Update app.py

app.py

@@ -1,75 +1,86 @@
 import streamlit as st
 import pandas as pd
 from utils.data_cleaning import handle_missing_values, remove_outliers_iqr, cap_extreme_values
-from utils.visualizations import plot_correlation_heatmap, save_plot_as_png
+from utils.visualizations import plot_correlation_heatmap
 from utils.model_training import train_all_models
 import io
 
 # Streamlit App Title
-st.title("Model Training with Metrics and Correlation Heatmap")
+st.title("Data Analysis, Model Training, and Visualization")
 
-# File uploader with unique keys
-uploaded_file = st.file_uploader("Upload a CSV file for data analysis", type=["csv"], key="file_uploader_1")
+# File Uploader
+uploaded_file = st.file_uploader("Upload a CSV file for data analysis", type=["csv"])
 
 if uploaded_file is not None:
-    # Read the uploaded file into a DataFrame
+    # Load dataset
     df = pd.read_csv(uploaded_file)
-    st.write("Dataset Uploaded Successfully!")
+    st.write("### Dataset Preview")
     st.dataframe(df)
 
-    # Clean Data: Missing values, outliers, and extreme values
-    st.subheader("Data Cleaning")
-    df_cleaned = handle_missing_values(df)
-    df_cleaned = remove_outliers_iqr(df_cleaned)
-    df_cleaned = cap_extreme_values(df_cleaned)
+    try:
+        # Data Cleaning
+        st.subheader("Data Cleaning")
+        st.write("Handling missing values, removing outliers, and capping extreme values...")
+        df_cleaned = handle_missing_values(df)
+        df_cleaned = remove_outliers_iqr(df_cleaned)
+        df_cleaned = cap_extreme_values(df_cleaned)
 
-    st.write("Cleaned Dataset:")
-    st.dataframe(df_cleaned)
+        st.write("### Cleaned Dataset")
+        st.dataframe(df_cleaned)
 
-    # Add download option for the cleaned dataset
-    st.subheader("Download Cleaned Dataset")
-    st.download_button(
-        label="Download Cleaned Dataset (CSV)",
-        data=df_cleaned.to_csv(index=False),
-        file_name="cleaned_dataset.csv",
-        mime="text/csv"
-    )
+        # Download option for cleaned dataset
+        st.download_button(
+            label="Download Cleaned Dataset (CSV)",
+            data=df_cleaned.to_csv(index=False),
+            file_name="cleaned_dataset.csv",
+            mime="text/csv"
+        )
 
-    # Correlation Heatmap
-    st.subheader("Correlation Heatmap")
-    corr_plot = plot_correlation_heatmap(df_cleaned)
-    st.pyplot(corr_plot)
+        # Correlation Heatmap
+        st.subheader("Correlation Heatmap")
+        st.write("Visualizing correlations between numeric features...")
+        heatmap_plot = plot_correlation_heatmap(df_cleaned)
+        st.pyplot(heatmap_plot)
 
-    # Save heatmap as PNG
-    heatmap_buffer = io.BytesIO()
-    corr_plot.savefig(heatmap_buffer, format='png')
-    heatmap_buffer.seek(0)
+        # Save and download heatmap as PNG
+        heatmap_buffer = io.BytesIO()
+        heatmap_plot.savefig(heatmap_buffer, format="png")
+        heatmap_buffer.seek(0)
 
-    # Download Button for Heatmap
-    st.download_button(
-        label="Download Correlation Heatmap as PNG",
-        data=heatmap_buffer,
-        file_name="correlation_heatmap.png",
-        mime="image/png"
-    )
+        st.download_button(
+            label="Download Correlation Heatmap (PNG)",
+            data=heatmap_buffer,
+            file_name="correlation_heatmap.png",
+            mime="image/png"
+        )
 
-    # Target and Feature Selection
-    st.subheader("Select Target and Features")
-    target = st.selectbox("Select Target Variable", df_cleaned.columns)
-    features = [col for col in df_cleaned.columns if col != target]
-    X = df_cleaned[features]
-    y = df_cleaned[target]
+        # Select Target and Features
+        st.subheader("Feature and Target Selection")
+        target = st.selectbox("Select Target Variable", df_cleaned.columns)
+        features = [col for col in df_cleaned.columns if col != target]
 
-    # Train and Evaluate Models
-    st.subheader("Model Training and Evaluation")
-    model_results = train_all_models(X, y)  # Train all models based on data type
-    st.write("Model Training Results:")
-    st.dataframe(model_results)
+        if not features:
+            st.warning("No features available after removing the target variable.")
+        else:
+            X = df_cleaned[features]
+            y = df_cleaned[target]
 
-    # Add download option for model results
-    st.download_button(
-        label="Download Model Results (CSV)",
-        data=model_results.to_csv(),
-        file_name="model_results.csv",
-        mime="text/csv"
-    )
+            # Train and Evaluate Models
+            st.subheader("Model Training and Evaluation")
+            st.write("Training models and calculating metrics...")
+            model_results = train_all_models(X, y)
+
+            st.write("### Model Training Results")
+            st.dataframe(model_results)
+
+            # Download option for model results
+            st.download_button(
+                label="Download Model Results (CSV)",
+                data=model_results.to_csv(index=False),
+                file_name="model_results.csv",
+                mime="text/csv"
+            )
+    except Exception as e:
+        st.error(f"An error occurred: {e}")
+else:
+    st.info("Please upload a CSV file to proceed.")