Create app.py

app.py

@@ -0,0 +1,76 @@
+import streamlit as st
+import json
+import numpy as np
+import tensorflow as tf
+import gdown
+import zipfile
+
+# Cache the model loading
+@st.cache(allow_output_mutation=True)
+def load_model():
+    # The shareable link to your Google Drive file
+    url = "https://drive.google.com/uc?id=1m9YVs0cBRT3-j98rn7d_0DT7jwB_EXPu"
+    output = 'model.zip'
+    gdown.download(url, output, quiet=False)
+    
+    with zipfile.ZipFile(output, 'r') as zip_ref:
+        zip_ref.extractall(".")
+    
+    model = tf.keras.models.load_model('best_model')
+    return model
+
+# Cache the JSON data loading
+@st.cache(allow_output_mutation=True)
+def load_json(filename):
+    with open(filename, 'r') as f:
+        return json.load(f)
+
+user_db = load_json("user_db.json")
+item_db = load_json("item_db.json")
+
+# Function to predict rating
+def predict_rating(reviewerID, itemID, model):
+    item_attributes = item_db.get(itemID, {})
+    user_attributes = user_db.get(reviewerID, {})
+    
+    category = item_attributes.get('category', 4)  # Assuming default values
+    price = item_attributes.get('price', 13.71)
+    userAvgRating = user_attributes.get('userAvgRating', 4)
+    itemAvgRating = item_attributes.get('itemAvgRating', 4)
+    review_time = user_attributes.get('unixReviewTime', 1285579290)
+
+    prediction_inputs = {
+        'reviewer_id': np.array([[reviewerID]]),
+        'item_id': np.array([[itemID]]),
+        'category': np.array([[category]]),
+        'price': np.array([[price]]),
+        'paid_price': np.array([[price]]),
+        "unixReviewTime": np.array([[review_time]]),
+        'userAvgRating': np.array([[userAvgRating]]),
+        'itemAvgRating': np.array([[itemAvgRating]]),
+        'review_text': np.array([["placeholder text"]]),
+        'summary': np.array([["placeholder summary"]]),
+    }
+    
+    # Transform the input dictionary as needed to match your model's expected input format
+    prediction = model.predict(prediction_inputs)
+    return prediction[0][0]
+
+# Load the model
+model = load_model()
+
+# Streamlit app interface
+st.title('Product Rating Prediction')
+
+# Example input values
+example_reviewerID = "A1YS9MDZP93857"  # Example reviewerID
+example_itemID = "B001GVISJM"  # Example itemID
+
+# User inputs with examples
+reviewerID = st.text_input('Reviewer ID', value=example_reviewerID)
+itemID = st.text_input('Item ID', value=example_itemID)
+
+# Predict button
+if st.button('Predict Rating'):
+    prediction = predict_rating(reviewerID, itemID, model)
+    st.write(f'Predicted Rating: {prediction:.2f}')