Create app.py

app.py

@@ -0,0 +1,71 @@
+import pandas as pd
+import numpy as np
+import joblib
+from sklearn.ensemble import RandomForestClassifier
+from sklearn.preprocessing import LabelEncoder
+from sklearn.model_selection import train_test_split
+
+# Load dataset
+df = pd.read_excel("mask_dataset.xlsx")
+
+# Encode categorical features
+label_encoder_tone = LabelEncoder()
+df["skin_tone"] = label_encoder_tone.fit_transform(df["skin_tone"])
+
+label_encoder_shape = LabelEncoder()
+df["face_shape"] = label_encoder_shape.fit_transform(df["face_shape"])
+
+label_encoder_mask = LabelEncoder()
+df["recommended_mask"] = label_encoder_mask.fit_transform(df["recommended_mask"])
+
+# Prepare data for training
+X = df[["skin_tone", "face_shape"]].values
+y = df["recommended_mask"].values
+
+# Split into training and testing sets
+X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
+
+# Train model
+model = RandomForestClassifier(n_estimators=100)
+model.fit(X_train, y_train)
+
+# Save the trained model
+joblib.dump(model, "mask_recommender.pkl")
+joblib.dump(label_encoder_mask, "label_encoder_mask.pkl")
+
+print("✅ Model trained & saved!")
+import joblib
+
+# Load trained model
+model = joblib.load("mask_recommender.pkl")
+label_encoder_mask = joblib.load("label_encoder_mask.pkl")
+
+def recommend_mask(face_image):
+    """Predict the best mask for the uploaded face."""
+    skin_tone = detect_skin_tone(face_image)
+    face_shape = detect_face_shape(face_image)
+
+    # Encode features
+    tone_encoded = label_encoder_tone.transform([skin_tone])[0]
+    shape_encoded = label_encoder_shape.transform([face_shape])[0]
+
+    # Predict mask
+    predicted_mask = model.predict([[tone_encoded, shape_encoded]])[0]
+    recommended_mask = label_encoder_mask.inverse_transform([predicted_mask])[0]
+
+    return recommended_mask
+    def apply_ai_mask(person_img, seed, randomize_seed):
+    """Predicts & applies AI-recommended mask to a face image."""
+    recommended_mask = recommend_mask(person_img)
+    
+    mask_path = f"masks/{recommended_mask}.png"
+    mask_img = Image.open(mask_path)
+
+    return apply_mask(person_img, mask_img, seed, randomize_seed)
+    gr.Interface(
+    fn=apply_ai_mask,
+    inputs=gr.Image(type="pil", label="Upload Your Face"),
+    outputs=[gr.Image(label="Masked Result"), gr.Textbox(label="Recommended Mask")],
+    title="AI Mask Suggestion",
+    description="Upload your face and let AI suggest the perfect party mask!"
+).launch()