farmacy-plant-disease

Farmacy Plant Disease Detection Model

This repository contains plant disease detection models for multiple crops.

Available Models

Tomato Model

Training Details

• Dataset: 1,000 images (100 per class) from PlantVillage
• Model: EfficientNet-B0 (pretrained on ImageNet)
• Epochs: 10
• Validation Accuracy: 86.5%
• Test Accuracy: 90.0%

Classes (10 Tomato Diseases)

Potato Model

Training Details

• Dataset: 300 images (100 per class) from PlantVillage
• Model: EfficientNet-B0 (pretrained on ImageNet)
• Epochs: 10
• Validation Accuracy: 98.3%
• Test Accuracy: 100% (10 random images)

Classes (3 Potato Conditions)

Maize Model

Training Details

• Dataset: 400 images (100 per class) from PlantVillage
• Model: EfficientNet-B0 (pretrained on ImageNet)
• Epochs: 10
• Validation Accuracy: 95.0%
• Test Accuracy: 100% (10 random images)

Classes (4 Maize Conditions)

Usage

Tomato Model

import torch
import torch.nn as nn
from torchvision import transforms, models
from PIL import Image
from huggingface_hub import hf_hub_download

# Download the tomato model
model_path = hf_hub_download(
    repo_id="uWandile/farmacy-plant-disease",
    filename="tomato/final_model.pth"
)

# Load the model
model = models.efficientnet_b0(pretrained=False)
model.classifier = nn.Sequential(
    nn.Dropout(p=0.2),
    nn.Linear(model.classifier[1].in_features, 10)
)
model.load_state_dict(torch.load(model_path, map_location=torch.device('cpu')))
model.eval()

# Preprocess image
transform = transforms.Compose([
    transforms.Resize(256),
    transforms.CenterCrop(224),
    transforms.ToTensor(),
    transforms.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])
])

# Predict
img = Image.open("tomato_leaf.jpg").convert('RGB')
img_tensor = transform(img).unsqueeze(0)

with torch.no_grad():
    outputs = model(img_tensor)
    _, predicted = outputs.max(1)
    print(f"Prediction: {predicted.item()}")