TimeMixer 在 PeMS03 上复现不出来论文中的效果，请问如何才能复现出论文中的效果？ #195
Description
TimeMixer fails to replicate the results reported in the paper on PeMS03. How can I reproduce the results from the paper?
Based on BasicTS. The configuration file is as follows:
import os
import sys
from easydict import EasyDict
sys.path.append(os.path.abspath(file + '/../../..'))
from basicts.metrics import masked_mae, masked_mse, masked_mape, masked_rmse
from basicts.data import TimeSeriesForecastingDataset
from basicts.runners import SimpleTimeSeriesForecastingRunner
from basicts.scaler import ZScoreScaler
from basicts.utils import get_regular_settings
from baselines.TimeMixer.arch import TimeMixer
############################## Hot Parameters ##############################
Dataset & Metrics configuration
DATA_NAME = 'PEMS03' # Dataset name
regular_settings = get_regular_settings(DATA_NAME)
INPUT_LEN = regular_settings['INPUT_LEN'] # Length of input sequence
OUTPUT_LEN = regular_settings['OUTPUT_LEN'] # Length of output sequence
TRAIN_VAL_TEST_RATIO = regular_settings['TRAIN_VAL_TEST_RATIO'] # Train/Validation/Test split ratios
NORM_EACH_CHANNEL = regular_settings['NORM_EACH_CHANNEL'] # Whether to normalize each channel of the data
RESCALE = regular_settings['RESCALE'] # Whether to rescale the data
NULL_VAL = regular_settings['NULL_VAL'] # Null value in the data
Model architecture and parameters
MODEL_ARCH = TimeMixer
NUM_NODES = 358
MODEL_PARAM = {
"enc_in": NUM_NODES, # num nodes
"dec_in": NUM_NODES,
"c_out": NUM_NODES,
"seq_len": INPUT_LEN,
"label_len": 0, # start token length used in decoder
"pred_len": OUTPUT_LEN, # prediction sequence length
"factor": 3, # attn factor
"down_sampling_window": 2,
"down_sampling_layers": 3,
"top_k":5,
"down_sampling_method": 'avg',
"channel_independence": 0,
"d_model": 128,
"moving_avg": 25, # window size of moving average. This is a CRUCIAL hyper-parameter.
"n_heads": 8,
"e_layers": 5, # num of encoder layers
"d_layers": 1, # num of decoder layers
"d_ff": 256,
"distil": True,
"sigma" : 0.2,
"dropout": 0.1,
"freq": 'h',
"use_norm" : 0,
"decomp_method" : 'moving_avg', # dft_decomp or moving_avg
"output_attention": False,
"embed": "timeF", # [timeF, fixed, learned]
"activation": "gelu",
"num_time_features": 3, # number of used time features
"time_of_day_size": 24,
"day_of_week_size": 7,
"day_of_month_size": 31,
# "day_of_year_size": 366
}
NUM_EPOCHS = 50
############################## General Configuration ##############################
CFG = EasyDict()
General settings
CFG.DESCRIPTION = 'TimeMixer 010701'
CFG.GPU_NUM = 1 # Number of GPUs to use (0 for CPU mode)
Runner
CFG.RUNNER = SimpleTimeSeriesForecastingRunner
############################## Dataset Configuration ##############################
CFG.DATASET = EasyDict()
Dataset settings
CFG.DATASET.NAME = DATA_NAME
CFG.DATASET.TYPE = TimeSeriesForecastingDataset
CFG.DATASET.PARAM = EasyDict({
'dataset_name': DATA_NAME,
'train_val_test_ratio': TRAIN_VAL_TEST_RATIO,
'input_len': INPUT_LEN,
'output_len': OUTPUT_LEN,
# 'mode' is automatically set by the runner
})
############################## Scaler Configuration ##############################
CFG.SCALER = EasyDict()
Scaler settings
CFG.SCALER.TYPE = ZScoreScaler # Scaler class
CFG.SCALER.PARAM = EasyDict({
'dataset_name': DATA_NAME,
'train_ratio': TRAIN_VAL_TEST_RATIO[0],
'norm_each_channel': NORM_EACH_CHANNEL,
'rescale': RESCALE,
})
############################## Model Configuration ##############################
CFG.MODEL = EasyDict()
Model settings
CFG.MODEL.NAME = MODEL_ARCH.name
CFG.MODEL.ARCH = MODEL_ARCH
CFG.MODEL.PARAM = MODEL_PARAM
CFG.MODEL.FORWARD_FEATURES = [0, 1, 2, 3]
CFG.MODEL.TARGET_FEATURES = [0]
############################## Metrics Configuration ##############################
CFG.METRICS = EasyDict()
Metrics settings
CFG.METRICS.FUNCS = EasyDict({
'MAE': masked_mae,
'RMSE': masked_rmse,
'MAPE': masked_mape
})
CFG.METRICS.TARGET = 'MAE'
CFG.METRICS.NULL_VAL = NULL_VAL
############################## Training Configuration ##############################
CFG.TRAIN = EasyDict()
CFG.TRAIN.NUM_EPOCHS = NUM_EPOCHS
CFG.TRAIN.CKPT_SAVE_DIR = os.path.join(
'checkpoints',
MODEL_ARCH.name,
'_'.join([DATA_NAME, str(CFG.TRAIN.NUM_EPOCHS), str(INPUT_LEN), str(OUTPUT_LEN)])
)
CFG.TRAIN.LOSS = masked_mae
Optimizer settings
CFG.TRAIN.OPTIM = EasyDict()
CFG.TRAIN.OPTIM.TYPE = "Adam"
CFG.TRAIN.OPTIM.PARAM = {
"lr": 0.001,
}
Learning rate scheduler settings
CFG.TRAIN.LR_SCHEDULER = EasyDict()
CFG.TRAIN.LR_SCHEDULER.TYPE = "MultiStepLR"
CFG.TRAIN.LR_SCHEDULER.PARAM = {
"milestones": [1, 25, 50],
"gamma": 0.5
}
CFG.TRAIN.CLIP_GRAD_PARAM = {
'max_norm': 5.0
}
Train data loader settings
CFG.TRAIN.DATA = EasyDict()
CFG.TRAIN.DATA.BATCH_SIZE = 16
CFG.TRAIN.DATA.SHUFFLE = True
############################## Validation Configuration ##############################
CFG.VAL = EasyDict()
CFG.VAL.INTERVAL = 1
CFG.VAL.DATA = EasyDict()
CFG.VAL.DATA.BATCH_SIZE = 32
############################## Test Configuration ##############################
CFG.TEST = EasyDict()
CFG.TEST.INTERVAL = 1
CFG.TEST.DATA = EasyDict()
CFG.TEST.DATA.BATCH_SIZE = 32
############################## Evaluation Configuration ##############################
CFG.EVAL = EasyDict()
评估参数
评估时的预测时间范围。默认值为 []。注意：HORIZONS[i] 指的是在 ”第 i 个时间片“ 上进行测试，表示该时间片的损失（Loss）。
这是时空预测中的常见配置。对于长序列预测，建议将 HORIZONS 保持为默认值 []，以避免引发误解。
CFG.EVAL.HORIZONS = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12]
CFG.EVAL.USE_GPU = True # 是否在评估时使用 GPU。默认值：True
The log file is as follows:
2026-01-07 19:05:27,929 - easytorch-training - INFO - Initializing training.
2026-01-07 19:05:27,930 - easytorch-training - INFO - Set clip grad, param: {'max_norm': 5.0}
2026-01-07 19:05:27,930 - easytorch-training - INFO - Building training data loader.
2026-01-07 19:05:27,983 - easytorch-training - INFO - Train dataset length: 15619
2026-01-07 19:05:28,558 - easytorch-training - INFO - Set optim: Adam (
Parameter Group 0
amsgrad: False
betas: (0.9, 0.999)
capturable: False
decoupled_weight_decay: False
differentiable: False
eps: 1e-08
foreach: None
fused: None
lr: 0.001
maximize: False
weight_decay: 0
)
2026-01-07 19:05:28,558 - easytorch-training - INFO - Set lr_scheduler: <torch.optim.lr_scheduler.MultiStepLR object at 0x000001D71E5AB8D0>
2026-01-07 19:05:28,559 - easytorch-training - INFO - Initializing validation.
2026-01-07 19:05:28,559 - easytorch-training - INFO - Building val data loader.
2026-01-07 19:05:28,576 - easytorch-training - INFO - Validation dataset length: 5134
2026-01-07 19:05:28,593 - easytorch-training - INFO - Test dataset length: 5134
2026-01-07 19:05:28,594 - easytorch-training - INFO - Number of parameters: 1002818
2026-01-07 19:05:28,594 - easytorch-training - INFO - Epoch 1 / 50
2026-01-07 19:06:04,554 - easytorch-training - INFO - Result : [train/time: 35.96 (s), train/lr: 1.00e-03, train/loss: 19.8201, train/MAE: 19.8201, train/RMSE: 29.9873, train/MAPE: 0.2512]
2026-01-07 19:06:04,556 - easytorch-training - INFO - Start validation.
2026-01-07 19:06:07,083 - easytorch-training - INFO - Result : [val/time: 2.53 (s), val/loss: 18.5760, val/MAE: 18.5760, val/RMSE: 26.9742, val/MAPE: 0.2503]
2026-01-07 19:06:07,125 - easytorch-training - INFO - Checkpoint checkpoints\TimeMixer\PEMS03_50_96_12\bd7b81eefca29cd030a5d8a903f93b6d\TimeMixer_best_val_MAE.pt saved
2026-01-07 19:06:09,444 - easytorch-training - INFO - Evaluate best model on test data for horizon 1, Test MAE: 13.7364, Test RMSE: 20.3010, Test MAPE: 0.1788
2026-01-07 19:06:09,447 - easytorch-training - INFO - Evaluate best model on test data for horizon 2, Test MAE: 15.0376, Test RMSE: 22.4322, Test MAPE: 0.2069
2026-01-07 19:06:09,449 - easytorch-training - INFO - Evaluate best model on test data for horizon 3, Test MAE: 15.8854, Test RMSE: 24.0931, Test MAPE: 0.1920
2026-01-07 19:06:09,452 - easytorch-training - INFO - Evaluate best model on test data for horizon 4, Test MAE: 17.4436, Test RMSE: 26.2315, Test MAPE: 0.2298
2026-01-07 19:06:09,455 - easytorch-training - INFO - Evaluate best model on test data for horizon 5, Test MAE: 18.5441, Test RMSE: 27.5700, Test MAPE: 0.3042
2026-01-07 19:06:09,457 - easytorch-training - INFO - Evaluate best model on test data for horizon 6, Test MAE: 19.2197, Test RMSE: 28.3666, Test MAPE: 0.3327
2026-01-07 19:06:09,460 - easytorch-training - INFO - Evaluate best model on test data for horizon 7, Test MAE: 19.2155, Test RMSE: 29.0167, Test MAPE: 0.2695
2026-01-07 19:06:09,462 - easytorch-training - INFO - Evaluate best model on test data for horizon 8, Test MAE: 19.3406, Test RMSE: 29.5341, Test MAPE: 0.1995
2026-01-07 19:06:09,464 - easytorch-training - INFO - Evaluate best model on test data for horizon 9, Test MAE: 20.9926, Test RMSE: 31.4945, Test MAPE: 0.3267
2026-01-07 19:06:09,467 - easytorch-training - INFO - Evaluate best model on test data for horizon 10, Test MAE: 20.7622, Test RMSE: 31.6503, Test MAPE: 0.2782
2026-01-07 19:06:09,470 - easytorch-training - INFO - Evaluate best model on test data for horizon 11, Test MAE: 21.7396, Test RMSE: 33.1682, Test MAPE: 0.2415
2026-01-07 19:06:09,472 - easytorch-training - INFO - Evaluate best model on test data for horizon 12, Test MAE: 21.9008, Test RMSE: 33.5001, Test MAPE: 0.2450
2026-01-07 19:06:09,498 - easytorch-training - INFO - Result : [test/time: 2.37 (s), test/loss: 18.6055, test/MAE: 18.6516, test/RMSE: 28.4004, test/MAPE: 0.2504]
2026-01-07 19:06:09,537 - easytorch-training - INFO - Checkpoint checkpoints\TimeMixer\PEMS03_50_96_12\bd7b81eefca29cd030a5d8a903f93b6d\TimeMixer_01.pt saved
2026-01-07 19:06:09,538 - easytorch-training - INFO - The estimated training finish time is 2026-01-07 19:39:35
2026-01-07 19:06:09,538 - easytorch-training - INFO - Epoch 2 / 50
2026-01-07 19:06:44,470 - easytorch-training - INFO - Result : [train/time: 34.93 (s), train/lr: 5.00e-04, train/loss: 15.3761, train/MAE: 15.3761, train/RMSE: 24.0199, train/MAPE: 0.1637]
2026-01-07 19:06:44,472 - easytorch-training - INFO - Start validation.
2026-01-07 19:06:47,069 - easytorch-training - INFO - Result : [val/time: 2.60 (s), val/loss: 15.2532, val/MAE: 15.2532, val/RMSE: 22.4800, val/MAPE: 0.1720]
2026-01-07 19:06:47,108 - easytorch-training - INFO - Checkpoint checkpoints\TimeMixer\PEMS03_50_96_12\bd7b81eefca29cd030a5d8a903f93b6d\TimeMixer_best_val_MAE.pt saved
2026-01-07 19:06:49,409 - easytorch-training - INFO - Evaluate best model on test data for horizon 1, Test MAE: 12.7025, Test RMSE: 19.2495, Test MAPE: 0.1290
2026-01-07 19:06:49,412 - easytorch-training - INFO - Evaluate best model on test data for horizon 2, Test MAE: 13.4885, Test RMSE: 20.7202, Test MAPE: 0.1424
2026-01-07 19:06:49,413 - easytorch-training - INFO - Evaluate best model on test data for horizon 3, Test MAE: 14.3366, Test RMSE: 22.0355, Test MAPE: 0.1650
2026-01-07 19:06:49,415 - easytorch-training - INFO - Evaluate best model on test data for horizon 4, Test MAE: 15.1585, Test RMSE: 23.2090, Test MAPE: 0.1829
2026-01-07 19:06:49,418 - easytorch-training - INFO - Evaluate best model on test data for horizon 5, Test MAE: 15.4154, Test RMSE: 23.8056, Test MAPE: 0.1656
2026-01-07 19:06:49,421 - easytorch-training - INFO - Evaluate best model on test data for horizon 6, Test MAE: 16.2049, Test RMSE: 24.7434, Test MAPE: 0.2082
2026-01-07 19:06:49,423 - easytorch-training - INFO - Evaluate best model on test data for horizon 7, Test MAE: 16.4726, Test RMSE: 25.3034, Test MAPE: 0.2051
2026-01-07 19:06:49,426 - easytorch-training - INFO - Evaluate best model on test data for horizon 8, Test MAE: 16.7977, Test RMSE: 25.8571, Test MAPE: 0.1793
2026-01-07 19:06:49,428 - easytorch-training - INFO - Evaluate best model on test data for horizon 9, Test MAE: 17.4111, Test RMSE: 26.5948, Test MAPE: 0.2301
2026-01-07 19:06:49,431 - easytorch-training - INFO - Evaluate best model on test data for horizon 10, Test MAE: 17.3795, Test RMSE: 26.7412, Test MAPE: 0.2050
2026-01-07 19:06:49,433 - easytorch-training - INFO - Evaluate best model on test data for horizon 11, Test MAE: 18.0088, Test RMSE: 27.5158, Test MAPE: 0.2002
2026-01-07 19:06:49,436 - easytorch-training - INFO - Evaluate best model on test data for horizon 12, Test MAE: 18.3735, Test RMSE: 28.0315, Test MAPE: 0.2057
2026-01-07 19:06:49,459 - easytorch-training - INFO - Result : [test/time: 2.35 (s), test/loss: 15.9420, test/MAE: 15.9792, test/RMSE: 24.6258, test/MAPE: 0.1849]
2026-01-07 19:06:49,498 - easytorch-training - INFO - Checkpoint checkpoints\TimeMixer\PEMS03_50_96_12\bd7b81eefca29cd030a5d8a903f93b6d\TimeMixer_02.pt saved
2026-01-07 19:06:49,498 - easytorch-training - INFO - The estimated training finish time is 2026-01-07 19:39:11
2026-01-07 19:06:49,498 - easytorch-training - INFO - Epoch 3 / 50
2026-01-07 19:07:24,264 - easytorch-training - INFO - Result : [train/time: 34.77 (s), train/lr: 5.00e-04, train/loss: 14.6529, train/MAE: 14.6529, train/RMSE: 23.0816, train/MAPE: 0.1546]
2026-01-07 19:07:24,266 - easytorch-training - INFO - Start validation.
2026-01-07 19:07:26,745 - easytorch-training - INFO - Result : [val/time: 2.48 (s), val/loss: 14.9212, val/MAE: 14.9212, val/RMSE: 22.0334, val/MAPE: 0.1655]
2026-01-07 19:07:26,783 - easytorch-training - INFO - Checkpoint checkpoints\TimeMixer\PEMS03_50_96_12\bd7b81eefca29cd030a5d8a903f93b6d\TimeMixer_best_val_MAE.pt saved
2026-01-07 19:07:29,016 - easytorch-training - INFO - Evaluate best model on test data for horizon 1, Test MAE: 12.7625, Test RMSE: 19.2749, Test MAPE: 0.1396
2026-01-07 19:07:29,019 - easytorch-training - INFO - Evaluate best model on test data for horizon 2, Test MAE: 13.8423, Test RMSE: 21.0103, Test MAPE: 0.1714
2026-01-07 19:07:29,021 - easytorch-training - INFO - Evaluate best model on test data for horizon 3, Test MAE: 14.2585, Test RMSE: 21.9664, Test MAPE: 0.1484
2026-01-07 19:07:29,024 - easytorch-training - INFO - Evaluate best model on test data for horizon 4, Test MAE: 14.8834, Test RMSE: 23.0107, Test MAPE: 0.1585
2026-01-07 19:07:29,026 - easytorch-training - INFO - Evaluate best model on test data for horizon 5, Test MAE: 15.4529, Test RMSE: 23.7783, Test MAPE: 0.1745
2026-01-07 19:07:29,028 - easytorch-training - INFO - Evaluate best model on test data for horizon 6, Test MAE: 15.6868, Test RMSE: 24.2600, Test MAPE: 0.1781
2026-01-07 19:07:29,031 - easytorch-training - INFO - Evaluate best model on test data for horizon 7, Test MAE: 16.2460, Test RMSE: 25.0347, Test MAPE: 0.1916
2026-01-07 19:07:29,033 - easytorch-training - INFO - Evaluate best model on test data for horizon 8, Test MAE: 16.5877, Test RMSE: 25.7080, Test MAPE: 0.1757
2026-01-07 19:07:29,035 - easytorch-training - INFO - Evaluate best model on test data for horizon 9, Test MAE: 17.2443, Test RMSE: 26.2653, Test MAPE: 0.2366
2026-01-07 19:07:29,038 - easytorch-training - INFO - Evaluate best model on test data for horizon 10, Test MAE: 17.2115, Test RMSE: 26.7233, Test MAPE: 0.1880
2026-01-07 19:07:29,040 - easytorch-training - INFO - Evaluate best model on test data for horizon 11, Test MAE: 17.4056, Test RMSE: 27.1142, Test MAPE: 0.1857
2026-01-07 19:07:29,042 - easytorch-training - INFO - Evaluate best model on test data for horizon 12, Test MAE: 18.0224, Test RMSE: 27.9163, Test MAPE: 0.2116
2026-01-07 19:07:29,065 - easytorch-training - INFO - Result : [test/time: 2.28 (s), test/loss: 15.7590, test/MAE: 15.8003, test/RMSE: 24.4695, test/MAPE: 0.1800]
2026-01-07 19:07:29,106 - easytorch-training - INFO - Checkpoint checkpoints\TimeMixer\PEMS03_50_96_12\bd7b81eefca29cd030a5d8a903f93b6d\TimeMixer_03.pt saved
2026-01-07 19:07:29,106 - easytorch-training - INFO - The estimated training finish time is 2026-01-07 19:38:57
2026-01-07 19:07:29,106 - easytorch-training - INFO - Epoch 4 / 50
2026-01-07 19:08:03,672 - easytorch-training - INFO - Result : [train/time: 34.57 (s), train/lr: 5.00e-04, train/loss: 14.3047, train/MAE: 14.3047, train/RMSE: 22.6800, train/MAPE: 0.1492]
2026-01-07 19:08:03,674 - easytorch-training - INFO - Start validation.
2026-01-07 19:08:06,168 - easytorch-training - INFO - Result : [val/time: 2.49 (s), val/loss: 14.5551, val/MAE: 14.5551, val/RMSE: 21.6882, val/MAPE: 0.1517]
2026-01-07 19:08:06,207 - easytorch-training - INFO - Checkpoint checkpoints\TimeMixer\PEMS03_50_96_12\bd7b81eefca29cd030a5d8a903f93b6d\TimeMixer_best_val_MAE.pt saved
2026-01-07 19:08:08,469 - easytorch-training - INFO - Evaluate best model on test data for horizon 1, Test MAE: 12.5864, Test RMSE: 19.0939, Test MAPE: 0.1310
2026-01-07 19:08:08,471 - easytorch-training - INFO - Evaluate best model on test data for horizon 2, Test MAE: 13.4045, Test RMSE: 20.6030, Test MAPE: 0.1476
2026-01-07 19:08:08,474 - easytorch-training - INFO - Evaluate best model on test data for horizon 3, Test MAE: 13.9241, Test RMSE: 21.6522, Test MAPE: 0.1407
2026-01-07 19:08:08,476 - easytorch-training - INFO - Evaluate best model on test data for horizon 4, Test MAE: 14.4562, Test RMSE: 22.5675, Test MAPE: 0.1483
2026-01-07 19:08:08,477 - easytorch-training - INFO - Evaluate best model on test data for horizon 5, Test MAE: 14.8921, Test RMSE: 23.2807, Test MAPE: 0.1591
2026-01-07 19:08:08,481 - easytorch-training - INFO - Evaluate best model on test data for horizon 6, Test MAE: 15.4379, Test RMSE: 24.0683, Test MAPE: 0.1687
2026-01-07 19:08:08,483 - easytorch-training - INFO - Evaluate best model on test data for horizon 7, Test MAE: 15.8855, Test RMSE: 24.7823, Test MAPE: 0.1760
2026-01-07 19:08:08,485 - easytorch-training - INFO - Evaluate best model on test data for horizon 8, Test MAE: 16.1653, Test RMSE: 25.2859, Test MAPE: 0.1740
2026-01-07 19:08:08,489 - easytorch-training - INFO - Evaluate best model on test data for horizon 9, Test MAE: 16.5497, Test RMSE: 25.8069, Test MAPE: 0.1771
2026-01-07 19:08:08,491 - easytorch-training - INFO - Evaluate best model on test data for horizon 10, Test MAE: 16.8464, Test RMSE: 26.2271, Test MAPE: 0.1978
2026-01-07 19:08:08,493 - easytorch-training - INFO - Evaluate best model on test data for horizon 11, Test MAE: 17.3944, Test RMSE: 26.9040, Test MAPE: 0.1897
2026-01-07 19:08:08,495 - easytorch-training - INFO - Evaluate best model on test data for horizon 12, Test MAE: 17.6207, Test RMSE: 27.4047, Test MAPE: 0.2005
2026-01-07 19:08:08,518 - easytorch-training - INFO - Result : [test/time: 2.31 (s), test/loss: 15.3888, test/MAE: 15.4303, test/RMSE: 24.1017, test/MAPE: 0.1676]
2026-01-07 19:08:08,556 - easytorch-training - INFO - Checkpoint checkpoints\TimeMixer\PEMS03_50_96_12\bd7b81eefca29cd030a5d8a903f93b6d\TimeMixer_04.pt saved
2026-01-07 19:08:08,557 - easytorch-training - INFO - The estimated training finish time is 2026-01-07 19:38:48
2026-01-07 19:08:08,557 - easytorch-training - INFO - Epoch 5 / 50
2026-01-07 19:08:43,185 - easytorch-training - INFO - Result : [train/time: 34.63 (s), train/lr: 5.00e-04, train/loss: 14.0180, train/MAE: 14.0180, train/RMSE: 22.3298, train/MAPE: 0.1445]
2026-01-07 19:08:43,187 - easytorch-training - INFO - Start validation.
2026-01-07 19:08:45,682 - easytorch-training - INFO - Result : [val/time: 2.49 (s), val/loss: 14.6339, val/MAE: 14.6339, val/RMSE: 21.7134, val/MAPE: 0.1711]
2026-01-07 19:08:47,929 - easytorch-training - INFO - Evaluate best model on test data for horizon 1, Test MAE: 12.6456, Test RMSE: 19.1812, Test MAPE: 0.1306
2026-01-07 19:08:47,931 - easytorch-training - INFO - Evaluate best model on test data for horizon 2, Test MAE: 13.4368, Test RMSE: 20.6576, Test MAPE: 0.1482
2026-01-07 19:08:47,933 - easytorch-training - INFO - Evaluate best model on test data for horizon 3, Test MAE: 14.4492, Test RMSE: 21.9230, Test MAPE: 0.2023
2026-01-07 19:08:47,936 - easytorch-training - INFO - Evaluate best model on test data for horizon 4, Test MAE: 14.9719, Test RMSE: 22.8305, Test MAPE: 0.1996
2026-01-07 19:08:47,938 - easytorch-training - INFO - Evaluate best model on test data for horizon 5, Test MAE: 15.4360, Test RMSE: 23.6333, Test MAPE: 0.1889
2026-01-07 19:08:47,941 - easytorch-training - INFO - Evaluate best model on test data for horizon 6, Test MAE: 15.8910, Test RMSE: 24.2835, Test MAPE: 0.2120
2026-01-07 19:08:47,943 - easytorch-training - INFO - Evaluate best model on test data for horizon 7, Test MAE: 15.9089, Test RMSE: 24.8655, Test MAPE: 0.1681
2026-01-07 19:08:47,945 - easytorch-training - INFO - Evaluate best model on test data for horizon 8, Test MAE: 16.7484, Test RMSE: 25.6160, Test MAPE: 0.2386
2026-01-07 19:08:47,947 - easytorch-training - INFO - Evaluate best model on test data for horizon 9, Test MAE: 16.5282, Test RMSE: 25.8628, Test MAPE: 0.1760
2026-01-07 19:08:47,950 - easytorch-training - INFO - Evaluate best model on test data for horizon 10, Test MAE: 16.7702, Test RMSE: 26.2413, Test MAPE: 0.1909
2026-01-07 19:08:47,952 - easytorch-training - INFO - Evaluate best model on test data for horizon 11, Test MAE: 17.1476, Test RMSE: 26.7471, Test MAPE: 0.2059
2026-01-07 19:08:47,954 - easytorch-training - INFO - Evaluate best model on test data for horizon 12, Test MAE: 17.3150, Test RMSE: 27.2062, Test MAPE: 0.1832
2026-01-07 19:08:47,979 - easytorch-training - INFO - Result : [test/time: 2.30 (s), test/loss: 15.5664, test/MAE: 15.6041, test/RMSE: 24.2079, test/MAPE: 0.1870]
2026-01-07 19:08:48,018 - easytorch-training - INFO - Checkpoint checkpoints\TimeMixer\PEMS03_50_96_12\bd7b81eefca29cd030a5d8a903f93b6d\TimeMixer_05.pt saved
2026-01-07 19:08:48,018 - easytorch-training - INFO - The estimated training finish time is 2026-01-07 19:38:42
2026-01-07 19:08:48,018 - easytorch-training - INFO - Epoch 6 / 50
2026-01-07 19:09:22,547 - easytorch-training - INFO - Result : [train/time: 34.53 (s), train/lr: 5.00e-04, train/loss: 13.8390, train/MAE: 13.8390, train/RMSE: 22.1017, train/MAPE: 0.1435]
2026-01-07 19:09:22,548 - easytorch-training - INFO - Start validation.
2026-01-07 19:09:25,027 - easytorch-training - INFO - Result : [val/time: 2.48 (s), val/loss: 14.6383, val/MAE: 14.6383, val/RMSE: 21.7693, val/MAPE: 0.1572]
2026-01-07 19:09:27,294 - easytorch-training - INFO - Evaluate best model on test data for horizon 1, Test MAE: 12.8246, Test RMSE: 19.3095, Test MAPE: 0.1594
2026-01-07 19:09:27,296 - easytorch-training - INFO - Evaluate best model on test data for horizon 2, Test MAE: 13.6429, Test RMSE: 20.9078, Test MAPE: 0.1666
2026-01-07 19:09:27,298 - easytorch-training - INFO - Evaluate best model on test data for horizon 3, Test MAE: 14.0806, Test RMSE: 21.8846, Test MAPE: 0.1464
2026-01-07 19:09:27,301 - easytorch-training - INFO - Evaluate best model on test data for horizon 4, Test MAE: 14.9161, Test RMSE: 23.0738, Test MAPE: 0.1826
2026-01-07 19:09:27,304 - easytorch-training - INFO - Evaluate best model on test data for horizon 5, Test MAE: 15.5514, Test RMSE: 24.0370, Test MAPE: 0.2006
2026-01-07 19:09:27,306 - easytorch-training - INFO - Evaluate best model on test data for horizon 6, Test MAE: 15.5179, Test RMSE: 24.3611, Test MAPE: 0.1613
2026-01-07 19:09:27,309 - easytorch-training - INFO - Evaluate best model on test data for horizon 7, Test MAE: 16.1144, Test RMSE: 25.3309, Test MAPE: 0.1642
2026-01-07 19:09:27,311 - easytorch-training - INFO - Evaluate best model on test data for horizon 8, Test MAE: 16.4900, Test RMSE: 25.8425, Test MAPE: 0.1815
2026-01-07 19:09:27,314 - easytorch-training - INFO - Evaluate best model on test data for horizon 9, Test MAE: 16.6709, Test RMSE: 26.1931, Test MAPE: 0.1777
2026-01-07 19:09:27,316 - easytorch-training - INFO - Evaluate best model on test data for horizon 10, Test MAE: 16.9514, Test RMSE: 26.6822, Test MAPE: 0.1731
2026-01-07 19:09:27,318 - easytorch-training - INFO - Evaluate best model on test data for horizon 11, Test MAE: 17.5456, Test RMSE: 27.4463, Test MAPE: 0.1991
2026-01-07 19:09:27,320 - easytorch-training - INFO - Evaluate best model on test data for horizon 12, Test MAE: 17.8450, Test RMSE: 28.0247, Test MAPE: 0.1898
2026-01-07 19:09:27,343 - easytorch-training - INFO - Result : [test/time: 2.31 (s), test/loss: 15.6358, test/MAE: 15.6793, test/RMSE: 24.5608, test/MAPE: 0.1752]
2026-01-07 19:09:27,382 - easytorch-training - INFO - Checkpoint checkpoints\TimeMixer\PEMS03_50_96_12\bd7b81eefca29cd030a5d8a903f93b6d\TimeMixer_06.pt saved
2026-01-07 19:09:27,383 - easytorch-training - INFO - The estimated training finish time is 2026-01-07 19:38:38
2026-01-07 19:09:27,383 - easytorch-training - INFO - Epoch 7 / 50
2026-01-07 19:10:02,086 - easytorch-training - INFO - Result : [train/time: 34.70 (s), train/lr: 5.00e-04, train/loss: 13.6530, train/MAE: 13.6530, train/RMSE: 21.8810, train/MAPE: 0.1411]
2026-01-07 19:10:02,088 - easytorch-training - INFO - Start validation.
2026-01-07 19:10:04,568 - easytorch-training - INFO - Result : [val/time: 2.48 (s), val/loss: 14.8153, val/MAE: 14.8153, val/RMSE: 22.0952, val/MAPE: 0.1547]
2026-01-07 19:10:06,834 - easytorch-training - INFO - Evaluate best model on test data for horizon 1, Test MAE: 13.0689, Test RMSE: 19.5593, Test MAPE: 0.1437
2026-01-07 19:10:06,837 - easytorch-training - INFO - Evaluate best model on test data for horizon 2, Test MAE: 13.6382, Test RMSE: 20.8934, Test MAPE: 0.1427
2026-01-07 19:10:06,839 - easytorch-training - INFO - Evaluate best model on test data for horizon 3, Test MAE: 14.3517, Test RMSE: 22.1204, Test MAPE: 0.1488
2026-01-07 19:10:06,841 - easytorch-training - INFO - Evaluate best model on test data for horizon 4, Test MAE: 15.2136, Test RMSE: 23.2358, Test MAPE: 0.1982
2026-01-07 19:10:06,843 - easytorch-training - INFO - Evaluate best model on test data for horizon 5, Test MAE: 15.3928, Test RMSE: 23.8671, Test MAPE: 0.1632
2026-01-07 19:10:06,846 - easytorch-training - INFO - Evaluate best model on test data for horizon 6, Test MAE: 16.2131, Test RMSE: 24.9993, Test MAPE: 0.1714
2026-01-07 19:10:06,848 - easytorch-training - INFO - Evaluate best model on test data for horizon 7, Test MAE: 16.6824, Test RMSE: 25.7293, Test MAPE: 0.1814
2026-01-07 19:10:06,851 - easytorch-training - INFO - Evaluate best model on test data for horizon 8, Test MAE: 16.9546, Test RMSE: 26.3023, Test MAPE: 0.1796
2026-01-07 19:10:06,852 - easytorch-training - INFO - Evaluate best model on test data for horizon 9, Test MAE: 17.0639, Test RMSE: 26.6038, Test MAPE: 0.1794
2026-01-07 19:10:06,854 - easytorch-training - INFO - Evaluate best model on test data for horizon 10, Test MAE: 17.2573, Test RMSE: 26.9127, Test MAPE: 0.1846
2026-01-07 19:10:06,856 - easytorch-training - INFO - Evaluate best model on test data for horizon 11, Test MAE: 17.9737, Test RMSE: 27.8002, Test MAPE: 0.2019
2026-01-07 19:10:06,858 - easytorch-training - INFO - Evaluate best model on test data for horizon 12, Test MAE: 18.4590, Test RMSE: 28.5181, Test MAPE: 0.2083
2026-01-07 19:10:06,881 - easytorch-training - INFO - Result : [test/time: 2.31 (s), test/loss: 15.9835, test/MAE: 16.0225, test/RMSE: 24.8578, test/MAPE: 0.1753]
2026-01-07 19:10:06,920 - easytorch-training - INFO - Checkpoint checkpoints\TimeMixer\PEMS03_50_96_12\bd7b81eefca29cd030a5d8a903f93b6d\TimeMixer_07.pt saved
2026-01-07 19:10:06,920 - easytorch-training - INFO - The estimated training finish time is 2026-01-07 19:38:36
2026-01-07 19:10:06,920 - easytorch-training - INFO - Epoch 8 / 50
2026-01-07 19:10:41,518 - easytorch-training - INFO - Result : [train/time: 34.60 (s), train/lr: 5.00e-04, train/loss: 13.5176, train/MAE: 13.5176, train/RMSE: 21.7169, train/MAPE: 0.1389]
2026-01-07 19:10:41,519 - easytorch-training - INFO - Start validation.
2026-01-07 19:10:44,037 - easytorch-training - INFO - Result : [val/time: 2.52 (s), val/loss: 14.3209, val/MAE: 14.3209, val/RMSE: 21.3935, val/MAPE: 0.1480]
2026-01-07 19:10:44,076 - easytorch-training - INFO - Checkpoint checkpoints\TimeMixer\PEMS03_50_96_12\bd7b81eefca29cd030a5d8a903f93b6d\TimeMixer_best_val_MAE.pt saved
2026-01-07 19:10:46,313 - easytorch-training - INFO - Evaluate best model on test data for horizon 1, Test MAE: 12.7495, Test RMSE: 19.3525, Test MAPE: 0.1350
2026-01-07 19:10:46,315 - easytorch-training - INFO - Evaluate best model on test data for horizon 2, Test MAE: 13.6730, Test RMSE: 20.9399, Test MAPE: 0.1618
2026-01-07 19:10:46,318 - easytorch-training - INFO - Evaluate best model on test data for horizon 3, Test MAE: 14.0689, Test RMSE: 21.9259, Test MAPE: 0.1487
2026-01-07 19:10:46,320 - easytorch-training - INFO - Evaluate best model on test data for horizon 4, Test MAE: 14.5862, Test RMSE: 22.8294, Test MAPE: 0.1586
2026-01-07 19:10:46,323 - easytorch-training - INFO - Evaluate best model on test data for horizon 5, Test MAE: 15.1581, Test RMSE: 23.6793, Test MAPE: 0.1845
2026-01-07 19:10:46,326 - easytorch-training - INFO - Evaluate best model on test data for horizon 6, Test MAE: 15.4732, Test RMSE: 24.2840, Test MAPE: 0.1804
2026-01-07 19:10:46,328 - easytorch-training - INFO - Evaluate best model on test data for horizon 7, Test MAE: 15.7739, Test RMSE: 24.9308, Test MAPE: 0.1623
2026-01-07 19:10:46,331 - easytorch-training - INFO - Evaluate best model on test data for horizon 8, Test MAE: 16.0329, Test RMSE: 25.3099, Test MAPE: 0.1675
2026-01-07 19:10:46,333 - easytorch-training - INFO - Evaluate best model on test data for horizon 9, Test MAE: 16.2671, Test RMSE: 25.7687, Test MAPE: 0.1688
2026-01-07 19:10:46,335 - easytorch-training - INFO - Evaluate best model on test data for horizon 10, Test MAE: 16.4487, Test RMSE: 26.0973, Test MAPE: 0.1701
2026-01-07 19:10:46,337 - easytorch-training - INFO - Evaluate best model on test data for horizon 11, Test MAE: 16.7895, Test RMSE: 26.6365, Test MAPE: 0.1750
2026-01-07 19:10:46,340 - easytorch-training - INFO - Evaluate best model on test data for horizon 12, Test MAE: 17.1333, Test RMSE: 27.1633, Test MAPE: 0.1827
2026-01-07 19:10:46,365 - easytorch-training - INFO - Result : [test/time: 2.29 (s), test/loss: 15.3043, test/MAE: 15.3462, test/RMSE: 24.1867, test/MAPE: 0.1663]
2026-01-07 19:10:46,404 - easytorch-training - INFO - Checkpoint checkpoints\TimeMixer\PEMS03_50_96_12\bd7b81eefca29cd030a5d8a903f93b6d\TimeMixer_08.pt saved
2026-01-07 19:10:46,404 - easytorch-training - INFO - The estimated training finish time is 2026-01-07 19:38:34
2026-01-07 19:10:46,404 - easytorch-training - INFO - Epoch 9 / 50
2026-01-07 19:11:21,882 - easytorch-training - INFO - Result : [train/time: 35.48 (s), train/lr: 5.00e-04, train/loss: 13.3796, train/MAE: 13.3796, train/RMSE: 21.5315, train/MAPE: 0.1371]
2026-01-07 19:11:21,884 - easytorch-training - INFO - Start validation.
2026-01-07 19:11:24,610 - easytorch-training - INFO - Result : [val/time: 2.73 (s), val/loss: 14.3782, val/MAE: 14.3782, val/RMSE: 21.5427, val/MAPE: 0.1480]
2026-01-07 19:11:27,076 - easytorch-training - INFO - Evaluate best model on test data for horizon 1, Test MAE: 12.7964, Test RMSE: 19.4435, Test MAPE: 0.1319
2026-01-07 19:11:27,078 - easytorch-training - INFO - Evaluate best model on test data for horizon 2, Test MAE: 13.9122, Test RMSE: 21.1956, Test MAPE: 0.1727
2026-01-07 19:11:27,081 - easytorch-training - INFO - Evaluate best model on test data for horizon 3, Test MAE: 14.2460, Test RMSE: 22.1031, Test MAPE: 0.1540
2026-01-07 19:11:27,084 - easytorch-training - INFO - Evaluate best model on test data for horizon 4, Test MAE: 14.7370, Test RMSE: 22.9894, Test MAPE: 0.1591
2026-01-07 19:11:27,086 - easytorch-training - INFO - Evaluate best model on test data for horizon 5, Test MAE: 15.2940, Test RMSE: 23.8528, Test MAPE: 0.1685
2026-01-07 19:11:27,089 - easytorch-training - INFO - Evaluate best model on test data for horizon 6, Test MAE: 15.5642, Test RMSE: 24.3959, Test MAPE: 0.1648
2026-01-07 19:11:27,091 - easytorch-training - INFO - Evaluate best model on test data for horizon 7, Test MAE: 15.9828, Test RMSE: 25.0793, Test MAPE: 0.1677
2026-01-07 19:11:27,094 - easytorch-training - INFO - Evaluate best model on test data for horizon 8, Test MAE: 16.3713, Test RMSE: 25.6868, Test MAPE: 0.1735
2026-01-07 19:11:27,097 - easytorch-training - INFO - Evaluate best model on test data for horizon 9, Test MAE: 16.5456, Test RMSE: 26.0761, Test MAPE: 0.1735
2026-01-07 19:11:27,101 - easytorch-training - INFO - Evaluate best model on test data for horizon 10, Test MAE: 17.1042, Test RMSE: 26.7770, Test MAPE: 0.1901
2026-01-07 19:11:27,104 - easytorch-training - INFO - Evaluate best model on test data for horizon 11, Test MAE: 17.3052, Test RMSE: 27.1968, Test MAPE: 0.1817
2026-01-07 19:11:27,106 - easytorch-training - INFO - Evaluate best model on test data for horizon 12, Test MAE: 17.8204, Test RMSE: 27.8189, Test MAPE: 0.2057
2026-01-07 19:11:27,131 - easytorch-training - INFO - Result : [test/time: 2.52 (s), test/loss: 15.5994, test/MAE: 15.6400, test/RMSE: 24.5082, test/MAPE: 0.1703]
2026-01-07 19:11:27,175 - easytorch-training - INFO - Checkpoint checkpoints\TimeMixer\PEMS03_50_96_12\bd7b81eefca29cd030a5d8a903f93b6d\TimeMixer_09.pt saved
2026-01-07 19:11:27,175 - easytorch-training - INFO - The estimated training finish time is 2026-01-07 19:38:40
2026-01-07 19:11:27,175 - easytorch-training - INFO - Epoch 10 / 50
2026-01-07 19:12:02,275 - easytorch-training - INFO - Result : [train/time: 35.10 (s), train/lr: 5.00e-04, train/loss: 13.2591, train/MAE: 13.2591, train/RMSE: 21.3691, train/MAPE: 0.1350]
2026-01-07 19:12:02,277 - easytorch-training - INFO - Start validation.
2026-01-07 19:12:04,784 - easytorch-training - INFO - Result : [val/time: 2.51 (s), val/loss: 14.2259, val/MAE: 14.2259, val/RMSE: 21.3574, val/MAPE: 0.1461]
2026-01-07 19:12:04,824 - easytorch-training - INFO - Checkpoint checkpoints\TimeMixer\PEMS03_50_96_12\bd7b81eefca29cd030a5d8a903f93b6d\TimeMixer_best_val_MAE.pt saved
2026-01-07 19:12:07,053 - easytorch-training - INFO - Evaluate best model on test data for horizon 1, Test MAE: 12.8465, Test RMSE: 19.5207, Test MAPE: 0.1354
2026-01-07 19:12:07,055 - easytorch-training - INFO - Evaluate best model on test data for horizon 2, Test MAE: 13.7947, Test RMSE: 21.1657, Test MAPE: 0.1534
2026-01-07 19:12:07,058 - easytorch-training - INFO - Evaluate best model on test data for horizon 3, Test MAE: 14.2182, Test RMSE: 22.1047, Test MAPE: 0.1526
2026-01-07 19:12:07,060 - easytorch-training - INFO - Evaluate best model on test data for horizon 4, Test MAE: 14.8001, Test RMSE: 23.0896, Test MAPE: 0.1648
2026-01-07 19:12:07,063 - easytorch-training - INFO - Evaluate best model on test data for horizon 5, Test MAE: 15.1666, Test RMSE: 23.7682, Test MAPE: 0.1688
2026-01-07 19:12:07,065 - easytorch-training - INFO - Evaluate best model on test data for horizon 6, Test MAE: 15.5303, Test RMSE: 24.4079, Test MAPE: 0.1673
2026-01-07 19:12:07,068 - easytorch-training - INFO - Evaluate best model on test data for horizon 7, Test MAE: 15.9319, Test RMSE: 25.0776, Test MAPE: 0.1707
2026-01-07 19:12:07,070 - easytorch-training - INFO - Evaluate best model on test data for horizon 8, Test MAE: 16.3033, Test RMSE: 25.6804, Test MAPE: 0.1753
2026-01-07 19:12:07,073 - easytorch-training - INFO - Evaluate best model on test data for horizon 9, Test MAE: 16.5896, Test RMSE: 26.1587, Test MAPE: 0.1766
2026-01-07 19:12:07,076 - easytorch-training - INFO - Evaluate best model on test data for horizon 10, Test MAE: 16.9335, Test RMSE: 26.5829, Test MAPE: 0.1935
2026-01-07 19:12:07,078 - easytorch-training - INFO - Evaluate best model on test data for horizon 11, Test MAE: 17.0625, Test RMSE: 26.9673, Test MAPE: 0.1807
2026-01-07 19:12:07,081 - easytorch-training - INFO - Evaluate best model on test data for horizon 12, Test MAE: 17.4959, Test RMSE: 27.5585, Test MAPE: 0.1935
2026-01-07 19:12:07,105 - easytorch-training - INFO - Result : [test/time: 2.28 (s), test/loss: 15.5142, test/MAE: 15.5561, test/RMSE: 24.4566, test/MAPE: 0.1694]
2026-01-07 19:12:07,144 - easytorch-training - INFO - Checkpoint checkpoints\TimeMixer\PEMS03_50_96_12\bd7b81eefca29cd030a5d8a903f93b6d\TimeMixer_10.pt saved
2026-01-07 19:12:07,155 - easytorch-training - INFO - The estimated training finish time is 2026-01-07 19:38:41
2026-01-07 19:12:07,155 - easytorch-training - INFO - Epoch 11 / 50
2026-01-07 19:12:41,536 - easytorch-training - INFO - Result : [train/time: 34.38 (s), train/lr: 5.00e-04, train/loss: 13.1287, train/MAE: 13.1287, train/RMSE: 21.1881, train/MAPE: 0.1337]
2026-01-07 19:12:41,537 - easytorch-training - INFO - Start validation.
2026-01-07 19:12:44,025 - easytorch-training - INFO - Result : [val/time: 2.49 (s), val/loss: 14.3087, val/MAE: 14.3087, val/RMSE: 21.4471, val/MAPE: 0.1493]
2026-01-07 19:12:46,248 - easytorch-training - INFO - Evaluate best model on test data for horizon 1, Test MAE: 12.8260, Test RMSE: 19.5014, Test MAPE: 0.1489
2026-01-07 19:12:46,251 - easytorch-training - INFO - Evaluate best model on test data for horizon 2, Test MAE: 13.5731, Test RMSE: 20.9974, Test MAPE: 0.1391
2026-01-07 19:12:46,253 - easytorch-training - INFO - Evaluate best model on test data for horizon 3, Test MAE: 14.0871, Test RMSE: 22.0301, Test MAPE: 0.1439
2026-01-07 19:12:46,256 - easytorch-training - INFO - Evaluate best model on test data for horizon 4, Test MAE: 14.7151, Test RMSE: 22.9821, Test MAPE: 0.1772
2026-01-07 19:12:46,258 - easytorch-training - INFO - Evaluate best model on test data for horizon 5, Test MAE: 15.2447, Test RMSE: 23.8183, Test MAPE: 0.1927
2026-01-07 19:12:46,261 - easytorch-training - INFO - Evaluate best model on test data for horizon 6, Test MAE: 15.3731, Test RMSE: 24.2482, Test MAPE: 0.1692
2026-01-07 19:12:46,263 - easytorch-training - INFO - Evaluate best model on test data for horizon 7, Test MAE: 15.7054, Test RMSE: 24.8586, Test MAPE: 0.1582
2026-01-07 19:12:46,265 - easytorch-training - INFO - Evaluate best model on test data for horizon 8, Test MAE: 16.1435, Test RMSE: 25.4887, Test MAPE: 0.1678
2026-01-07 19:12:46,267 - easytorch-training - INFO - Evaluate best model on test data for horizon 9, Test MAE: 16.4034, Test RMSE: 26.0040, Test MAPE: 0.1693
2026-01-07 19:12:46,269 - easytorch-training - INFO - Evaluate best model on test data for horizon 10, Test MAE: 16.6278, Test RMSE: 26.3629, Test MAPE: 0.1808
2026-01-07 19:12:46,272 - easytorch-training - INFO - Evaluate best model on test data for horizon 11, Test MAE: 16.8934, Test RMSE: 26.8261, Test MAPE: 0.1729
2026-01-07 19:12:46,274 - easytorch-training - INFO - Evaluate best model on test data for horizon 12, Test MAE: 17.2548, Test RMSE: 27.4082, Test MAPE: 0.1733
2026-01-07 19:12:46,298 - easytorch-training - INFO - Result : [test/time: 2.27 (s), test/loss: 15.3596, test/MAE: 15.4040, test/RMSE: 24.3237, test/MAPE: 0.1661]
2026-01-07 19:12:46,338 - easytorch-training - INFO - Checkpoint checkpoints\TimeMixer\PEMS03_50_96_12\bd7b81eefca29cd030a5d8a903f93b6d\TimeMixer_11.pt saved
2026-01-07 19:12:46,338 - easytorch-training - INFO - The estimated training finish time is 2026-01-07 19:38:38
2026-01-07 19:12:46,338 - easytorch-training - INFO - Epoch 12 / 50
2026-01-07 19:13:21,090 - easytorch-training - INFO - Result : [train/time: 34.75 (s), train/lr: 5.00e-04, train/loss: 13.0211, train/MAE: 13.0211, train/RMSE: 21.0159, train/MAPE: 0.1322]
2026-01-07 19:13:21,091 - easytorch-training - INFO - Start validation.
2026-01-07 19:13:23,602 - easytorch-training - INFO - Result : [val/time: 2.51 (s), val/loss: 14.3420, val/MAE: 14.3420, val/RMSE: 21.4577, val/MAPE: 0.1445]
2026-01-07 19:13:25,852 - easytorch-training - INFO - Evaluate best model on test data for horizon 1, Test MAE: 12.8395, Test RMSE: 19.5296, Test MAPE: 0.1341
2026-01-07 19:13:25,855 - easytorch-training - INFO - Evaluate best model on test data for horizon 2, Test MAE: 13.5845, Test RMSE: 20.9785, Test MAPE: 0.1434
2026-01-07 19:13:25,857 - easytorch-training - INFO - Evaluate best model on test data for horizon 3, Test MAE: 14.2103, Test RMSE: 22.0695, Test MAPE: 0.1508
2026-01-07 19:13:25,859 - easytorch-training - INFO - Evaluate best model on test data for horizon 4, Test MAE: 14.7907, Test RMSE: 23.0505, Test MAPE: 0.1629
2026-01-07 19:13:25,861 - easytorch-training - INFO - Evaluate best model on test data for horizon 5, Test MAE: 15.1481, Test RMSE: 23.7260, Test MAPE: 0.1600
2026-01-07 19:13:25,863 - easytorch-training - INFO - Evaluate best model on test data for horizon 6, Test MAE: 15.6439, Test RMSE: 24.4323, Test MAPE: 0.1770
2026-01-07 19:13:25,865 - easytorch-training - INFO - Evaluate best model on test data for horizon 7, Test MAE: 16.0116, Test RMSE: 25.0433, Test MAPE: 0.1706
2026-01-07 19:13:25,867 - easytorch-training - INFO - Evaluate best model on test data for horizon 8, Test MAE: 16.2928, Test RMSE: 25.5582, Test MAPE: 0.1737
2026-01-07 19:13:25,869 - easytorch-training - INFO - Evaluate best model on test data for horizon 9, Test MAE: 16.5717, Test RMSE: 26.0088, Test MAPE: 0.1749
2026-01-07 19:13:25,871 - easytorch-training - INFO - Evaluate best model on test data for horizon 10, Test MAE: 16.8006, Test RMSE: 26.4294, Test MAPE: 0.1770
2026-01-07 19:13:25,873 - easytorch-training - INFO - Evaluate best model on test data for horizon 11, Test MAE: 17.0889, Test RMSE: 26.9133, Test MAPE: 0.1797
2026-01-07 19:13:25,876 - easytorch-training - INFO - Evaluate best model on test data for horizon 12, Test MAE: 17.4757, Test RMSE: 27.4821, Test MAPE: 0.1898
2026-01-07 19:13:25,901 - easytorch-training - INFO - Result : [test/time: 2.30 (s), test/loss: 15.4970, test/MAE: 15.5382, test/RMSE: 24.3836, test/MAPE: 0.1662]
2026-01-07 19:13:25,940 - easytorch-training - INFO - Checkpoint checkpoints\TimeMixer\PEMS03_50_96_12\bd7b81eefca29cd030a5d8a903f93b6d\TimeMixer_12.pt saved
2026-01-07 19:13:25,941 - easytorch-training - INFO - The estimated training finish time is 2026-01-07 19:38:37
2026-01-07 19:13:25,941 - easytorch-training - INFO - Epoch 13 / 50
2026-01-07 19:14:01,141 - easytorch-training - INFO - Result : [train/time: 35.20 (s), train/lr: 5.00e-04, train/loss: 12.8958, train/MAE: 12.8958, train/RMSE: 20.8180, train/MAPE: 0.1304]
2026-01-07 19:14:01,143 - easytorch-training - INFO - Start validation.
2026-01-07 19:14:03,694 - easytorch-training - INFO - Result : [val/time: 2.55 (s), val/loss: 14.5344, val/MAE: 14.5344, val/RMSE: 21.6193, val/MAPE: 0.1717]
2026-01-07 19:14:06,077 - easytorch-training - INFO - Evaluate best model on test data for horizon 1, Test MAE: 13.1918, Test RMSE: 19.8370, Test MAPE: 0.1794
2026-01-07 19:14:06,079 - easytorch-training - INFO - Evaluate best model on test data for horizon 2, Test MAE: 13.7032, Test RMSE: 21.1692, Test MAPE: 0.1476
2026-01-07 19:14:06,081 - easytorch-training - INFO - Evaluate best model on test data for horizon 3, Test MAE: 14.4051, Test RMSE: 22.2845, Test MAPE: 0.1743
2026-01-07 19:14:06,084 - easytorch-training - INFO - Evaluate best model on test data for horizon 4, Test MAE: 15.2741, Test RMSE: 23.4224, Test MAPE: 0.2230
2026-01-07 19:14:06,086 - easytorch-training - INFO - Evaluate best model on test data for horizon 5, Test MAE: 15.3308, Test RMSE: 23.9840, Test MAPE: 0.1759
2026-01-07 19:14:06,089 - easytorch-training - INFO - Evaluate best model on test data for horizon 6, Test MAE: 15.7924, Test RMSE: 24.6776, Test MAPE: 0.1870
2026-01-07 19:14:06,091 - easytorch-training - INFO - Evaluate best model on test data for horizon 7, Test MAE: 16.1999, Test RMSE: 25.2595, Test MAPE: 0.2110
2026-01-07 19:14:06,094 - easytorch-training - INFO - Evaluate best model on test data for horizon 8, Test MAE: 16.2291, Test RMSE: 25.5820, Test MAPE: 0.1767
2026-01-07 19:14:06,096 - easytorch-training - INFO - Evaluate best model on test data for horizon 9, Test MAE: 16.5528, Test RMSE: 26.0515, Test MAPE: 0.1887
2026-01-07 19:14:06,098 - easytorch-training - INFO - Evaluate best model on test data for horizon 10, Test MAE: 16.8517, Test RMSE: 26.5060, Test MAPE: 0.1825
2026-01-07 19:14:06,101 - easytorch-training - INFO - Evaluate best model on test data for horizon 11, Test MAE: 17.5740, Test RMSE: 27.2066, Test MAPE: 0.2515
2026-01-07 19:14:06,104 - easytorch-training - INFO - Evaluate best model on test data for horizon 12, Test MAE: 17.5690, Test RMSE: 27.6020, Test MAPE: 0.2145
2026-01-07 19:14:06,130 - easytorch-training - INFO - Result : [test/time: 2.43 (s), test/loss: 15.6798, test/MAE: 15.7228, test/RMSE: 24.5745, test/MAPE: 0.1927]
2026-01-07 19:14:06,170 - easytorch-training - INFO - Checkpoint checkpoints\TimeMixer\PEMS03_50_96_12\bd7b81eefca29cd030a5d8a903f93b6d\TimeMixer_13.pt saved
2026-01-07 19:14:06,170 - easytorch-training - INFO - The estimated training finish time is 2026-01-07 19:38:39
2026-01-07 19:14:06,170 - easytorch-training - INFO - Epoch 14 / 50