limbxy_pose_2heads_2layers_8embeddim

This model is a fine-tuned version of c14kevincardenas/beit-large-patch16-384-limb on the c14kevincardenas/beta_caller_284_limbxy_pose dataset. It achieves the following results on the evaluation set:

• Loss: 0.1417
• Rmse: 0.3765

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: 5e-05
• train_batch_size: 64
• eval_batch_size: 64
• seed: 2014
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 250
• num_epochs: 20.0
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Rmse
0.3414	1.0	89	0.3311	0.5754
0.3313	2.0	178	0.3311	0.5754
0.2277	3.0	267	0.1668	0.4085
0.1756	4.0	356	0.1516	0.3893
0.1761	5.0	445	0.1556	0.3945
0.1632	6.0	534	0.1460	0.3821
0.1488	7.0	623	0.1453	0.3812
0.1502	8.0	712	0.1449	0.3807
0.1512	9.0	801	0.1509	0.3885
0.1491	10.0	890	0.1451	0.3809
0.1615	11.0	979	0.1451	0.3810
0.1549	12.0	1068	0.1453	0.3812
0.1511	13.0	1157	0.1451	0.3810
0.146	14.0	1246	0.1448	0.3806
0.1495	15.0	1335	0.1442	0.3797
0.1463	16.0	1424	0.1433	0.3785
0.1455	17.0	1513	0.1428	0.3779
0.144	18.0	1602	0.1422	0.3772
0.1461	19.0	1691	0.1427	0.3778
0.1425	20.0	1780	0.1417	0.3765

Framework versions

• Transformers 4.45.2
• Pytorch 2.5.0+cu124
• Datasets 3.0.1
• Tokenizers 0.20.1