metadata

license: other
tags:
  - generated_from_trainer
datasets:
  - image_folder
metrics:
  - accuracy
model-index:
  - name: mobilenet_v2_1.0_224-plant-disease-identification
    results:
      - task:
          name: Image Classification
          type: image-classification
        dataset:
          name: image_folder
          type: image_folder
          config: default
          split: train
          args: default
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.7857752489331437

mobilenet_v2_1.0_224-plant-disease-identification

This model is a fine-tuned version of google/mobilenet_v2_1.0_224 on the Kaggle version of the Plant Village dataset. It achieves the following results on the evaluation set:

Cross Entropy Loss: 1.0461
Accuracy: 0.7858

Will be further training it (such as finding optimal hyperparameters) better to achieve much better accuracy.

Intended uses & limitations

For identifying common diseases in crops and assessing plant health.

Training and evaluation data

The plant village dataset consists of 38 classes of diseases in common crops (including healthy/normal crops).

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 0.0003
train_batch_size: 256
eval_batch_size: 256
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
lr_scheduler_warmup_ratio: 0.15
num_epochs: 8

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
2.9265	1.0	248	2.7159	0.4703
1.9734	2.0	496	1.7668	0.6649
1.7206	3.0	744	1.4012	0.7206
1.6406	4.0	992	1.2514	0.7644
1.6075	5.0	1240	1.2934	0.7094
1.5932	6.0	1488	1.2093	0.7257
1.5203	7.0	1736	1.0461	0.7858
1.5076	8.0	1984	1.0580	0.7848

Framework versions

Transformers 4.27.3
Pytorch 1.13.0
Datasets 2.1.0
Tokenizers 0.13.2