Update app.py

app.py

@@ -3,58 +3,72 @@ import pandas as pd
 import numpy as np
 from sklearn.model_selection import train_test_split
 from sklearn.ensemble import RandomForestRegressor
-from sklearn.metrics import mean_absolute_error, mean_squared_error, r2_score
-import matplotlib.pyplot as plt
+from sklearn.metrics import mean_squared_error
 
-# Title
-st.title("AI-Powered Load Forecasting")
+# Title of the Streamlit app
+st.title("Load Forecasting Application")
+
+# File upload section
+uploaded_file = st.file_uploader("Upload a CSV file containing historical load data", type=["csv"])
 
-# Upload Dataset
-st.sidebar.header("Upload Data")
-uploaded_file = st.sidebar.file_uploader("Upload your CSV file", type=["csv"])
 if uploaded_file is not None:
+    # Load the dataset
     data = pd.read_csv(uploaded_file)
-    st.write("Data Preview")
-    st.write(data.head())
-else:
-    st.info("Awaiting CSV file upload. You can use the sample dataset.")
-    # Load sample data
-    data = pd.read_csv("sample_data.csv")
-    st.write("Using Sample Data")
+    st.write("Preview of the uploaded data:")
     st.write(data.head())
 
-# Feature Selection
-st.sidebar.header("Feature Selection")
-target_variable = st.sidebar.selectbox("Select Target Variable", options=data.columns, index=len(data.columns) - 1)
-predictors = st.sidebar.multiselect("Select Predictor Variables", options=[col for col in data.columns if col != target_variable], default=data.columns[:-1])
+    # Ensure the date column is in datetime format
+    if 'date' in data.columns:
+        data['date'] = pd.to_datetime(data['date'])
+        
+        # Extract useful features from the date column
+        data['year'] = data['date'].dt.year
+        data['month'] = data['date'].dt.month
+        data['day'] = data['date'].dt.day
+        data['day_of_week'] = data['date'].dt.dayofweek
+        
+        # Drop the original date column
+        data = data.drop(columns=['date'])
 
-# Model Training
-if st.sidebar.button("Train Model"):
-    st.subheader("Training the Model...")
+    # Check for missing values
+    if data.isnull().sum().sum() > 0:
+        st.write("The dataset contains missing values. They will be filled with the mean.")
+        data = data.fillna(data.mean())
 
-    # Split data
-    X = data[predictors]
-    y = data[target_variable]
+    # Define features and target variable
+    X = data.drop(columns=['load'])  # Features (exclude the target 'load')
+    y = data['load']  # Target variable
+
+    # Split the data 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 Random Forest Regressor
+    # Model training
     model = RandomForestRegressor(n_estimators=100, random_state=42)
     model.fit(X_train, y_train)
+
+    # Model prediction
     y_pred = model.predict(X_test)
 
-    # Model Evaluation
-    st.write("**Evaluation Metrics:**")
-    st.write(f"Mean Absolute Error (MAE): {mean_absolute_error(y_test, y_pred):.2f}")
-    st.write(f"Mean Squared Error (MSE): {mean_squared_error(y_test, y_pred):.2f}")
-    st.write(f"R² Score: {r2_score(y_test, y_pred):.2f}")
-
-    # Plot Results
-    fig, ax = plt.subplots()
-    ax.plot(y_test.values, label="Actual", marker="o")
-    ax.plot(y_pred, label="Predicted", marker="x")
-    ax.legend()
-    ax.set_title("Actual vs. Predicted Load")
-    st.pyplot(fig)
-
-# Footer
-st.sidebar.markdown("Developed by [Sunny Nazir](https://huggingface.co/spaces)")
+    # Calculate and display performance metrics
+    mse = mean_squared_error(y_test, y_pred)
+    st.write(f"Mean Squared Error (MSE): {mse:.2f}")
+
+    # Feature importance
+    feature_importance = pd.DataFrame({
+        'Feature': X.columns,
+        'Importance': model.feature_importances_
+    }).sort_values(by='Importance', ascending=False)
+
+    st.write("Feature Importance:")
+    st.write(feature_importance)
+
+    # Future prediction
+    st.write("## Predict Future Load")
+    user_input = {}
+    for feature in X.columns:
+        user_input[feature] = st.number_input(f"Enter value for {feature}")
+
+    if st.button("Predict"):
+        input_data = np.array([list(user_input.values())]).reshape(1, -1)
+        prediction = model.predict(input_data)
+        st.write(f"Predicted Load: {prediction[0]:.2f}")