SignBart-ASL-2730

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

• Loss: 1.0663
• Accuracy: 0.7522
• Precision: 0.7748
• Recall: 0.7522

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
7.9504	1.0	316	7.5910	0.0015	0.0000	0.0015
7.3419	2.0	632	6.8014	0.0120	0.0012	0.0120
6.7466	3.0	948	6.2123	0.0324	0.0085	0.0324
6.2532	4.0	1264	5.7091	0.0574	0.0246	0.0574
5.8427	5.0	1580	5.2449	0.1003	0.0630	0.1003
5.4563	6.0	1896	4.8599	0.1334	0.0990	0.1334
5.1075	7.0	2212	4.4728	0.1887	0.1664	0.1887
4.8626	8.0	2528	4.1315	0.2347	0.2114	0.2347
4.5077	9.0	2844	3.8084	0.2820	0.2686	0.2820
4.3056	10.0	3160	3.5184	0.3342	0.3262	0.3342
4.0527	11.0	3476	3.2573	0.3712	0.3816	0.3712
3.9347	12.0	3792	3.0366	0.4050	0.4086	0.4050
3.6534	13.0	4108	2.7821	0.4473	0.4689	0.4473
3.4758	14.0	4424	2.6075	0.4692	0.4862	0.4692
3.3033	15.0	4740	2.4175	0.5104	0.5303	0.5104
3.1366	16.0	5056	2.2547	0.5277	0.5506	0.5277
2.9912	17.0	5372	2.1465	0.5388	0.5582	0.5388
2.9211	18.0	5688	2.0410	0.5608	0.5852	0.5608
2.8221	19.0	6004	1.9460	0.5741	0.5967	0.5741
2.728	20.0	6320	1.8412	0.5935	0.6158	0.5935
2.6948	21.0	6636	1.7873	0.6009	0.6276	0.6009
2.4883	22.0	6952	1.7102	0.6154	0.6438	0.6154
2.517	23.0	7268	1.6517	0.6253	0.6496	0.6253
2.4333	24.0	7584	1.6176	0.6298	0.6548	0.6298
2.353	25.0	7900	1.5679	0.6401	0.6688	0.6401
2.2862	26.0	8216	1.5282	0.6463	0.6745	0.6463
2.3052	27.0	8532	1.4695	0.6574	0.6824	0.6574
2.1156	28.0	8848	1.4635	0.6595	0.6846	0.6595
2.126	29.0	9164	1.4364	0.6598	0.6859	0.6598
2.1957	30.0	9480	1.4060	0.6706	0.6948	0.6706
2.0033	31.0	9796	1.3580	0.6820	0.7072	0.6820
2.0278	32.0	10112	1.3474	0.6809	0.7038	0.6809
2.0128	33.0	10428	1.3391	0.6837	0.7113	0.6837
2.0176	34.0	10744	1.3162	0.6871	0.7125	0.6871
1.9308	35.0	11060	1.2937	0.6901	0.7150	0.6901
2.0256	36.0	11376	1.2763	0.6960	0.7251	0.6960
1.8052	37.0	11692	1.2545	0.7003	0.7252	0.7003
1.8055	38.0	12008	1.2543	0.6990	0.7256	0.6990
1.8889	39.0	12324	1.2371	0.7026	0.7292	0.7026
1.8205	40.0	12640	1.2237	0.7053	0.7289	0.7053
1.7679	41.0	12956	1.1919	0.7129	0.7368	0.7129
1.7532	42.0	13272	1.1966	0.7101	0.7334	0.7101
1.7701	43.0	13588	1.1882	0.7133	0.7379	0.7133
1.7272	44.0	13904	1.1786	0.7179	0.7425	0.7179
1.6642	45.0	14220	1.1660	0.7199	0.7445	0.7199
1.6463	46.0	14536	1.1609	0.7197	0.7431	0.7197
1.6629	47.0	14852	1.1418	0.7228	0.7481	0.7228
1.6612	48.0	15168	1.1496	0.7228	0.7462	0.7228
1.6032	49.0	15484	1.1317	0.7280	0.7503	0.7280
1.5918	50.0	15800	1.1229	0.7315	0.7571	0.7315
1.5362	51.0	16116	1.1341	0.7269	0.7510	0.7269
1.6448	52.0	16432	1.1116	0.7335	0.7572	0.7335
1.4685	53.0	16748	1.1258	0.7312	0.7549	0.7312
1.6576	54.0	17064	1.1095	0.7327	0.7572	0.7327
1.6566	55.0	17380	1.1104	0.7330	0.7553	0.7330
1.4888	56.0	17696	1.1040	0.7331	0.7535	0.7331
1.4835	57.0	18012	1.0956	0.7387	0.7606	0.7387
1.5224	58.0	18328	1.0980	0.7407	0.7634	0.7407
1.5324	59.0	18644	1.1112	0.7336	0.7577	0.7336
1.5162	60.0	18960	1.0971	0.7354	0.7619	0.7354
1.4942	61.0	19276	1.0899	0.7436	0.7672	0.7436
1.4682	62.0	19592	1.0822	0.7429	0.7671	0.7429
1.5173	63.0	19908	1.0962	0.7403	0.7626	0.7403
1.4293	64.0	20224	1.0856	0.7412	0.7651	0.7412
1.4511	65.0	20540	1.0834	0.7422	0.7671	0.7422
1.5094	66.0	20856	1.0783	0.7457	0.7697	0.7457
1.3379	67.0	21172	1.0804	0.7469	0.7713	0.7469
1.375	68.0	21488	1.0876	0.7438	0.7680	0.7438
1.3573	69.0	21804	1.0713	0.7474	0.7733	0.7474
1.2803	70.0	22120	1.0862	0.7429	0.7681	0.7429
1.331	71.0	22436	1.0800	0.7466	0.7722	0.7466
1.3896	72.0	22752	1.0616	0.7491	0.7732	0.7491
1.3718	73.0	23068	1.0678	0.7513	0.7762	0.7513
1.3099	74.0	23384	1.0672	0.7541	0.7770	0.7541
1.2683	75.0	23700	1.0631	0.7502	0.7743	0.7502
1.2928	76.0	24016	1.0622	0.7505	0.7739	0.7505
1.3567	77.0	24332	1.0507	0.7531	0.7754	0.7531
1.3422	78.0	24648	1.0574	0.7498	0.7728	0.7498
1.2405	79.0	24964	1.0649	0.7544	0.7786	0.7544
1.281	80.0	25280	1.0635	0.7525	0.7753	0.7525
1.3208	81.0	25596	1.0596	0.7540	0.7775	0.7540
1.3675	82.0	25912	1.0506	0.7545	0.7776	0.7545
1.3179	83.0	26228	1.0621	0.7524	0.7755	0.7524
1.2691	84.0	26544	1.0617	0.7535	0.7752	0.7535
1.3058	85.0	26860	1.0624	0.7548	0.7775	0.7548
1.2305	86.0	27176	1.0663	0.7546	0.7766	0.7546
1.2468	87.0	27492	1.0663	0.7522	0.7748	0.7522

Framework versions

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