beit-base-patch16-384_alpha0.5_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: 0.5753
• Accuracy: 0.8547

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
0.8115	1.0	164	1.4100	0.2549
0.7144	2.0	328	1.1924	0.4328
0.5506	3.0	492	0.8301	0.7364
0.4594	4.0	656	0.7320	0.7852
0.3955	5.0	820	0.6395	0.8124
0.3511	6.0	984	0.6268	0.8167
0.3292	7.0	1148	0.6236	0.8254
0.3197	8.0	1312	0.6250	0.8265
0.3173	9.0	1476	0.6289	0.8351
0.2963	10.0	1640	0.6381	0.8200
0.2915	11.0	1804	0.6562	0.8189
0.2948	12.0	1968	0.5872	0.8330
0.2848	13.0	2132	0.6199	0.8373
0.2873	14.0	2296	0.5970	0.8373
0.2885	15.0	2460	0.5928	0.8373
0.2829	16.0	2624	0.5842	0.8514
0.2895	17.0	2788	0.5851	0.8579
0.2759	18.0	2952	0.5753	0.8547
0.2825	19.0	3116	0.5779	0.8514
0.2758	20.0	3280	0.5770	0.8525

Framework versions

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