lung-cancer-image-classification

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

• Loss: 0.0177
• Precision: 0.9963
• Recall: 0.9963
• F1: 0.9963
• Accuracy: 0.9963
• Confusion matrix: 1245 1 4 0 1250 0 9 0 1241

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: 0.0002
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 3

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy	Confusion matrix
0.3173	0.21	100	0.1952	0.9371	0.9331	0.9339	0.9331	1186 0 64
90 1160 0
97 0 1153
0.1312	0.43	200	0.0752	0.9786	0.9779	0.9778	0.9779	1178 1 71
2 1248 0
9 0 1241
0.1453	0.64	300	0.0688	0.9759	0.9752	0.9752	0.9752	1232 1 17
8 1242 0
67 0 1183
0.0146	0.85	400	0.0485	0.9854	0.9853	0.9853	0.9853	1212 2 36
0 1250 0
17 0 1233
0.0075	1.07	500	0.0376	0.9897	0.9896	0.9896	0.9896	1220 1 29
5 1245 0
4 0 1246
0.054	1.28	600	0.0233	0.9939	0.9939	0.9939	0.9939	1241 1 8
0 1250 0
14 0 1236
0.0272	1.49	700	0.0156	0.9950	0.9949	0.9949	0.9949	1235 1 14
0 1250 0
4 0 1246
0.0307	1.71	800	0.0172	0.9949	0.9949	0.9949	0.9949	1244 1 5
0 1250 0
13 0 1237
0.0022	1.92	900	0.0144	0.9963	0.9963	0.9963	0.9963	1237 1 12
0 1250 0
1 0 1249
0.0015	2.13	1000	0.0156	0.9963	0.9963	0.9963	0.9963	1238 1 11
0 1250 0
2 0 1248
0.0014	2.35	1100	0.0138	0.9971	0.9971	0.9971	0.9971	1243 1 6
0 1250 0
4 0 1246
0.0317	2.56	1200	0.0110	0.9973	0.9973	0.9973	0.9973	1244 1 5
0 1250 0
4 0 1246
0.0011	2.77	1300	0.0159	0.9963	0.9963	0.9963	0.9963	1236 1 13
0 1250 0
0 0 1250
0.0012	2.99	1400	0.0120	0.9971	0.9971	0.9971	0.9971	1239 1 10
0 1250 0
0 0 1250

Framework versions

• Transformers 4.38.2
• Pytorch 2.2.1+cu121
• Datasets 2.19.0
• Tokenizers 0.15.2