Update app.py

app.py

@@ -20,6 +20,7 @@ import requests
 import asyncio
 from io import BytesIO
 import base64
+import time
 
 # Configuration
 st.set_page_config(page_title="Data Wizard Pro", layout="wide", page_icon="🧙")
@@ -545,102 +546,221 @@ elif app_mode == "Advanced EDA":
     if st.session_state.cleaned_data is not None:
         df = st.session_state.cleaned_data
 
-        # Data Filtering Section (Move to the top for more control)
+        # Initialize session state for plot configuration
+        if 'plot_config' not in st.session_state:
+            st.session_state.plot_config = {
+                'plot_type': "Histogram",
+                'x_col': None,
+                'y_col': None,
+                'z_col': None,
+                'color_col': None,
+                'size_col': None,
+                'time_col': None,
+                'value_col': None,
+                'scatter_matrix_cols': None,
+                'color_palette': "#00f7ff",
+                'color_continuous_scale': "Viridis",
+                'hover_data_cols': [],
+                'filter_col': None,
+                'filter_options': []
+            }
+
+        # Data Filtering Section
         with st.expander("🔎 Data Filtering", expanded=False):
-            filter_col = st.selectbox("Filter Column", [None] + list(df.columns), help="Choose a column to filter the data.")
-            if filter_col:
-                unique_values = df[filter_col].unique()
-                filter_options = st.multiselect("Filter Values", unique_values, default=unique_values, help=f"Select the values to include from the '{filter_col}' column.")
-                df = df[df[filter_col].isin(filter_options)]
+            filter_col = st.selectbox(
+                "Filter Column", 
+                [None] + list(df.columns), 
+                key='filter_col',
+                help="Choose a column to filter the data."
+            )
+            if st.session_state.plot_config['filter_col']:
+                unique_values = df[st.session_state.plot_config['filter_col']].unique()
+                filter_options = st.multiselect(
+                    "Filter Values", 
+                    unique_values, 
+                    default=unique_values, 
+                    key='filter_options',
+                    help=f"Select the values to include from the '{st.session_state.plot_config['filter_col']}' column."
+                )
+                df = df[df[st.session_state.plot_config['filter_col']].isin(st.session_state.plot_config['filter_options'])]
 
         # Visualization Selection and Configuration
-        st.sidebar.header("📊 Plot Configuration")  # Sidebar for settings
-
-        plot_type = st.sidebar.selectbox("Choose Visualization", [
-            "Histogram",
-            "Scatter Plot",
-            "Box Plot",
-            "Correlation Heatmap",
-            "3D Scatter",
-            "Violin Plot",
-            "Time Series",
-            "Scatter Matrix"  # ADDED
-        ], help="Select the type of plot to generate.")
-
-        x_col = None  # Initialize x_col
-        y_col = None  # Initialize y_col
-        z_col = None  # Initialize z_col
-        color_col = "#FF6347" # Intialize color column
-        size_col = None # Initailize size column
-        time_col = None
-        value_col = None
-        scatter_matrix_cols = None
-        color_palette = ["#FF6347", "#4682B4", "#32CD32", "#FFD700"]
-
-
-        if plot_type != "Correlation Heatmap":
-            x_col = st.sidebar.selectbox("X Axis", df.columns, help="Select the column for the x-axis.")
-
-        if plot_type in ["Scatter Plot", "Box Plot", "Violin Plot", "Time Series", "3D Scatter", "Histogram"]:
-            y_col = st.sidebar.selectbox("Y Axis", df.columns, help="Select the column for the y-axis.")
-
-        if plot_type == "3D Scatter":
-            z_col = st.sidebar.selectbox("Z Axis", df.columns, help="Select the column for the z-axis.")
-            color_col = st.sidebar.selectbox("Color by", [None] + list(df.columns), help="Optional column to color the data points.")
-
-        # Add a parameter for Color Scales in Heatmap (and make palette choice for other plots)
-        if plot_type == "Correlation Heatmap":
-          color_continuous_scale = st.sidebar.selectbox("Color Scale", ['Viridis', 'Plasma', 'Magma', 'Cividis', 'RdBu'], help="Select the color scale for the heatmap.")
-        else:
-            color_palette = st.sidebar.selectbox("Color Palette", ['#00f7ff', '#ff00ff', '#f70000', '#0000f7'], help="Select the color for the plot.")
-
-        if plot_type == "Scatter Plot":
-            size_col = st.sidebar.selectbox("Size by", [None] + list(df.columns), help="Optional column to size the data points.")
-            hover_data_cols = st.sidebar.multiselect("Hover Data", df.columns, help="Optional columns to display on hover.")
-
-        if plot_type == "Time Series":
-            time_col = st.sidebar.selectbox("Time Column", df.columns, help="Column representing time.")
-            value_col = st.sidebar.selectbox("Value Column", df.columns, help="Column representing the value to plot over time.")
-
-        if plot_type == "Scatter Matrix":
-            scatter_matrix_cols = st.multiselect("Columns for Scatter Matrix", df.columns, default=df.columns.tolist()[:5], help="Select the columns to include in the scatter matrix.")
-
-        # Generate Plot
-        if st.button("Generate Visualization"):
-            try:  # add try-except block for potential errors
-                fig = None  # Initialize fig
-                if plot_type == "Histogram":
-                    fig = px.histogram(df, x=x_col, y=y_col, nbins=30, template="plotly_dark", color_discrete_sequence=[color_palette])
-                elif plot_type == "Scatter Plot":
-                    fig = px.scatter(df, x=x_col, y=y_col, color_discrete_sequence=[color_palette], size=size_col, hover_data=hover_data_cols)
-                elif plot_type == "3D Scatter":
-                    fig = px.scatter_3d(df, x=x_col, y=y_col, z=z_col, color=color_col, color_discrete_sequence=[color_palette])
-                elif plot_type == "Correlation Heatmap":
-                    corr = df.corr(numeric_only=True) #handle non-numeric cols
-                    fig = px.imshow(corr, text_auto=True, color_continuous_scale=color_continuous_scale)
-                elif plot_type == "Box Plot":
-                    fig = px.box(df,x=x_col, y=y_col, color_discrete_sequence=[color_palette])
-                elif plot_type == "Violin Plot":
-                    fig = px.violin(df, x=x_col, y=y_col, color_discrete_sequence=[color_palette])
-                elif plot_type == "Time Series":
-                     fig = px.line(df, x=time_col, y=value_col, color_discrete_sequence=[color_palette])
-                elif plot_type == "Scatter Matrix":
-                    if len(scatter_matrix_cols) > 1:
-                        fig = px.scatter_matrix(df[scatter_matrix_cols], color_discrete_sequence=[color_palette])
-                    else:
-                        st.error("Please select at least two columns for the scatter matrix.")
+        st.sidebar.header("📊 Plot Configuration")
+
+        # Plot type selection
+        plot_type = st.sidebar.selectbox(
+            "Choose Visualization", 
+            [
+                "Histogram", "Scatter Plot", "Box Plot", "Correlation Heatmap", 
+                "3D Scatter", "Violin Plot", "Time Series", "Scatter Matrix"
+            ], 
+            key='plot_type',
+            help="Select the type of plot to generate."
+        )
 
-                if fig:
-                    fig.update_layout(
-                        plot_bgcolor="#1e1e30",
-                        paper_bgcolor="#1e1e30",
-                        font_color="#e0e0ff"
+        # Dynamic axis and configuration options
+        if st.session_state.plot_config['plot_type'] != "Correlation Heatmap":
+            x_col = st.sidebar.selectbox(
+                "X Axis", 
+                df.columns, 
+                key='x_col',
+                help="Select the column for the x-axis."
+            )
+
+        if st.session_state.plot_config['plot_type'] in ["Scatter Plot", "Box Plot", "Violin Plot", "Time Series", "3D Scatter", "Histogram"]:
+            y_col = st.sidebar.selectbox(
+                "Y Axis", 
+                df.columns, 
+                key='y_col',
+                help="Select the column for the y-axis."
+            )
+
+        if st.session_state.plot_config['plot_type'] == "3D Scatter":
+            z_col = st.sidebar.selectbox(
+                "Z Axis", 
+                df.columns, 
+                key='z_col',
+                help="Select the column for the z-axis."
+            )
+            color_col = st.sidebar.selectbox(
+                "Color by", 
+                [None] + list(df.columns), 
+                key='color_col',
+                help="Optional column to color the data points."
+            )
+
+        # Color configuration
+        if st.session_state.plot_config['plot_type'] == "Correlation Heatmap":
+            color_continuous_scale = st.sidebar.selectbox(
+                "Color Scale", 
+                ['Viridis', 'Plasma', 'Magma', 'Cividis', 'RdBu'], 
+                key='color_continuous_scale',
+                help="Select the color scale for the heatmap."
+            )
+        else:
+            color_palette = st.sidebar.selectbox(
+                "Color Palette", 
+                ['#00f7ff', '#ff00ff', '#f70000', '#0000f7'], 
+                key='color_palette',
+                help="Select the color for the plot."
+            )
+
+        # Additional plot-specific configurations
+        if st.session_state.plot_config['plot_type'] == "Scatter Plot":
+            size_col = st.sidebar.selectbox(
+                "Size by", 
+                [None] + list(df.columns), 
+                key='size_col',
+                help="Optional column to size the data points."
+            )
+            hover_data_cols = st.sidebar.multiselect(
+                "Hover Data", 
+                df.columns, 
+                key='hover_data_cols',
+                help="Optional columns to display on hover."
+            )
+
+        if st.session_state.plot_config['plot_type'] == "Time Series":
+            time_col = st.sidebar.selectbox(
+                "Time Column", 
+                df.columns, 
+                key='time_col',
+                help="Column representing time."
+            )
+            value_col = st.sidebar.selectbox(
+                "Value Column", 
+                df.columns, 
+                key='value_col',
+                help="Column representing the value to plot over time."
+            )
+
+        if st.session_state.plot_config['plot_type'] == "Scatter Matrix":
+            scatter_matrix_cols = st.multiselect(
+                "Columns for Scatter Matrix", 
+                df.columns, 
+                default=df.columns.tolist()[:5], 
+                key='scatter_matrix_cols',
+                help="Select the columns to include in the scatter matrix."
+            )
+
+        # Generate and update plot in real-time
+        try:
+            fig = None
+            if st.session_state.plot_config['plot_type'] == "Histogram":
+                fig = px.histogram(
+                    df, 
+                    x=st.session_state.plot_config['x_col'], 
+                    y=st.session_state.plot_config['y_col'], 
+                    nbins=30, 
+                    template="plotly_dark", 
+                    color_discrete_sequence=[st.session_state.plot_config['color_palette']]
+                )
+            elif st.session_state.plot_config['plot_type'] == "Scatter Plot":
+                fig = px.scatter(
+                    df, 
+                    x=st.session_state.plot_config['x_col'], 
+                    y=st.session_state.plot_config['y_col'], 
+                    color_discrete_sequence=[st.session_state.plot_config['color_palette']], 
+                    size=st.session_state.plot_config['size_col'], 
+                    hover_data=st.session_state.plot_config['hover_data_cols']
+                )
+            elif st.session_state.plot_config['plot_type'] == "3D Scatter":
+                fig = px.scatter_3d(
+                    df, 
+                    x=st.session_state.plot_config['x_col'], 
+                    y=st.session_state.plot_config['y_col'], 
+                    z=st.session_state.plot_config['z_col'], 
+                    color=st.session_state.plot_config['color_col'], 
+                    color_discrete_sequence=[st.session_state.plot_config['color_palette']]
+                )
+            elif st.session_state.plot_config['plot_type'] == "Correlation Heatmap":
+                corr = df.corr(numeric_only=True)
+                fig = px.imshow(
+                    corr, 
+                    text_auto=True, 
+                    color_continuous_scale=st.session_state.plot_config['color_continuous_scale']
+                )
+            elif st.session_state.plot_config['plot_type'] == "Box Plot":
+                fig = px.box(
+                    df, 
+                    x=st.session_state.plot_config['x_col'], 
+                    y=st.session_state.plot_config['y_col'], 
+                    color_discrete_sequence=[st.session_state.plot_config['color_palette']]
+                )
+            elif st.session_state.plot_config['plot_type'] == "Violin Plot":
+                fig = px.violin(
+                    df, 
+                    x=st.session_state.plot_config['x_col'], 
+                    y=st.session_state.plot_config['y_col'], 
+                    color_discrete_sequence=[st.session_state.plot_config['color_palette']]
+                )
+            elif st.session_state.plot_config['plot_type'] == "Time Series":
+                fig = px.line(
+                    df, 
+                    x=st.session_state.plot_config['time_col'], 
+                    y=st.session_state.plot_config['value_col'], 
+                    color_discrete_sequence=[st.session_state.plot_config['color_palette']]
+                )
+            elif st.session_state.plot_config['plot_type'] == "Scatter Matrix":
+                if len(st.session_state.plot_config['scatter_matrix_cols']) > 1:
+                    fig = px.scatter_matrix(
+                        df[st.session_state.plot_config['scatter_matrix_cols']], 
+                        color_discrete_sequence=[st.session_state.plot_config['color_palette']]
                     )
-                    st.plotly_chart(fig, use_container_width=True)
-
-            except Exception as e:
-                st.error(f"Error generating plot: {e}")
+                else:
+                    st.error("Please select at least two columns for the scatter matrix.")
+
+            if fig:
+                fig.update_layout(
+                    plot_bgcolor="#1e1e30",
+                    paper_bgcolor="#1e1e30",
+                    font_color="#e0e0ff"
+                )
+                st.plotly_chart(fig, use_container_width=True)
 
+        except Exception as e:
+            st.error(f"Error generating plot: {e}")
+            
 elif app_mode == "Model Training":
     st.title("🤖 Model Training Studio")