Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+variables/variables.data-00000-of-00001 filter=lfs diff=lfs merge=lfs -text

README.md

@@ -1,3 +1,131 @@
----
-license: mit
----
+# Maize Disease Detection Model
+
+This repository contains a TensorFlow model for detecting diseases in maize (corn) using camera images.
+
+## Model Description
+
+This model is designed to classify maize leaf images into four categories:
+
+0. Healthy
+1. Gray Leaf Spot
+2. Blight
+3. Common Rust
+
+The model uses computer vision techniques to analyze images of maize leaves and identify the presence of common diseases.
+
+## Usage
+
+To use this model for inference, follow these steps:
+
+1. Clone this repository:
+   ```
+   git clone https://huggingface.co/eligapris/maize-diseases-detection
+   cd maize-diseases-detection
+   ```
+
+2. Install the required dependencies (assuming you have Python and pip installed):
+   ```
+   pip install tensorflow pillow numpy
+   ```
+
+3. Use the `example.py` script as a reference for how to load and use the model. Here's a basic example:
+
+   ```python
+   import tensorflow as tf
+   from PIL import Image
+   import numpy as np
+   import json
+
+   # Load the model
+   model = tf.keras.models.load_model('saved_model.pb')
+
+   # Load and preprocess the image
+   img = Image.open('image.jpg')
+   img = img.resize((224, 224))  # Adjust size as needed
+   img_array = np.array(img) / 255.0  # Normalize pixel values
+   img_array = np.expand_dims(img_array, axis=0)  # Add batch dimension
+
+   # Make prediction
+   prediction = model.predict(img_array)
+
+   # Load class names
+   with open('classes.json', 'r') as f:
+       class_names = json.load(f)
+
+   # Get the predicted class
+   predicted_class = class_names[str(np.argmax(prediction[0]))]
+   print(f"Predicted class: {predicted_class}")
+   ```
+
+   Make sure to replace 'image.jpg' with the path to your input image.
+
+## Model Architecture
+
+The model architecture is based on a convolutional neural network (CNN) designed for image classification. The exact architecture details are not provided in the repository, but you can infer some information from the `saved_model.pb` file:
+
+- The model is saved in TensorFlow's SavedModel format, which preserves the entire tf.keras model, including its architecture and weights.
+- Input images are likely resized to a specific dimension (e.g., 224x224 pixels) before being fed into the network.
+- The output layer corresponds to the four classes defined in `classes.json`.
+
+To get more detailed information about the model architecture, you can load the model and use TensorFlow's built-in summary function:
+
+```python
+import tensorflow as tf
+
+model = tf.keras.models.load_model('saved_model.pb')
+model.summary()
+```
+
+This will print out the layers and parameters of the model.
+
+## Training Data
+
+The model was trained on a dataset derived from the popular PlantVillage and PlantDoc datasets. The dataset consists of images of corn (maize) leaves in various health states:
+
+- Common Rust: 1306 images
+- Gray Leaf Spot: 574 images
+- Blight: 1146 images
+- Healthy: 1162 images
+
+Total images: 4188
+
+The original dataset can be found on Kaggle: [Corn or Maize Leaf Disease Dataset](https://www.kaggle.com/datasets/smaranjitghose/corn-or-maize-leaf-disease-dataset)
+
+Note: During the formation of this dataset, certain images from the original sources were removed if they were not found to be useful.
+
+## Performance
+
+[Include information about the model's performance, such as accuracy, precision, recall, or any other relevant metrics once available.]
+
+## Limitations
+
+- The model's performance may vary on images that significantly differ from the training dataset in terms of lighting conditions, camera angles, or image quality.
+- The model is trained to detect four specific conditions (Healthy, Gray Leaf Spot, Blight, and Common Rust). It may not accurately classify other maize diseases or conditions not included in these categories.
+- The model's performance on images with multiple diseases present has not been extensively tested.
+
+## License
+
+This model is released under the MIT License.
+
+
+Additionally, please credit the original authors of the PlantVillage and PlantDoc datasets, as this model's training data is derived from their work.
+
+## Contact
+
+Grey
+Eligapris
+
+## Acknowledgements
+
+- The dataset used for training this model is derived from the PlantVillage and PlantDoc datasets.
+- Special thanks to the creators and contributors of the [Corn or Maize Leaf Disease Dataset](https://www.kaggle.com/datasets/smaranjitghose/corn-or-maize-leaf-disease-dataset) on Kaggle.
+
+## File Structure
+
+- `assets/`: Contains additional resources for the model
+- `classes.json`: JSON file containing the mapping of class indices to disease names
+- `example.py`: Python script demonstrating how to use the model
+- `image.jpg`: Sample image for testing the model
+- `README.md`: This file, containing information about the model and how to use it
+- `saved_model.pb`: The saved TensorFlow model file
+- `variables/`: Directory containing model variables

classes.json

@@ -0,0 +1 @@
+{"Healthy": 0, "Gray_Leaf_Spot": 1, "Blight": 2, "Common_Rust": 3}

example.py

@@ -0,0 +1,23 @@
+
+# TF for image classification model
+
+import tensorflow
+import numpy
+from PIL import Image
+
+model = tensorflow.saved_model.load('./')
+classes = [ "Healthy" ,  "Gray_Leaf_Spot" ,  "Blight" ,  "Common_Rust" , ]
+
+img = Image.open("image.jpg").convert('RGB')
+img = img.resize((300, 300 * img.size[1] // img.size[0]), Image.ANTIALIAS)
+inp_numpy = numpy.array(img)[None]
+
+
+inp = tensorflow.constant(inp_numpy, dtype='float32')
+
+class_scores = model(inp)[0].numpy()
+
+
+print("")
+print("class_scores", class_scores)
+print("Class : ", classes[class_scores.argmax()])

saved_model.pb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0e36d12919e120792bdc0439994cc725f4a047fc3c1911a9b50dc35d90703709
+size 6428971

variables/variables.data-00000-of-00001

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bc4f206c90fb449f8987a2111dbbd7b8dbcf9b6d22bfed47b1f6b8a15fee4999
+size 18309528