run_longExp.py

import argparse
import os
os.environ["CUDA_VISIBLE_DEVICES"] = "0"
os.environ["KMP_DUPLICATE_LIB_OK"]="TRUE"
import torch
from exp.exp_MSN import Exp_MSN
from exp.exp_EvoMSN import Exp_EvoMSN
from exp.exp_EvoMSN_FSNet import Exp_EvoMSN_FSNet
import random
import numpy as np

fix_seed = 2021
random.seed(fix_seed)
torch.manual_seed(fix_seed)
np.random.seed(fix_seed)
torch.set_num_threads(6)

parser = argparse.ArgumentParser(description='Autoformer & Transformer family for Time Series Forecasting')

# basic config
parser.add_argument('--is_training', type=int, default=1, help='status')
parser.add_argument('--model_id', type=str, default='test', help='model id')
parser.add_argument('--model', type=str, default='FEDformer',
                    help='model name, options: [Autoformer, Informer, Transformer]')
parser.add_argument('--exp', type=str, default='Exp_Main', help='experiment type, options: [Exp_Main, Exp_Norm, Exp_Bilevel, Exp_Koopa, Exp_Minmax]')

# statistics prediction module config
parser.add_argument('--station_type', type=str, default='adaptive', help='type of SAN, options: [none, adaptive, minmax_adaptive]')
parser.add_argument('--station_lr', type=float, default=0.0001)
parser.add_argument('--top_k', type=int, default=4, help='find top-k period')
parser.add_argument('--num_kernels', type=int, default=3, help='for Inception')
parser.add_argument('--station_pretrain_epochs', type=int, default=5, help='statistics model train epochs')

# data loader
parser.add_argument('--data', type=str, default='ETTh2', help='dataset type')
parser.add_argument('--root_path', type=str, default='./datasets/ETT-small', help='root path of the data file')
parser.add_argument('--data_path', type=str, default='ETTh2.csv', help='data file')
parser.add_argument('--features', type=str, default='M',
                    help='forecasting task, options:[M, S, MS]; M:multivariate predict multivariate, S:univariate predict univariate, MS:multivariate 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('--scale', type=bool, default=True, help='Scale the whole data with StandardScaler()')
parser.add_argument('--checkpoints', type=str, default='./checkpoints/', help='location of forecast model checkpoints')
parser.add_argument('--stat_checkpoints', type=str, default='./stat_checkpoints/', help='location of stat model checkpoints')

# normalization
parser.add_argument('--revin', type=int, default=1, help='RevIN; True 1 False 0')
parser.add_argument('--affine', type=int, default=0, help='RevIN-affine; True 1 False 0')
parser.add_argument('--norm', type=str, default='none', help='normalization method, options:[none, revin, dishts]')
parser.add_argument('--dish_init', type=str, default='standard', help='ways to initialize DishTS parameters, options:[standard, avg, uniform]') 

# forecasting task
parser.add_argument('--seq_len', type=int, default=96, help='input sequence length')
parser.add_argument('--label_len', type=int, default=48, help='start token length')
parser.add_argument('--pred_len', type=int, default=96, help='prediction sequence length')
parser.add_argument('--n_series', type=int, default=1, help='number of series')
parser.add_argument('--test_freq', type=int, default=1, help='online test frequency')

# optimization
parser.add_argument('--num_workers', type=int, default=0, help='data loader num workers')
parser.add_argument('--itr', type=int, default=1, help='experiments times')
parser.add_argument('--train_epochs', type=int, default=10, help='train epochs')
parser.add_argument('--batch_size', type=int, default=32, help='batch size of train input data')
parser.add_argument('--test_batch_size', type=int, default=32, help='batch size of test input data')
parser.add_argument('--patience', type=int, default=3, help='early stopping patience')
parser.add_argument('--learning_rate', type=float, default=0.0001, help='optimizer learning rate')
parser.add_argument('--des', type=str, default='test', help='exp description')
parser.add_argument('--loss', type=str, default='mse', help='loss function')
parser.add_argument('--lradj', type=str, default='type1', help='adjust learning rate')
parser.add_argument('--use_amp', action='store_true', help='use automatic mixed precision training', default=False)
parser.add_argument('--online_learning', type=str, default='backbone', help='online strategy, options: [none, stat, backbone, stat_backbone]')
parser.add_argument('--buffer_size', type=int, default=1, help='Buffer size of test input data for backbone')

# supplementary config for FEDformer model
parser.add_argument('--version', type=str, default='Fourier',
                    help='for FEDformer, there are two versions to choose, options: [Fourier, Wavelets]')
parser.add_argument('--mode_select', type=str, default='random',
                    help='for FEDformer, there are two mode selection method, options: [random, low]')
parser.add_argument('--modes', type=int, default=64, help='modes to be selected random 64')
parser.add_argument('--L', type=int, default=3, help='ignore level')
parser.add_argument('--base', type=str, default='legendre', help='mwt base')
parser.add_argument('--cross_activation', type=str, default='tanh',
                    help='mwt cross atention activation function tanh or softmax')
# PatchTST
parser.add_argument('--fc_dropout', type=float, default=0.05, help='fully connected dropout')
parser.add_argument('--head_dropout', type=float, default=0.0, help='head dropout')
parser.add_argument('--patch_len', type=int, default=16, help='patch length')
parser.add_argument('--stride', type=int, default=8, help='stride')
parser.add_argument('--padding_patch', default='end', help='None: None; end: padding on the end')
parser.add_argument('--subtract_last', type=int, default=0, help='0: subtract mean; 1: subtract last')
parser.add_argument('--decomposition', type=int, default=0, help='decomposition; True 1 False 0')
parser.add_argument('--kernel_size', type=int, default=25, help='decomposition-kernel')
parser.add_argument('--individual_head', type=int, default=0, help='individual head; True 1 False 0')

# CrossFormer
parser.add_argument('--win_size', type=int, default=2, help='window size for segment merge')
parser.add_argument('--cross_factor', type=int, default=10, help='num of routers in Cross-Dimension Stage of TSA (c)')

# DLinear
parser.add_argument('--individual', action='store_true', default=False,
                    help='DLinear: a linear layer for each variate(channel) individually')
# Formers 
parser.add_argument('--embed_type', type=int, default=0,
                    help='0: default 1: value embedding + temporal embedding + positional embedding 2: value embedding + temporal embedding 3: value embedding + positional embedding 4: value embedding')
parser.add_argument('--enc_in', type=int, default=7,
                    help='encoder input size')  # DLinear with --individual, use this hyperparameter as the number of channels
parser.add_argument('--dec_in', type=int, default=7, help='decoder input size')
parser.add_argument('--c_out', type=int, default=7, help='output size')
parser.add_argument('--d_model', type=int, default=512, help='dimension of model')
parser.add_argument('--n_heads', type=int, default=8, help='num of heads')
parser.add_argument('--e_layers', type=int, default=2, help='num of encoder layers')
parser.add_argument('--d_layers', type=int, default=1, help='num of decoder layers')
parser.add_argument('--d_ff', type=int, default=2048, help='dimension of fcn')
parser.add_argument('--moving_avg', type=int, default=25, help='window size of moving average')
parser.add_argument('--factor', type=int, default=3, help='attn factor')
parser.add_argument('--distil', action='store_false',
                    help='whether to use distilling in encoder, using this argument means not using distilling',
                    default=True)
parser.add_argument('--dropout', type=float, default=0.05, help='dropout')
parser.add_argument('--embed', type=str, default='timeF',
                    help='time features encoding, options:[timeF, fixed, learned]')
parser.add_argument('--activation', type=str, default='gelu', help='activation')
parser.add_argument('--output_attention', action='store_true', help='whether to output attention in ecoder')
parser.add_argument('--do_predict', action='store_true', help='whether to predict unseen future data')

# 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='0,1,2,3', 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()
if args.features == 'S':
    args.enc_in, args.dec_in, args.c_out = 1, 1, 1

args.use_gpu = True if torch.cuda.is_available() and args.use_gpu else False

# print('gpu:', torch.cuda.is_available())
# print('use_gpu:', args.use_gpu)

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]

print('Args in experiment:')
print(args)

exp_dict = {
    'Exp_EvoMSN': Exp_EvoMSN,
    'Exp_MSN': Exp_MSN,
    'Exp_EvoMSN_FSNet': Exp_EvoMSN_FSNet
    }

Exp = exp_dict[args.exp]
mses = []
maes = []

if args.is_training:
    for ii in range(args.itr):
        # setting record of experiments
        setting = '{}_{}_{}_ft{}_sl{}_ll{}_pl{}_dm{}_epoch{}_station{}_online{}_{}_{}'.format(
            args.model_id,
            args.model,
            args.data,
            args.features,
            args.seq_len,
            args.label_len,
            args.pred_len,
            args.d_model,
            args.train_epochs,
            args.station_type,
            args.online_learning,
            args.exp, ii)

        exp = Exp(args)  # set experiments
        print('>>>>>>>start training : {}>>>>>>>>>>>>>>>>>>>>>>>>>>'.format(setting))
        exp.train(setting)

        print('>>>>>>>testing : {}<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<'.format(setting))
        mse, mae = exp.test(setting)
        mses.append(mse)
        maes.append(mae)

        if args.do_predict:
            print('>>>>>>>predicting : {}<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<'.format(setting))
            exp.predict(setting, True)

        torch.cuda.empty_cache()
    print('average mse:{0:.3f}±{1:.3f}, mae:{2:.3f}±{3:.3f}'.format(np.mean(mses), np.std(mses), np.mean(maes),
                                                                    np.std(maes)))
else:
    ii = 0
    setting = '{}_{}_{}_ft{}_sl{}_ll{}_pl{}_dm{}_epoch{}_scale{}_station{}_cotrain{}_online{}_{}_{}_{}'.format(
            args.model_id,
            args.model,
            args.data,
            args.features,
            args.seq_len,
            args.label_len,
            args.pred_len,
            args.d_model,
            args.train_epochs,
            args.scale,
            args.station_type,
            args.cotrain,
            args.online_learning,
            args.norm,
            args.exp, ii)
    
    exp = Exp(args)  # set experiments
    print('>>>>>>>testing : {}<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<'.format(setting))
    exp.test(setting, test=1)
    torch.cuda.empty_cache()