger.info(f"Saving null_odds to {null_odds_file}.")
            with open(null_odds_file, "w") as writer:
                writer.write(json.dumps(scores_diff_json, ensure_ascii=False, indent=4) + "\n")

    return all_predictions


class QuestionAnsweringTrainer(Trainer):
    def __init__(self, *args, eval_examples=None, post_process_function=None, **kwargs):
        super().__init__(*args, **kwargs)
        self.eval_examples = eval_examples
        self.post_process_function = post_process_function

    def evaluate(self, eval_dataset=None, eval_examples=None, ignore_keys=None, metric_key_prefix: str = "eval"):
        eval_dataset = self.eval_dataset if eval_dataset is None else eval_dataset
        eval_dataloader = self.get_eval_dataloader(eval_dataset)
        eval_examples = self.eval_examples if eval_examples is None else eval_examples

        # Temporarily disable metric computation, we will do it in the loop here.
        compute_metrics = self.compute_metrics
        self.compute_metrics = None
        eval_loop = self.prediction_loop if self.args.use_legacy_prediction_loop else self.evaluation_loop
        try:
            output = eval_loop(
                eval_dataloader,
                description="Evaluation",
                # No point gathering the predictions if there are no metrics, otherwise we defer to
                # self.args.prediction_loss_only
                prediction_loss_only=True if compute_metrics is None else None,
                ignore_keys=ignore_keys,
            )
        finally:
            self.compute_metrics = compute_metrics

        if self.post_process_function is not None and self.compute_metrics is not None:
            eval_preds = self.post_process_function(eval_examples, eval_dataset, output.predictions)
            metrics = self.compute_metrics(eval_preds)

            # Prefix all keys with metric_key_prefix + '_'
            for key in list(metrics.keys()):
                if not key.startswith(f"{metric_key_prefix}_"):
                    metrics[f"{metric_key_prefix}_{key}"] = metrics.pop(key)

            self.log(metrics)
        else:
            metrics = {}

        if self.args.tpu_metrics_debug or self.args.debug:
            # tpu-comment: Logging debug metrics for PyTorch/XLA (compile, execute times, ops, etc.)
            xm.master_print(met.metrics_report())

        self.control = self.callback_handler.on_evaluate(self.args, self.state, self.control, metrics)
        return metrics

    def predict(self, predict_dataset, predict_examples, ignore_keys=None, metric_key_prefix: str = "test"):
        predict_dataloader = self.get_test_dataloader(predict_dataset)

        # Temporarily disable metric computation, we will do it in the loop here.
        compute_metrics = self.compute_metrics
        self.compute_metrics = None
        eval_loop = self.prediction_loop if self.args.use_legacy_prediction_loop else self.evaluation_loop
        try:
            output = eval_loop(
                predict_dataloader,
                description="Prediction",
                # No point gathering the predictions if there are no metrics, otherwise we defer to
                # self.args.prediction_loss_only
                prediction_loss_only=True if compute_metrics is None else None,
                ignore_keys=ignore_keys,
            )
        finally:
            self.compute_metrics = compute_metrics

        if self.post_process_function is None or self.compute_metrics is None:
            return output

        predictions = self.post_process_function(predict_examples, predict_dataset, output.predictions, "predict")
        metrics = self.compute_metrics(predictions)

        # Prefix all keys with metric_key_prefix + '_'
        for key in list(metrics.keys()):
            if not key.startswith(f"{metric_key_prefix}_"):
                metrics[f"{metric_key_prefix}_{key}"] = metrics.pop(key)

        return PredictionOutput(predictions=predictions.predictions, label_ids=predictions.label_ids, metrics=metrics)