pneumonia-classification-model

This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the chestxrayclassification dataset. It achieves the following results on the evaluation set:

Loss: 0.1143
Accuracy: 0.9657

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: 32

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.6297	0.98	25	0.5258	0.7230
0.3781	2.0	51	0.3011	0.9154
0.2268	2.98	76	0.1981	0.9314
0.1827	4.0	102	0.1602	0.9473
0.1765	4.98	127	0.1446	0.9461
0.1508	6.0	153	0.1449	0.9510
0.1332	6.98	178	0.1510	0.9375
0.1187	8.0	204	0.1169	0.9596
0.131	8.98	229	0.1315	0.9559
0.1043	10.0	255	0.1114	0.9571
0.1022	10.98	280	0.1633	0.9375
0.0893	12.0	306	0.1167	0.9596
0.0848	12.98	331	0.0936	0.9694
0.0885	14.0	357	0.1074	0.9608
0.0928	14.98	382	0.1052	0.9645
0.0776	16.0	408	0.1116	0.9608
0.0895	16.98	433	0.1060	0.9645
0.0817	18.0	459	0.1107	0.9632
0.0766	18.98	484	0.0993	0.9669
0.0697	20.0	510	0.0938	0.9681
0.0626	20.98	535	0.1199	0.9620
0.0665	22.0	561	0.1100	0.9657
0.0613	22.98	586	0.1246	0.9620
0.054	24.0	612	0.1066	0.9645
0.0474	24.98	637	0.1100	0.9669
0.0456	26.0	663	0.1118	0.9645
0.0473	26.98	688	0.1137	0.9645
0.0543	28.0	714	0.0955	0.9632
0.0493	28.98	739	0.1300	0.9559
0.043	30.0	765	0.1229	0.9669
0.039	30.98	790	0.1125	0.9608
0.0398	31.37	800	0.1143	0.9657

Framework versions

Transformers 4.38.1
Pytorch 2.1.2
Datasets 2.1.0
Tokenizers 0.15.2