SignBart-ASL-400

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

• Loss: 0.8715
• Accuracy: 0.7804
• Precision: 0.7968
• Recall: 0.7804

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: 0.0002
• train_batch_size: 256
• eval_batch_size: 256
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• num_epochs: 1000

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall
6.0768	1.0	35	5.9977	0.0034	0.0001	0.0034
6.0349	2.0	70	5.9374	0.0046	0.0004	0.0046
5.9294	3.0	105	5.6893	0.0154	0.0034	0.0154
5.6867	4.0	140	5.3946	0.0212	0.0094	0.0212
5.4328	5.0	175	5.1652	0.0298	0.0124	0.0298
5.2349	6.0	210	4.9791	0.0588	0.0395	0.0588
5.0643	7.0	245	4.8018	0.0742	0.0398	0.0742
4.9482	8.0	280	4.6428	0.1032	0.0799	0.1032
4.8503	9.0	315	4.4834	0.1393	0.0978	0.1393
4.6618	10.0	350	4.3149	0.1723	0.1382	0.1723
4.5032	11.0	385	4.1480	0.2108	0.2118	0.2108
4.378	12.0	420	4.0017	0.2533	0.2448	0.2533
4.313	13.0	455	3.8613	0.2794	0.2851	0.2794
4.2141	14.0	490	3.7279	0.3126	0.3257	0.3126
4.0236	15.0	525	3.5737	0.3448	0.3708	0.3448
4.0748	16.0	560	3.4735	0.3638	0.3895	0.3638
3.8336	17.0	595	3.3687	0.3763	0.3983	0.3763
3.7442	18.0	630	3.2452	0.4139	0.4401	0.4139
3.6543	19.0	665	3.1658	0.4131	0.4391	0.4131
3.5757	20.0	700	3.0657	0.4412	0.4715	0.4412
3.5742	21.0	735	2.9638	0.4561	0.4941	0.4561
3.4422	22.0	770	2.8810	0.4741	0.5117	0.4741
3.3174	23.0	805	2.7804	0.4868	0.5277	0.4868
3.1678	24.0	840	2.7162	0.5002	0.5339	0.5002
3.3057	25.0	875	2.6508	0.5010	0.5425	0.5010
3.0912	26.0	910	2.5528	0.5266	0.5597	0.5266
3.0476	27.0	945	2.4603	0.5566	0.5920	0.5566
2.9213	28.0	980	2.4284	0.5564	0.5957	0.5564
2.9069	29.0	1015	2.3612	0.5612	0.5947	0.5612
2.7931	30.0	1050	2.2954	0.5776	0.6164	0.5776
2.9095	31.0	1085	2.2388	0.5764	0.6103	0.5764
2.8603	32.0	1120	2.1720	0.5903	0.6279	0.5903
2.7008	33.0	1155	2.1129	0.5993	0.6244	0.5993
2.7811	34.0	1190	2.0643	0.6042	0.6385	0.6042
2.6529	35.0	1225	2.0183	0.6069	0.6361	0.6069
2.5363	36.0	1260	1.9997	0.6054	0.6484	0.6054
2.4029	37.0	1295	1.9467	0.6130	0.6404	0.6130
2.4787	38.0	1330	1.8683	0.6379	0.6574	0.6379
2.5173	39.0	1365	1.8616	0.6327	0.6602	0.6327
2.6548	40.0	1400	1.8232	0.6349	0.6638	0.6349
2.3104	41.0	1435	1.7686	0.6464	0.6701	0.6464
2.3891	42.0	1470	1.7388	0.6528	0.6840	0.6528
2.4407	43.0	1505	1.7080	0.6506	0.6763	0.6506
2.2347	44.0	1540	1.7102	0.6398	0.6742	0.6398
2.2116	45.0	1575	1.6484	0.6672	0.6908	0.6672
2.0694	46.0	1610	1.6012	0.6684	0.6917	0.6684
2.0869	47.0	1645	1.5758	0.6696	0.6900	0.6696
2.1581	48.0	1680	1.5490	0.6837	0.7098	0.6837
2.0666	49.0	1715	1.5222	0.6776	0.7003	0.6776
1.9592	50.0	1750	1.4896	0.6881	0.7135	0.6881
2.1065	51.0	1785	1.5038	0.6708	0.6935	0.6708
1.8903	52.0	1820	1.4422	0.6935	0.7182	0.6935
1.8887	53.0	1855	1.4312	0.6886	0.7062	0.6886
2.028	54.0	1890	1.3897	0.7020	0.7218	0.7020
1.937	55.0	1925	1.3826	0.6862	0.7075	0.6862
1.9876	56.0	1960	1.3687	0.6967	0.7185	0.6967
1.8689	57.0	1995	1.3580	0.6950	0.7192	0.6950
1.7962	58.0	2030	1.3743	0.6901	0.7156	0.6901
1.8528	59.0	2065	1.3034	0.7047	0.7323	0.7047
1.7134	60.0	2100	1.2906	0.7128	0.7316	0.7128
2.0335	61.0	2135	1.3115	0.7011	0.7241	0.7011
1.8349	62.0	2170	1.2628	0.7133	0.7368	0.7133
1.6479	63.0	2205	1.2390	0.7174	0.7385	0.7174
1.6653	64.0	2240	1.2174	0.7206	0.7364	0.7206
1.6352	65.0	2275	1.1998	0.7230	0.7418	0.7230
2.1608	66.0	2310	1.1977	0.7247	0.7449	0.7247
1.6582	67.0	2345	1.2085	0.7174	0.7425	0.7174
1.6359	68.0	2380	1.1698	0.7252	0.7462	0.7252
1.6276	69.0	2415	1.1569	0.7282	0.7459	0.7282
1.6379	70.0	2450	1.1647	0.7262	0.7458	0.7262
1.6273	71.0	2485	1.1557	0.7269	0.7490	0.7269
1.6997	72.0	2520	1.1177	0.7304	0.7473	0.7304
1.3425	73.0	2555	1.1140	0.7418	0.7602	0.7418
1.3719	74.0	2590	1.0945	0.7399	0.7522	0.7399
1.6558	75.0	2625	1.0909	0.7457	0.7651	0.7457
1.3146	76.0	2660	1.0984	0.7399	0.7606	0.7399
1.4204	77.0	2695	1.0874	0.7347	0.7523	0.7347
1.4834	78.0	2730	1.0677	0.7430	0.7604	0.7430
1.318	79.0	2765	1.0695	0.7367	0.7563	0.7367
1.5943	80.0	2800	1.0558	0.7428	0.7638	0.7428
1.4331	81.0	2835	1.0573	0.7396	0.7588	0.7396
1.7275	82.0	2870	1.0503	0.7457	0.7649	0.7457
1.4237	83.0	2905	1.0315	0.7491	0.7676	0.7491
1.348	84.0	2940	1.0464	0.7413	0.7632	0.7413
1.4183	85.0	2975	1.0038	0.7543	0.7710	0.7543
1.3382	86.0	3010	1.0062	0.7535	0.7726	0.7535
1.3279	87.0	3045	1.0036	0.7484	0.7659	0.7484
1.2822	88.0	3080	1.0072	0.7501	0.7665	0.7501
1.205	89.0	3115	0.9785	0.7609	0.7743	0.7609
1.1973	90.0	3150	0.9826	0.7521	0.7703	0.7521
1.3498	91.0	3185	0.9904	0.7533	0.7704	0.7533
1.3605	92.0	3220	0.9947	0.7526	0.7682	0.7526
1.4065	93.0	3255	0.9897	0.7562	0.7747	0.7562
1.1482	94.0	3290	0.9745	0.7552	0.7744	0.7552
1.1124	95.0	3325	0.9712	0.7560	0.7751	0.7560
1.2142	96.0	3360	0.9556	0.7599	0.7732	0.7599
1.2644	97.0	3395	0.9534	0.7604	0.7761	0.7604
1.2449	98.0	3430	0.9559	0.7565	0.7730	0.7565
1.42	99.0	3465	0.9568	0.7516	0.7692	0.7516
1.5439	100.0	3500	0.9487	0.7579	0.7741	0.7579
1.2883	101.0	3535	0.9348	0.7628	0.7784	0.7628
1.2313	102.0	3570	0.9478	0.7638	0.7804	0.7638
1.2525	103.0	3605	0.9521	0.7621	0.7801	0.7621
1.0693	104.0	3640	0.9183	0.7672	0.7854	0.7672
1.2183	105.0	3675	0.9488	0.7606	0.7810	0.7606
1.0968	106.0	3710	0.9239	0.7640	0.7790	0.7640
1.146	107.0	3745	0.9285	0.7606	0.7779	0.7606
1.061	108.0	3780	0.9320	0.7618	0.7801	0.7618
1.4022	109.0	3815	0.9174	0.7640	0.7802	0.7640
1.1946	110.0	3850	0.9197	0.7604	0.7747	0.7604
1.0868	111.0	3885	0.9124	0.7662	0.7819	0.7662
0.9491	112.0	3920	0.9253	0.7689	0.7839	0.7689
1.1374	113.0	3955	0.9254	0.7643	0.7797	0.7643
1.1835	114.0	3990	0.9142	0.7653	0.7821	0.7653
1.1594	115.0	4025	0.9279	0.7611	0.7808	0.7611
1.0603	116.0	4060	0.9025	0.7770	0.7927	0.7770
1.174	117.0	4095	0.8895	0.7726	0.7891	0.7726
1.0453	118.0	4130	0.8904	0.7748	0.7892	0.7748
1.2013	119.0	4165	0.9123	0.7672	0.7831	0.7672
1.1358	120.0	4200	0.9235	0.7701	0.7880	0.7701
1.1468	121.0	4235	0.8970	0.7723	0.7871	0.7723
0.9599	122.0	4270	0.8889	0.7738	0.7893	0.7738
0.9898	123.0	4305	0.8958	0.7743	0.7898	0.7743
0.9008	124.0	4340	0.8854	0.7757	0.7901	0.7757
0.9634	125.0	4375	0.8892	0.7775	0.7925	0.7775
1.1803	126.0	4410	0.8921	0.7687	0.7828	0.7687
0.9796	127.0	4445	0.8758	0.7777	0.7911	0.7777
0.979	128.0	4480	0.8849	0.7753	0.7919	0.7753
0.9416	129.0	4515	0.8756	0.7816	0.7974	0.7816
1.1879	130.0	4550	0.8878	0.7755	0.7905	0.7755
0.9852	131.0	4585	0.8923	0.7748	0.7918	0.7748
0.8054	132.0	4620	0.8682	0.7836	0.7980	0.7836
0.8866	133.0	4655	0.8755	0.7777	0.7913	0.7777
1.0883	134.0	4690	0.8875	0.7740	0.7884	0.7740
1.0614	135.0	4725	0.8807	0.7770	0.7930	0.7770
0.9038	136.0	4760	0.8823	0.7831	0.7977	0.7831
0.8942	137.0	4795	0.8715	0.7804	0.7968	0.7804

Framework versions

• Transformers 4.45.1
• Pytorch 2.4.0
• Datasets 3.0.1
• Tokenizers 0.20.0