SignBart-WLASL-300

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

• Loss: 0.9651
• Accuracy: 0.785
• Precision: 0.7465
• Recall: 0.785

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: 2e-05
• train_batch_size: 256
• eval_batch_size: 256
• 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: cosine
• num_epochs: 1000

Training results

Training Loss	Epoch	Step	Accuracy	Validation Loss	Precision	Recall
5.7718	1.0	17	0.0067	5.7282	0.0004	0.0067
5.6955	2.0	34	0.0033	5.6754	0.0001	0.0033
5.6077	3.0	51	0.005	5.5546	0.0003	0.005
5.4955	4.0	68	0.0117	5.4315	0.0010	0.0117
5.3592	5.0	85	0.0083	5.3055	0.0004	0.0083
5.2644	6.0	102	0.0317	5.1698	0.0095	0.0317
5.1057	7.0	119	0.05	5.0218	0.0319	0.05
5.0073	8.0	136	0.065	4.8797	0.0362	0.065
4.8617	9.0	153	0.0783	4.7455	0.0418	0.0783
4.7956	10.0	170	0.1083	4.6018	0.0644	0.1083
4.675	11.0	187	0.13	4.4889	0.0760	0.13
4.5547	12.0	204	0.1717	4.3711	0.0962	0.1717
4.4234	13.0	221	0.22	4.2526	0.1269	0.22
4.3897	14.0	238	0.2367	4.1556	0.1548	0.2367
4.1534	15.0	255	0.2833	4.0489	0.1746	0.2833
4.1545	16.0	272	0.2517	3.9673	0.1660	0.2517
4.1049	17.0	289	0.32	3.8363	0.2142	0.32
3.8591	18.0	306	0.3383	3.7406	0.2265	0.3383
3.884	19.0	323	0.34	3.6559	0.2461	0.34
3.8226	20.0	340	0.3683	3.5522	0.2731	0.3683
3.7225	21.0	357	0.39	3.4836	0.2909	0.39
3.4983	22.0	374	0.4	3.3859	0.2879	0.4
3.5758	23.0	391	0.4217	3.3075	0.3171	0.4217
3.4614	24.0	408	0.4283	3.2179	0.3154	0.4283
3.4703	25.0	425	0.4583	3.1577	0.3592	0.4583
3.3732	26.0	442	0.4383	3.0920	0.3232	0.4383
3.3544	27.0	459	0.4483	3.0366	0.3458	0.4483
3.0549	28.0	476	0.4817	2.9235	0.3756	0.4817
3.0809	29.0	493	0.48	2.8708	0.3780	0.48
2.9662	30.0	510	0.5017	2.7895	0.4013	0.5017
2.855	31.0	527	0.5217	2.7227	0.4279	0.5217
3.1957	32.0	544	0.505	2.6715	0.4102	0.505
2.7941	33.0	561	0.5183	2.6059	0.4218	0.5183
2.6172	34.0	578	0.5317	2.5320	0.4380	0.5317
2.7093	35.0	595	0.535	2.4864	0.4312	0.535
2.4898	36.0	612	0.5633	2.4015	0.4586	0.5633
2.5333	37.0	629	0.535	2.3635	0.4341	0.535
2.673	38.0	646	0.575	2.2933	0.4807	0.575
2.474	39.0	663	0.5767	2.2527	0.4745	0.5767
2.4882	40.0	680	0.5683	2.2024	0.4716	0.5683
2.29	41.0	697	0.5967	2.1558	0.5034	0.5967
2.4192	42.0	714	0.5933	2.1114	0.5002	0.5933
2.5482	43.0	731	0.5967	2.0733	0.4935	0.5967
2.3896	44.0	748	0.6133	2.0100	0.5276	0.6133
2.2779	45.0	765	0.5867	1.9905	0.4992	0.5867
2.1224	46.0	782	0.6367	1.9394	0.5563	0.6367
2.1622	47.0	799	0.62	1.8911	0.5379	0.62
1.9595	48.0	816	0.6283	1.8558	0.5392	0.6283
2.2248	49.0	833	0.6433	1.8164	0.5624	0.6433
2.0088	50.0	850	0.635	1.7991	0.5599	0.635
2.2999	51.0	867	0.6267	1.7537	0.5499	0.6267
2.0076	52.0	884	0.6583	1.7174	0.5900	0.6583
1.8275	53.0	901	0.65	1.6969	0.5755	0.65
1.8442	54.0	918	0.6717	1.6591	0.6079	0.6717
1.9328	55.0	935	0.67	1.6218	0.6033	0.67
1.6616	56.0	952	0.665	1.5984	0.5955	0.665
1.8712	57.0	969	0.66	1.5906	0.5922	0.66
1.8647	58.0	986	0.6633	1.5589	0.6003	0.6633
1.6898	59.0	1003	0.65	1.6114	0.5822	0.65
1.7192	60.0	1020	0.67	1.5186	0.6077	0.67
1.7742	61.0	1037	0.68	1.5255	0.6218	0.68
1.8068	62.0	1054	0.7017	1.4744	0.6395	0.7017
1.8291	63.0	1071	0.6967	1.4518	0.6370	0.6967
1.7829	64.0	1088	0.71	1.4397	0.6553	0.71
1.4763	65.0	1105	0.7083	1.4246	0.6457	0.7083
1.8288	66.0	1122	0.6933	1.4138	0.6355	0.6933
1.8425	67.0	1139	0.72	1.3842	0.6737	0.72
1.4383	68.0	1156	0.7033	1.3605	0.6375	0.7033
1.647	69.0	1173	0.7283	1.3421	0.6651	0.7283
1.5617	70.0	1190	0.7217	1.3300	0.6641	0.7217
1.3654	71.0	1207	0.7183	1.3078	0.6631	0.7183
1.3424	72.0	1224	0.715	1.2971	0.6632	0.715
1.3684	73.0	1241	0.735	1.2699	0.6816	0.735
1.3537	74.0	1258	0.7383	1.2655	0.6839	0.7383
1.5039	75.0	1275	0.7367	1.2530	0.6923	0.7367
1.3729	76.0	1292	0.715	1.2332	0.6677	0.715
1.291	77.0	1309	0.7317	1.2334	0.6763	0.7317
1.2466	78.0	1326	0.745	1.2081	0.7004	0.745
1.5836	79.0	1343	0.7217	1.2080	0.6622	0.7217
1.7043	80.0	1360	0.7333	1.1704	0.6881	0.7333
1.3191	81.0	1377	0.7433	1.1777	0.6960	0.7433
1.2723	82.0	1394	0.7433	1.1591	0.6859	0.7433
1.5445	83.0	1411	0.7633	1.1395	0.7151	0.7633
1.3321	84.0	1428	0.76	1.1584	0.7225	0.76
1.3126	85.0	1445	0.7417	1.1845	0.6845	0.7417
1.1269	86.0	1462	0.735	1.1707	0.6831	0.735
1.5785	87.0	1479	0.7283	1.1432	0.6838	0.7283
1.2552	88.0	1496	0.7267	1.1437	0.6718	0.7267
1.0932	89.0	1513	1.1219	0.7533	0.7077	0.7533
1.3984	90.0	1530	1.1110	0.7433	0.6960	0.7433
1.408	91.0	1547	1.1082	0.7617	0.7164	0.7617
1.2296	92.0	1564	1.1055	0.74	0.6989	0.74
1.2381	93.0	1581	1.1077	0.7483	0.7045	0.7483
1.489	94.0	1598	1.1064	0.7433	0.6992	0.7433
1.4369	95.0	1615	1.0628	0.7533	0.7072	0.7533
1.2974	96.0	1632	1.0806	0.77	0.7303	0.77
1.1192	97.0	1649	1.0643	0.755	0.7184	0.755
1.1608	98.0	1666	1.0573	0.7633	0.7271	0.7633
0.9436	99.0	1683	1.0878	0.7617	0.7130	0.7617
1.1929	100.0	1700	1.0632	0.76	0.7237	0.76
1.082	101.0	1717	1.0523	0.7417	0.7001	0.7417
1.0668	102.0	1734	1.0365	0.75	0.7066	0.75
1.2324	103.0	1751	1.0778	0.7667	0.7191	0.7667
1.1573	104.0	1768	1.0174	0.765	0.7265	0.765
0.9963	105.0	1785	1.0099	0.7917	0.7519	0.7917
0.9011	106.0	1802	1.0014	0.7783	0.7304	0.7783
1.3193	107.0	1819	1.0096	0.7683	0.7364	0.7683
0.9542	108.0	1836	1.0233	0.7717	0.7317	0.7717
1.0171	109.0	1853	0.9941	0.7767	0.7429	0.7767
1.1018	110.0	1870	0.9953	0.7917	0.7634	0.7917
1.2737	111.0	1887	0.9910	0.7917	0.7565	0.7917
1.1907	112.0	1904	0.9899	0.78	0.7469	0.78
0.8852	113.0	1921	1.0077	0.7817	0.7471	0.7817
0.8181	114.0	1938	1.0278	0.7767	0.7465	0.7767
1.0496	115.0	1955	0.9822	0.7817	0.7508	0.7817
0.9526	116.0	1972	0.9816	0.7833	0.7515	0.7833
1.0634	117.0	1989	0.9972	0.7733	0.7316	0.7733
0.9541	118.0	2006	0.9709	0.7867	0.7540	0.7867
1.0232	119.0	2023	0.9822	0.7833	0.7516	0.7833
1.0437	120.0	2040	0.9658	0.7817	0.7411	0.7817
0.7996	121.0	2057	0.9629	0.7883	0.7557	0.7883
0.8277	122.0	2074	0.9565	0.775	0.7430	0.775
1.0182	123.0	2091	0.9729	0.7867	0.7500	0.7867
0.7932	124.0	2108	0.9586	0.79	0.7457	0.79
1.1399	125.0	2125	0.9782	0.785	0.7528	0.785
0.7441	126.0	2142	0.9470	0.7867	0.755	0.7867
0.8847	127.0	2159	0.9526	0.7817	0.7489	0.7817
0.978	128.0	2176	0.9600	0.7817	0.7385	0.7817
0.7465	129.0	2193	0.9367	0.7833	0.7516	0.7833
0.9107	130.0	2210	0.9397	0.7817	0.7428	0.7817
1.0219	131.0	2227	0.9556	0.78	0.7495	0.78
0.8883	132.0	2244	0.9778	0.775	0.7428	0.775
1.1886	133.0	2261	0.9555	0.79	0.7453	0.79
0.7394	134.0	2278	0.9651	0.785	0.7465	0.785

Framework versions

• Transformers 4.47.0
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0