深度学习模型训练-argparse传参

方式一：argparse

适合服务器运行时，直接修改

import argparse
from collections import defaultdict
import json


def parse_args():
    parser = argparse.ArgumentParser()
    parser.add_argument(
        "--use_model", type=str, default="seq2seq", help="model for train, seq2seq, wavenet, transformer"
    )
    parser.add_argument(
        "--data_dir", type=str, default="../data/international-airline-passengers.csv", help="data directory"
    )
    parser.add_argument("--model_dir", type=str, default="../weights/checkpoint", help="saved checkpoint directory")
    parser.add_argument("--saved_model_dir", type=str, default="../weights", help="saved pb directory")
    parser.add_argument("--log_dir", type=str, default="../data/logs", help="saved pb directory")
    parser.add_argument("--input_seq_length", type=int, default=20, help="sequence length for input")
    parser.add_argument("--output_seq_length", type=int, default=5, help="sequence length for output")
    parser.add_argument("--n_epochs", type=int, default=10, help="Number of training epochs")
    parser.add_argument("--batch_size", type=int, default=16, help="Batch size for training")
    parser.add_argument("--learning_rate", type=float, default=3e-4, help="learning rate for training")

    args = parser.parse_args()
    return args


class Config(object):
    def __init__(self):
        self.params = defaultdict()

    def from_json_file(self, json_file):
        with open(json_file, "r") as f:
            self.params = json.load(f)

    def to_json_string(self, json_file, params):
        with open(json_file, "w") as f:
            json.dump(params, f)


if __name__ == "__main__":
    args = parse_args()
    params = vars(args)

    config = Config()
    config.to_json_string("./config.json", params)
    # config.from_json_file('./config.json')