limbxy_ext_crop_hands

This model is a fine-tuned version of c14kevincardenas/beit-large-patch16-384-limb on the c14kevincardenas/beta_caller_284_limbxy_ext_crop_hands dataset. It achieves the following results on the evaluation set:

• Loss: 0.0130
• Rmse: 0.1140

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
0.2234	1.0	47	0.1538	0.3922
0.1462	2.0	94	0.0762	0.2759
0.0446	3.0	141	0.0488	0.2209
0.0436	4.0	188	0.0176	0.1329
0.0326	5.0	235	0.0322	0.1796
0.0212	6.0	282	0.0186	0.1365
0.0334	7.0	329	0.0302	0.1739
0.0247	8.0	376	0.0362	0.1904
0.0238	9.0	423	0.0153	0.1236
0.0342	10.0	470	0.0152	0.1236
0.0188	11.0	517	0.0162	0.1272
0.0165	12.0	564	0.0228	0.1509
0.0163	13.0	611	0.0145	0.1205
0.0154	14.0	658	0.0140	0.1182
0.0168	15.0	705	0.0147	0.1215
0.0156	16.0	752	0.0146	0.1208
0.0159	17.0	799	0.0157	0.1253
0.0145	18.0	846	0.0142	0.1191
0.0133	19.0	893	0.0132	0.1150
0.0127	20.0	940	0.0130	0.1140

Framework versions

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