vit-base-brain-tumor-detection

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

• Loss: 0.5832
• Accuracy: 0.785

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: 20
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.9535	0.4	100	0.8966	0.618
0.862	0.8	200	1.1149	0.561
0.7373	1.2	300	0.8543	0.605
0.6476	1.6	400	0.7307	0.666
0.6712	2.0	500	0.6954	0.694
0.4892	2.4	600	0.6391	0.707
0.5801	2.8	700	0.6247	0.708
0.3505	3.2	800	0.6056	0.778
0.3503	3.6	900	0.6264	0.743
0.3416	4.0	1000	0.5832	0.785
0.1427	4.4	1100	0.7297	0.769
0.1982	4.8	1200	0.7761	0.73
0.193	5.2	1300	0.8467	0.741
0.1831	5.6	1400	0.6975	0.774
0.2612	6.0	1500	0.8719	0.775
0.102	6.4	1600	0.9045	0.788
0.1029	6.8	1700	0.9655	0.783
0.0735	7.2	1800	0.9906	0.78
0.0715	7.6	1900	0.8893	0.787
0.1254	8.0	2000	1.1221	0.761
0.021	8.4	2100	1.1648	0.779
0.0133	8.8	2200	0.9857	0.806
0.0086	9.2	2300	1.0365	0.799
0.0223	9.6	2400	0.9826	0.812
0.0023	10.0	2500	1.0697	0.795
0.0021	10.4	2600	1.0490	0.815
0.0401	10.8	2700	1.1594	0.8
0.0012	11.2	2800	1.0811	0.817
0.0034	11.6	2900	1.0956	0.825
0.0012	12.0	3000	1.2010	0.808
0.0011	12.4	3100	1.1712	0.81
0.0092	12.8	3200	1.1814	0.813
0.0007	13.2	3300	1.1677	0.818
0.0007	13.6	3400	1.1723	0.818
0.0006	14.0	3500	1.1852	0.821
0.0005	14.4	3600	1.1928	0.82
0.0005	14.8	3700	1.2030	0.819
0.0005	15.2	3800	1.2093	0.818
0.0005	15.6	3900	1.2160	0.818
0.0004	16.0	4000	1.2232	0.819
0.0004	16.4	4100	1.2302	0.819
0.0004	16.8	4200	1.2350	0.819
0.0004	17.2	4300	1.2400	0.82
0.0004	17.6	4400	1.2442	0.821
0.0004	18.0	4500	1.2483	0.821
0.0004	18.4	4600	1.2518	0.821
0.0004	18.8	4700	1.2546	0.821
0.0004	19.2	4800	1.2561	0.821
0.0004	19.6	4900	1.2574	0.82
0.0004	20.0	5000	1.2577	0.82

Framework versions

• Transformers 4.41.2
• Pytorch 2.3.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1