patchtst-tsmixup-two-layer

This model is a fine-tuned version of on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.1497
• Mse: 258.6847
• Mae: 0.6232
• Rmse: 16.0837
• Smape: 70.2567

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.0001
• train_batch_size: 448
• eval_batch_size: 896
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 896
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 1000
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Mse	Mae	Rmse	Smape
0.1786	0.1666	1000	0.1714	461.4028	0.7236	21.4803	100.4842
0.1661	0.3333	2000	0.1660	413.4870	0.7017	20.3344	76.6379
0.1675	0.4999	3000	0.1632	444.1673	0.6908	21.0753	144.1648
0.1641	0.6666	4000	0.1617	368.5459	0.6717	19.1976	187.8992
0.1631	0.8332	5000	0.1598	329.2304	0.6614	18.1447	77.3954
0.163	0.9998	6000	0.1594	376.1231	0.6678	19.3939	2604.0836
0.1631	1.1665	7000	0.1582	396.0430	0.6601	19.9008	79.8599
0.1604	1.3331	8000	0.1572	331.4156	0.6539	18.2048	89.4498
0.1552	1.4998	9000	0.1571	272.9918	0.6467	16.5225	88.8514
0.1569	1.6664	10000	0.1568	301.4679	0.6509	17.3628	88.6194
0.1613	1.8330	11000	0.1568	287.0067	0.6522	16.9413	137.6847
0.1564	1.9997	12000	0.1558	321.1451	0.6502	17.9205	126.8916
0.1572	2.1663	13000	0.1554	304.7481	0.6503	17.4570	190.2351
0.1565	2.3329	14000	0.1556	276.5483	0.6449	16.6297	103.9599
0.1571	2.4996	15000	0.1546	303.2998	0.6458	17.4155	144.8287
0.156	2.6662	16000	0.1544	278.1052	0.6376	16.6765	78.1067
0.1553	2.8329	17000	0.1543	274.6881	0.6390	16.5737	81.9482
0.1542	2.9995	18000	0.1547	239.9231	0.6360	15.4895	83.6961
0.1549	3.1661	19000	0.1541	277.2799	0.6419	16.6517	375.4867
0.1542	3.3328	20000	0.1542	275.0111	0.6375	16.5835	86.7323
0.1572	3.4994	21000	0.1536	264.1418	0.6371	16.2524	111.1521
0.1559	3.6661	22000	0.1539	271.3185	0.6393	16.4717	83.0078
0.154	3.8327	23000	0.1533	253.9782	0.6338	15.9367	96.5289
0.1542	3.9993	24000	0.1532	267.4779	0.6425	16.3548	68.1349
0.1534	4.1660	25000	0.1532	262.3679	0.6358	16.1978	83.8336
0.1533	4.3326	26000	0.1528	317.2105	0.6429	17.8104	85.2472
0.1556	4.4993	27000	0.1528	266.3440	0.6333	16.3200	116.4548
0.1537	4.6659	28000	0.1527	259.0167	0.6342	16.0940	91.6244
0.1541	4.8325	29000	0.1524	281.7036	0.6396	16.7840	75.8411
0.1527	4.9992	30000	0.1523	304.5508	0.6393	17.4514	86.2233
0.1522	5.1658	31000	0.1522	261.6904	0.6314	16.1768	77.8058
0.1538	5.3324	32000	0.1522	284.4175	0.6336	16.8647	97.7625
0.1537	5.4991	33000	0.1520	309.5190	0.6375	17.5932	134.9614
0.1516	5.6657	34000	0.1519	252.8892	0.6305	15.9025	119.1109
0.1536	5.8324	35000	0.1520	293.8005	0.6377	17.1406	84.0213
0.1528	5.9990	36000	0.1515	291.5611	0.6328	17.0752	353.8475
0.1515	6.1656	37000	0.1518	254.8325	0.6315	15.9635	85.4300
0.1529	6.3323	38000	0.1513	254.4357	0.6292	15.9510	112.0245
0.1526	6.4989	39000	0.1516	265.3687	0.6320	16.2901	86.3409
0.1526	6.6656	40000	0.1512	254.5356	0.6289	15.9542	87.9753
0.1518	6.8322	41000	0.1511	233.5401	0.6244	15.2820	122.1964
0.1518	6.9988	42000	0.1512	249.4746	0.6250	15.7948	86.1394
0.1501	7.1655	43000	0.1512	285.0164	0.6310	16.8824	94.1700
0.1515	7.3321	44000	0.1509	266.2695	0.6274	16.3178	96.5646
0.1523	7.4988	45000	0.1507	256.3644	0.6250	16.0114	223.2191
0.1524	7.6654	46000	0.1508	269.4569	0.6292	16.4151	88.9256
0.1513	7.8320	47000	0.1507	247.6273	0.6247	15.7362	77.1014
0.1517	7.9987	48000	0.1505	251.7547	0.6256	15.8668	83.9576
0.1512	8.1653	49000	0.1504	245.7318	0.6245	15.6758	79.3953
0.1502	8.3319	50000	0.1502	275.0424	0.6281	16.5844	90.1375
0.1523	8.4986	51000	0.1501	252.4250	0.6235	15.8879	89.7133
0.1521	8.6652	52000	0.1502	247.0445	0.6230	15.7176	78.6296
0.1529	8.8319	53000	0.1502	258.8629	0.6248	16.0892	nan
0.1511	8.9985	54000	0.1501	274.1158	0.6279	16.5564	90.8373
0.1489	9.1651	55000	0.1499	264.4435	0.6254	16.2617	80.3961
0.1511	9.3318	56000	0.1500	267.5066	0.6259	16.3556	76.1822
0.1536	9.4984	57000	0.1499	257.8295	0.6236	16.0571	107.1616
0.1511	9.6651	58000	0.1497	265.9769	0.6247	16.3088	66.7921
0.1521	9.8317	59000	0.1497	261.9222	0.6244	16.1840	79.6868
0.1471	9.9983	60000	0.1497	258.6847	0.6232	16.0837	70.2567

Framework versions

• Transformers 4.51.3
• Pytorch 2.7.1+cu126
• Datasets 2.17.1
• Tokenizers 0.21.1