vit-msn-small-corect_deepcleaned_dataset_lateral_flow_ivalidation

This model is a fine-tuned version of facebook/vit-msn-small on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.2229
• Accuracy: 0.9194

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: 64
• eval_batch_size: 64
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 256
• 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
No log	0.9231	3	0.6175	0.7216
No log	1.8462	6	0.4141	0.8352
No log	2.7692	9	0.7408	0.5788
0.5817	4.0	13	0.2757	0.9158
0.5817	4.9231	16	0.2847	0.8791
0.5817	5.8462	19	0.2456	0.9011
0.3724	6.7692	22	0.2547	0.9121
0.3724	8.0	26	0.3007	0.8828
0.3724	8.9231	29	0.3043	0.9011
0.3155	9.8462	32	0.2603	0.9048
0.3155	10.7692	35	0.2481	0.9158
0.3155	12.0	39	0.2229	0.9194
0.2844	12.9231	42	0.3036	0.8791
0.2844	13.8462	45	0.2579	0.9084
0.2844	14.7692	48	0.2434	0.9158
0.2517	16.0	52	0.2718	0.9048
0.2517	16.9231	55	0.2513	0.9121
0.2517	17.8462	58	0.2503	0.9121
0.2468	18.4615	60	0.2491	0.9121

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 3.2.0
• Tokenizers 0.19.1