distill-beans-vit-224-to-MobileNetV2-on-beans

This model is a fine-tuned version of on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.4276
• Accuracy: 0.7656

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: 16
• eval_batch_size: 16
• seed: 42
• 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: cosine
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 30
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.9476	1.0	65	0.8270	0.3308
0.9024	2.0	130	0.7717	0.4887
0.8263	3.0	195	0.7346	0.5414
0.7279	4.0	260	0.5934	0.6090
0.6607	5.0	325	0.6376	0.6466
0.5868	6.0	390	0.5602	0.6316
0.5896	7.0	455	0.5396	0.6541
0.5419	8.0	520	0.5851	0.6316
0.5087	9.0	585	0.5418	0.6617
0.5222	10.0	650	0.6134	0.6617
0.4928	11.0	715	0.5752	0.6992
0.4822	12.0	780	0.8421	0.6316
0.4542	13.0	845	0.5489	0.7068
0.4517	14.0	910	0.5498	0.6842
0.4499	15.0	975	0.5248	0.6767
0.4347	16.0	1040	0.4689	0.6992
0.4199	17.0	1105	0.5158	0.7293
0.4183	18.0	1170	0.4212	0.7669
0.3957	19.0	1235	0.4291	0.7143
0.3632	20.0	1300	0.5485	0.7068
0.3751	21.0	1365	0.6471	0.6466
0.3413	22.0	1430	0.5275	0.7218
0.3265	23.0	1495	0.4950	0.7218
0.325	24.0	1560	0.4674	0.7594
0.3312	25.0	1625	0.4459	0.7594
0.3195	26.0	1690	0.5959	0.7143
0.3119	27.0	1755	0.5471	0.6992
0.3049	28.0	1820	0.4409	0.7669
0.3327	29.0	1885	0.6083	0.6391
0.3348	30.0	1950	0.5056	0.7669

Framework versions

• Transformers 4.56.1
• Pytorch 2.6.0+cu124
• Datasets 3.6.0
• Tokenizers 0.22.0