#!/usr/bin/env python
from transformers import AutoModelForSequenceClassification, Trainer, TrainingArguments
from sklearn.metrics import accuracy_score, precision_recall_fscore_support
from datasets import load_from_disk
from transformers import AutoTokenizer
import random
import logging
import sys
import argparse
import os
import torch
import pandas as pd
import pathlib
import json

if __name__ == "__main__":

    parser = argparse.ArgumentParser()

    # hyperparameters sent by the client are passed as command-line arguments to the script.
    parser.add_argument("--epochs", type=int, default=3)
    parser.add_argument("--train_batch_size", type=int, default=32)
    parser.add_argument("--eval_batch_size", type=int, default=64)
    parser.add_argument("--warmup_steps", type=int, default=500)
    parser.add_argument("--model_name", type=str)
    parser.add_argument("--learning_rate", type=str, default=5e-5)

    # Data, model, and output directories
    parser.add_argument("--output-data-dir", type=str, default=os.environ["SM_OUTPUT_DATA_DIR"])
    parser.add_argument("--model-dir", type=str, default=os.environ["SM_MODEL_DIR"])
    parser.add_argument("--n_gpus", type=str, default=os.environ["SM_NUM_GPUS"])
    parser.add_argument("--training_dir", type=str, default=os.environ["SM_CHANNEL_TRAIN"])
    parser.add_argument("--test_dir", type=str, default=os.environ["SM_CHANNEL_TEST"])

    args, _ = parser.parse_known_args()

    # Set up logging
    logger = logging.getLogger(__name__)

    logging.basicConfig(
        level=logging.getLevelName("INFO"),
        handlers=[logging.StreamHandler(sys.stdout)],
        format="%(asctime)s - %(name)s - %(levelname)s - %(message)s",
    )
    
    print('output_data_dir location',args.output_data_dir)
    print('model_dir location',args.model_dir)

    # load datasets
    train_dataset = load_from_disk(args.training_dir)
    test_dataset = load_from_disk(args.test_dir)

    logger.info(f" loaded train_dataset length is: {len(train_dataset)}")
    logger.info(f" loaded test_dataset length is: {len(test_dataset)}")

    # compute metrics function for binary classification
    def compute_metrics(pred):
        labels = pred.label_ids
        preds = pred.predictions.argmax(-1)
        precision, recall, f1, _ = precision_recall_fscore_support(labels, preds, average="binary")
        acc = accuracy_score(labels, preds)
        return {"accuracy": acc, "f1": f1, "precision": precision, "recall": recall}

    # download model from model hub
    model = AutoModelForSequenceClassification.from_pretrained(args.model_name)
    tokenizer = AutoTokenizer.from_pretrained(args.model_name)

    # define training args
    training_args = TrainingArguments(
        output_dir=args.model_dir,
        num_train_epochs=args.epochs,
        per_device_train_batch_size=args.train_batch_size,
        per_device_eval_batch_size=args.eval_batch_size,
        warmup_steps=args.warmup_steps,
        evaluation_strategy="epoch",
        logging_dir=f"{args.output_data_dir}/logs",
        learning_rate=float(args.learning_rate),
    )

    # create Trainer instance
    trainer = Trainer(
        model=model,
        args=training_args,
        compute_metrics=compute_metrics,
        train_dataset=train_dataset,
        eval_dataset=test_dataset,
        tokenizer=tokenizer,
    )

    # train model
    trainer.train()

    # evaluate model
    eval_result = trainer.evaluate(eval_dataset=test_dataset)
    
    # writes eval result to file which can be accessed later in s3 ouput
    with open(os.path.join(args.output_data_dir, "eval_results_s3.txt"), "w") as writer:
        print(f"***** Eval results *****")
        for key, value in sorted(eval_result.items()):
            writer.write(f"{key} = {value}\n")

    evaluation_path = "/opt/ml/model/evaluation.json"
    with open(evaluation_path, "w+") as f:
        f.write(json.dumps(eval_result))
        
    # Saves the model to s3
    trainer.save_model(args.model_dir)
    
    
    # #Predictions on each row of the test dataset
    print('--- Inference For Downstream Analysis---')
    
    arr_text = [tokenizer.decode(ele["input_ids"]) for ele in test_dataset]
        
    predictions = trainer.predict(test_dataset)
    arr_preds = predictions.predictions.argmax(-1)
    arr_labels = predictions.label_ids
    
    df_res = pd.DataFrame({'text':arr_text, 'preds':arr_preds, 'labels':arr_labels})
    
    df_res.to_csv(args.output_data_dir + '/preds_data.csv', index=False)
#    df_res.to_csv(args.model_dir + '/preds_data.csv', index=False)
    
#     default_bucket = 'sagemaker-us-east-1-114175483951'
#     s3_prefix = "RegMLNB"

#     output_destination = "s3://{}/{}/data".format(default_bucket, s3_prefix)
#     print('output_destination:',output_destination)

#     df_res.to_csv(output_destination + '/preds_data.csv', index=False)
    
#     print('Save model to S3..', output_destination + '/model/')

#    trainer.save_model(output_destination + '/model/model.tar.gz')