Will you share the source code for TimerMixer++?

[weituo2002]

After reading the author's latest paper, will the source code of TimerMixer++ be shared?

[kwuking]

Many thanks for your interest in our latest work. As per our usual practice, we will release our code when the camera-ready version of the paper is published. You can keep an eye on the TimeMixer repository, and we will notify you as soon as the code is released.

[Vincent131499]

感谢杰出的 work ！
TimeMixer++论文已经放出了，代码请问已经更新了吗？

[kwuking]

感谢杰出的 work ！ TimeMixer++论文已经放出了，代码请问已经更新了吗？

非常感谢您对于我们工作的关注，按照我们的惯例，我们将在正式接收发表的camera-ready版本中公布我们的代码，您可以持续关注我们的代码仓库，更新后的第一时间会和您通知，谢谢对于我们工作的支持。

[taro22666]

感谢优秀的 work ！
TimeMixer++论文已经放出了，但是代码似乎撤回了？请问什么时候会重新放出呢？

[kwuking]

感谢优秀的 work ！ TimeMixer++论文已经放出了，但是代码似乎撤回了？请问什么时候会重新放出呢？

你好，非常感谢对我们的关注，目前我们还没有更新代码，我们将在正式接收发表的camera-ready版本中公布我们的代码，您可以持续关注我们的代码仓库，再次感谢您对于我们的支持。

[VahidNaghashi]

When will you release the source code of TimeMixer++?
By the way, it's really a very impressive model!

[Vincent131499]

iclr25已公开论文，代码请问什么时候放出呢？

[OneStep758]

Hello, we continue to pay attention to your outstanding work. When can you provide the source code of TimeMixer++?

[LeeJarvis996]

Thank you all for your interest and support for our work! At the moment, the relevant code has not yet received full approval for open-sourcing. In light of this, we have organized and released the available code and model weights in an anonymous GitHub repository for your convenience. Once we obtain full open-source approval, we will also work towards releasing the project on Hugging Face. Please note, however, that we have not received company authorization to release under the Apache license. As such, the current release is not permitted for use in commercial projects. We appreciate your understanding and continued support!

[a1473838623]

The training part is not given in the anonymous Github repository. Can you share the offical code sooner? The paper and the model are good.

[Miller-Xu]

Thank you all for your interest and support for our work! At the moment, the relevant code has not yet received full approval for open-sourcing. In light of this, we have organized and released the available code and model weights in an anonymous GitHub repository for your convenience. Once we obtain full open-source approval, we will also work towards releasing the project on Hugging Face. Please note, however, that we have not received company authorization to release under the Apache license. As such, the current release is not permitted for use in commercial projects. We appreciate your understanding and continued support!

感谢杰出的工作，可以提供PEM数据集的配置参数吗？还是说直接参考TimeMixer的？还有，我复现过程发现模型训练的显存占用和训练速度并不像论文figure14所示的和TimeMixer差不多。可能和我设置的参数相关？目前主要是参考Traffic数据集设置的。

[JasonAlexTan]

Thank you all for your interest and support for our work! At the moment, the relevant code has not yet received full approval for open-sourcing. In light of this, we have organized and released the available code and model weights in an anonymous GitHub repository for your convenience. Once we obtain full open-source approval, we will also work towards releasing the project on Hugging Face. Please note, however, that we have not received company authorization to release under the Apache license. As such, the current release is not permitted for use in commercial projects. We appreciate your understanding and continued support!

Thank you for the excellent work and for sharing the code! I'm currently using this model for electricity load forecasting, so I tried training timemixer++ on the ECL dataset with the timemixer code. I have to admit, the results are really impressive. However, there's something that's been bothering me.

I noticed that the model's final output and the metric calculations are all done on normalized data. For my electricity forecasting application, this doesn't make much sense. The difference between the original data and the normalized data is too significant, and the results in this form are not usable in real-world electricity forecasting scenarios. Also, I noticed that in your paper, you cited models like Informer, Autoformer, and Time-Series-Library, and they all seem to follow the same approach. I don’t quite understand why the outputs aren't converted back to the original scale or why the metrics aren’t calculated on the original scale. Could you help explain this? By the way, in my forecasting scenario, I only split the data into training and testing sets without creating a separate validation set. Would this be okay for this model?

Additionally, I observed that when running your scripts, the gpu utilization was quite low, and the prediction process was very slow. However, when I used the timemixer code to train timemixer++, it ran much faster. Here are my training logs. Could you help me check if there’s something wrong with my setup?

start training : long_term_forecast_ECL_96_96_none_TimeMixerPP_custom_sl96_pl96_dm16_nh8_el3_dl1_df64_fc3_ebtimeF_dtTrue_Exp_0>>>>>>>>>>>>>>>>>>>>>>>>>>
train 18221
val 2537
test 5165
Epoch: 1 cost time: 756.7272894382477
Epoch: 1, Steps: 569 | Train Loss: 0.2647970 Vali Loss: 0.1805777 Test Loss: 0.2024910
Validation loss decreased (inf --> 0.202491). Saving model ...
Updating learning rate to 0.0004361031845833939
iters: 100, epoch: 2 | loss: 0.2019584
speed: 3.3137s/iter; left time: 239130.4876s
iters: 200, epoch: 2 | loss: 0.2064125
speed: 0.9925s/iter; left time: 71527.0733s
iters: 300, epoch: 2 | loss: 0.1806236
speed: 0.9967s/iter; left time: 71725.1869s
iters: 400, epoch: 2 | loss: 0.1718689
speed: 1.0334s/iter; left time: 74266.1321s
iters: 500, epoch: 2 | loss: 0.1790082
speed: 1.0203s/iter; left time: 73222.5566s
Epoch: 2 cost time: 735.3920855522156
Epoch: 2, Steps: 569 | Train Loss: 0.1913048 Vali Loss: 0.1612632 Test Loss: 0.1810411
Validation loss decreased (0.202491 --> 0.181041). Saving model ...
Updating learning rate to 0.0005438696383598014
iters: 100, epoch: 3 | loss: 0.1737544
speed: 3.3320s/iter; left time: 238557.6399s
iters: 200, epoch: 3 | loss: 0.1789310
speed: 1.0318s/iter; left time: 73767.7909s
iters: 300, epoch: 3 | loss: 0.1665391
speed: 1.0309s/iter; left time: 73603.6197s
iters: 400, epoch: 3 | loss: 0.1848269
speed: 1.0256s/iter; left time: 73123.6398s
iters: 500, epoch: 3 | loss: 0.1668039
speed: 1.0318s/iter; left time: 73461.2600s
Epoch: 3 cost time: 751.6240499019623
Epoch: 3, Steps: 569 | Train Loss: 0.1762433 Vali Loss: 0.1515081 Test Loss: 0.1697921
Validation loss decreased (0.181041 --> 0.169792). Saving model ...
Updating learning rate to 0.0007216782312573101
iters: 100, epoch: 4 | loss: 0.1836338
speed: 3.4291s/iter; left time: 243557.9046s
iters: 200, epoch: 4 | loss: 0.1758635
speed: 1.0429s/iter; left time: 73970.9727s
iters: 300, epoch: 4 | loss: 0.1549026
speed: 1.0467s/iter; left time: 74133.3897s
iters: 400, epoch: 4 | loss: 0.1693477
speed: 1.0495s/iter; left time: 74225.1523s
iters: 500, epoch: 4 | loss: 0.1675291
speed: 1.0631s/iter; left time: 75079.4604s
Epoch: 4 cost time: 776.1883296966553
Epoch: 4, Steps: 569 | Train Loss: 0.1669124 Vali Loss: 0.1438198 Test Loss: 0.1633192
Validation loss decreased (0.169792 --> 0.163319). Saving model ...
Updating learning rate to 0.0009668541897551314
iters: 100, epoch: 5 | loss: 0.2054136
speed: 3.5833s/iter; left time: 252470.1488s
iters: 200, epoch: 5 | loss: 0.1449289
speed: 1.0766s/iter; left time: 75745.1565s
iters: 300, epoch: 5 | loss: 0.1641962
speed: 1.0796s/iter; left time: 75852.3986s
iters: 400, epoch: 5 | loss: 0.1423285
speed: 1.0942s/iter; left time: 76763.0595s
iters: 500, epoch: 5 | loss: 0.1692078
speed: 1.1129s/iter; left time: 77965.2895s
Epoch: 5 cost time: 810.8060607910156
Epoch: 5, Steps: 569 | Train Loss: 0.1600826 Vali Loss: 0.1375290 Test Loss: 0.1579647
Validation loss decreased (0.163319 --> 0.157965). Saving model ...
Updating learning rate to 0.00127570933348449
iters: 100, epoch: 6 | loss: 0.1811412
speed: 3.7949s/iter; left time: 265215.6239s
iters: 200, epoch: 6 | loss: 0.1560985
speed: 1.1357s/iter; left time: 79256.7592s
iters: 300, epoch: 6 | loss: 0.1489340
speed: 1.1283s/iter; left time: 78625.7490s
iters: 400, epoch: 6 | loss: 0.1682404
speed: 1.1230s/iter; left time: 78150.5873s
iters: 500, epoch: 6 | loss: 0.1598627
speed: 1.1313s/iter; left time: 78610.1398s
Epoch: 6 cost time: 835.6478009223938
Epoch: 6, Steps: 569 | Train Loss: 0.1534857 Vali Loss: 0.1326768 Test Loss: 0.1532239
Validation loss decreased (0.157965 --> 0.153224). Saving model ...
Updating learning rate to 0.0016435975565022264
iters: 100, epoch: 7 | loss: 0.1480751
speed: 3.8564s/iter; left time: 267322.8725s
iters: 200, epoch: 7 | loss: 0.1680434
speed: 1.1435s/iter; left time: 79149.8896s
iters: 300, epoch: 7 | loss: 0.1503858
speed: 1.1010s/iter; left time: 76102.4247s
iters: 400, epoch: 7 | loss: 0.1449037
speed: 1.0821s/iter; left time: 74688.2330s
iters: 500, epoch: 7 | loss: 0.1438249
speed: 1.0718s/iter; left time: 73866.1101s
Epoch: 7 cost time: 804.6097757816315
Epoch: 7, Steps: 569 | Train Loss: 0.1466978 Vali Loss: 0.1248935 Test Loss: 0.1462551
Validation loss decreased (0.153224 --> 0.146255). Saving model ...
Updating learning rate to 0.0020649847186326436
iters: 100, epoch: 8 | loss: 0.1424668
speed: 3.5710s/iter; left time: 245505.2574s
iters: 200, epoch: 8 | loss: 0.1597175
speed: 1.0626s/iter; left time: 72950.9013s
iters: 300, epoch: 8 | loss: 0.1518050
speed: 1.0612s/iter; left time: 72746.0483s
iters: 400, epoch: 8 | loss: 0.1528331
speed: 1.0613s/iter; left time: 72646.9084s
iters: 500, epoch: 8 | loss: 0.1401727
speed: 1.0633s/iter; left time: 72675.0088s
Epoch: 8 cost time: 782.5107827186584
Epoch: 8, Steps: 569 | Train Loss: 0.1416359 Vali Loss: 0.1227273 Test Loss: 0.1446138
Validation loss decreased (0.146255 --> 0.144614). Saving model ...
Updating learning rate to 0.0025335318955026316
iters: 100, epoch: 9 | loss: 0.1640212
speed: 3.5917s/iter; left time: 244883.5439s
iters: 200, epoch: 9 | loss: 0.1424782
speed: 1.0803s/iter; left time: 73550.7044s
iters: 300, epoch: 9 | loss: 0.1406053
speed: 1.0691s/iter; left time: 72681.7942s
iters: 400, epoch: 9 | loss: 0.1344687
speed: 1.0734s/iter; left time: 72863.9119s
iters: 500, epoch: 9 | loss: 0.1344212
speed: 1.0729s/iter; left time: 72720.7651s
Epoch: 9 cost time: 790.8796648979187
Epoch: 9, Steps: 569 | Train Loss: 0.1383337 Vali Loss: 0.1234406 Test Loss: 0.1445752
Validation loss decreased (0.144614 --> 0.144575). Saving model ...
Updating learning rate to 0.0030421907349402868
iters: 100, epoch: 10 | loss: 0.1221671
speed: 3.6158s/iter; left time: 244474.7717s
iters: 200, epoch: 10 | loss: 0.1388021
speed: 1.0803s/iter; left time: 72931.8338s
iters: 300, epoch: 10 | loss: 0.1417881
speed: 1.0787s/iter; left time: 72718.5007s
iters: 400, epoch: 10 | loss: 0.1355302
speed: 1.0773s/iter; left time: 72517.9725s
iters: 500, epoch: 10 | loss: 0.1564880
speed: 1.0761s/iter; left time: 72323.5529s
Epoch: 10 cost time: 788.7694547176361
Epoch: 10, Steps: 569 | Train Loss: 0.1358557 Vali Loss: 0.1225463 Test Loss: 0.1448935
EarlyStopping counter: 1 out of 15
Updating learning rate to 0.0035833094852913607
iters: 100, epoch: 11 | loss: 0.1306558
speed: 3.5695s/iter; left time: 239309.6703s
iters: 200, epoch: 11 | loss: 0.1292739
speed: 1.0730s/iter; left time: 71832.1571s
iters: 300, epoch: 11 | loss: 0.1189102
speed: 1.0767s/iter; left time: 71971.7447s
iters: 400, epoch: 11 | loss: 0.1148966
speed: 1.0708s/iter; left time: 71470.4896s
iters: 500, epoch: 11 | loss: 0.1668994
speed: 1.0813s/iter; left time: 72061.8719s
Epoch: 11 cost time: 788.3503956794739
Epoch: 11, Steps: 569 | Train Loss: 0.1347841 Vali Loss: 0.1217528 Test Loss: 0.1436831
Validation loss decreased (0.144575 --> 0.143683). Saving model ...
Updating learning rate to 0.004148748100670248
iters: 100, epoch: 12 | loss: 0.1256359
speed: 3.5872s/iter; left time: 238456.5907s
iters: 200, epoch: 12 | loss: 0.1215385
speed: 1.0943s/iter; left time: 72636.3414s
iters: 300, epoch: 12 | loss: 0.1364799
speed: 1.0889s/iter; left time: 72162.5911s
iters: 400, epoch: 12 | loss: 0.1285805
speed: 1.0904s/iter; left time: 72153.0903s
iters: 500, epoch: 12 | loss: 0.1385314
speed: 1.1060s/iter; left time: 73080.3693s
Epoch: 12 cost time: 803.489027261734
Epoch: 12, Steps: 569 | Train Loss: 0.1334861 Vali Loss: 0.1211780 Test Loss: 0.1425641
Validation loss decreased (0.143683 --> 0.142564). Saving model ...
Updating learning rate to 0.004730000691617947
iters: 100, epoch: 13 | loss: 0.1310799
speed: 3.6822s/iter; left time: 242675.1602s
iters: 200, epoch: 13 | loss: 0.1252051
speed: 1.1022s/iter; left time: 72531.4118s
iters: 300, epoch: 13 | loss: 0.1324320
speed: 1.1022s/iter; left time: 72420.4552s
iters: 400, epoch: 13 | loss: 0.1329231
speed: 1.1098s/iter; left time: 72811.3946s
iters: 500, epoch: 13 | loss: 0.1390626
speed: 1.0981s/iter; left time: 71928.7995s
Epoch: 13 cost time: 808.8779151439667
Epoch: 13, Steps: 569 | Train Loss: 0.1323428 Vali Loss: 0.1211742 Test Loss: 0.1422590
Validation loss decreased (0.142564 --> 0.142259). Saving model ...
Updating learning rate to 0.005318323479142555
iters: 100, epoch: 14 | loss: 0.1322173
speed: 3.6603s/iter; left time: 239146.1068s
iters: 200, epoch: 14 | loss: 0.1258045
speed: 1.1045s/iter; left time: 72056.4090s
iters: 300, epoch: 14 | loss: 0.1308311
speed: 1.1126s/iter; left time: 72471.0808s
iters: 400, epoch: 14 | loss: 0.1096189
speed: 1.1031s/iter; left time: 71742.5051s
iters: 500, epoch: 14 | loss: 0.1165087
speed: 1.1097s/iter; left time: 72057.6367s
Epoch: 14 cost time: 813.2106020450592
Epoch: 14, Steps: 569 | Train Loss: 0.1318251 Vali Loss: 0.1196903 Test Loss: 0.1417156
Validation loss decreased (0.142259 --> 0.141716). Saving model ...
Updating learning rate to 0.005904866327330485
iters: 100, epoch: 15 | loss: 0.1328065
speed: 3.7229s/iter; left time: 241119.7179s
iters: 200, epoch: 15 | loss: 0.1321158
speed: 1.1187s/iter; left time: 72343.7692s
iters: 300, epoch: 15 | loss: 0.1203708
speed: 1.1039s/iter; left time: 71276.5483s
iters: 400, epoch: 15 | loss: 0.1346501
speed: 1.1211s/iter; left time: 72276.8856s
iters: 500, epoch: 15 | loss: 0.1562731
speed: 1.1225s/iter; left time: 72250.1356s
Epoch: 15 cost time: 814.9609882831573
Epoch: 15, Steps: 569 | Train Loss: 0.1314835 Vali Loss: 0.1189521 Test Loss: 0.1438855
EarlyStopping counter: 1 out of 15
Updating learning rate to 0.006480805875884149
iters: 100, epoch: 16 | loss: 0.1382683
speed: 3.6578s/iter; left time: 234824.5235s
iters: 200, epoch: 16 | loss: 0.1208789
speed: 1.0888s/iter; left time: 69787.8019s
iters: 300, epoch: 16 | loss: 0.1287526
speed: 1.1095s/iter; left time: 71006.8119s
iters: 400, epoch: 16 | loss: 0.1200616
speed: 1.1173s/iter; left time: 71395.5879s
iters: 500, epoch: 16 | loss: 0.1363845
speed: 1.1028s/iter; left time: 70358.7361s
Epoch: 16 cost time: 806.88671708107
Epoch: 16, Steps: 569 | Train Loss: 0.1308369 Vali Loss: 0.1195413 Test Loss: 0.1423072
EarlyStopping counter: 2 out of 15
Updating learning rate to 0.007037478269874255
iters: 100, epoch: 17 | loss: 0.1160005
speed: 3.6684s/iter; left time: 233418.1647s
iters: 200, epoch: 17 | loss: 0.1320223
speed: 1.1077s/iter; left time: 70373.7366s
iters: 300, epoch: 17 | loss: 0.1400796
speed: 1.1114s/iter; left time: 70496.7972s
iters: 400, epoch: 17 | loss: 0.1349611
speed: 1.1089s/iter; left time: 70224.1652s
iters: 500, epoch: 17 | loss: 0.1367357
speed: 1.1122s/iter; left time: 70322.0230s
Epoch: 17 cost time: 812.928195476532
Epoch: 17, Steps: 569 | Train Loss: 0.1301353 Vali Loss: 0.1217013 Test Loss: 0.1443913
EarlyStopping counter: 3 out of 15
Updating learning rate to 0.007566509490053841
iters: 100, epoch: 18 | loss: 0.1203674
speed: 3.6941s/iter; left time: 232950.2027s
iters: 200, epoch: 18 | loss: 0.1253801
speed: 1.1099s/iter; left time: 69881.0709s
iters: 300, epoch: 18 | loss: 0.1170303
speed: 1.1100s/iter; left time: 69775.0307s
iters: 400, epoch: 18 | loss: 0.1261112
speed: 1.1110s/iter; left time: 69723.7878s
iters: 500, epoch: 18 | loss: 0.1464421
speed: 1.1179s/iter; left time: 70049.4050s
Epoch: 18 cost time: 813.0010011196136
Epoch: 18, Steps: 569 | Train Loss: 0.1296439 Vali Loss: 0.1200742 Test Loss: 0.1446345
EarlyStopping counter: 4 out of 15
Updating learning rate to 0.008059941323176025
iters: 100, epoch: 19 | loss: 0.1214533
speed: 3.6898s/iter; left time: 230579.2640s
iters: 200, epoch: 19 | loss: 0.1245885
speed: 1.1129s/iter; left time: 69436.6002s
iters: 300, epoch: 19 | loss: 0.1245924
speed: 1.1132s/iter; left time: 69341.4095s
iters: 400, epoch: 19 | loss: 0.1425354
speed: 1.1173s/iter; left time: 69483.6497s
iters: 500, epoch: 19 | loss: 0.1240206
speed: 1.1194s/iter; left time: 69502.0105s
Epoch: 19 cost time: 822.0454754829407
Epoch: 19, Steps: 569 | Train Loss: 0.1295379 Vali Loss: 0.1200511 Test Loss: 0.1428440
EarlyStopping counter: 5 out of 15
Updating learning rate to 0.008510351077344748
iters: 100, epoch: 20 | loss: 0.1263539
speed: 3.7566s/iter; left time: 232613.7427s
iters: 200, epoch: 20 | loss: 0.1187757
speed: 1.1186s/iter; left time: 69154.3312s
iters: 300, epoch: 20 | loss: 0.1373482
speed: 1.1196s/iter; left time: 69105.0846s
iters: 400, epoch: 20 | loss: 0.1567149
speed: 1.1282s/iter; left time: 69522.0716s
iters: 500, epoch: 20 | loss: 0.1515603
speed: 1.1275s/iter; left time: 69364.4968s
Epoch: 20 cost time: 823.766086101532
Epoch: 20, Steps: 569 | Train Loss: 0.1281382 Vali Loss: 0.1174305 Test Loss: 0.1411486
Validation loss decreased (0.141716 --> 0.141149). Saving model ...
Updating learning rate to 0.008910963241521299
iters: 100, epoch: 21 | loss: 0.1266839
speed: 3.7462s/iter; left time: 229842.8568s
iters: 200, epoch: 21 | loss: 0.1243480
speed: 1.1236s/iter; left time: 68825.3901s
iters: 300, epoch: 21 | loss: 0.1199915
speed: 1.1093s/iter; left time: 67839.6759s
iters: 400, epoch: 21 | loss: 0.1314764
speed: 1.1203s/iter; left time: 68399.7094s
iters: 500, epoch: 21 | loss: 0.1308424
speed: 1.1226s/iter; left time: 68422.9836s
Epoch: 21 cost time: 826.518247127533
Epoch: 21, Steps: 569 | Train Loss: 0.1287808 Vali Loss: 0.1187790 Test Loss: 0.1415707
EarlyStopping counter: 1 out of 15
Updating learning rate to 0.009255751409493335
iters: 100, epoch: 22 | loss: 0.1549387
speed: 3.8115s/iter; left time: 231678.8401s
iters: 200, epoch: 22 | loss: 0.1276367
speed: 1.1414s/iter; left time: 69262.2688s
iters: 300, epoch: 22 | loss: 0.1412580
speed: 1.1378s/iter; left time: 68930.3101s
iters: 400, epoch: 22 | loss: 0.1388814
speed: 1.1405s/iter; left time: 68979.3570s
iters: 500, epoch: 22 | loss: 0.1211233
speed: 1.1367s/iter; left time: 68640.2472s
Epoch: 22 cost time: 839.4935293197632
Epoch: 22, Steps: 569 | Train Loss: 0.1275341 Vali Loss: 0.1179090 Test Loss: 0.1409055
Validation loss decreased (0.141149 --> 0.140906). Saving model ...
Updating learning rate to 0.00953952893506483
iters: 100, epoch: 23 | loss: 0.1430758
speed: 3.8458s/iter; left time: 231572.4882s
iters: 200, epoch: 23 | loss: 0.1514685
speed: 1.1313s/iter; left time: 68006.0719s
iters: 300, epoch: 23 | loss: 0.1548619
speed: 1.1460s/iter; left time: 68778.4242s
iters: 400, epoch: 23 | loss: 0.1287902
speed: 1.1476s/iter; left time: 68759.8925s
iters: 500, epoch: 23 | loss: 0.1221040
speed: 1.1298s/iter; left time: 67581.6119s
Epoch: 23 cost time: 836.227929353714
Epoch: 23, Steps: 569 | Train Loss: 0.1379215 Vali Loss: 0.1189515 Test Loss: 0.1409961
EarlyStopping counter: 1 out of 15
Updating learning rate to 0.00975802695474148
iters: 100, epoch: 24 | loss: 0.1394719
speed: 3.8070s/iter; left time: 227072.6245s
iters: 200, epoch: 24 | loss: 0.1276447
speed: 1.1391s/iter; left time: 67828.8806s
iters: 300, epoch: 24 | loss: 0.1340853
speed: 1.1153s/iter; left time: 66297.3793s
iters: 400, epoch: 24 | loss: 0.1531461
speed: 1.1226s/iter; left time: 66619.2120s
iters: 500, epoch: 24 | loss: 0.1259017
speed: 1.1287s/iter; left time: 66868.8516s
Epoch: 24 cost time: 834.0026700496674
Epoch: 24, Steps: 569 | Train Loss: 0.1323397 Vali Loss: 0.1247473 Test Loss: 0.1474355
EarlyStopping counter: 2 out of 15
Updating learning rate to 0.00990795860421683
iters: 100, epoch: 25 | loss: 0.1494782
speed: 3.8226s/iter; left time: 225829.6084s
iters: 200, epoch: 25 | loss: 0.1317555
speed: 1.1274s/iter; left time: 66487.7582s
iters: 300, epoch: 25 | loss: 0.1249609
speed: 1.1341s/iter; left time: 66771.2953s
iters: 400, epoch: 25 | loss: 0.1275704
speed: 1.1353s/iter; left time: 66728.0288s
iters: 500, epoch: 25 | loss: 0.1231593
speed: 1.1186s/iter; left time: 65634.8291s
Epoch: 25 cost time: 829.4594376087189
Epoch: 25, Steps: 569 | Train Loss: 0.1303625 Vali Loss: 0.1200667 Test Loss: 0.1412977
EarlyStopping counter: 3 out of 15
Updating learning rate to 0.009987068462650923
iters: 100, epoch: 26 | loss: 0.1184397
speed: 3.7719s/iter; left time: 220685.1751s
iters: 200, epoch: 26 | loss: 0.1477326
speed: 1.1195s/iter; left time: 65390.6223s
iters: 300, epoch: 26 | loss: 0.1457703
speed: 1.1236s/iter; left time: 65516.3195s
iters: 400, epoch: 26 | loss: 0.1186529
speed: 1.1191s/iter; left time: 65140.9817s
iters: 500, epoch: 26 | loss: 0.1353677
speed: 1.1232s/iter; left time: 65269.4359s
Epoch: 26 cost time: 823.9760458469391
Epoch: 26, Steps: 569 | Train Loss: 0.1283893 Vali Loss: 0.1177257 Test Loss: 0.1404042
Validation loss decreased (0.140906 --> 0.140404). Saving model ...
Updating learning rate to 0.009999620195133964
iters: 100, epoch: 27 | loss: 0.1568359
speed: 3.7479s/iter; left time: 217150.0734s
iters: 200, epoch: 27 | loss: 0.1162058
speed: 1.1172s/iter; left time: 64618.7169s
iters: 300, epoch: 27 | loss: 0.1212554
speed: 1.1190s/iter; left time: 64611.8196s
iters: 400, epoch: 27 | loss: 0.1313380
speed: 1.1318s/iter; left time: 65237.2783s
iters: 500, epoch: 27 | loss: 0.1280621
speed: 1.1249s/iter; left time: 64727.4381s
Epoch: 27 cost time: 825.2092733383179
Epoch: 27, Steps: 569 | Train Loss: 0.1272240 Vali Loss: 0.1192657 Test Loss: 0.1416844
EarlyStopping counter: 1 out of 15
Updating learning rate to 0.009995377074165727
iters: 100, epoch: 28 | loss: 0.1252169
speed: 3.7786s/iter; left time: 216777.0777s
iters: 200, epoch: 28 | loss: 0.1274050
speed: 1.1331s/iter; left time: 64890.1740s
iters: 300, epoch: 28 | loss: 0.1205370
speed: 1.1397s/iter; left time: 65155.8838s
iters: 400, epoch: 28 | loss: 0.1329421
speed: 1.1363s/iter; left time: 64851.2820s
iters: 500, epoch: 28 | loss: 0.1263644
speed: 1.1273s/iter; left time: 64220.7642s
Epoch: 28 cost time: 830.3269376754761
Epoch: 28, Steps: 569 | Train Loss: 0.1243257 Vali Loss: 0.1175510 Test Loss: 0.1400369
Validation loss decreased (0.140404 --> 0.140037). Saving model ...
Updating learning rate to 0.009986432497967705
iters: 100, epoch: 29 | loss: 0.1320557
speed: 3.7669s/iter; left time: 213965.3757s
iters: 200, epoch: 29 | loss: 0.1540249
speed: 1.1384s/iter; left time: 64548.0865s
iters: 300, epoch: 29 | loss: 0.1324210
speed: 1.1400s/iter; left time: 64522.4873s
iters: 400, epoch: 29 | loss: 0.1272685
speed: 1.1382s/iter; left time: 64307.4729s
iters: 500, epoch: 29 | loss: 0.1147796
speed: 1.1305s/iter; left time: 63761.4617s
Epoch: 29 cost time: 831.008050441742
Epoch: 29, Steps: 569 | Train Loss: 0.1239177 Vali Loss: 0.1161080 Test Loss: 0.1388716
Validation loss decreased (0.140037 --> 0.138872). Saving model ...
Updating learning rate to 0.009972794884861934
iters: 100, epoch: 30 | loss: 0.1183965
speed: 3.7607s/iter; left time: 211474.4543s
iters: 200, epoch: 30 | loss: 0.1224973
speed: 1.1463s/iter; left time: 64344.3692s
iters: 300, epoch: 30 | loss: 0.1187710
speed: 1.1449s/iter; left time: 64148.4668s
iters: 400, epoch: 30 | loss: 0.1147176
speed: 1.1366s/iter; left time: 63572.9679s
iters: 500, epoch: 30 | loss: 0.1360426
speed: 1.1338s/iter; left time: 63304.7836s
Epoch: 30 cost time: 836.4768686294556
Epoch: 30, Steps: 569 | Train Loss: 0.1222085 Vali Loss: 0.1158515 Test Loss: 0.1406534
EarlyStopping counter: 1 out of 15
Updating learning rate to 0.009954477070092525
iters: 100, epoch: 31 | loss: 0.1412870
speed: 3.7973s/iter; left time: 211368.0744s
iters: 200, epoch: 31 | loss: 0.1159413
speed: 1.1271s/iter; left time: 62623.4226s
iters: 300, epoch: 31 | loss: 0.1257505
speed: 1.1407s/iter; left time: 63266.4546s
iters: 400, epoch: 31 | loss: 0.1185998
speed: 1.1349s/iter; left time: 62834.2223s
iters: 500, epoch: 31 | loss: 0.1407108
speed: 1.1227s/iter; left time: 62042.6172s
Epoch: 31 cost time: 826.9412605762482
Epoch: 31, Steps: 569 | Train Loss: 0.1217030 Vali Loss: 0.1146795 Test Loss: 0.1355571
Validation loss decreased (0.138872 --> 0.138555). Saving model ...
Updating learning rate to 0.009931496293745574
iters: 100, epoch: 32 | loss: 0.1284591
speed: 3.7371s/iter; left time: 205891.5067s
iters: 200, epoch: 32 | loss: 0.1244491
speed: 1.1200s/iter; left time: 61592.2875s
iters: 300, epoch: 32 | loss: 0.1287348
speed: 1.1250s/iter; left time: 61753.3241s
iters: 400, epoch: 32 | loss: 0.1156265
speed: 1.1422s/iter; left time: 62587.8117s
iters: 500, epoch: 32 | loss: 0.1143681
speed: 1.1391s/iter; left time: 62300.3049s
Epoch: 32 cost time: 830.485196352005
Epoch: 32, Steps: 569 | Train Loss: 0.1215707 Vali Loss: 0.1191571 Test Loss: 0.1409026
EarlyStopping counter: 1 out of 15
Updating learning rate to 0.009903874184523402
iters: 100, epoch: 33 | loss: 0.1251111
speed: 3.7921s/iter; left time: 206766.9690s
iters: 200, epoch: 33 | loss: 0.1226198
speed: 1.1444s/iter; left time: 62281.4583s
iters: 300, epoch: 33 | loss: 0.1282259
speed: 1.1384s/iter; left time: 61843.6806s
iters: 400, epoch: 33 | loss: 0.1163408
speed: 1.1391s/iter; left time: 61768.0832s
iters: 500, epoch: 33 | loss: 0.1259831
speed: 1.1410s/iter; left time: 61756.3180s
Epoch: 33 cost time: 837.5469632148743
Epoch: 33, Steps: 569 | Train Loss: 0.1200218 Vali Loss: 0.1201527 Test Loss: 0.1422589
EarlyStopping counter: 2 out of 15
Updating learning rate to 0.009871636739388375
iters: 100, epoch: 34 | loss: 0.1173344
speed: 3.7980s/iter; left time: 204922.5265s
iters: 200, epoch: 34 | loss: 0.1316442
speed: 1.1288s/iter; left time: 60790.9463s
iters: 300, epoch: 34 | loss: 0.1053582
speed: 1.1327s/iter; left time: 60889.2737s
iters: 400, epoch: 34 | loss: 0.1193904
speed: 1.1202s/iter; left time: 60104.7573s
iters: 500, epoch: 34 | loss: 0.1155428
speed: 1.1219s/iter; left time: 60085.5878s
Epoch: 34 cost time: 826.464114189148
Epoch: 34, Steps: 569 | Train Loss: 0.1196060 Vali Loss: 0.1162458 Test Loss: 0.1405911
EarlyStopping counter: 3 out of 15
Updating learning rate to 0.009834814299095471
iters: 100, epoch: 35 | loss: 0.1159900
speed: 3.7653s/iter; left time: 201015.5638s
iters: 200, epoch: 35 | loss: 0.1259470
speed: 1.1118s/iter; left time: 59242.7752s
iters: 300, epoch: 35 | loss: 0.1054239
speed: 1.1265s/iter; left time: 59913.6273s
iters: 400, epoch: 35 | loss: 0.1217524
speed: 1.1297s/iter; left time: 59973.9272s
iters: 500, epoch: 35 | loss: 0.1293151
speed: 1.1325s/iter; left time: 60009.1380s
Epoch: 35 cost time: 833.0298178195953
Epoch: 35, Steps: 569 | Train Loss: 0.1182280 Vali Loss: 0.1157054 Test Loss: 0.1397015
EarlyStopping counter: 4 out of 15
Updating learning rate to 0.00979344151963663
iters: 100, epoch: 36 | loss: 0.1082402
speed: 3.8603s/iter; left time: 203895.1430s
iters: 200, epoch: 36 | loss: 0.1055611
speed: 1.1545s/iter; left time: 60862.1874s
iters: 300, epoch: 36 | loss: 0.1189032
speed: 1.1503s/iter; left time: 60525.1116s
iters: 400, epoch: 36 | loss: 0.1234522
speed: 1.1417s/iter; left time: 59958.0502s
iters: 500, epoch: 36 | loss: 0.1142284
speed: 1.1351s/iter; left time: 59497.2293s
Epoch: 36 cost time: 839.5045502185822
Epoch: 36, Steps: 569 | Train Loss: 0.1166528 Vali Loss: 0.1160769 Test Loss: 0.1409409
EarlyStopping counter: 5 out of 15
Updating learning rate to 0.00974755733962374
iters: 100, epoch: 37 | loss: 0.1153315
speed: 3.7988s/iter; left time: 198483.4486s
iters: 200, epoch: 37 | loss: 0.1229952
speed: 1.1401s/iter; left time: 59454.6619s
iters: 300, epoch: 37 | loss: 0.1108699
speed: 1.1306s/iter; left time: 58845.9782s
iters: 400, epoch: 37 | loss: 0.1177943
speed: 1.1351s/iter; left time: 58968.7842s
iters: 500, epoch: 37 | loss: 0.1190817
speed: 1.1344s/iter; left time: 58818.3932s
Epoch: 37 cost time: 832.9059233665466
Epoch: 37, Steps: 569 | Train Loss: 0.1156981 Vali Loss: 0.1178569 Test Loss: 0.1419994
EarlyStopping counter: 6 out of 15
Updating learning rate to 0.009697204943640982
iters: 100, epoch: 38 | loss: 0.1276972
speed: 3.7777s/iter; left time: 195229.9105s
iters: 200, epoch: 38 | loss: 0.1204563
speed: 1.1270s/iter; left time: 58130.4872s
iters: 300, epoch: 38 | loss: 0.1175427
speed: 1.1236s/iter; left time: 57841.8129s
iters: 400, epoch: 38 | loss: 0.1106953
speed: 1.1199s/iter; left time: 57540.0563s
iters: 500, epoch: 38 | loss: 0.1014386
speed: 1.1220s/iter; left time: 57537.1089s
Epoch: 38 cost time: 828.7733871936798
Epoch: 38, Steps: 569 | Train Loss: 0.1160038 Vali Loss: 0.1170599 Test Loss: 0.1412026
EarlyStopping counter: 7 out of 15
Updating learning rate to 0.009642431721601015
iters: 100, epoch: 39 | loss: 0.1168443
speed: 3.7986s/iter; left time: 194147.7026s
iters: 200, epoch: 39 | loss: 0.1294523
speed: 1.1485s/iter; left time: 58584.5877s
iters: 300, epoch: 39 | loss: 0.1090681
speed: 1.1357s/iter; left time: 57819.8470s
iters: 400, epoch: 39 | loss: 0.1095421
speed: 1.1332s/iter; left time: 57577.7899s
iters: 500, epoch: 39 | loss: 0.1296429
speed: 1.1455s/iter; left time: 58088.1493s
Epoch: 39 cost time: 840.1748130321503
Epoch: 39, Steps: 569 | Train Loss: 0.1148623 Vali Loss: 0.1194504 Test Loss: 0.1422009
EarlyStopping counter: 8 out of 15
Updating learning rate to 0.009583289224143234
iters: 100, epoch: 40 | loss: 0.1109730
speed: 3.8569s/iter; left time: 194935.6053s
iters: 200, epoch: 40 | loss: 0.1062282
speed: 1.1297s/iter; left time: 56984.4237s
iters: 300, epoch: 40 | loss: 0.1315950
speed: 1.1213s/iter; left time: 56447.9984s
iters: 400, epoch: 40 | loss: 0.1106779
speed: 1.1160s/iter; left time: 56070.1257s
iters: 500, epoch: 40 | loss: 0.1079388
speed: 1.1207s/iter; left time: 56195.2142s
Epoch: 40 cost time: 824.3057656288147
Epoch: 40, Steps: 569 | Train Loss: 0.1152440 Vali Loss: 0.1166266 Test Loss: 0.1412615
EarlyStopping counter: 9 out of 15
Updating learning rate to 0.009519833114116133
iters: 100, epoch: 41 | loss: 0.1173238
speed: 3.7279s/iter; left time: 186293.9348s
iters: 200, epoch: 41 | loss: 0.1118353
speed: 1.1264s/iter; left time: 56176.7757s
iters: 300, epoch: 41 | loss: 0.1118452
speed: 1.1205s/iter; left time: 55771.8775s
iters: 400, epoch: 41 | loss: 0.1172003
speed: 1.1260s/iter; left time: 55932.6919s
iters: 500, epoch: 41 | loss: 0.1048472
speed: 1.1322s/iter; left time: 56128.7458s
Epoch: 41 cost time: 831.1303684711456
Epoch: 41, Steps: 569 | Train Loss: 0.1129380 Vali Loss: 0.1166707 Test Loss: 0.1425917
EarlyStopping counter: 10 out of 15
Updating learning rate to 0.009452123114189366
iters: 100, epoch: 42 | loss: 0.1119412
speed: 3.8204s/iter; left time: 188742.6803s
iters: 200, epoch: 42 | loss: 0.1091013
speed: 1.1324s/iter; left time: 55831.6080s
iters: 300, epoch: 42 | loss: 0.1103965
speed: 1.1441s/iter; left time: 56295.0167s
iters: 400, epoch: 42 | loss: 0.1083918
speed: 1.1332s/iter; left time: 55643.2145s
iters: 500, epoch: 42 | loss: 0.1068891
speed: 1.1351s/iter; left time: 55624.6720s
Epoch: 42 cost time: 832.8055894374847
Epoch: 42, Steps: 569 | Train Loss: 0.1122377 Vali Loss: 0.1158231 Test Loss: 0.1420120
EarlyStopping counter: 11 out of 15
Updating learning rate to 0.009380222950644869
iters: 100, epoch: 43 | loss: 0.1044593
speed: 3.7866s/iter; left time: 184920.2291s
iters: 200, epoch: 43 | loss: 0.1141709
speed: 1.1321s/iter; left time: 55173.6204s
iters: 300, epoch: 43 | loss: 0.1073963
speed: 1.1445s/iter; left time: 55660.8742s
iters: 400, epoch: 43 | loss: 0.1161310
speed: 1.1276s/iter; left time: 54728.2759s
iters: 500, epoch: 43 | loss: 0.1060538
speed: 1.1213s/iter; left time: 54310.5074s
Epoch: 43 cost time: 829.2817466259003
Epoch: 43, Steps: 569 | Train Loss: 0.1117694 Vali Loss: 0.1188707 Test Loss: 0.1432596
EarlyStopping counter: 12 out of 15
Updating learning rate to 0.009304200293399904
iters: 100, epoch: 44 | loss: 0.1143510
speed: 3.7405s/iter; left time: 180536.6860s
iters: 200, epoch: 44 | loss: 0.1145173
speed: 1.1054s/iter; left time: 53244.1001s
iters: 300, epoch: 44 | loss: 0.1076961
speed: 1.1026s/iter; left time: 52995.4158s
iters: 400, epoch: 44 | loss: 0.1168368
speed: 1.1128s/iter; left time: 53376.7964s
iters: 500, epoch: 44 | loss: 0.1042197
speed: 1.1145s/iter; left time: 53349.0329s
Epoch: 44 cost time: 816.4163084030151
Epoch: 44, Steps: 569 | Train Loss: 0.1108055 Vali Loss: 0.1175551 Test Loss: 0.1425284
EarlyStopping counter: 13 out of 15
Updating learning rate to 0.009224126692318512
iters: 100, epoch: 45 | loss: 0.1124947
speed: 3.7410s/iter; left time: 178434.2272s
iters: 200, epoch: 45 | loss: 0.1003242
speed: 1.1232s/iter; left time: 53461.9835s
iters: 300, epoch: 45 | loss: 0.1141060
speed: 1.1231s/iter; left time: 53342.3200s
iters: 400, epoch: 45 | loss: 0.1117448
speed: 1.1300s/iter; left time: 53556.5923s
iters: 500, epoch: 45 | loss: 0.1143763
speed: 1.1155s/iter; left time: 52758.3714s
Epoch: 45 cost time: 824.5720844268799
Epoch: 45, Steps: 569 | Train Loss: 0.1108459 Vali Loss: 0.1149000 Test Loss: 0.1405220
EarlyStopping counter: 14 out of 15
Updating learning rate to 0.009140077509871279
iters: 100, epoch: 46 | loss: 0.1117209
speed: 3.7541s/iter; left time: 176921.6896s
iters: 200, epoch: 46 | loss: 0.1118171
speed: 1.1321s/iter; left time: 53239.7828s
iters: 300, epoch: 46 | loss: 0.1136606
speed: 1.1223s/iter; left time: 52668.0477s
iters: 400, epoch: 46 | loss: 0.1194230
speed: 1.1199s/iter; left time: 52443.1803s
iters: 500, epoch: 46 | loss: 0.1014266
speed: 1.1368s/iter; left time: 53118.3237s
Epoch: 46 cost time: 831.8894190788269
Epoch: 46, Steps: 569 | Train Loss: 0.1105391 Vali Loss: 0.1172196 Test Loss: 0.1433370
EarlyStopping counter: 15 out of 15
Early stopping
testing : long_term_forecast_ECL_96_96_none_TimeMixerPP_custom_sl96_pl96_dm16_nh8_el3_dl1_df64_fc3_ebtimeF_dtTrue_Exp_0<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
test 5165
test shape: (5165, 96, 321) (5165, 96, 321)
test shape: (5165, 96, 321) (5165, 96, 321)
mse:0.1352925808429718, mae:0.23428012430667877

[jelly1130]

Thank you very much for your valuable work and contributions. I would appreciate it if you could kindly inform me when the training code will be made publicly available. Your efforts are sincerely appreciated, and I look forward to your response.