beit-base-patch16-224-pt22k-ft22k-finetuned-mobile-eye-tracking-dataset-v2

This model is a fine-tuned version of microsoft/beit-base-patch16-224-pt22k-ft22k on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.2672
• Accuracy: 0.8281

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
• gradient_accumulation_steps: 4
• total_train_batch_size: 128
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 25

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	1.0	9	0.9670	0.625
1.31	2.0	18	0.6063	0.7422
0.7874	3.0	27	0.5690	0.7344
0.6566	4.0	36	0.4792	0.7422
0.587	5.0	45	0.4839	0.8047
0.5418	6.0	54	0.5108	0.75
0.4702	7.0	63	0.3821	0.8047
0.4439	8.0	72	0.3471	0.7969
0.4292	9.0	81	0.4176	0.7891
0.4472	10.0	90	0.4166	0.8047
0.4472	11.0	99	0.3206	0.8125
0.4243	12.0	108	0.3537	0.8281
0.3744	13.0	117	0.3523	0.8125
0.3409	14.0	126	0.3425	0.8281
0.3537	15.0	135	0.3124	0.7891
0.3532	16.0	144	0.3125	0.8203
0.3175	17.0	153	0.3535	0.8125
0.3205	18.0	162	0.3072	0.8125
0.3043	19.0	171	0.2680	0.8281
0.2832	20.0	180	0.2917	0.8125
0.2832	21.0	189	0.2942	0.7969
0.2755	22.0	198	0.2930	0.8203
0.2774	23.0	207	0.2962	0.7734
0.2823	24.0	216	0.2715	0.8281
0.2606	25.0	225	0.2672	0.8281

Framework versions

• Transformers 4.33.0
• Pytorch 2.6.0+cu124
• Datasets 3.5.0
• Tokenizers 0.13.3