first

app.py

@@ -0,0 +1,70 @@
+import gradio as gr
+import pandas as pd
+import joblib
+
+# Load scaler, encoder, and model
+scaler = joblib.load(r'toolkit\scaler.joblib')
+model = joblib.load(r'toolkit\model_final.joblib')
+encoder = joblib.load(r'toolkit\encoder.joblib')
+
+# Prediction function
+def predict(Gender, Urea, Cr, HbA1c, Chol, TG, HDL, LDL, VLDL, BMI):
+    # Create a DataFrame
+    input_df = pd.DataFrame({
+        'Gender': [Gender],
+        'Urea': [float(Urea)],
+        'Cr': [float(Cr)],
+        'HbA1c': [float(HbA1c)],
+        'Chol': [float(Chol)],
+        'TG': [float(TG)],
+        'HDL': [float(HDL)],
+        'LDL': [float(LDL)],
+        'VLDL': [float(VLDL)],
+        'BMI': [float(BMI)],
+    })
+
+    # Transform gender using the encoder
+    input_df['Gender'] = encoder.transform(input_df['Gender'])
+    # Apply scaler if necessary (uncomment if scaling is required)
+    # input_df = scaler.transform(input_df)
+
+    # Make prediction
+    prediction = model.predict(input_df)
+
+    # Prediction label mapping
+    prediction_label = {0: "No Diabetes", 1: "Prediabetic", 2: "Diabetic"}
+    return prediction_label[int(prediction[0])]
+
+# Gradio app
+with gr.Blocks(theme=gr.themes.Monochrome()) as app:
+    gr.Image("diabete1.jpg", label="Diabetes Prediction App")
+    gr.Markdown("# Diabetes Prediction App")
+    gr.Markdown(
+        "This app predicts a patient's diabetes status based on their health parameters. "
+        "Please provide the following inputs:"
+    )
+    
+    with gr.Row():
+        with gr.Column():
+            Gender = gr.Radio(['M', 'F'], label="Gender")
+            Urea = gr.Slider(0, 100, step=1, label="Urea (mg/dL)")
+            Cr = gr.Number(label="Creatinine (mg/dL)")
+            HbA1c = gr.Number(label="HbA1c (%)")
+            Chol = gr.Number(label="Cholesterol (mg/dL)")
+        with gr.Column():
+            TG = gr.Number(label="Triglycerides (mg/dL)")
+            HDL = gr.Slider(0, 100, step=1, label="HDL Cholesterol (mg/dL)")
+            LDL = gr.Number(label="LDL Cholesterol (mg/dL)")
+            VLDL = gr.Number(label="VLDL (mg/dL)")
+            BMI = gr.Number(label="Body Mass Index (BMI)")
+    
+    predict_btn = gr.Button("Predict")
+    output = gr.Label(label="Prediction Result")
+
+    predict_btn.click(
+        fn=predict,
+        inputs=[Gender, Urea, Cr, HbA1c, Chol, TG, HDL, LDL, VLDL, BMI],
+        outputs=output
+    )
+
+app.launch(share=True)

requirements.txt

@@ -0,0 +1,6 @@
+pandas
+numpy
+seaborn
+joblib
+gradio
+scikit-learn

toolkit/encoder.joblib

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a773365438957038e0df5904e618d7a28b7278b726f912ec705927efa28f2ba3
+size 534

toolkit/model_final.joblib

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a458b1320736be446b50fdb02ac9c530da95407970b402f09c4ad19b41d157cc
+size 776385

toolkit/scaler.joblib

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd31b5cd4ef3a901412794f8db4de41983e8847af2dce4902eaf373223d6da10
+size 1463