import streamlit as st import pandas as pd import plotly.express as px import plotly.graph_objects as go import numpy as np import google.generativeai as genai import os from io import StringIO import json st.set_page_config(layout="wide", page_title="Dynamic Data Dashboard") def main(): st.title("Dynamic Data Dashboard Generator") st.markdown(""" Upload your CSV file to generate an interactive dashboard tailored to your data. The application uses AI to analyze your data and create relevant visualizations. """) # API key input with validation api_key_input = st.sidebar.text_input("Enter your Gemini API key for more power", type="password") api_key = api_key_input or os.getenv("GEMINI_API_KEY") uploaded_file = st.file_uploader("Choose a CSV file", type="csv") if uploaded_file is not None: try: # Read and display data df = pd.read_csv(uploaded_file) with st.expander("Preview Data", expanded=True): st.dataframe(df.head(10)) # Basic data info st.subheader("Data Overview") col1, col2 = st.columns(2) with col1: st.metric("Rows", df.shape[0]) st.metric("Columns", df.shape[1]) with col2: st.metric("Numerical Columns", len(df.select_dtypes(include=np.number).columns)) st.metric("Categorical Columns", len(df.select_dtypes(exclude=np.number).columns)) # If API key is provided, use Gemini for analysis if api_key: st.subheader("AI-Powered Dashboard") with st.spinner("Analyzing your data and generating visualizations..."): try: generate_ai_dashboard(df, api_key) except Exception as e: st.error(f"Error generating AI dashboard: {e}") # Standard visualizations st.subheader("Standard Visualizations") generate_standard_dashboard(df) except Exception as e: st.error(f"Error processing your file: {e}") def generate_standard_dashboard(df): """Generate standard visualizations based on data types""" # Identify numerical and categorical columns numerical_cols = df.select_dtypes(include=np.number).columns.tolist() categorical_cols = df.select_dtypes(exclude=np.number).columns.tolist() # Data completeness st.subheader("Data Completeness") missing_data = pd.DataFrame({'column': df.columns, 'missing_values': df.isnull().sum(), 'percentage': (df.isnull().sum() / len(df) * 100).round(2)}) fig = px.bar(missing_data, x='column', y='percentage', title='Missing Values Percentage', labels={'percentage': 'Missing Values (%)', 'column': 'Column'}) st.plotly_chart(fig, use_container_width=True) # Distribution of numerical columns if numerical_cols: st.subheader("Numerical Distributions") selected_num_col = st.selectbox("Select a numerical column", numerical_cols) col1, col2 = st.columns(2) with col1: fig = px.histogram(df, x=selected_num_col, title=f'Distribution of {selected_num_col}') st.plotly_chart(fig, use_container_width=True) with col2: fig = px.box(df, y=selected_num_col, title=f'Box Plot of {selected_num_col}') st.plotly_chart(fig, use_container_width=True) # Distribution of categorical columns if categorical_cols: st.subheader("Categorical Distributions") selected_cat_col = st.selectbox("Select a categorical column", categorical_cols) # Limit to top 10 categories for readability value_counts = df[selected_cat_col].value_counts().nlargest(10) fig = px.bar(x=value_counts.index, y=value_counts.values, title=f'Top 10 Categories in {selected_cat_col}', labels={'x': selected_cat_col, 'y': 'Count'}) st.plotly_chart(fig, use_container_width=True) # Correlation heatmap for numerical data if len(numerical_cols) > 1: st.subheader("Correlation Between Numerical Variables") corr = df[numerical_cols].corr() fig = px.imshow(corr, text_auto=True, aspect="auto", title="Correlation Heatmap") st.plotly_chart(fig, use_container_width=True) # Scatter plot for exploring relationships if len(numerical_cols) >= 2: st.subheader("Explore Relationships") col1, col2 = st.columns(2) with col1: x_col = st.selectbox("X-axis", numerical_cols, index=0) with col2: y_col = st.selectbox("Y-axis", numerical_cols, index=min(1, len(numerical_cols)-1)) color_col = None if categorical_cols: color_col = st.selectbox("Color by (optional)", ['None'] + categorical_cols) if color_col == 'None': color_col = None fig = px.scatter(df, x=x_col, y=y_col, color=color_col, title=f'{y_col} vs {x_col}', opacity=0.7) st.plotly_chart(fig, use_container_width=True) def generate_ai_dashboard(df, api_key): """Use Gemini AI to analyze data and generate dashboard recommendations""" # Configure Gemini genai.configure(api_key=api_key) model = genai.GenerativeModel('gemini-2.0-flash') # Generate data summary column_info = {col: { 'dtype': str(df[col].dtype), 'unique_values': int(df[col].nunique()), 'missing_values': int(df[col].isna().sum()), 'sample': [str(x) for x in df[col].dropna().sample(min(5, len(df))).tolist()] } for col in df.columns} # Prepare prompt full_prompt = f""" Analyze the following dataset and suggest visualizations that would be insightful: Dataset Summary: - Rows: {df.shape[0]} - Columns: {df.shape[1]} Column Information: {json.dumps(column_info, indent=2)} Please provide visualization recommendations in the following JSON format: {{ "insights": [ "Key insight about the data", "Another insight about the data" ], "visualizations": [ {{ "title": "Visualization Title", "description": "What this visualization shows", "type": "bar|line|scatter|pie|histogram|box|heatmap", "x_column": "column_name_for_x_axis", "y_column": "column_name_for_y_axis", "color_column": "optional_column_for_color", "facet_column": "optional_column_for_faceting" }} ] }} Return ONLY the JSON, no other text. """ # Call Gemini API response = model.generate_content( full_prompt, generation_config={"temperature": 0.3} ) try: # Try to parse the response as JSON response_text = response.text # Clean the response if it contains markdown code blocks if "```json" in response_text: response_text = response_text.split("```json")[1].split("```")[0].strip() elif "```" in response_text: response_text = response_text.split("```")[1].split("```")[0].strip() recommendations = json.loads(response_text) # Display AI insights st.subheader("AI Insights") for insight in recommendations.get("insights", []): st.info(insight) # Create visualizations st.subheader("AI Recommended Visualizations") for viz in recommendations.get("visualizations", []): with st.expander(viz["title"], expanded=True): st.write(viz["description"]) try: x_col = viz.get("x_column") y_col = viz.get("y_column") color_col = viz.get("color_column") viz_type = viz.get("type", "bar").lower() if viz_type == "bar": fig = px.bar(df, x=x_col, y=y_col, color=color_col, title=viz["title"]) elif viz_type == "line": fig = px.line(df, x=x_col, y=y_col, color=color_col, title=viz["title"]) elif viz_type == "scatter": fig = px.scatter(df, x=x_col, y=y_col, color=color_col, title=viz["title"]) elif viz_type == "pie": fig = px.pie(df, names=x_col, values=y_col, title=viz["title"]) elif viz_type == "histogram": fig = px.histogram(df, x=x_col, color=color_col, title=viz["title"]) elif viz_type == "box": fig = px.box(df, y=y_col, x=x_col, color=color_col, title=viz["title"]) elif viz_type == "heatmap": pivot_table = pd.pivot_table(df, values=y_col, index=x_col, columns=color_col, aggfunc='mean') fig = px.imshow(pivot_table, title=viz["title"]) else: fig = px.bar(df, x=x_col, y=y_col, title=viz["title"]) st.plotly_chart(fig, use_container_width=True) except Exception as e: st.error(f"Could not create this visualization: {e}") except Exception as e: st.error(f"Could not parse AI recommendations: {e}") st.code(response.text, language="json") if __name__ == "__main__": main()