mobilevit-x-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: 0.9618
• Accuracy: 0.6562

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
0.5533	1.0	164	1.3960	0.2744
0.5292	2.0	328	1.3454	0.2972
0.4766	3.0	492	1.2178	0.4848
0.4393	4.0	656	1.1268	0.5488
0.4237	5.0	820	1.0611	0.5846
0.3992	6.0	984	1.0341	0.6074
0.3827	7.0	1148	1.0388	0.5976
0.3843	8.0	1312	1.0189	0.6150
0.3707	9.0	1476	0.9944	0.6247
0.3523	10.0	1640	1.0051	0.6150
0.351	11.0	1804	1.0020	0.6106
0.3541	12.0	1968	0.9797	0.6410
0.3336	13.0	2132	0.9624	0.6443
0.3365	14.0	2296	0.9929	0.6345
0.3333	15.0	2460	0.9805	0.6345
0.3305	16.0	2624	0.9753	0.6399
0.3274	17.0	2788	0.9668	0.6518
0.3167	18.0	2952	0.9618	0.6562
0.3228	19.0	3116	0.9693	0.6388
0.3163	20.0	3280	0.9693	0.6443

Framework versions

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