run_glue.py

#!/usr/bin/env python
# coding=utf-8
# Copyright 2020 The HuggingFace Inc. team. All rights reserved.
# Copyright (c) Meta Platforms, Inc. All rights reserved.
# This source code is licensed under the license found in the LICENSE file in
# the root directory of this source tree.

""" Finetuning the library models for sequence classification on GLUE."""
# You can also adapt this script on your own text classification task. Pointers for this are left as comments.

import logging
import os
import pickle
import random
import sys
from dataclasses import dataclass, field
from pathlib import Path
from typing import Optional

sys.path.append('../')

import numpy as np
from datasets import load_dataset, load_metric

import transformers
from transformers import (
    AdapterConfig,
    AutoConfig,
    AutoModelForSequenceClassification,
    AutoTokenizer,
    DataCollatorWithPadding,
    EvalPrediction,
    HfArgumentParser,
    MultiLingAdapterArguments,
    PretrainedConfig,
    Trainer,
    TrainingArguments,
    default_data_collator,
    set_seed,
)
from transformers.trainer_utils import get_last_checkpoint, is_main_process
from transformers.utils import check_min_version

# Will error if the minimal version of Transformers is not installed. Remove at your own risks.
from utils import freeze_params, choose_gpu, freeze_params_by_layers

check_min_version("4.5.0")

task_to_keys = {
    "cola": ("sentence", None),
    "mnli": ("premise", "hypothesis"),
    "mrpc": ("sentence1", "sentence2"),
    "qnli": ("question", "sentence"),
    "qqp": ("question1", "question2"),
    "rte": ("sentence1", "sentence2"),
    "sst2": ("sentence", None),
    "stsb": ("sentence1", "sentence2"),
    "wnli": ("sentence1", "sentence2"),
}

logger = logging.getLogger(__name__)


@dataclass
class DataTrainingArguments:
    """
    Arguments pertaining to what data we are going to input our model for training and eval.
    Using `HfArgumentParser` we can turn this class
    into argparse arguments to be able to specify them on
    the command line.
    """

    task_name: Optional[str] = field(
        default=None,
        metadata={"help": "The name of the task to train on: " + ", ".join(task_to_keys.keys())},
    )
    max_seq_length: int = field(
        default=128,
        metadata={
            "help": "The maximum total input sequence length after tokenization. Sequences longer "
                    "than this will be truncated, sequences shorter will be padded."
        },
    )
    overwrite_cache: bool = field(
        default=False, metadata={"help": "Overwrite the cached preprocessed datasets or not."}
    )
    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."
        },
    )
    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_val_samples: Optional[int] = field(
        default=None,
        metadata={
            "help": "For debugging purposes or quicker training, truncate the number of validation examples to this "
                    "value if set."
        },
    )
    max_test_samples: Optional[int] = field(
        default=None,
        metadata={
            "help": "For debugging purposes or quicker training, truncate the number of test examples to this "
                    "value if set."
        },
    )
    train_file: Optional[str] = field(
        default=None, metadata={"help": "A csv or a json file containing the training data."}
    )
    validation_file: Optional[str] = field(
        default=None, metadata={"help": "A csv or a json file containing the validation data."}
    )
    test_file: Optional[str] = field(default=None, metadata={"help": "A csv or a json file containing the test data."})
    data_seed: Optional[int] = field(
        default=42,
        metadata={
            "help": "seed for selecting subset of the dataset if not using all."
        },
    )
    train_as_val: bool = field(
        default=True,
        metadata={"help": "if True, sample 1k from train as val"},
    )
    early_stopping_patience: Optional[int] = field(
        default=10,
    )

    def __post_init__(self):
        if self.task_name is not None:
            self.task_name = self.task_name.lower()
            if self.task_name not in task_to_keys.keys():
                raise ValueError("Unknown task, you should pick one in " + ",".join(task_to_keys.keys()))
        elif self.train_file is None or self.validation_file is None:
            raise ValueError("Need either a GLUE task or a training/validation file.")
        else:
            train_extension = self.train_file.split(".")[-1]
            assert train_extension in ["csv", "json"], "`train_file` should be a csv or a json file."
            validation_extension = self.validation_file.split(".")[-1]
            assert (
                    validation_extension == train_extension
            ), "`validation_file` should have the same extension (csv or json) as `train_file`."


@dataclass
class ModelArguments:
    """
    Arguments pertaining to which model/config/tokenizer we are going to fine-tune from.
    """

    model_name_or_path: str = field(
        metadata={"help": "Path to pretrained model or model identifier from huggingface.co/models"}
    )
    config_name: Optional[str] = field(
        default=None, metadata={"help": "Pretrained config name or path if not the same as model_name"}
    )
    tokenizer_name: Optional[str] = field(
        default=None, metadata={"help": "Pretrained tokenizer name or path if not the same as model_name"}
    )
    cache_dir: Optional[str] = field(
        default=None,
        metadata={"help": "Where do you want to store the pretrained models downloaded from huggingface.co"},
    )
    use_fast_tokenizer: bool = field(
        default=True,
        metadata={"help": "Whether to use one of the fast tokenizer (backed by the tokenizers library) or not."},
    )
    model_revision: str = field(
        default="main",
        metadata={"help": "The specific model version to use (can be a branch name, tag name or commit id)."},
    )
    use_auth_token: bool = field(
        default=False,
        metadata={
            "help": "Will use the token generated when running `transformers-cli login` (necessary to use this script "
                    "with private models)."
        },
    )
    # prefix-tuning parameters
    add_enc_prefix: bool = field(
        default=False,
        metadata={"help": "Whether use prefix tuning"},
    )
    add_dec_prefix: bool = field(
        default=False,
        metadata={"help": "Whether use prefix tuning"},
    )
    add_cross_prefix: bool = field(
        default=False,
        metadata={"help": "Whether use prefix tuning"},
    )
    prefix_len: Optional[int] = field(
        default=10,
        metadata={"help": "length of prefix tokens"},
    )
    mid_dim: Optional[int] = field(
        default=512,
        metadata={"help": "dim of middle layer"},
    )
    # bitfit parameters
    tune_bias: bool = field(
        default=False,
        metadata={"help": "Whether tune bias terms"},
    )
    # LoRA parameters
    add_lora: bool = field(
        default=False,
        metadata={"help": "Whether to use lora for linear layers"},
    )
    lora_r: Optional[int] = field(
        default=8,
        metadata={"help": "rank of lora"},
    )
    lora_alpha: Optional[int] = field(
        default=16,
        metadata={"help": "scaling = alpha / r"},
    )

    drop_first_layers: Optional[int] = field(
        default=0,
        metadata={
            "help": "drop first k layers, work for both prefix and adapter, freeze transformer layers if fine-tuning"},
    )
    drop_first_adapter_layers: Optional[int] = field(
        default=0,
        metadata={"help": "drop first k adapter layers"},
    )
    drop_first_prefix_layers_enc: Optional[int] = field(
        default=0,
        metadata={"help": "drop first k prefix layers"},
    )
    drop_first_prefix_layers_dec: Optional[int] = field(
        default=0,
        metadata={"help": "drop first k prefix layers"},
    )
    drop_first_prefix_layers_cross: Optional[int] = field(
        default=0,
        metadata={"help": "drop first k prefix layers"},
    )
    add_adapter_gate: bool = field(
        default=True,
        metadata={"help": "add a gate to the adapter"},
    )
    add_prefix_gate: bool = field(
        default=True,
        metadata={"help": "add a gate to the prefix"},
    )
    add_lora_gate: bool = field(
        default=True,
        metadata={"help": "add a gate to the lora"},
    )
    add_central_gate: bool = field(
        default=False,
        metadata={"help": "add a shared gate"},
    )


def main():
    # See all possible arguments in src/transformers/training_args.py
    # or by passing the --help flag to this script.
    # We now keep distinct sets of args, for a cleaner separation of concerns.

    parser = HfArgumentParser((ModelArguments, DataTrainingArguments, TrainingArguments, MultiLingAdapterArguments))

    if len(sys.argv) == 2 and sys.argv[1].endswith(".json"):
        # If we pass only one argument to the script and it's the path to a json file,
        # let's parse it to get our arguments.
        model_args, data_args, training_args, adapter_args = parser.parse_json_file(
            json_file=os.path.abspath(sys.argv[1])
        )
    else:
        model_args, data_args, training_args, adapter_args = parser.parse_args_into_dataclasses()

    # Detecting last checkpoint.
    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:
            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."
            )

    # 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)],
    )
    logger.setLevel(logging.INFO if is_main_process(training_args.local_rank) else logging.WARN)

    # Log on each process the small summary:
    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}"
    )
    # Set the verbosity to info of the Transformers logger (on main process only):
    if is_main_process(training_args.local_rank):
        transformers.utils.logging.set_verbosity_info()
        transformers.utils.logging.enable_default_handler()
        transformers.utils.logging.enable_explicit_format()
    logger.info(f"Training/evaluation parameters {training_args}")

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

    # Get the datasets: you can either provide your own CSV/JSON training and evaluation files (see below)
    # or specify a GLUE benchmark task (the dataset will be downloaded automatically from the datasets Hub).
    #
    # For CSV/JSON files, this script will use as labels the column called 'label' and as pair of sentences the
    # sentences in columns called 'sentence1' and 'sentence2' if such column exists or the first two columns not named
    # label if at least two columns are provided.
    #
    # If the CSVs/JSONs contain only one non-label column, the script does single sentence classification on this
    # single column. 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 data_args.task_name is not None:
        # Downloading and loading a dataset from the hub.
        datasets = load_dataset("glue", data_args.task_name)
    else:
        # Loading a dataset from your local files.
        # CSV/JSON training and evaluation files are needed.
        data_files = {"train": data_args.train_file, "validation": data_args.validation_file}

        # Get the test dataset: you can provide your own CSV/JSON test file (see below)
        # when you use `do_predict` without specifying a GLUE benchmark task.
        if training_args.do_predict:
            if data_args.test_file is not None:
                train_extension = data_args.train_file.split(".")[-1]
                test_extension = data_args.test_file.split(".")[-1]
                assert (
                        test_extension == train_extension
                ), "`test_file` should have the same extension (csv or json) as `train_file`."
                data_files["test"] = data_args.test_file
            else:
                raise ValueError("Need either a GLUE task or a test file for `do_predict`.")

        for key in data_files.keys():
            logger.info(f"load a local file for {key}: {data_files[key]}")

        if data_args.train_file.endswith(".csv"):
            # Loading a dataset from local csv files
            datasets = load_dataset("csv", data_files=data_files)
        else:
            # Loading a dataset from local json files
            datasets = load_dataset("json", data_files=data_files)
    # See more about loading any type of standard or custom dataset at
    # https://huggingface.co/docs/datasets/loading_datasets.html.

    # Labels
    if data_args.task_name is not None:
        is_regression = data_args.task_name == "stsb"
        if not is_regression:
            label_list = datasets["train"].features["label"].names
            num_labels = len(label_list)
        else:
            num_labels = 1
    else:
        # Trying to have good defaults here, don't hesitate to tweak to your needs.
        is_regression = datasets["train"].features["label"].dtype in ["float32", "float64"]
        if is_regression:
            num_labels = 1
        else:
            # A useful fast method:
            # https://huggingface.co/docs/datasets/package_reference/main_classes.html#datasets.Dataset.unique
            label_list = datasets["train"].unique("label")
            label_list.sort()  # Let's sort it for determinism
            num_labels = len(label_list)

    # Load pretrained model and tokenizer
    #
    # In distributed training, the .from_pretrained methods guarantee that only one local process can concurrently
    # download model & vocab.
    config = AutoConfig.from_pretrained(
        model_args.config_name if model_args.config_name else model_args.model_name_or_path,
        num_labels=num_labels,
        finetuning_task=data_args.task_name,
        cache_dir=model_args.cache_dir,
        revision=model_args.model_revision,
        use_auth_token=True if model_args.use_auth_token else None,
    )
    # additional arguments
    config.add_enc_prefix = model_args.add_enc_prefix
    config.add_dec_prefix = model_args.add_dec_prefix
    config.add_cross_prefix = model_args.add_cross_prefix
    config.prefix_len = model_args.prefix_len
    config.mid_dim = model_args.mid_dim
    if 'bert' in model_args.model_name_or_path:
        num_layers = config.num_hidden_layers
    elif 'bart' in model_args.model_name_or_path:
        num_layers = config.encoder_layers
    config.add_adapter_gate = model_args.add_adapter_gate
    config.add_prefix_gate = model_args.add_prefix_gate
    config.tune_bias = model_args.tune_bias
    config.add_lora = model_args.add_lora
    config.lora_r = model_args.lora_r
    config.lora_alpha = model_args.lora_alpha
    config.add_lora_gate = model_args.add_lora_gate
    config.add_central_gate = model_args.add_central_gate
    config.early_stopping_patience = data_args.early_stopping_patience

    if model_args.drop_first_layers == 0:
        config.drop_first_prefix_layers_enc = list(range(model_args.drop_first_prefix_layers_enc))
        config.drop_first_prefix_layers_dec = list(range(model_args.drop_first_prefix_layers_dec))
        config.drop_first_prefix_layers_cross = list(range(model_args.drop_first_prefix_layers_cross))
    else:
        # override by drop_first_layers
        model_args.drop_first_adapter_layers = model_args.drop_first_layers
        config.drop_first_prefix_layers_enc = list(range(model_args.drop_first_layers))
        config.drop_first_prefix_layers_dec = list(range(model_args.drop_first_layers - num_layers))
        config.drop_first_prefix_layers_cross = list(range(model_args.drop_first_layers - num_layers))

    tokenizer = AutoTokenizer.from_pretrained(
        model_args.tokenizer_name if model_args.tokenizer_name else model_args.model_name_or_path,
        cache_dir=model_args.cache_dir,
        use_fast=model_args.use_fast_tokenizer,
        revision=model_args.model_revision,
        use_auth_token=True if model_args.use_auth_token else None,
    )
    model = AutoModelForSequenceClassification.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,
        revision=model_args.model_revision,
        use_auth_token=True if model_args.use_auth_token else None,
    )

    # Setup adapters
    if adapter_args.train_adapter:
        task_name = data_args.task_name or "glue"
        # check if adapter already exists, otherwise add it
        if task_name not in model.config.adapters:
            # resolve the adapter config
            adapter_config = AdapterConfig.load(
                adapter_args.adapter_config,
                non_linearity=adapter_args.adapter_non_linearity,
                reduction_factor=adapter_args.adapter_reduction_factor,
                leave_out=list(range(model_args.drop_first_adapter_layers))
            )
            # load a pre-trained from Hub if specified
            if adapter_args.load_adapter:
                model.load_adapter(
                    adapter_args.load_adapter,
                    config=adapter_config,
                    load_as=task_name,
                )
            # otherwise, add a fresh adapter
            else:
                model.add_adapter(task_name, config=adapter_config)
        # optionally load a pre-trained language adapter
        if adapter_args.load_lang_adapter:
            # resolve the language adapter config
            lang_adapter_config = AdapterConfig.load(
                adapter_args.lang_adapter_config,
                non_linearity=adapter_args.lang_adapter_non_linearity,
                reduction_factor=adapter_args.lang_adapter_reduction_factor,
            )
            # load the language adapter from Hub
            lang_adapter_name = model.load_adapter(
                adapter_args.load_lang_adapter,
                config=lang_adapter_config,
                load_as=adapter_args.language,
            )
        else:
            lang_adapter_name = None
        # Freeze all model weights except of those of this adapter
        model.train_adapter([task_name])
        # Set the adapters to be used in every forward pass
        if lang_adapter_name:
            model.set_active_adapters([lang_adapter_name, task_name])
        else:
            model.set_active_adapters([task_name])
    else:
        except_para_l = []
        if config.tune_bias:
            except_para_l.append('bias')
        if config.add_lora:
            except_para_l.append('lora')
        if any([config.add_enc_prefix, config.add_dec_prefix, config.add_cross_prefix]):
            except_para_l.append('prefix')
        if len(except_para_l) > 0:
            freeze_params(model, except_para_l=except_para_l)
        elif model_args.drop_first_layers > 0:
            freeze_params_by_layers(model, num_layers, num_frozen_layers=model_args.drop_first_layers)

        if adapter_args.load_adapter or adapter_args.load_lang_adapter:
            raise ValueError(
                "Adapters can only be loaded in adapters training mode."
                "Use --train_adapter to enable adapter training"
            )

    total_params = sum(p.numel() for p in model.parameters())
    trainable_params = sum(p.numel() for p in model.parameters() if p.requires_grad)
    logger.info(f"trainable_params: {trainable_params}, total_params: {total_params}")

    # Preprocessing the datasets
    if data_args.task_name is not None:
        sentence1_key, sentence2_key = task_to_keys[data_args.task_name]
    else:
        # Again, we try to have some nice defaults but don't hesitate to tweak to your use case.
        non_label_column_names = [name for name in datasets["train"].column_names if name != "label"]
        if "sentence1" in non_label_column_names and "sentence2" in non_label_column_names:
            sentence1_key, sentence2_key = "sentence1", "sentence2"
        else:
            if len(non_label_column_names) >= 2:
                sentence1_key, sentence2_key = non_label_column_names[:2]
            else:
                sentence1_key, sentence2_key = non_label_column_names[0], None

    # Padding strategy
    if data_args.pad_to_max_length:
        padding = "max_length"
    else:
        # We will pad later, dynamically at batch creation, to the max sequence length in each batch
        padding = False

    # Some models have set the order of the labels to use, so let's make sure we do use it.
    label_to_id = None
    if (
            model.config.label2id != PretrainedConfig(num_labels=num_labels).label2id
            and data_args.task_name is not None
            and not is_regression
    ):
        # Some have all caps in their config, some don't.
        label_name_to_id = {k.lower(): v for k, v in model.config.label2id.items()}
        if list(sorted(label_name_to_id.keys())) == list(sorted(label_list)):
            label_to_id = {i: int(label_name_to_id[label_list[i]]) for i in range(num_labels)}
        else:
            logger.warn(
                "Your model seems to have been trained with labels, but they don't match the dataset: ",
                f"model labels: {list(sorted(label_name_to_id.keys()))}, dataset labels: {list(sorted(label_list))}."
                "\nIgnoring the model labels as a result.",
            )
    elif data_args.task_name is None and not is_regression:
        label_to_id = {v: i for i, v in enumerate(label_list)}

    if data_args.max_seq_length > tokenizer.model_max_length:
        logger.warn(
            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 preprocess_function(examples):
        # Tokenize the texts
        args = (
            (examples[sentence1_key],) if sentence2_key is None else (examples[sentence1_key], examples[sentence2_key])
        )
        result = tokenizer(*args, padding=padding, max_length=max_seq_length, truncation=True)

        # Map labels to IDs (not necessary for GLUE tasks)
        if label_to_id is not None and "label" in examples:
            result["label"] = [(label_to_id[l] if l != -1 else -1) for l in examples["label"]]
        return result

    datasets = datasets.map(preprocess_function, batched=True, load_from_cache_file=not data_args.overwrite_cache)
    if training_args.do_train:
        if "train" not in datasets:
            raise ValueError("--do_train requires a train dataset")
        train_dataset = datasets["train"]
        if data_args.max_train_samples is not None:
            logger.warning(f'shuffling training set w. seed {data_args.data_seed}!')
            train_dataset_all = train_dataset.shuffle(seed=data_args.data_seed)
            train_dataset = train_dataset_all.select(range(data_args.max_train_samples))

    if training_args.do_eval:
        if "validation" not in datasets and "validation_matched" not in datasets:
            raise ValueError("--do_eval requires a validation dataset")
        eval_dataset = datasets["validation_matched" if data_args.task_name == "mnli" else "validation"]
        if data_args.max_val_samples is not None:
            eval_dataset = eval_dataset.select(range(data_args.max_val_samples))

    if training_args.do_predict or data_args.task_name is not None or data_args.test_file is not None:
        if "test" not in datasets and "test_matched" not in datasets:
            raise ValueError("--do_predict requires a test dataset")
        test_dataset = datasets["test_matched" if data_args.task_name == "mnli" else "test"]
        if data_args.max_test_samples is not None:
            test_dataset = test_dataset.select(range(data_args.max_test_samples))

    if data_args.train_as_val:
        test_dataset = eval_dataset
        eval_dataset = train_dataset_all.select(range(data_args.max_train_samples, data_args.max_train_samples + 1000))

    # 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]}.")

    # Get the metric function
    if data_args.task_name is not None:
        metric = load_metric("glue", data_args.task_name, keep_in_memory=True)

    # You can define your custom compute_metrics function. It takes an `EvalPrediction` object (a namedtuple with a
    # predictions and label_ids field) and has to return a dictionary string to float.
    def compute_metrics(p: EvalPrediction):
        preds = p.predictions[0] if isinstance(p.predictions, tuple) else p.predictions
        preds = np.squeeze(preds) if is_regression else np.argmax(preds, axis=1)
        if data_args.task_name is not None:
            result = metric.compute(predictions=preds, references=p.label_ids)
            if len(result) > 1:
                result["combined_score"] = np.mean(list(result.values())).item()
            return result
        elif is_regression:
            return {"mse": ((preds - p.label_ids) ** 2).mean().item()}
        else:
            return {"accuracy": (preds == p.label_ids).astype(np.float32).mean().item()}

    # Data collator will default to DataCollatorWithPadding, so we change it if we already did the padding.
    if data_args.pad_to_max_length:
        data_collator = default_data_collator
    elif training_args.fp16:
        data_collator = DataCollatorWithPadding(tokenizer, pad_to_multiple_of=8)
    else:
        data_collator = None

    # Initialize our Trainer
    trainer = Trainer(
        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,
        compute_metrics=compute_metrics,
        tokenizer=tokenizer,
        data_collator=data_collator,
        do_save_full_model=True,  # otherwise, P in AP may not be saved
        do_save_adapters=adapter_args.train_adapter,
    )

    # Training
    if training_args.do_train:
        checkpoint = None
        if last_checkpoint is not None:
            checkpoint = last_checkpoint
        elif os.path.isdir(model_args.model_name_or_path):
            # Check the config from that potential checkpoint has the right number of labels before using it as a
            # checkpoint.
            if AutoConfig.from_pretrained(model_args.model_name_or_path).num_labels == num_labels:
                checkpoint = model_args.model_name_or_path

        train_result = trainer.train(resume_from_checkpoint=checkpoint)
        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.save_model()  # Saves the tokenizer too for easy upload

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

    # Evaluation
    if training_args.do_eval:
        logger.info("*** Evaluate ***")

        # Loop to handle MNLI double evaluation (matched, mis-matched)
        tasks = [data_args.task_name]
        eval_datasets = [eval_dataset]
        if data_args.task_name == "mnli":
            tasks.append("mnli-mm")
            eval_datasets.append(datasets["validation_mismatched"])

        for eval_dataset, task in zip(eval_datasets, tasks):
            metrics = trainer.evaluate(eval_dataset=eval_dataset)

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

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

    if training_args.do_predict:
        logger.info("*** Test ***")

        # Loop to handle MNLI double evaluation (matched, mis-matched)
        tasks = [data_args.task_name]
        test_datasets = [test_dataset]
        # not evaluating test_mismatched
        # if data_args.task_name == "mnli":
        #     tasks.append("mnli-mm")
        #     test_datasets.append(datasets["test_mismatched"])

        for test_dataset, task in zip(test_datasets, tasks):
            # only do_predict if train_as_val
            # Removing the `label` columns because it contains -1 and Trainer won't like that.
            # test_dataset.remove_columns_("label")
            metrics = trainer.evaluate(eval_dataset=test_dataset, metric_key_prefix='test')
            trainer.log_metrics("test", metrics)
            trainer.save_metrics("test", metrics)

            predictions = trainer.predict(test_dataset=test_dataset).predictions
            predictions = np.squeeze(predictions) if is_regression else np.argmax(predictions, axis=1)

            output_test_file = os.path.join(training_args.output_dir, f"test_results_{task}.txt")
            if trainer.is_world_process_zero():
                with open(output_test_file, "w") as writer:
                    logger.info(f"***** Test results {task} *****")
                    writer.write("index\tprediction\n")
                    for index, item in enumerate(predictions):
                        if is_regression:
                            writer.write(f"{index}\t{item:3.3f}\n")
                        else:
                            item = label_list[item]
                            writer.write(f"{index}\t{item}\n")


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


if __name__ == "__main__":
    choose_gpu(min_gpu_memory=5000, retry=True)
    main()