import torch import numpy as np import random from exp.exp_main import Exp_Main import argparse import time fix_seed = 1024 random.seed(fix_seed) torch.manual_seed(fix_seed) np.random.seed(fix_seed) if __name__ == '__main__': parser = argparse.ArgumentParser(description='Multivariate Time Series Forecasting') # basic config parser.add_argument('--is_training', type=int, default=1, help='status') parser.add_argument('--model', type=str, default='PathFormer', help='model name, options: [PathFormer]') parser.add_argument('--model_id', type=str, default="ETT.sh") # data loader parser.add_argument('--data', type=str, default='custom', help='dataset type') parser.add_argument('--root_path', type=str, default='./dataset/weather', help='root path of the data file') parser.add_argument('--data_path', type=str, default='weather.csv', help='data file') parser.add_argument('--features', type=str, default='M', help='forecasting task, options:[M, S]; M:multivariate predict multivariate, S:univariate predict univariate') parser.add_argument('--target', type=str, default='OT', help='target feature in S or MS task') parser.add_argument('--freq', type=str, default='h', help='freq for time features encoding, options:[s:secondly, t:minutely, h:hourly, d:daily, b:business days, w:weekly, m:monthly], you can also use more detailed freq like 15min or 3h') parser.add_argument('--checkpoints', type=str, default='./checkpoints/', help='location of model checkpoints') # forecasting task parser.add_argument('--seq_len', type=int, default=96, help='input sequence length') parser.add_argument('--pred_len', type=int, default=96, help='prediction sequence length') parser.add_argument('--individual', action='store_true', default=False, help='DLinear: a linear layer for each variate(channel) individually') # model parser.add_argument('--d_model', type=int, default=16) parser.add_argument('--d_ff', type=int, default=64) parser.add_argument('--num_nodes', type=int, default=21) parser.add_argument('--layer_nums', type=int, default=3) parser.add_argument('--k', type=int, default=2, help='choose the Top K patch size at the every layer ') parser.add_argument('--num_experts_list', type=list, default=[4, 4, 4]) parser.add_argument('--patch_size_list', nargs='+', type=int, default=[16,12,8,32,12,8,6,4,8,6,4,2]) parser.add_argument('--do_predict', action='store_true', help='whether to predict unseen future data') parser.add_argument('--revin', type=int, default=1, help='whether to apply RevIN') parser.add_argument('--drop', type=float, default=0.1, help='dropout ratio') parser.add_argument('--embed', type=str, default='timeF', help='time features encoding, options:[timeF, fixed, learned]') parser.add_argument('--residual_connection', type=int, default=0) parser.add_argument('--metric',type=str, default='mae') # optimization parser.add_argument('--num_workers', type=int, default=10, help='data loader num workers') parser.add_argument('--itr', type=int, default=1, help='experiments times') parser.add_argument('--train_epochs', type=int, default=20, help='train epochs') parser.add_argument('--batch_size', type=int, default=64, help='batch size of train input data') parser.add_argument('--patience', type=int, default=5, help='early stopping patience') parser.add_argument('--learning_rate', type=float, default=0.001, help='optimizer learning rate') parser.add_argument('--lradj', type=str, default='TST', help='adjust learning rate') parser.add_argument('--use_amp', action='store_true', help='use automatic mixed precision training', default=False) parser.add_argument('--pct_start', type=float, default=0.4, help='pct_start') # GPU parser.add_argument('--use_gpu', type=bool, default=True, help='use gpu') parser.add_argument('--gpu', type=int, default=0, help='gpu') parser.add_argument('--use_multi_gpu', action='store_true', help='use multiple gpus', default=False) parser.add_argument('--devices', type=str, default='2', help='device ids of multile gpus') parser.add_argument('--test_flop', action='store_true', default=False, help='See utils/tools for usage') args = parser.parse_args() args.use_gpu = True if torch.cuda.is_available() and args.use_gpu else False if args.use_gpu and args.use_multi_gpu: args.dvices = args.devices.replace(' ', '') device_ids = args.devices.split(',') args.device_ids = [int(id_) for id_ in device_ids] args.gpu = args.device_ids[0] args.patch_size_list = np.array(args.patch_size_list).reshape(args.layer_nums, -1).tolist() print('Args in experiment:') print(args) Exp = Exp_Main if args.is_training: for ii in range(args.itr): # setting record of experiments setting = '{}_{}_ft{}_sl{}_pl{}_{}'.format( args.model_id, args.model, args.data_path[:-4], args.features, args.seq_len, args.pred_len, ii) exp = Exp(args) # set experiments print('>>>>>>>start training : {}>>>>>>>>>>>>>>>>>>>>>>>>>>'.format(setting)) exp.train(setting) time_now = time.time() print('>>>>>>>testing : {}<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<'.format(setting)) exp.test(setting) print('Inference time: ', time.time() - time_now) if args.do_predict: print('>>>>>>>predicting : {}<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<'.format(setting)) exp.predict(setting, True) torch.cuda.empty_cache() else: ii = 0 setting = '{}_{}_ft{}_sl{}_pl{}_{}'.format( args.model_id, args.model, args.data_path[:-4], args.features, args.seq_len, args.pred_len, ii) exp = Exp(args) # set experiments print('>>>>>>>testing : {}<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<'.format(setting)) exp.test(setting, test=1) torch.cuda.empty_cache()