vit-base-beans

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

• Loss: 0.0027
• Accuracy: 1.0

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: 2e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 1337
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 50.0

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.2628	1.0	130	0.1949	0.9624
0.118	2.0	260	0.1251	0.9699
0.1361	3.0	390	0.0617	0.9925
0.0528	4.0	520	0.0738	0.9774
0.1193	5.0	650	0.0450	0.9925
0.0533	6.0	780	0.0440	0.9850
0.112	7.0	910	0.0817	0.9850
0.1805	8.0	1040	0.0566	0.9850
0.0257	9.0	1170	0.0193	0.9925
0.0132	10.0	1300	0.0122	1.0
0.0138	11.0	1430	0.0113	1.0
0.0702	12.0	1560	0.0733	0.9850
0.0631	13.0	1690	0.1681	0.9624
0.0234	14.0	1820	0.0080	1.0
0.088	15.0	1950	0.0077	1.0
0.0502	16.0	2080	0.0069	1.0
0.007	17.0	2210	0.0070	1.0
0.0787	18.0	2340	0.0159	0.9925
0.0322	19.0	2470	0.0927	0.9699
0.0051	20.0	2600	0.0704	0.9774
0.0053	21.0	2730	0.0051	1.0
0.0056	22.0	2860	0.0311	0.9925
0.0763	23.0	2990	0.0043	1.0
0.0045	24.0	3120	0.0045	1.0
0.0039	25.0	3250	0.0042	1.0
0.0041	26.0	3380	0.0038	1.0
0.0038	27.0	3510	0.0038	1.0
0.0732	28.0	3640	0.0368	0.9925
0.003	29.0	3770	0.0618	0.9774
0.003	30.0	3900	0.0770	0.9774
0.0029	31.0	4030	0.0280	0.9925
0.0027	32.0	4160	0.0055	1.0
0.0027	33.0	4290	0.0046	1.0
0.0073	34.0	4420	0.0027	1.0
0.0325	35.0	4550	0.0102	0.9925
0.003	36.0	4680	0.0334	0.9925
0.0023	37.0	4810	0.0319	0.9925
0.0042	38.0	4940	0.0031	1.0
0.0024	39.0	5070	0.0191	0.9925
0.0022	40.0	5200	0.0036	1.0
0.0029	41.0	5330	0.0101	0.9925
0.0021	42.0	5460	0.0144	0.9925
0.0021	43.0	5590	0.0069	1.0
0.065	44.0	5720	0.0103	0.9925
0.0022	45.0	5850	0.0109	0.9925
0.002	46.0	5980	0.0076	1.0
0.0021	47.0	6110	0.0104	0.9925
0.0034	48.0	6240	0.0231	0.9850
0.0578	49.0	6370	0.0278	0.9925
0.0041	50.0	6500	0.0286	0.9925

Framework versions

• Transformers 4.54.1
• Pytorch 2.7.1+cu126
• Datasets 4.0.0
• Tokenizers 0.21.4