ht-stmini-cls-v6_ftis_noPretrain

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

• Loss: 3.7306
• Accuracy: 0.9077
• Macro F1: 0.7713

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
34.0309	0.0013	174	54.3149	0.0208	0.0149
13.9676	1.0013	348	106.7151	0.2082	0.0707
6.6221	2.0013	522	162.7861	0.4882	0.1244
5.6829	3.0013	696	147.2379	0.5382	0.1328
4.7257	4.0013	870	126.3738	0.5631	0.1399
4.1961	5.0013	1044	90.0849	0.5861	0.1449
3.393	6.0012	1218	74.7840	0.5754	0.1495
2.97	7.0012	1392	59.0515	0.6010	0.1596
2.6074	8.0012	1566	41.9846	0.6152	0.1598
2.4519	9.0012	1740	31.2251	0.6339	0.1744
2.3446	10.0012	1914	27.2994	0.6257	0.1711
2.1896	11.0012	2088	24.9494	0.6354	0.1897
2.1176	12.0012	2262	18.0234	0.6286	0.1901
2.166	13.0012	2436	16.7247	0.6481	0.2038
1.9853	14.0012	2610	16.4197	0.6614	0.2220
1.9643	15.0012	2784	14.9593	0.6594	0.2237
1.8193	16.0012	2958	10.3868	0.6682	0.2492
1.7405	17.0012	3132	9.2300	0.6515	0.2544
1.7794	18.0012	3306	10.9663	0.6843	0.3047
1.6996	19.0012	3480	8.8828	0.6839	0.2899
1.5995	20.0011	3654	9.0213	0.6987	0.3046
1.5234	21.0011	3828	6.5229	0.7215	0.3458
1.4337	22.0011	4002	7.9392	0.7445	0.3797
1.4328	23.0011	4176	7.1796	0.7483	0.3963
1.2907	24.0011	4350	6.5769	0.7434	0.4137
1.2665	25.0011	4524	8.2191	0.7450	0.3911
1.1392	26.0011	4698	6.4740	0.7661	0.4338
1.1307	27.0011	4872	8.9443	0.7740	0.4551
1.0104	28.0011	5046	9.2499	0.7699	0.4467
1.0111	29.0011	5220	8.1054	0.7693	0.4691
0.9978	30.0011	5394	7.6890	0.7674	0.4611
0.9727	31.0011	5568	8.9909	0.7854	0.4751
0.8926	32.0011	5742	8.0058	0.7853	0.4826
0.8245	33.0010	5916	11.8066	0.7745	0.4663
0.782	34.0010	6090	11.8401	0.7803	0.5065
0.8448	35.0010	6264	10.3797	0.7824	0.4931
0.7597	36.0010	6438	9.6193	0.8003	0.5112
0.7055	37.0010	6612	10.2706	0.7959	0.5151
0.7622	38.0010	6786	10.5250	0.7970	0.5187
0.703	39.0010	6960	12.6031	0.7886	0.4959
0.5646	40.0010	7134	12.4053	0.8063	0.5345
0.5838	41.0010	7308	11.5444	0.8031	0.5242
0.5668	42.0010	7482	13.1968	0.8198	0.5566
0.5412	43.0010	7656	13.3207	0.8259	0.5638
0.4823	44.0010	7830	14.3239	0.8365	0.5842
0.4444	45.0010	8004	16.5850	0.8259	0.5701
0.4614	46.0010	8178	18.1794	0.8416	0.5988
0.4054	47.0009	8352	16.7647	0.8504	0.6067
0.3694	48.0009	8526	18.4008	0.8434	0.5968
0.3453	49.0009	8700	17.3898	0.8481	0.6201
0.3473	50.0009	8874	14.6636	0.8508	0.6095
0.3727	51.0009	9048	19.7604	0.8547	0.6144
0.302	52.0009	9222	18.8168	0.8524	0.6177
0.2825	53.0009	9396	14.5673	0.8530	0.6279
0.2729	54.0009	9570	16.7929	0.8580	0.6365
0.2548	55.0009	9744	18.6582	0.8650	0.6466
0.2389	56.0009	9918	17.1777	0.8654	0.6428
0.2084	57.0009	10092	14.7710	0.8711	0.6617
0.2217	58.0009	10266	12.9179	0.8708	0.6684
0.2174	59.0009	10440	13.1419	0.8651	0.6493
0.2062	60.0008	10614	12.6303	0.8676	0.6625
0.1818	61.0008	10788	13.4220	0.8757	0.6703
0.1786	62.0008	10962	16.0360	0.8694	0.6652
0.169	63.0008	11136	12.0630	0.8746	0.6673
0.1628	64.0008	11310	10.7680	0.8771	0.6783
0.1554	65.0008	11484	10.6072	0.8713	0.6717
0.1496	66.0008	11658	11.1266	0.8718	0.6814
0.1345	67.0008	11832	10.2608	0.8726	0.6733
0.1383	68.0008	12006	11.4513	0.8828	0.6932
0.1291	69.0008	12180	8.7030	0.8822	0.6921
0.1161	70.0008	12354	10.1977	0.8826	0.6980
0.1147	71.0008	12528	7.5840	0.8830	0.7010
0.1275	72.0008	12702	8.0253	0.8860	0.7090
0.1077	73.0007	12876	7.8158	0.8818	0.6936
0.1088	74.0007	13050	7.5541	0.8804	0.6942
0.1031	75.0007	13224	6.8490	0.8842	0.7081
0.112	76.0007	13398	7.3048	0.8848	0.7055
0.096	77.0007	13572	6.4591	0.8795	0.7054
0.0815	78.0007	13746	7.0254	0.8880	0.7174
0.08	79.0007	13920	7.1399	0.8852	0.7151
0.0929	80.0007	14094	6.8153	0.8869	0.7080
0.0803	81.0007	14268	6.3389	0.8869	0.7125
0.0775	82.0007	14442	5.7511	0.8881	0.7109
0.0752	83.0007	14616	6.5417	0.8855	0.7037
0.072	84.0007	14790	5.8264	0.8916	0.7258
0.0692	85.0007	14964	6.6974	0.8874	0.7194
0.0672	86.0007	15138	5.4084	0.8919	0.7243
0.0612	87.0006	15312	5.1521	0.8820	0.7298
0.0595	88.0006	15486	5.0756	0.8953	0.7371
0.0613	89.0006	15660	5.0541	0.8852	0.7144
0.0628	90.0006	15834	4.8922	0.8931	0.7316
0.0577	91.0006	16008	6.0652	0.8884	0.7243
0.057	92.0006	16182	4.4136	0.8899	0.7258
0.0554	93.0006	16356	5.2064	0.8945	0.7380
0.0501	94.0006	16530	4.6064	0.8914	0.7328
0.0503	95.0006	16704	4.4629	0.8916	0.7319
0.0524	96.0006	16878	4.6768	0.8987	0.7339
0.0512	97.0006	17052	4.0132	0.8948	0.7299
0.0456	98.0006	17226	4.4474	0.8955	0.7344
0.0466	99.0006	17400	4.3961	0.8931	0.7310
0.0514	100.0005	17574	4.1822	0.9001	0.7445
0.0574	101.0005	17748	4.5937	0.8937	0.7324
0.0441	102.0005	17922	4.6609	0.8972	0.7485
0.0393	103.0005	18096	4.8007	0.8954	0.7391
0.0407	104.0005	18270	4.8574	0.8979	0.7442
0.0398	105.0005	18444	4.0337	0.8940	0.7465
0.0387	106.0005	18618	4.3228	0.8995	0.7521
0.0426	107.0005	18792	3.8549	0.8953	0.7543
0.0425	108.0005	18966	4.3286	0.8909	0.7309
0.0405	109.0005	19140	3.8524	0.8971	0.7428
0.0377	110.0005	19314	4.2679	0.9014	0.7452
0.0375	111.0005	19488	4.2141	0.9026	0.7470
0.0342	112.0005	19662	4.0151	0.8986	0.7478
0.0367	113.0005	19836	4.1929	0.8993	0.7605
0.0335	114.0004	20010	4.4448	0.8964	0.7481
0.0409	115.0004	20184	3.7438	0.8960	0.7454
0.0342	116.0004	20358	3.6396	0.9024	0.7668
0.0302	117.0004	20532	3.7275	0.9014	0.7517
0.035	118.0004	20706	3.8500	0.8972	0.7461
0.03	119.0004	20880	4.0516	0.8987	0.7510
0.0314	120.0004	21054	3.8933	0.8948	0.7503
0.0273	121.0004	21228	3.5428	0.8940	0.7518
0.0312	122.0004	21402	3.3716	0.8998	0.7599
0.0311	123.0004	21576	3.9390	0.8972	0.7526
0.0317	124.0004	21750	3.5682	0.9077	0.7713
0.0268	125.0004	21924	4.2324	0.8947	0.7434
0.0268	126.0004	22098	4.2147	0.8973	0.7512
0.0287	127.0003	22272	4.2243	0.8992	0.7497
0.0245	128.0003	22446	4.1498	0.8944	0.7443
0.0296	129.0003	22620	3.3685	0.9004	0.7562
0.0285	130.0003	22794	3.5653	0.8927	0.7431
0.0259	131.0003	22968	4.1205	0.8981	0.7534
0.0251	132.0003	23142	3.4222	0.9015	0.7534
0.0259	133.0003	23316	3.7601	0.8948	0.7487
0.0266	134.0003	23490	3.6945	0.9017	0.7602
0.0259	135.0003	23664	3.3098	0.9052	0.7668
0.0238	136.0003	23838	4.4286	0.9004	0.7511
0.026	137.0003	24012	3.9345	0.9023	0.7686
0.0268	138.0003	24186	3.9251	0.9024	0.7606
0.0209	139.0003	24360	3.8137	0.9043	0.7673
0.0222	140.0003	24534	3.6423	0.9039	0.7657
0.0243	141.0002	24708	4.1213	0.9033	0.7616
0.0212	142.0002	24882	3.6288	0.8978	0.7558
0.0215	143.0002	25056	3.3427	0.9033	0.7558
0.0243	144.0002	25230	3.2026	0.9048	0.7630

Framework versions

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