Update app.py

app.py

@@ -1,13 +1,16 @@
+import os
+import cv2
 import joblib
 import numpy as np
 import pandas as pd
-import cv2
+import gradio as gr
 from skimage.color import rgb2hsv
 from skimage.measure import shannon_entropy
 from scipy.ndimage import generic_filter
 
-# Extract features from the image (same as your previous code)
+# Function to extract features from the image
 def extract_features(image_path):
+    # Load image and convert to RGB
     image = cv2.imread(image_path)
     image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
     
@@ -55,7 +58,7 @@ def extract_features(image_path):
     g4 = generic_filter(gray_image, g4_filter, size=3).mean()
     g5 = generic_filter(gray_image, g5_filter, size=3).mean()
 
-    # Return features
+    # Return extracted features
     return {
         "meanr": meanr,
         "meang": meang,
@@ -70,10 +73,10 @@ def extract_features(image_path):
         "g5": g5,
     }
 
-# Function to make predictions
-def predict_hemoglobin(age, gender, image_path):
+# Function to predict hemoglobin
+def predict_hemoglobin(age, gender, image):
     # Extract features from the image
-    features = extract_features(image_path)
+    features = extract_features(image.name)  # Get the image file path
     
     # Add age and gender to the features
     features['age'] = age
@@ -82,17 +85,37 @@ def predict_hemoglobin(age, gender, image_path):
     # Convert features to DataFrame
     features_df = pd.DataFrame([features])
     
-    # Load the pre-trained models
+    # Load pre-trained models
     svr_model = joblib.load('svr_model.pkl')  # SVR model
     scaler = joblib.load('minmax_scaler.pkl')        # MinMaxScaler
     label_encoder = joblib.load('label_encoder.pkl')  # LabelEncoder
 
-    # Apply MinMaxScaler and LabelEncoder transformations
-    # For age and gender, you can scale them or leave them as-is, depending on your training procedure
+    # Apply MinMaxScaler transformation
     features_df_scaled = scaler.transform(features_df)
     
-    # Make the prediction
+    # Predict hemoglobin value
     hemoglobin = svr_model.predict(features_df_scaled)[0]
     
-    return hemoglobin
+    return f"Predicted Hemoglobin Value: {hemoglobin:.2f}"
+
+# Gradio Interface
+def create_gradio_interface():
+    # Define the input components (age, gender, and image)
+    age_input = gr.Number(label="Age", value=25, interactive=True)
+    gender_input = gr.Dropdown(choices=["Male", "Female"], label="Gender", value="Male", interactive=True)
+    image_input = gr.Image(type="file", label="Image (Path to Image)", interactive=True)
+    
+    # Define the output component (prediction result)
+    output = gr.Textbox(label="Hemoglobin Prediction")
+    
+    # Create the Gradio interface
+    gr.Interface(
+        fn=predict_hemoglobin,
+        inputs=[age_input, gender_input, image_input],
+        outputs=output,
+        live=True
+    ).launch()
 
+# Start the Gradio app
+if __name__ == "__main__":
+    create_gradio_interface()