vit-base-beans

This model is a fine-tuned version of timm/resnet18.a1_in1k on the beans dataset. It achieves the following results on the evaluation set:

• Loss: 0.7389
• Accuracy: 0.8045

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 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: 15.0

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.0863	1.0	130	1.0882	0.4286
1.063	2.0	260	1.0590	0.5414
1.0447	3.0	390	1.0229	0.6992
1.0223	4.0	520	0.9968	0.6917
1.0	5.0	650	0.9575	0.7519
0.9726	6.0	780	0.9298	0.7744
0.9258	7.0	910	0.8871	0.8045
0.9203	8.0	1040	0.8487	0.8346
0.9038	9.0	1170	0.8330	0.8120
0.8112	10.0	1300	0.8084	0.8346
0.8335	11.0	1430	0.7785	0.8346
0.8062	12.0	1560	0.7569	0.8346
0.8141	13.0	1690	0.7536	0.8496
0.8172	14.0	1820	0.7532	0.8271
0.7896	15.0	1950	0.7389	0.8045

Framework versions

• Transformers 4.47.0.dev0
• Pytorch 2.4.1+cu118
• Datasets 2.21.0
• Tokenizers 0.20.0