mobilevit-x-small_alpha0.7_temp3.0_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: 0.9598
• Accuracy: 0.6453

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
0.5532	1.0	164	1.3950	0.2766
0.5286	2.0	328	1.3456	0.3015
0.4764	3.0	492	1.2156	0.4881
0.4397	4.0	656	1.1208	0.5499
0.4225	5.0	820	1.0595	0.5900
0.3976	6.0	984	1.0301	0.6085
0.3816	7.0	1148	1.0354	0.5987
0.3825	8.0	1312	1.0171	0.6182
0.3696	9.0	1476	0.9928	0.6269
0.3513	10.0	1640	1.0023	0.6226
0.3489	11.0	1804	1.0007	0.6095
0.3529	12.0	1968	0.9781	0.6377
0.3323	13.0	2132	0.9598	0.6453
0.3355	14.0	2296	0.9915	0.6388
0.3328	15.0	2460	0.9801	0.6377
0.3305	16.0	2624	0.9745	0.6388
0.3268	17.0	2788	0.9656	0.6497
0.3165	18.0	2952	0.9608	0.6529
0.3222	19.0	3116	0.9698	0.6421
0.3161	20.0	3280	0.9689	0.6421

Framework versions

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