Upload 6 files

SVR_model.joblib

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:04f78b940f119368c8d63d13453c7bd677071249ac85714f5efd4d8e6e645b65
+size 2099780

app.py

@@ -0,0 +1,100 @@
+# -*- coding: utf-8 -*-
+"""app.ipynb
+
+Automatically generated by Colab.
+
+Original file is located at
+    https://colab.research.google.com/drive/1oJP09Coya1D16dQ_7fpVGlbuLnCWxHTN
+"""
+
+import gradio as gr
+import joblib
+import numpy as np
+from tensorflow.keras.models import load_model
+
+# Load models
+svr_model = joblib.load('SVR_model.joblib')
+rf_model = joblib.load('RandomForestRegressor_model.joblib')
+dt_model = joblib.load('best_DecisionTreeRegressor_model.joblib')
+dl_model = load_model('best_DeepLearning_model.h5')
+
+# Load preprocessing objects
+label_encoder = joblib.load('label_encoder.joblib')
+column_transformer = joblib.load('column_transformer.joblib')
+
+def predict_warfarin_dose(gender, race, age, height, weight, diabetes,
+                         simvastatin, amiodarone, inr_reported,
+                         cyp2c9, vkorc1, model_choice):
+    try:
+        # Encode Age
+        age_encoded = label_encoder.transform([age])
+
+        # Create input list
+        inputs = [
+            str(gender),
+            str(race),
+            str(age),
+            float(height) if height is not None else 0.0,
+            float(weight) if weight is not None else 0.0,
+            float(diabetes),
+            float(simvastatin),
+            float(amiodarone),
+            float(inr_reported) if inr_reported is not None else 0.0,
+            str(cyp2c9),
+            str(vkorc1)
+        ]
+
+        # Transform inputs
+        inputs_transformed = column_transformer.transform([inputs])
+        inputs_transformed[0][-7] = age_encoded[0]
+        input_data = np.array(inputs_transformed, dtype=np.float32)
+
+        # Make prediction based on model choice
+        if model_choice == 'Deep Learning':
+            prediction = dl_model.predict(input_data)[0][0]
+        elif model_choice == 'Support Vector Regression':
+            prediction = svr_model.predict(input_data)[0]
+        elif model_choice == 'Random Forest':
+            prediction = rf_model.predict(input_data)[0]
+        else:
+            prediction = dt_model.predict(input_data)[0]
+
+        return f"Predicted Warfarin Dose: {prediction:.2f} mg/week"
+
+    except Exception as e:
+        return f"Error in prediction: {str(e)}"
+
+# Create the Gradio interface
+iface = gr.Interface(
+    fn=predict_warfarin_dose,
+    inputs=[
+        gr.Radio(["male", "female"], label="Gender"),
+        gr.Dropdown(["Asian", "Black", "White", "Unknown", "Mixed or Missing"], label="Race"),
+        gr.Dropdown(["0-9", "10-19", "20-29", "30-39", "40-49", "50-59", "60-69",
+                    "70-79", "80-89", "90+"], label="Age"),
+        gr.Number(label="Height (cm)"),
+        gr.Number(label="Weight (kg)"),
+        gr.Radio([0.0, 1.0], label="Diabetes"),
+        gr.Radio([0.0, 1.0], label="Simvastatin (Zocor)"),
+        gr.Radio([0.0, 1.0], label="Amiodarone (Cordarone)"),
+        gr.Number(label="INR on Reported Therapeutic Dose of Warfarin"),
+        gr.Dropdown(["*1/*1", "*1/*2", "*1/*3", "*2/*2", "*2/*3", "*3/*3"],
+                   label="Cyp2C9 genotypes"),
+        gr.Radio(["A/A", "A/G", "G/G"], label="VKORC1 genotypes"),
+        gr.Dropdown(['Decision Tree', 'Support Vector Regression',
+                    'Random Forest', 'Deep Learning'], label="Model Selection")
+    ],
+    outputs=gr.Textbox(label="Prediction Result"),
+    title="Warfarin Dosage Prediction System",
+    description="""This system predicts the optimal warfarin dosage based on patient characteristics.
+                   Enter the required information below and select a model for prediction.""",
+    examples=[
+        ["male", "Asian", "50-59", 170, 70, 0.0, 0.0, 0.0, 2.5, "*1/*1", "A/G", "Random Forest"],
+        ["female", "White", "60-69", 165, 65, 1.0, 1.0, 0.0, 2.8, "*1/*2", "G/G", "Deep Learning"]
+    ],
+    theme="default"
+)
+
+# Launch the interface
+if __name__ == "__main__":
+    iface.launch()

best_DecisionTreeRegressor_model.joblib

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d64a6f6fe40203c967075341ee946923648853e2cc88da6e99e6b518f351f046
+size 165137

best_DeepLearning_model.h5

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4937dd6061b79dac101bd26ea27b5b2bd1216ace8a3c8da7244df95a8904e392
+size 102568

column_transformer.joblib

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b4d987360470bb30eaa77602946506da66261028184ca164d1d333debe7d900
+size 4719

label_encoder.joblib

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5c20f850cce07c65075ca30ae733aa76faa2068edd9e1d10e6a0d969fb40f88e
+size 640