ht-stmini-cls-v6_ftis_noPretrain-smlo-bml

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

• Loss: 1.1492
• Accuracy: 0.8919
• Macro F1: 0.7363

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: 8
• eval_batch_size: 4
• seed: 42
• 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: 6733
• training_steps: 134674

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Macro F1
35.4515	0.0013	174	21.9070	0.0690	0.0299
16.8987	1.0013	348	9.9771	0.3375	0.0794
7.651	2.0013	522	6.7869	0.4999	0.1251
6.1382	3.0013	696	5.6561	0.5481	0.1349
5.1599	4.0013	870	4.9033	0.5618	0.1382
4.5752	5.0013	1044	4.2459	0.5792	0.1428
3.7321	6.0012	1218	3.4640	0.5679	0.1419
3.2741	7.0012	1392	3.2347	0.5943	0.1502
2.8305	8.0012	1566	2.9392	0.6110	0.1533
2.6049	9.0012	1740	2.6772	0.6255	0.1637
2.487	10.0012	1914	2.5802	0.6207	0.1634
2.32	11.0012	2088	2.4366	0.6287	0.1810
2.2465	12.0012	2262	2.4188	0.6218	0.1825
2.2835	13.0012	2436	2.3836	0.6427	0.1947
2.1001	14.0012	2610	2.2597	0.6568	0.2169
2.063	15.0012	2784	2.2488	0.6593	0.2216
1.9306	16.0012	2958	2.1703	0.6634	0.2415
1.8336	17.0012	3132	2.2613	0.6549	0.2434
1.8524	18.0012	3306	2.0121	0.6844	0.2899
1.8061	19.0012	3480	2.0636	0.6829	0.2673
1.683	20.0011	3654	1.9940	0.7029	0.2963
1.6015	21.0011	3828	1.9928	0.7082	0.3174
1.558	22.0011	4002	1.9387	0.7167	0.3336
1.5793	23.0011	4176	1.9036	0.7294	0.3545
1.4352	24.0011	4350	1.8033	0.7261	0.3568
1.4207	25.0011	4524	1.8566	0.7269	0.3552
1.2726	26.0011	4698	1.6691	0.7496	0.3957
1.2905	27.0011	4872	1.6683	0.7490	0.3987
1.1671	28.0011	5046	1.6512	0.7552	0.4098
1.2042	29.0011	5220	1.5627	0.7555	0.4283
1.1664	30.0011	5394	1.7746	0.7414	0.4084
1.1403	31.0011	5568	1.6726	0.7605	0.4230
1.0731	32.0011	5742	1.6963	0.7375	0.4301
1.0289	33.0010	5916	1.7578	0.7467	0.4306
0.967	34.0010	6090	1.6874	0.7475	0.4541
0.9965	35.0010	6264	1.6983	0.7525	0.4285
0.9682	36.0010	6438	1.5561	0.7678	0.4434
0.9182	37.0010	6612	1.5473	0.7617	0.4456
0.9415	38.0010	6786	1.5597	0.7697	0.4666
0.8798	39.0010	6960	1.5203	0.7824	0.4783
0.7888	40.0010	7134	1.4801	0.7731	0.4821
0.7844	41.0010	7308	1.4949	0.7833	0.4843
0.807	42.0010	7482	1.4205	0.7919	0.4976
0.7701	43.0010	7656	1.4601	0.7967	0.4940
0.6837	44.0010	7830	1.3816	0.7932	0.5038
0.7004	45.0010	8004	1.4924	0.7917	0.5024
0.6791	46.0010	8178	1.4133	0.7944	0.5093
0.6205	47.0009	8352	1.4418	0.8002	0.5106
0.6017	48.0009	8526	1.5872	0.7908	0.5066
0.5733	49.0009	8700	1.6913	0.7806	0.5124
0.553	50.0009	8874	1.4380	0.8078	0.5167
0.5911	51.0009	9048	1.4504	0.8050	0.5274
0.5374	52.0009	9222	1.4591	0.8098	0.5357
0.4927	53.0009	9396	1.5309	0.8057	0.5260
0.4839	54.0009	9570	1.3519	0.8119	0.5380
0.4482	55.0009	9744	1.3275	0.8187	0.5372
0.437	56.0009	9918	1.5105	0.8060	0.5351
0.4146	57.0009	10092	1.3840	0.8119	0.5448
0.435	58.0009	10266	1.4256	0.8163	0.5585
0.4028	59.0009	10440	1.5280	0.8154	0.5466
0.396	60.0008	10614	1.3369	0.8234	0.5608
0.347	61.0008	10788	1.3442	0.8288	0.5646
0.3664	62.0008	10962	1.4856	0.8198	0.5577
0.3537	63.0008	11136	1.3081	0.8298	0.5761
0.3253	64.0008	11310	1.3074	0.8333	0.5790
0.3318	65.0008	11484	1.3272	0.8311	0.5880
0.295	66.0008	11658	1.2614	0.8353	0.5856
0.3052	67.0008	11832	1.3376	0.8341	0.5865
0.3088	68.0008	12006	1.3082	0.8393	0.5880
0.2873	69.0008	12180	1.3254	0.8343	0.5973
0.2719	70.0008	12354	1.3320	0.8363	0.5987
0.2629	71.0008	12528	1.4144	0.8323	0.5860
0.2846	72.0008	12702	1.3451	0.8374	0.6041
0.2509	73.0007	12876	1.3525	0.8423	0.6013
0.2502	74.0007	13050	1.1525	0.8497	0.6330
0.239	75.0007	13224	1.2372	0.8465	0.6130
0.2332	76.0007	13398	1.2791	0.8502	0.6106
0.2288	77.0007	13572	1.2336	0.8543	0.6314
0.2123	78.0007	13746	1.3242	0.8526	0.6183
0.1992	79.0007	13920	1.3304	0.8521	0.6198
0.2154	80.0007	14094	1.3772	0.8480	0.6123
0.2118	81.0007	14268	1.2787	0.8544	0.6254
0.1925	82.0007	14442	1.3153	0.8584	0.6294
0.1979	83.0007	14616	1.3995	0.8488	0.6184
0.1852	84.0007	14790	1.2301	0.8528	0.6261
0.1828	85.0007	14964	1.3435	0.8570	0.6325
0.1857	86.0007	15138	1.2396	0.8535	0.6315
0.1774	87.0006	15312	1.2718	0.8553	0.6334
0.1597	88.0006	15486	1.1920	0.8660	0.6454
0.1642	89.0006	15660	1.3240	0.8553	0.6362
0.1639	90.0006	15834	1.1937	0.8607	0.6496
0.1616	91.0006	16008	1.2719	0.8584	0.6436
0.1624	92.0006	16182	1.2800	0.8610	0.6447
0.1545	93.0006	16356	1.2380	0.8618	0.6457
0.1609	94.0006	16530	1.2556	0.8646	0.6511
0.1504	95.0006	16704	1.2569	0.8577	0.6402
0.1429	96.0006	16878	1.2041	0.8683	0.6553
0.1509	97.0006	17052	1.2054	0.8659	0.6582
0.1412	98.0006	17226	1.3651	0.8623	0.6515
0.1337	99.0006	17400	1.1862	0.8639	0.6561
0.1405	100.0005	17574	1.4219	0.8595	0.6436
0.151	101.0005	17748	1.3142	0.8664	0.6533
0.1348	102.0005	17922	1.1308	0.8699	0.6644
0.1288	103.0005	18096	1.2989	0.8637	0.6593
0.1262	104.0005	18270	1.2686	0.8663	0.6631
0.1274	105.0005	18444	1.3115	0.8654	0.6581
0.1245	106.0005	18618	1.1777	0.8689	0.6671
0.1258	107.0005	18792	1.2439	0.8679	0.6642
0.1199	108.0005	18966	1.2911	0.8699	0.6654
0.1176	109.0005	19140	1.3485	0.8686	0.6661
0.1146	110.0005	19314	1.3156	0.8737	0.6803
0.121	111.0005	19488	1.2584	0.8708	0.6652
0.1095	112.0005	19662	1.2118	0.8753	0.6769
0.1077	113.0005	19836	1.1833	0.8737	0.6822
0.1061	114.0004	20010	1.1777	0.8723	0.6776
0.1139	115.0004	20184	1.2363	0.8717	0.6760
0.1045	116.0004	20358	1.2453	0.8728	0.6793
0.1056	117.0004	20532	1.3777	0.8662	0.6679
0.1107	118.0004	20706	1.4315	0.8743	0.6757
0.1044	119.0004	20880	1.1999	0.8747	0.6816
0.1058	120.0004	21054	1.2723	0.8757	0.6763
0.0916	121.0004	21228	1.1777	0.8798	0.6888
0.1	122.0004	21402	1.3443	0.8769	0.6831
0.0984	123.0004	21576	1.2729	0.8758	0.6804
0.0947	124.0004	21750	1.1473	0.8794	0.6984
0.0915	125.0004	21924	1.3647	0.8716	0.6784
0.0968	126.0004	22098	1.3699	0.8744	0.6781
0.096	127.0003	22272	1.3357	0.8726	0.6767
0.0931	128.0003	22446	1.3305	0.8777	0.6867
0.0917	129.0003	22620	1.3569	0.8750	0.6878
0.0912	130.0003	22794	1.4142	0.8766	0.6820
0.0887	131.0003	22968	1.4611	0.8720	0.6875
0.0878	132.0003	23142	1.3584	0.8784	0.6867
0.0937	133.0003	23316	1.4110	0.8759	0.6867
0.0844	134.0003	23490	1.4329	0.8727	0.6833
0.0821	135.0003	23664	1.3382	0.8779	0.6926
0.0876	136.0003	23838	1.3005	0.8809	0.6907
0.0886	137.0003	24012	1.2828	0.8776	0.6867
0.0864	138.0003	24186	1.3294	0.8808	0.6923
0.0823	139.0003	24360	1.3551	0.8765	0.6888
0.089	140.0003	24534	1.4637	0.8783	0.6892
0.0814	141.0002	24708	1.4108	0.8686	0.6872
0.0802	142.0002	24882	1.2838	0.8810	0.6954
0.0793	143.0002	25056	1.2916	0.8821	0.6988
0.0836	144.0002	25230	1.4550	0.8785	0.6948
0.0801	145.0002	25404	1.2169	0.8828	0.7010
0.0813	146.0002	25578	1.2913	0.8787	0.6948
0.0755	147.0002	25752	1.4690	0.8776	0.6931
0.08	148.0002	25926	1.2524	0.8842	0.7010
0.073	149.0002	26100	1.4146	0.8849	0.7005
0.0754	150.0002	26274	1.3797	0.8806	0.7009
0.0703	151.0002	26448	1.4607	0.8731	0.6912
0.0693	152.0002	26622	1.2254	0.8833	0.7042
0.0718	153.0002	26796	1.4364	0.8726	0.6913
0.0768	154.0001	26970	1.2333	0.8826	0.7142
0.0713	155.0001	27144	1.3738	0.8786	0.7012
0.0729	156.0001	27318	1.3060	0.8831	0.7057
0.0725	157.0001	27492	1.3991	0.8795	0.6936
0.0693	158.0001	27666	1.4688	0.8800	0.7071
0.0694	159.0001	27840	1.3043	0.8863	0.7083
0.0672	160.0001	28014	1.3511	0.8812	0.7073
0.0696	161.0001	28188	1.2976	0.8834	0.7030
0.0676	162.0001	28362	1.3276	0.8820	0.7002
0.0678	163.0001	28536	1.2867	0.8822	0.7045
0.0664	164.0001	28710	1.3584	0.8837	0.7090
0.0659	165.0001	28884	1.3738	0.8785	0.7029
0.0615	166.0001	29058	1.2903	0.8810	0.7045
0.0619	167.0001	29232	1.4785	0.8813	0.7062
0.0618	168.0000	29406	1.2958	0.8841	0.7064
0.0635	169.0000	29580	1.2425	0.8865	0.7103
0.065	170.0000	29754	1.3259	0.8875	0.7182
0.0635	171.0000	29928	1.4124	0.8812	0.7105
0.0617	172.0000	30102	1.3992	0.8831	0.7068
0.0618	173.0000	30276	1.4231	0.8825	0.7101
0.062	173.0013	30450	1.5441	0.8832	0.7027
0.0645	174.0013	30624	1.2789	0.8891	0.7191
0.0588	175.0013	30798	1.3923	0.8851	0.7036
0.0603	176.0013	30972	1.2946	0.8896	0.7177
0.0628	177.0013	31146	1.3610	0.8867	0.7143
0.0581	178.0013	31320	1.3410	0.8854	0.7121
0.057	179.0013	31494	1.3122	0.8895	0.7209
0.0586	180.0012	31668	1.5097	0.8839	0.7117
0.0625	181.0012	31842	1.4167	0.8847	0.7152
0.0563	182.0012	32016	1.1503	0.8925	0.7368
0.0609	183.0012	32190	1.1861	0.8900	0.7281
0.0582	184.0012	32364	1.2710	0.8894	0.7202
0.056	185.0012	32538	1.3732	0.8867	0.7167
0.0595	186.0012	32712	1.4251	0.8845	0.7144
0.0566	187.0012	32886	1.3183	0.8923	0.7213
0.0538	188.0012	33060	1.5077	0.8781	0.7138
0.0555	189.0012	33234	1.3157	0.8900	0.7201
0.0559	190.0012	33408	1.2063	0.8940	0.7299
0.0519	191.0012	33582	1.2676	0.8857	0.7239
0.0555	192.0012	33756	1.3803	0.8861	0.7147
0.0535	193.0012	33930	1.3908	0.8854	0.7259
0.0544	194.0011	34104	1.4638	0.8867	0.7217
0.0533	195.0011	34278	1.3818	0.8871	0.7271
0.0501	196.0011	34452	1.3062	0.8905	0.7247
0.0531	197.0011	34626	1.4616	0.8816	0.7163
0.0625	198.0011	34800	1.5393	0.8812	0.7069
0.0528	199.0011	34974	1.3661	0.8857	0.7184
0.0537	200.0011	35148	1.1742	0.8909	0.7270
0.0502	201.0011	35322	1.3423	0.8876	0.7256
0.0515	202.0011	35496	1.3618	0.8885	0.7228

Framework versions

• Transformers 4.46.0
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.20.1