mobilevit-x-small_alpha0.5_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: 0.8649
• Accuracy: 0.6833

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
0.8021	1.0	164	1.3864	0.2787
0.7632	2.0	328	1.3358	0.2983
0.684	3.0	492	1.1592	0.4718
0.6194	4.0	656	1.0484	0.5640
0.5802	5.0	820	0.9864	0.6247
0.5481	6.0	984	0.9503	0.6584
0.5327	7.0	1148	0.9417	0.6421
0.5193	8.0	1312	0.9287	0.6584
0.5071	9.0	1476	0.8931	0.6822
0.4696	10.0	1640	0.9343	0.6356
0.4724	11.0	1804	0.9316	0.6475
0.4661	12.0	1968	0.8823	0.6811
0.4464	13.0	2132	0.8741	0.6931
0.4466	14.0	2296	0.8901	0.6735
0.4406	15.0	2460	0.8990	0.6725
0.4395	16.0	2624	0.9002	0.6638
0.4411	17.0	2788	0.8779	0.6768
0.4257	18.0	2952	0.8649	0.6833
0.4325	19.0	3116	0.8692	0.6909
0.4237	20.0	3280	0.8844	0.6779

Framework versions

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