patchtst-tsmixup

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

• Loss: 0.2566
• Mse: 4643.9019
• Mae: 1.1228
• Rmse: 68.1462
• Smape: nan

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: 256
• eval_batch_size: 512
• seed: 42
• gradient_accumulation_steps: 2
• total_train_batch_size: 512
• 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
0.3084	0.0980	1000	0.3024	5346.5903	1.2307	73.1204	153.0007
0.2929	0.1960	2000	0.2897	5068.2759	1.1988	71.1918	nan
0.2907	0.2940	3000	0.2848	4958.8062	1.1856	70.4188	nan
0.2839	0.3920	4000	0.2819	4832.9458	1.1795	69.5194	nan
0.2839	0.4901	5000	0.2792	4758.4775	1.1657	68.9817	173.3474
0.2799	0.5881	6000	0.2773	4732.7681	1.1655	68.7951	154.0682
0.2812	0.6861	7000	0.2756	4791.2764	1.1631	69.2190	inf
0.2762	0.7841	8000	0.2742	4711.9121	1.1609	68.6434	150.8088
0.2775	0.8821	9000	0.2745	4661.4844	1.1561	68.2751	nan
0.2739	0.9801	10000	0.2726	4697.7852	1.1541	68.5404	nan
0.2763	1.0781	11000	0.2718	4672.8022	1.1511	68.3579	nan
0.2736	1.1761	12000	0.2712	4646.1968	1.1528	68.1630	144.7422
0.2727	1.2741	13000	0.2699	4704.0283	1.1525	68.5859	nan
0.2705	1.3721	14000	0.2694	4732.0859	1.1485	68.7902	nan
0.2718	1.4702	15000	0.2691	4672.9360	1.1495	68.3589	nan
0.2714	1.5682	16000	0.2687	4653.2021	1.1452	68.2144	nan
0.2721	1.6662	17000	0.2677	4653.2695	1.1435	68.2149	263.4695
0.2694	1.7642	18000	0.2687	4744.7812	1.1485	68.8824	nan
0.2707	1.8622	19000	0.2680	4563.2783	1.1418	67.5520	nan
0.2701	1.9602	20000	0.2670	4582.1411	1.1379	67.6915	nan
0.2687	2.0582	21000	0.2666	4580.3374	1.1404	67.6782	203.8995
0.2698	2.1562	22000	0.2663	4663.5762	1.1435	68.2904	nan
0.2684	2.2542	23000	0.2663	4542.9971	1.1381	67.4018	160.5439
0.2692	2.3522	24000	0.2656	4628.6294	1.1399	68.0340	143.5725
0.2692	2.4503	25000	0.2654	4611.9399	1.1401	67.9113	nan
0.2674	2.5483	26000	0.2650	4631.6226	1.1373	68.0560	nan
0.2676	2.6463	27000	0.2652	4656.6133	1.1404	68.2394	261.2106
0.2701	2.7443	28000	0.2645	4583.7627	1.1386	67.7035	nan
0.2676	2.8423	29000	0.2645	4596.8721	1.1331	67.8002	147.3792
0.2671	2.9403	30000	0.2642	4660.9785	1.1388	68.2714	nan
0.2666	3.0383	31000	0.2646	4729.0498	1.1405	68.7681	nan
0.267	3.1363	32000	0.2638	4698.4321	1.1409	68.5451	nan
0.2663	3.2343	33000	0.2641	4621.0	1.1369	67.9779	nan
0.2642	3.3324	34000	0.2632	4717.3887	1.1363	68.6832	nan
0.2686	3.4304	35000	0.2633	4576.2842	1.1339	67.6482	nan
0.2639	3.5284	36000	0.2635	4640.9990	1.1401	68.1249	145.8278
0.2668	3.6264	37000	0.2631	4652.9268	1.1371	68.2124	nan
0.2685	3.7244	38000	0.2627	4674.4717	1.1338	68.3701	nan
0.2651	3.8224	39000	0.2626	4667.7871	1.1355	68.3212	nan
0.2637	3.9204	40000	0.2626	4603.3242	1.1337	67.8478	128.9741
0.2655	4.0184	41000	0.2623	4671.7549	1.1358	68.3502	nan
0.2657	4.1164	42000	0.2621	4639.7461	1.1318	68.1157	nan
0.2637	4.2144	43000	0.2619	4658.6704	1.1330	68.2545	nan
0.2649	4.3125	44000	0.2621	4717.4502	1.1337	68.6837	nan
0.2651	4.4105	45000	0.2617	4616.3667	1.1287	67.9438	nan
0.2648	4.5085	46000	0.2615	4641.1528	1.1327	68.1260	164.7345
0.2646	4.6065	47000	0.2616	4634.9507	1.1343	68.0805	nan
0.2619	4.7045	48000	0.2612	4717.5820	1.1327	68.6847	nan
0.2641	4.8025	49000	0.2612	4671.6055	1.1355	68.3491	nan
0.264	4.9005	50000	0.2613	4625.3916	1.1307	68.0102	nan
0.264	4.9985	51000	0.2607	4600.4443	1.1309	67.8266	nan
0.2646	5.0965	52000	0.2607	4653.5298	1.1327	68.2168	nan
0.2643	5.1946	53000	0.2604	4582.6050	1.1316	67.6949	nan
0.263	5.2926	54000	0.2607	4624.2041	1.1305	68.0015	nan
0.264	5.3906	55000	0.2604	4654.0234	1.1305	68.2204	nan
0.2621	5.4886	56000	0.2601	4626.9565	1.1290	68.0217	nan
0.2641	5.5866	57000	0.2604	4636.6865	1.1318	68.0932	nan
0.2649	5.6846	58000	0.2600	4662.0747	1.1305	68.2794	nan
0.2637	5.7826	59000	0.2599	4631.6445	1.1295	68.0562	nan
0.2632	5.8806	60000	0.2598	4632.5400	1.1311	68.0628	146.8735
0.263	5.9786	61000	0.2596	4634.8896	1.1297	68.0800	nan
0.2626	6.0766	62000	0.2598	4677.4688	1.1307	68.3920	nan
0.2623	6.1747	63000	0.2596	4674.2075	1.1313	68.3682	nan
0.2646	6.2727	64000	0.2595	4665.5918	1.1310	68.3051	nan
0.2631	6.3707	65000	0.2593	4672.3618	1.1285	68.3547	nan
0.2623	6.4687	66000	0.2593	4666.8711	1.1299	68.3145	nan
0.2636	6.5667	67000	0.2594	4603.8647	1.1279	67.8518	nan
0.262	6.6647	68000	0.2590	4614.9053	1.1276	67.9331	nan
0.2616	6.7627	69000	0.2591	4621.3652	1.1286	67.9806	nan
0.2623	6.8607	70000	0.2587	4653.7485	1.1297	68.2184	nan
0.2606	6.9587	71000	0.2588	4616.5127	1.1265	67.9449	nan
0.2625	7.0567	72000	0.2588	4605.3052	1.1267	67.8624	nan
0.2616	7.1548	73000	0.2586	4632.1304	1.1258	68.0598	nan
0.261	7.2528	74000	0.2584	4635.8457	1.1264	68.0870	nan
0.2624	7.3508	75000	0.2583	4700.9951	1.1278	68.5638	nan
0.2604	7.4488	76000	0.2583	4718.2422	1.1279	68.6895	nan
0.2611	7.5468	77000	0.2583	4681.1753	1.1288	68.4191	nan
0.2603	7.6448	78000	0.2581	4658.5015	1.1270	68.2532	nan
0.263	7.7428	79000	0.2581	4655.0098	1.1287	68.2276	nan
0.2593	7.8408	80000	0.2579	4674.6558	1.1277	68.3715	nan
0.2616	7.9388	81000	0.2580	4662.5923	1.1278	68.2832	nan
0.2623	8.0369	82000	0.2578	4622.4644	1.1260	67.9887	nan
0.2599	8.1349	83000	0.2578	4693.6475	1.1265	68.5102	nan
0.2606	8.2329	84000	0.2578	4664.7837	1.1271	68.2992	nan
0.2604	8.3309	85000	0.2576	4674.7075	1.1273	68.3718	nan
0.2605	8.4289	86000	0.2576	4615.8818	1.1244	67.9403	nan
0.2598	8.5269	87000	0.2575	4681.8462	1.1254	68.4240	nan
0.2606	8.6249	88000	0.2575	4662.5674	1.1251	68.2830	nan
0.2602	8.7229	89000	0.2573	4656.4746	1.1238	68.2384	nan
0.2588	8.8209	90000	0.2573	4647.8911	1.1249	68.1754	nan
0.2617	8.9189	91000	0.2573	4655.3872	1.1245	68.2304	nan
0.2617	9.0170	92000	0.2573	4648.3506	1.1258	68.1788	nan
0.2601	9.1150	93000	0.2571	4630.1865	1.1232	68.0455	nan
0.2598	9.2130	94000	0.2570	4677.0566	1.1245	68.3890	nan
0.2609	9.3110	95000	0.2570	4628.4395	1.1244	68.0326	nan
0.2596	9.4090	96000	0.2569	4638.3027	1.1245	68.1051	nan
0.2588	9.5070	97000	0.2568	4655.1597	1.1249	68.2287	nan
0.2613	9.6050	98000	0.2567	4639.6914	1.1242	68.1153	nan
0.2593	9.7030	99000	0.2568	4654.0952	1.1250	68.2209	nan
0.2593	9.8010	100000	0.2567	4644.1465	1.1229	68.1480	165.2788
0.261	9.8990	101000	0.2567	4641.4990	1.1234	68.1285	nan
0.2587	9.9971	102000	0.2566	4643.9019	1.1228	68.1462	nan

Framework versions

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