ht-stmini-cls-v6_ftis_noPretrain-gtsp-m1drp0.0trp0.0

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

• Loss: 4.4060
• Accuracy: 0.8929
• Macro F1: 0.7414

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
19.9958	0.0013	174	25.5174	0.0448	0.0328
9.0928	1.0013	348	59.1623	0.2134	0.0800
3.8475	2.0013	522	102.7102	0.4798	0.1244
3.3121	3.0013	696	84.3994	0.5223	0.1308
2.9094	4.0013	870	89.3619	0.5615	0.1361
2.6139	5.0013	1044	61.8611	0.5858	0.1460
2.1768	6.0012	1218	43.5596	0.5704	0.1476
2.0359	7.0012	1392	34.0368	0.6067	0.1562
1.6432	8.0012	1566	25.2031	0.6183	0.1598
1.6239	9.0012	1740	17.7458	0.6001	0.1620
1.5318	10.0012	1914	14.1748	0.6228	0.1634
1.3959	11.0012	2088	11.2831	0.6288	0.1830
1.3754	12.0012	2262	8.2148	0.6375	0.1978
1.326	13.0012	2436	7.3165	0.6301	0.2108
1.2529	14.0012	2610	5.8729	0.6471	0.2193
1.263	15.0012	2784	4.6390	0.6674	0.2515
1.1709	16.0012	2958	3.7234	0.6715	0.2507
1.1163	17.0012	3132	3.4561	0.6998	0.2850
1.0908	18.0012	3306	3.4270	0.7057	0.2896
1.089	19.0012	3480	2.7688	0.7130	0.3019
1.0303	20.0011	3654	2.5266	0.7266	0.3278
1.0225	21.0011	3828	2.1207	0.7391	0.3410
0.9721	22.0011	4002	2.3522	0.7427	0.3699
0.9075	23.0011	4176	2.5133	0.7322	0.3799
0.8896	24.0011	4350	1.8887	0.7493	0.3955
0.8802	25.0011	4524	1.9004	0.7607	0.4188
0.8528	26.0011	4698	2.1815	0.7545	0.4161
0.8697	27.0011	4872	2.4849	0.7456	0.4201
0.8201	28.0011	5046	2.3108	0.7690	0.4442
0.8203	29.0011	5220	1.9740	0.7602	0.4417
0.8	30.0011	5394	2.4098	0.7636	0.4522
0.7721	31.0011	5568	2.0346	0.7650	0.4363
0.7545	32.0011	5742	2.1536	0.7615	0.4543
0.7604	33.0010	5916	2.1326	0.7759	0.4762
0.7291	34.0010	6090	2.0523	0.7763	0.4918
0.7208	35.0010	6264	2.3761	0.7856	0.4916
0.7012	36.0010	6438	2.5768	0.7733	0.4749
0.6973	37.0010	6612	2.6120	0.7721	0.4798
0.6787	38.0010	6786	2.7726	0.7891	0.4942
0.659	39.0010	6960	4.0918	0.7846	0.4886
0.6445	40.0010	7134	3.3845	0.7945	0.5200
0.6361	41.0010	7308	3.7149	0.7977	0.5260
0.6193	42.0010	7482	4.3992	0.8050	0.5209
0.6124	43.0010	7656	2.9952	0.8017	0.5341
0.6113	44.0010	7830	4.5391	0.8094	0.5391
0.5983	45.0010	8004	4.8019	0.8012	0.5274
0.5816	46.0010	8178	4.7313	0.8085	0.5455
0.5697	47.0009	8352	4.6318	0.8094	0.5499
0.5659	48.0009	8526	5.8999	0.8044	0.5431
0.5516	49.0009	8700	5.9779	0.8136	0.5576
0.5421	50.0009	8874	6.4143	0.8129	0.5605
0.5414	51.0009	9048	7.5933	0.8122	0.5526
0.5213	52.0009	9222	7.8657	0.8125	0.5680
0.5074	53.0009	9396	7.6091	0.8166	0.5684
0.5041	54.0009	9570	7.9929	0.8221	0.5799
0.5018	55.0009	9744	9.0274	0.8245	0.5757
0.5	56.0009	9918	8.6587	0.8306	0.5870
0.4869	57.0009	10092	7.6765	0.8275	0.5901
0.4693	58.0009	10266	8.8191	0.8326	0.5965
0.4627	59.0009	10440	10.0637	0.8315	0.6004
0.4596	60.0008	10614	8.1679	0.8370	0.6042
0.4518	61.0008	10788	8.4349	0.8359	0.6036
0.4385	62.0008	10962	9.2130	0.8369	0.6038
0.4378	63.0008	11136	9.3141	0.8403	0.6101
0.4336	64.0008	11310	9.0886	0.8385	0.6147
0.4221	65.0008	11484	8.8874	0.8412	0.6167
0.418	66.0008	11658	8.3751	0.8392	0.6154
0.4143	67.0008	11832	8.4800	0.8473	0.6248
0.4222	68.0008	12006	8.5978	0.8522	0.6383
0.3991	69.0008	12180	8.9944	0.8506	0.6304
0.4036	70.0008	12354	10.3317	0.8483	0.6353
0.401	71.0008	12528	8.8000	0.8547	0.6419
0.3967	72.0008	12702	10.1595	0.8592	0.6489
0.3919	73.0007	12876	7.7002	0.8537	0.6402
0.3801	74.0007	13050	7.3951	0.8569	0.6521
0.3841	75.0007	13224	8.1720	0.8530	0.6475
0.3697	76.0007	13398	7.3588	0.8532	0.6478
0.3756	77.0007	13572	7.8917	0.8565	0.6525
0.376	78.0007	13746	6.6505	0.8564	0.6524
0.3696	79.0007	13920	7.3225	0.8669	0.6651
0.3655	80.0007	14094	7.7997	0.8628	0.6598
0.3671	81.0007	14268	6.8883	0.8652	0.6684
0.3511	82.0007	14442	7.2238	0.8610	0.6628
0.3467	83.0007	14616	7.2216	0.8718	0.6778
0.3451	84.0007	14790	5.8581	0.8646	0.6654
0.3479	85.0007	14964	6.7665	0.8710	0.6779
0.3427	86.0007	15138	6.9367	0.8635	0.6597
0.3357	87.0006	15312	6.8716	0.8669	0.6714
0.3347	88.0006	15486	7.3971	0.8720	0.6845
0.3363	89.0006	15660	7.4414	0.8675	0.6734
0.3349	90.0006	15834	7.6525	0.8723	0.6831
0.3328	91.0006	16008	7.3987	0.8677	0.6816
0.339	92.0006	16182	6.7756	0.8691	0.6776
0.3244	93.0006	16356	7.5267	0.8681	0.6821
0.3246	94.0006	16530	6.1880	0.8721	0.6851
0.3181	95.0006	16704	6.6994	0.8756	0.6883
0.3209	96.0006	16878	6.3240	0.8685	0.6768
0.3266	97.0006	17052	6.5154	0.8770	0.6915
0.3167	98.0006	17226	5.9771	0.8722	0.6943
0.3194	99.0006	17400	5.5076	0.8760	0.6924
0.3201	100.0005	17574	5.8461	0.8798	0.7009
0.3167	101.0005	17748	6.4426	0.8754	0.7015
0.3087	102.0005	17922	6.5166	0.8750	0.6961
0.3146	103.0005	18096	5.5320	0.8811	0.7013
0.3088	104.0005	18270	5.7319	0.8779	0.6971
0.3116	105.0005	18444	5.8537	0.8770	0.7026
0.3117	106.0005	18618	5.7541	0.8801	0.7103
0.3032	107.0005	18792	5.9563	0.8817	0.7082
0.3051	108.0005	18966	5.4975	0.8825	0.7052
0.3022	109.0005	19140	4.7743	0.8846	0.7058
0.3025	110.0005	19314	5.4460	0.8819	0.7073
0.2993	111.0005	19488	5.4186	0.8780	0.7033
0.3009	112.0005	19662	5.1009	0.8819	0.7075
0.2974	113.0005	19836	6.2867	0.8826	0.7066
0.2926	114.0004	20010	5.3547	0.8857	0.7144
0.2917	115.0004	20184	4.6978	0.8822	0.7113
0.2961	116.0004	20358	5.6941	0.8848	0.7138
0.2946	117.0004	20532	5.7652	0.8830	0.7154
0.2855	118.0004	20706	5.2779	0.8810	0.7127
0.294	119.0004	20880	5.4494	0.8830	0.7174
0.2949	120.0004	21054	4.9102	0.8863	0.7207
0.2953	121.0004	21228	5.4646	0.8819	0.7089
0.2885	122.0004	21402	5.2363	0.8852	0.7265
0.2953	123.0004	21576	5.3327	0.8848	0.7209
0.2905	124.0004	21750	4.6292	0.8860	0.7214
0.2872	125.0004	21924	4.7491	0.8840	0.7196
0.2883	126.0004	22098	5.4074	0.8891	0.7215
0.2867	127.0003	22272	5.2755	0.8835	0.7228
0.2914	128.0003	22446	4.4630	0.8900	0.7268
0.289	129.0003	22620	4.6869	0.8842	0.7226
0.2873	130.0003	22794	4.5551	0.8854	0.7192
0.2829	131.0003	22968	4.4850	0.8913	0.7287
0.2829	132.0003	23142	4.1853	0.8880	0.7264
0.2836	133.0003	23316	4.4679	0.8886	0.7237
0.282	134.0003	23490	5.3415	0.8836	0.7225
0.278	135.0003	23664	5.0775	0.8909	0.7267
0.2784	136.0003	23838	5.0476	0.8917	0.7322
0.2797	137.0003	24012	4.5069	0.8936	0.7322
0.2758	138.0003	24186	4.3858	0.8898	0.7293
0.2819	139.0003	24360	5.0124	0.8890	0.7335
0.2821	140.0003	24534	4.4299	0.8929	0.7414
0.2752	141.0002	24708	4.6434	0.8898	0.7315
0.2763	142.0002	24882	4.8096	0.8912	0.7262
0.2743	143.0002	25056	4.8188	0.8923	0.7330
0.2756	144.0002	25230	4.1003	0.8911	0.7322
0.2754	145.0002	25404	4.1141	0.8877	0.7264
0.2701	146.0002	25578	4.1455	0.8923	0.7350
0.2701	147.0002	25752	4.2590	0.8955	0.7348
0.2706	148.0002	25926	4.9214	0.8918	0.7321
0.2689	149.0002	26100	4.4422	0.8938	0.7345
0.2709	150.0002	26274	4.4412	0.8848	0.7243
0.2714	151.0002	26448	3.6271	0.8878	0.7280
0.2708	152.0002	26622	4.6959	0.8954	0.7410
0.268	153.0002	26796	4.9066	0.8936	0.7398
0.2655	154.0001	26970	4.6026	0.8961	0.7385
0.2643	155.0001	27144	4.3143	0.8950	0.7380
0.2661	156.0001	27318	4.2562	0.8916	0.7354
0.2684	157.0001	27492	4.1381	0.8909	0.7354
0.2605	158.0001	27666	4.0520	0.8945	0.7373
0.2606	159.0001	27840	4.4116	0.8917	0.7390
0.2623	160.0001	28014	3.8049	0.8912	0.7342

Framework versions

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