Akshay0706/Cinnamon-Plant-20-Epochs-Model

Rice-Plant-Disease-Detection-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: 0.2929
• Accuracy: 0.8958
• F1: 0.8965

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
0.5517	1.0	18	0.5222	0.875	0.8754
0.2996	2.0	36	0.3833	0.8542	0.8564
0.1529	3.0	54	0.3152	0.875	0.8763
0.0843	4.0	72	0.2929	0.8958	0.8965
0.0549	5.0	90	0.2756	0.875	0.8754
0.0402	6.0	108	0.2765	0.875	0.8754
0.0327	7.0	126	0.2875	0.875	0.8754
0.0277	8.0	144	0.2938	0.875	0.8754
0.0244	9.0	162	0.2992	0.875	0.8754
0.0222	10.0	180	0.2996	0.8958	0.8960
0.0203	11.0	198	0.3052	0.8958	0.8960
0.019	12.0	216	0.3087	0.8958	0.8960
0.018	13.0	234	0.3143	0.8958	0.8960
0.0171	14.0	252	0.3206	0.8958	0.8960
0.0164	15.0	270	0.3227	0.8958	0.8960
0.0158	16.0	288	0.3250	0.8958	0.8960
0.0155	17.0	306	0.3257	0.8958	0.8960
0.0152	18.0	324	0.3264	0.8958	0.8960
0.015	19.0	342	0.3276	0.8958	0.8960
0.0149	20.0	360	0.3275	0.8958	0.8960

Framework versions

• Transformers 4.34.0
• Pytorch 2.1.0+cpu
• Datasets 2.14.5
• Tokenizers 0.14.0