This model is a fine-tuned version of google/vit-base-patch16-224 on the medmnist-v2 dataset. It achieves the following results on the evaluation set:

Loss: 0.2659
Accuracy: 0.9065
F1: 0.8909

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: 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	Accuracy	F1
0.6113	0.0595	200	0.8908	0.6846	0.5917
0.3578	0.1189	400	0.5958	0.7956	0.7548
0.3118	0.1784	600	0.5688	0.7810	0.7132
0.2815	0.2378	800	0.5227	0.7961	0.7645
0.266	0.2973	1000	0.6554	0.7687	0.7229
0.2353	0.3567	1200	0.3328	0.8838	0.8615
0.2297	0.4162	1400	0.4696	0.8592	0.7990
0.2267	0.4756	1600	0.4362	0.8493	0.8117
0.2266	0.5351	1800	0.3286	0.8838	0.8407
0.2047	0.5945	2000	0.3614	0.8697	0.8382
0.1948	0.6540	2200	0.3144	0.8843	0.8546
0.1953	0.7134	2400	0.3805	0.8657	0.8180
0.1728	0.7729	2600	0.3364	0.8820	0.8339
0.1658	0.8323	2800	0.2873	0.8978	0.8743
0.1594	0.8918	3000	0.3062	0.8914	0.8580
0.1649	0.9512	3200	0.3313	0.8867	0.8577
0.1508	1.0107	3400	0.2117	0.9217	0.9133
0.1062	1.0702	3600	0.2978	0.8919	0.8756
0.1091	1.1296	3800	0.2832	0.9019	0.8831
0.0993	1.1891	4000	0.3275	0.8943	0.8718
0.1001	1.2485	4200	0.3420	0.8896	0.8568
0.1092	1.3080	4400	0.2594	0.9130	0.8909
0.092	1.3674	4600	0.3181	0.8966	0.8753
0.1036	1.4269	4800	0.2721	0.9048	0.8852
0.0896	1.4863	5000	0.3795	0.8820	0.8617
0.0904	1.5458	5200	0.2382	0.9171	0.8980
0.0864	1.6052	5400	0.3845	0.8814	0.8499
0.0809	1.6647	5600	0.3189	0.8984	0.8758
0.0764	1.7241	5800	0.3952	0.8843	0.8522
0.0796	1.7836	6000	0.3656	0.8867	0.8460
0.0695	1.8430	6200	0.3266	0.8925	0.8597
0.0682	1.9025	6400	0.3247	0.8960	0.8647
0.06	1.9620	6600	0.2349	0.9223	0.9055
0.0498	2.0214	6800	0.2578	0.9176	0.8952
0.0296	2.0809	7000	0.2592	0.9211	0.9070
0.0251	2.1403	7200	0.3249	0.9048	0.8797
0.02	2.1998	7400	0.2977	0.9165	0.8973
0.0274	2.2592	7600	0.3411	0.9013	0.8730
0.0241	2.3187	7800	0.3916	0.9013	0.8752
0.0253	2.3781	8000	0.2919	0.9136	0.8939
0.0197	2.4376	8200	0.3294	0.9077	0.8835
0.0209	2.4970	8400	0.3709	0.8966	0.8652
0.0175	2.5565	8600	0.3639	0.9001	0.8733
0.0191	2.6159	8800	0.3706	0.9048	0.8790
0.0167	2.6754	9000	0.3120	0.9171	0.8993
0.0224	2.7348	9200	0.3493	0.9048	0.8799
0.015	2.7943	9400	0.3398	0.9130	0.8889
0.0155	2.8537	9600	0.3707	0.9036	0.8758
0.0129	2.9132	9800	0.3467	0.9118	0.8909
0.0126	2.9727	10000	0.3470	0.9095	0.8874

Framework versions

Transformers 4.45.1
Pytorch 2.4.0
Datasets 3.0.1
Tokenizers 0.20.0

KiViDrag/ViT_bloodmnist_std_45

ViT_bloodmnist_std_45

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

KiViDrag/
ViT_bloodmnist_std_45