finetuned-beit-limb-seq-t3-8heads-2layers-1e-4lr

This model is a fine-tuned version of c14kevincardenas/beit-large-patch16-384-limb-person-crop on the c14kevincardenas/beta_caller_284_person_crop_seq_withlimb_3 dataset. It achieves the following results on the evaluation set:

• Loss: 0.3899
• Accuracy: 0.9262

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: 0.0001
• train_batch_size: 32
• eval_batch_size: 32
• 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
• label_smoothing_factor: 0.05

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.4783	1.0	164	0.4094	0.9165
0.4028	2.0	328	0.4224	0.9143
0.4222	3.0	492	0.4298	0.9121
0.4306	4.0	656	0.3900	0.9089
0.4038	5.0	820	0.4329	0.9067
0.4246	6.0	984	0.3950	0.9208
0.3861	7.0	1148	0.3914	0.9230
0.4031	8.0	1312	0.4069	0.9121
0.4023	9.0	1476	0.4183	0.9132
0.403	10.0	1640	0.3899	0.9262
0.3508	11.0	1804	0.4037	0.9176
0.3671	12.0	1968	0.4058	0.9100
0.3649	13.0	2132	0.4015	0.9230
0.3661	14.0	2296	0.4008	0.9230
0.3531	15.0	2460	0.4220	0.9100
0.3633	16.0	2624	0.4009	0.9219
0.3236	17.0	2788	0.3954	0.9284
0.3217	18.0	2952	0.3985	0.9230
0.3339	19.0	3116	0.3967	0.9219
0.3217	20.0	3280	0.3981	0.9252

Framework versions

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