Rice-Plant-20-Epochs-Model

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

• Loss: 0.1306
• Accuracy: 0.9720
• F1: 0.9719

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: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	F1
1.0906	1.0	116	0.6826	0.8660	0.8676
0.3792	2.0	232	0.3327	0.9470	0.9474
0.1934	3.0	348	0.2876	0.9283	0.9285
0.1157	4.0	464	0.2187	0.9470	0.9470
0.0849	5.0	580	0.1614	0.9688	0.9689
0.0675	6.0	696	0.1326	0.9688	0.9688
0.0561	7.0	812	0.1227	0.9688	0.9688
0.0485	8.0	928	0.1306	0.9720	0.9719
0.0416	9.0	1044	0.1356	0.9720	0.9719
0.0369	10.0	1160	0.1184	0.9688	0.9688
0.0335	11.0	1276	0.1281	0.9720	0.9720
0.0308	12.0	1392	0.1129	0.9720	0.9719
0.0285	13.0	1508	0.1074	0.9720	0.9719
0.0267	14.0	1624	0.1061	0.9720	0.9719
0.0253	15.0	1740	0.1049	0.9720	0.9719
0.0241	16.0	1856	0.1048	0.9720	0.9719
0.0232	17.0	1972	0.1045	0.9720	0.9719
0.0225	18.0	2088	0.1036	0.9720	0.9719
0.0221	19.0	2204	0.1035	0.9720	0.9719
0.0219	20.0	2320	0.1036	0.9720	0.9719

Framework versions

• Transformers 4.35.0
• Pytorch 2.1.0+cu118
• Datasets 2.14.6
• Tokenizers 0.14.1