limbxy_feet

This model is a fine-tuned version of c14kevincardenas/beit-large-patch16-384-limb on the c14kevincardenas/beta_caller_284_limbxy_feet dataset. It achieves the following results on the evaluation set:

• Loss: 0.0047
• Rmse: 0.0683

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
0.1975	1.0	47	0.1440	0.3794
0.1345	2.0	94	0.0639	0.2527
0.0329	3.0	141	0.0165	0.1284
0.0295	4.0	188	0.0178	0.1334
0.0272	5.0	235	0.0153	0.1237
0.0156	6.0	282	0.0164	0.1282
0.0254	7.0	329	0.0241	0.1551
0.0254	8.0	376	0.0679	0.2606
0.0268	9.0	423	0.0060	0.0773
0.0174	10.0	470	0.0081	0.0899
0.0073	11.0	517	0.0144	0.1200
0.0101	12.0	564	0.0094	0.0968
0.0064	13.0	611	0.0080	0.0894
0.0074	14.0	658	0.0055	0.0743
0.0037	15.0	705	0.0064	0.0803
0.0045	16.0	752	0.0055	0.0742
0.0022	17.0	799	0.0057	0.0754
0.0027	18.0	846	0.0047	0.0683
0.0019	19.0	893	0.0047	0.0684
0.0018	20.0	940	0.0047	0.0684

Framework versions

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