sign-language-gestures

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

• Loss: 0.5662
• Accuracy: 0.9446

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: 0.001
• train_batch_size: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 150
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	143	0.6546	0.9094
No log	2.0	286	0.8703	0.8901
No log	3.0	429	0.7896	0.9006
0.2236	4.0	572	0.8478	0.8927
0.2236	5.0	715	0.7310	0.8997
0.2236	6.0	858	0.9632	0.8725
0.2107	7.0	1001	1.0763	0.8698
0.2107	8.0	1144	0.7950	0.8918
0.2107	9.0	1287	0.6954	0.8945
0.2107	10.0	1430	0.7618	0.8980
0.1654	11.0	1573	0.6823	0.8927
0.1654	12.0	1716	0.6213	0.8971
0.1654	13.0	1859	0.8228	0.8716
0.1496	14.0	2002	0.9967	0.8637
0.1496	15.0	2145	0.7127	0.8980
0.1496	16.0	2288	1.0634	0.8769
0.1496	17.0	2431	0.6832	0.8971
0.1408	18.0	2574	0.5724	0.9068
0.1408	19.0	2717	0.6036	0.9182
0.1408	20.0	2860	0.9746	0.8716
0.1102	21.0	3003	0.9812	0.8795
0.1102	22.0	3146	0.5909	0.9182
0.1102	23.0	3289	0.5876	0.9077
0.1102	24.0	3432	0.7260	0.8927
0.1230	25.0	3575	0.6923	0.9050
0.1230	26.0	3718	0.5820	0.9077
0.1230	27.0	3861	0.6786	0.8971
0.1118	28.0	4004	0.6982	0.9033
0.1118	29.0	4147	0.6097	0.9208
0.1118	30.0	4290	0.6823	0.9085
0.1118	31.0	4433	0.6218	0.9191
0.0905	32.0	4576	0.6234	0.9103
0.0905	33.0	4719	0.7955	0.9077
0.0905	34.0	4862	0.5156	0.9261
0.0949	35.0	5005	0.9479	0.8901
0.0949	36.0	5148	0.9709	0.8716
0.0949	37.0	5291	0.5912	0.9182
0.0949	38.0	5434	0.6688	0.8980
0.0962	39.0	5577	0.6089	0.9208
0.0962	40.0	5720	0.8047	0.8962
0.0962	41.0	5863	0.5960	0.9138
0.0964	42.0	6006	0.6224	0.9261
0.0964	43.0	6149	0.5652	0.9129
0.0964	44.0	6292	0.4764	0.9367
0.0964	45.0	6435	0.5569	0.9164
0.0708	46.0	6578	0.7211	0.9033
0.0708	47.0	6721	0.6608	0.9112
0.0708	48.0	6864	0.5730	0.9252
0.0688	49.0	7007	0.6919	0.9156
0.0688	50.0	7150	0.6260	0.9191
0.0688	51.0	7293	0.6363	0.9156
0.0688	52.0	7436	0.7481	0.9059
0.0622	53.0	7579	0.5443	0.9235
0.0622	54.0	7722	0.6905	0.9050
0.0622	55.0	7865	0.6591	0.9200
0.0752	56.0	8008	0.7804	0.9173
0.0752	57.0	8151	0.6097	0.9226
0.0752	58.0	8294	0.6544	0.9182
0.0752	59.0	8437	0.7227	0.9156
0.0492	60.0	8580	0.6098	0.9244
0.0492	61.0	8723	0.6606	0.9173
0.0492	62.0	8866	0.5990	0.9305
0.0420	63.0	9009	0.6210	0.9226
0.0420	64.0	9152	0.7217	0.9138
0.0420	65.0	9295	0.6458	0.9156
0.0420	66.0	9438	0.6457	0.9147
0.0600	67.0	9581	0.6832	0.9156
0.0600	68.0	9724	0.4882	0.9296
0.0600	69.0	9867	0.6837	0.9235
0.0464	70.0	10010	0.6410	0.9200
0.0464	71.0	10153	0.7314	0.9112
0.0464	72.0	10296	0.5722	0.9358
0.0464	73.0	10439	0.6004	0.9235
0.0390	74.0	10582	0.6213	0.9235
0.0390	75.0	10725	0.6337	0.9173
0.0390	76.0	10868	0.6541	0.9208
0.0350	77.0	11011	0.7316	0.9217
0.0350	78.0	11154	0.6058	0.9244
0.0350	79.0	11297	0.6321	0.9279
0.0350	80.0	11440	0.6537	0.9200
0.0233	81.0	11583	0.6731	0.9191
0.0233	82.0	11726	0.6427	0.9244
0.0233	83.0	11869	0.6146	0.9244
0.0260	84.0	12012	0.5702	0.9279
0.0260	85.0	12155	0.5586	0.9305
0.0260	86.0	12298	0.8679	0.9077
0.0260	87.0	12441	0.6251	0.9305
0.0282	88.0	12584	0.5804	0.9367
0.0282	89.0	12727	0.5944	0.9349
0.0282	90.0	12870	0.5992	0.9340
0.0194	91.0	13013	0.6331	0.9296
0.0194	92.0	13156	0.6080	0.9314
0.0194	93.0	13299	0.5649	0.9296
0.0194	94.0	13442	0.6037	0.9340
0.0222	95.0	13585	0.5670	0.9296
0.0222	96.0	13728	0.6802	0.9226
0.0222	97.0	13871	0.6618	0.9244
0.0237	98.0	14014	0.5798	0.9332
0.0237	99.0	14157	0.6216	0.9279
0.0237	100.0	14300	0.6225	0.9296
0.0237	101.0	14443	0.6154	0.9235
0.0193	102.0	14586	0.6037	0.9340
0.0193	103.0	14729	0.6031	0.9340
0.0193	104.0	14872	0.6299	0.9252
0.0174	105.0	15015	0.6269	0.9279
0.0174	106.0	15158	0.6024	0.9279
0.0174	107.0	15301	0.6212	0.9420
0.0174	108.0	15444	0.6584	0.9252
0.0124	109.0	15587	0.6326	0.9323
0.0124	110.0	15730	0.5837	0.9367
0.0124	111.0	15873	0.6815	0.9261
0.0112	112.0	16016	0.6938	0.9226
0.0112	113.0	16159	0.7040	0.9270
0.0112	114.0	16302	0.6588	0.9252
0.0112	115.0	16445	0.5878	0.9323
0.0161	116.0	16588	0.6024	0.9323
0.0161	117.0	16731	0.6956	0.9226
0.0161	118.0	16874	0.6083	0.9332
0.0099	119.0	17017	0.5956	0.9376
0.0099	120.0	17160	0.6417	0.9323
0.0099	121.0	17303	0.6533	0.9349
0.0099	122.0	17446	0.6004	0.9367
0.0090	123.0	17589	0.5912	0.9402
0.0090	124.0	17732	0.6371	0.9314
0.0090	125.0	17875	0.6745	0.9279
0.0061	126.0	18018	0.5803	0.9411
0.0061	127.0	18161	0.5656	0.9411
0.0061	128.0	18304	0.5493	0.9358
0.0061	129.0	18447	0.5467	0.9411
0.0071	130.0	18590	0.5920	0.9305
0.0071	131.0	18733	0.6412	0.9296
0.0071	132.0	18876	0.6630	0.9252
0.0077	133.0	19019	0.5808	0.9384
0.0077	134.0	19162	0.6036	0.9358
0.0077	135.0	19305	0.6178	0.9332
0.0077	136.0	19448	0.6358	0.9288
0.0076	137.0	19591	0.6046	0.9367
0.0076	138.0	19734	0.5878	0.9411
0.0076	139.0	19877	0.5617	0.9358
0.0117	140.0	20020	0.5840	0.9402
0.0117	141.0	20163	0.5782	0.9384
0.0117	142.0	20306	0.5671	0.9393
0.0117	143.0	20449	0.6211	0.9384
0.0171	144.0	20592	0.5861	0.9428
0.0171	145.0	20735	0.5639	0.9411
0.0171	146.0	20878	0.5662	0.9446
0.0027	147.0	21021	0.5783	0.9402
0.0027	148.0	21164	0.5672	0.9376
0.0027	149.0	21307	0.5792	0.9420
0.0027	150.0	21450	0.5622	0.9393

Framework versions

• Transformers 5.0.0.dev0
• Pytorch 2.9.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.1