4-classifier-finetuned-padchest

This model is a fine-tuned version of nickmuchi/vit-finetuned-chest-xray-pneumonia on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.9186
• Accuracy: 0.7124

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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.0441	1.0	14	1.9084	0.3164
1.8716	2.0	28	1.6532	0.4484
1.4727	3.0	42	1.4218	0.5228
1.3452	4.0	56	1.3037	0.5736
1.2518	5.0	70	1.2799	0.5584
1.1646	6.0	84	1.1892	0.6244
1.1358	7.0	98	1.1543	0.6074
1.0664	8.0	112	1.1060	0.6277
1.041	9.0	126	1.0434	0.6667
1.002	10.0	140	1.0337	0.6582
0.9867	11.0	154	1.0373	0.6582
0.9485	12.0	168	0.9866	0.6887
0.9121	13.0	182	0.9827	0.6785
0.918	14.0	196	0.9588	0.7039
0.8882	15.0	210	0.9576	0.7005
0.873	16.0	224	0.9450	0.7022
0.8469	17.0	238	0.9266	0.7090
0.814	18.0	252	0.9463	0.6971
0.8206	19.0	266	0.9201	0.7090
0.8078	20.0	280	0.9186	0.7124

Framework versions

• Transformers 4.28.0.dev0
• Pytorch 2.0.0+cu117
• Datasets 2.18.0
• Tokenizers 0.13.3