potato-disesase

Potato Disease Classification Model

This is a Convolutional Neural Network (CNN) model built using TensorFlow/Keras to classify potato leaf diseases into three categories:

1. Early Blight (Potato___Early_blight)
2. Late Blight (Potato___Late_blight)
3. Healthy (Potato___healthy)

Model Metrics

• Training Accuracy: 89.12%
• Validation Accuracy: 88.54%
• Test Accuracy (Unseen Data): 87.50%
• Epochs trained: 5 epochs (using Keras 3 / TensorFlow 2.18 on CPU)

Dataset

The model is trained on a dataset containing 2,152 images of potato leaves belonging to the three categories mentioned above.

How to Load and Use the Model

Keras 3 / TensorFlow 2.16+

To load the native Keras format model:

import tensorflow as tf

model = tf.keras.models.load_model("1.keras")

Legacy SavedModel Format

You can also use the SavedModel folder files to load the model:

import tensorflow as tf

# Load as an inference-only layer in Keras 3
model_layer = tf.keras.layers.TFSMLayer("./1", call_endpoint='serving_default')