File size: 3,490 Bytes
92495fb |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 |
"""
AgriBot Maize Disease Diagnosis - Inference Example
Ekip Crusaders - AYITI IA 2025 Hackathon
"""
import joblib
import torch
import numpy as np
from PIL import Image
from torchvision import models, transforms
import sys
from pathlib import Path
class AgribotInference:
"""Wrapper class for AgriBot model inference"""
def __init__(self, model_path='agribot_models.pkl'):
print("Loading AgriBot model...")
self.model = joblib.load(model_path)
print("Loading MobileNetV2 feature extractor...")
self.mobilenet = models.mobilenet_v2(pretrained=True)
self.mobilenet.classifier = torch.nn.Identity()
self.mobilenet.eval()
self.class_labels = [
"Cercospora Leaf Spot (Gray Leaf Spot)",
"Common Rust",
"Northern Leaf Blight",
"Healthy",
"Other"
]
self.transform = transforms.Compose([
transforms.Resize((224, 224)),
transforms.ToTensor(),
transforms.Normalize(
mean=[0.485, 0.456, 0.406],
std=[0.229, 0.224, 0.225]
)
])
print("Model loaded successfully!")
def preprocess_image(self, image_path):
image = Image.open(image_path).convert('RGB')
return self.transform(image).unsqueeze(0)
def extract_features(self, img_tensor):
with torch.no_grad():
features = self.mobilenet(img_tensor)
return features.numpy()
def predict(self, image_path):
img_tensor = self.preprocess_image(image_path)
features = self.extract_features(img_tensor)
prediction_idx = self.model.predict(features)[0]
probabilities = self.model.predict_proba(features)[0]
confidence = float(np.max(probabilities) * 100)
diagnosis = self.class_labels[prediction_idx]
prob_dist = {
label: float(prob * 100)
for label, prob in zip(self.class_labels, probabilities)
}
return {
'diagnosis': diagnosis,
'confidence': round(confidence, 2),
'prediction_index': int(prediction_idx),
'probabilities': prob_dist,
'is_healthy': diagnosis == "Healthy",
'is_maize': diagnosis != "Other"
}
def main():
if len(sys.argv) < 2:
print("Usage: python inference_example.py <image_path>")
print("Example: python inference_example.py maize_leaf.jpg")
sys.exit(1)
image_path = sys.argv[1]
if not Path(image_path).exists():
print(f"Error: Image file '{image_path}' not found!")
sys.exit(1)
predictor = AgribotInference()
print(f"\nAnalyzing image: {image_path}")
print("-" * 50)
result = predictor.predict(image_path)
print(f"\n🌽 DIAGNOSIS: {result['diagnosis']}")
print(f"📊 Confidence: {result['confidence']:.2f}%")
print(f"🔍 Is Maize: {'Yes' if result['is_maize'] else 'No'}")
print(f"✅ Is Healthy: {'Yes' if result['is_healthy'] else 'No'}")
print("\n📈 Probability Distribution:")
for disease, prob in result['probabilities'].items():
bar_length = int(prob / 2)
bar = "█" * bar_length
print(f" {disease:40s} {prob:6.2f}% {bar}")
print("\n" + "-" * 50)
print("Analysis complete!")
if __name__ == "__main__":
main()
|