mobilevit-small_alpha0.7_temp5.0_t3

This model is a fine-tuned version of c14kevincardenas/ClimBEiT-t3 on the c14kevincardenas/beta_caller_284_person_crop_seq_withlimb dataset. It achieves the following results on the evaluation set:

• Loss: 0.9566
• Accuracy: 0.6584

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: 32
• eval_batch_size: 32
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.5605	1.0	164	1.4026	0.2625
0.5309	2.0	328	1.3395	0.3210
0.4735	3.0	492	1.1833	0.4967
0.4298	4.0	656	1.1390	0.5260
0.4036	5.0	820	1.0785	0.5748
0.3936	6.0	984	1.0463	0.5922
0.3746	7.0	1148	1.0220	0.6074
0.3644	8.0	1312	1.0024	0.6312
0.3514	9.0	1476	0.9987	0.6161
0.3366	10.0	1640	1.0162	0.6247
0.3377	11.0	1804	1.0073	0.6074
0.3291	12.0	1968	0.9801	0.6323
0.3156	13.0	2132	0.9602	0.6584
0.3177	14.0	2296	0.9742	0.6443
0.31	15.0	2460	0.9683	0.6486
0.3071	16.0	2624	0.9692	0.6508
0.3077	17.0	2788	0.9566	0.6573
0.2957	18.0	2952	0.9566	0.6584
0.3017	19.0	3116	0.9575	0.6649
0.2911	20.0	3280	0.9593	0.6573

Framework versions

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