ht-stmini-cls-v6_ftis_noPretrain-cssl-nps1nns3

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

• Loss: 18.9225
• Accuracy: 0.9405
• Macro F1: 0.8528

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: 134675

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Macro F1
26.8647	0.0015	202	55.4966	0.0225	0.0155
9.2483	1.0015	404	121.0748	0.3563	0.0997
6.6701	2.0015	606	153.3219	0.5231	0.1305
5.4916	3.0015	808	131.9640	0.5588	0.1404
4.7168	4.0015	1010	88.1905	0.5915	0.1473
3.9645	5.0015	1212	68.1442	0.6104	0.1592
3.4277	6.0015	1414	49.7885	0.6007	0.1553
3.1421	7.0015	1616	35.4737	0.6270	0.1692
2.9329	8.0015	1818	31.9072	0.6212	0.1737
2.7947	9.0015	2020	22.8596	0.6235	0.1857
2.5925	10.0015	2222	22.2309	0.6278	0.1877
2.4856	11.0015	2424	17.1074	0.6461	0.2075
2.4777	12.0015	2626	14.7842	0.6552	0.2285
2.3333	13.0015	2828	15.2059	0.6586	0.2431
2.2437	14.0015	3030	13.8892	0.6693	0.2707
2.1221	15.0015	3232	14.1571	0.6917	0.2912
1.9194	16.0015	3434	12.4244	0.7036	0.3182
1.9427	17.0015	3636	12.1604	0.7193	0.3340
1.8698	18.0015	3838	10.2386	0.7326	0.3615
1.7689	19.0015	4040	12.6714	0.7423	0.3859
1.5606	20.0015	4242	11.5203	0.7450	0.3894
1.5834	21.0015	4444	11.3107	0.7765	0.4315
1.3881	22.0015	4646	13.5044	0.7732	0.4383
1.347	23.0015	4848	11.7744	0.7769	0.4513
1.1804	24.0015	5050	11.7618	0.7867	0.4718
1.3527	25.0015	5252	13.5098	0.7954	0.4797
1.2322	26.0015	5454	13.4435	0.8068	0.5028
1.0932	27.0015	5656	13.2345	0.8073	0.5004
1.0589	28.0015	5858	14.5219	0.8104	0.5087
1.0667	29.0015	6060	17.0437	0.8129	0.5314
1.0117	30.0015	6262	18.5900	0.7988	0.5150
0.9144	31.0015	6464	19.0106	0.8271	0.5745
0.9311	32.0015	6666	16.4530	0.8245	0.5540
0.8036	33.0015	6868	17.4781	0.8476	0.5882
0.8184	34.0015	7070	20.5845	0.8434	0.5854
0.7435	35.0015	7272	21.0411	0.8499	0.6143
0.6798	36.0015	7474	22.5522	0.8611	0.6409
0.7025	37.0015	7676	21.6592	0.8739	0.6514
0.626	38.0015	7878	24.2972	0.8721	0.6610
0.5851	39.0015	8080	24.1912	0.8742	0.6562
0.5403	40.0015	8282	22.8676	0.8721	0.6645
0.5437	41.0015	8484	23.8102	0.8757	0.6587
0.5141	42.0015	8686	26.6472	0.8860	0.6830
0.474	43.0015	8888	23.6877	0.8748	0.6749
0.4614	44.0015	9090	27.1110	0.8920	0.6960
0.3745	45.0015	9292	29.8287	0.8826	0.6733
0.3788	46.0015	9494	23.6845	0.8902	0.7008
0.3729	47.0015	9696	28.5393	0.8855	0.6929
0.3247	48.0015	9898	21.9949	0.8951	0.7127
0.3231	49.0015	10100	27.8040	0.8968	0.7088
0.2806	50.0015	10302	22.4685	0.8979	0.7250
0.2667	51.0015	10504	21.8615	0.9035	0.7282
0.2871	52.0015	10706	20.7951	0.8947	0.7139
0.3013	53.0015	10908	23.8296	0.9062	0.7352
0.2336	54.0015	11110	20.9612	0.9021	0.7354
0.224	55.0015	11312	25.4931	0.9066	0.7334
0.2177	56.0015	11514	27.1630	0.9054	0.7366
0.1938	57.0015	11716	21.6787	0.9056	0.7261
0.2128	58.0015	11918	20.1527	0.9064	0.7386
0.1769	59.0015	12120	22.9292	0.9031	0.7476
0.1967	60.0015	12322	20.8235	0.9160	0.7579
0.1858	61.0015	12524	19.2070	0.9160	0.7610
0.1639	62.0015	12726	18.2907	0.9146	0.7611
0.162	63.0015	12928	17.4853	0.9168	0.7624
0.1424	64.0015	13130	19.0293	0.9168	0.7664
0.1433	65.0015	13332	17.5929	0.9167	0.7709
0.1287	66.0015	13534	18.2924	0.9155	0.7707
0.1538	67.0015	13736	16.9960	0.9201	0.7738
0.131	68.0015	13938	15.4065	0.9165	0.7661
0.1306	69.0015	14140	16.7293	0.9143	0.7686
0.1173	70.0015	14342	16.8151	0.9142	0.7726
0.1025	71.0015	14544	17.5942	0.9188	0.7773
0.1044	72.0015	14746	17.2179	0.9211	0.7805
0.1197	73.0015	14948	15.8603	0.9225	0.7910
0.1055	74.0015	15150	13.0298	0.9223	0.7835
0.1119	75.0015	15352	15.4385	0.9247	0.7864
0.1062	76.0015	15554	14.2398	0.9263	0.7962
0.0906	77.0015	15756	17.8967	0.9251	0.7918
0.0907	78.0015	15958	16.7705	0.9210	0.7922
0.0977	79.0015	16160	15.2437	0.9278	0.7953
0.0885	80.0015	16362	15.5348	0.9258	0.7951
0.0835	81.0015	16564	15.5831	0.9263	0.7952
0.0756	82.0015	16766	15.3867	0.9285	0.8037
0.0811	83.0015	16968	13.2699	0.9263	0.8024
0.0702	84.0015	17170	12.5821	0.9238	0.7844
0.074	85.0015	17372	13.3951	0.9310	0.8086
0.0682	86.0015	17574	12.8296	0.9276	0.7985
0.0743	87.0015	17776	12.4840	0.9321	0.8129
0.0682	88.0015	17978	13.8928	0.9288	0.8049
0.0639	89.0015	18180	17.4799	0.9313	0.8146
0.0745	90.0015	18382	15.3871	0.9338	0.8214
0.0561	91.0015	18584	14.8037	0.9325	0.8100
0.0626	92.0015	18786	16.0367	0.9321	0.7977
0.0561	93.0015	18988	15.5910	0.9294	0.8142
0.0537	94.0015	19190	15.4496	0.9341	0.8174
0.0574	95.0015	19392	16.8282	0.9322	0.8142
0.052	96.0015	19594	18.6193	0.9338	0.8190
0.0514	97.0015	19796	17.9728	0.9372	0.8245
0.0626	98.0015	19998	17.0515	0.9355	0.8255
0.061	99.0015	20200	17.2241	0.9320	0.7921
0.0487	100.0015	20402	14.3454	0.9306	0.8201
0.0499	101.0015	20604	14.5126	0.9365	0.8300
0.0489	102.0015	20806	16.4347	0.9314	0.8219
0.0465	103.0015	21008	15.5455	0.9338	0.8291
0.0433	104.0015	21210	13.5703	0.9361	0.8081
0.0468	105.0015	21412	15.9294	0.9378	0.8124
0.0475	106.0015	21614	15.1766	0.9365	0.8245
0.0406	107.0015	21816	15.3008	0.9340	0.8040
0.0498	108.0015	22018	15.7133	0.9385	0.8172
0.0438	109.0015	22220	17.0641	0.9349	0.8127
0.0372	110.0015	22422	14.7757	0.9340	0.8159
0.0398	111.0015	22624	14.4051	0.9374	0.8137
0.045	112.0015	22826	17.2541	0.9391	0.8169
0.0464	113.0015	23028	15.3776	0.9353	0.8065
0.0366	114.0015	23230	16.5125	0.9365	0.8133
0.0405	115.0015	23432	18.0453	0.9383	0.8148
0.0482	116.0015	23634	16.2310	0.9366	0.8330
0.0357	117.0015	23836	17.8717	0.9336	0.8092
0.0373	118.0015	24038	18.1503	0.9371	0.8373
0.0319	119.0015	24240	16.6249	0.9361	0.8335
0.041	120.0015	24442	19.1009	0.9374	0.8174
0.037	121.0015	24644	17.4892	0.9338	0.8136
0.0339	122.0015	24846	18.7711	0.9397	0.8415
0.0303	123.0015	25048	15.1847	0.9350	0.8324
0.0352	124.0015	25250	17.2595	0.9380	0.8222
0.0306	125.0015	25452	17.8348	0.9351	0.8191
0.0345	126.0015	25654	18.9903	0.9336	0.8118
0.0287	127.0015	25856	20.0766	0.9378	0.8221
0.0288	128.0015	26058	17.1600	0.9386	0.8261
0.0284	129.0015	26260	18.2565	0.9360	0.8240
0.0319	130.0015	26462	18.3116	0.9382	0.8136
0.0327	131.0015	26664	15.9438	0.9374	0.8161
0.0267	132.0015	26866	16.7417	0.9372	0.8379
0.0246	133.0015	27068	21.2340	0.9391	0.8195
0.0293	134.0015	27270	17.1422	0.9389	0.8389
0.0294	135.0015	27472	15.6383	0.9412	0.8213
0.0263	136.0015	27674	18.0431	0.9416	0.8259
0.0292	137.0015	27876	15.9478	0.9401	0.8444
0.0237	138.0015	28078	14.9965	0.9406	0.8286
0.0246	139.0015	28280	20.8328	0.9425	0.8336
0.0237	140.0015	28482	20.1271	0.9389	0.8279
0.027	141.0015	28684	17.0921	0.9373	0.8263
0.0263	142.0015	28886	18.7980	0.9338	0.8154
0.0292	143.0015	29088	17.2120	0.9423	0.8522
0.0239	144.0015	29290	19.3687	0.9405	0.8465
0.0259	145.0015	29492	17.4470	0.9391	0.8274
0.0229	146.0015	29694	17.4125	0.9403	0.8437
0.0317	147.0015	29896	19.9252	0.9412	0.8285
0.0218	148.0015	30098	14.9289	0.9420	0.8275
0.0237	149.0015	30300	18.3929	0.9414	0.8341
0.0234	150.0015	30502	19.4984	0.9385	0.8448
0.0223	151.0015	30704	19.4938	0.9405	0.8528
0.0221	152.0015	30906	18.3872	0.9398	0.8254
0.0219	153.0015	31108	16.8159	0.9380	0.8278
0.0231	154.0015	31310	18.0745	0.9382	0.8386
0.0219	155.0015	31512	18.9114	0.9446	0.8354
0.0193	156.0015	31714	15.2099	0.9412	0.8462
0.0205	157.0015	31916	22.1138	0.9403	0.8500
0.02	158.0015	32118	18.1379	0.9426	0.8338
0.0197	159.0015	32320	18.6837	0.9424	0.8346
0.0207	160.0015	32522	20.5484	0.9403	0.8332
0.0197	161.0015	32724	21.4245	0.9351	0.8239
0.0227	162.0015	32926	17.5369	0.9427	0.8302
0.0208	163.0015	33128	16.4532	0.9430	0.8381
0.0244	164.0015	33330	18.2460	0.9403	0.8296
0.0206	165.0015	33532	19.0137	0.9429	0.8431
0.0179	166.0015	33734	18.3660	0.9337	0.8493
0.0198	167.0015	33936	17.4318	0.9428	0.8361
0.0191	168.0015	34138	18.2683	0.9422	0.8375
0.0193	169.0015	34340	18.6608	0.9405	0.8358
0.0233	170.0015	34542	15.7378	0.9438	0.8522
0.023	171.0015	34744	17.3552	0.9416	0.8356

Framework versions

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