Create app.py

app.py

@@ -0,0 +1,69 @@
+import streamlit as st
+import pandas as pd
+from langchain_experimental.agents.agent_toolkits import create_pandas_dataframe_agent
+from langchain_google_genai import ChatGoogleGenerativeAI
+import os
+
+def main():
+    st.set_page_config(page_title="Data Analysis Agent 🤖")
+    st.title("🤖 Data Analysis Agent")
+    st.write("Upload a CSV file and ask questions about your data.")
+
+    # Get Gemini API Key from Streamlit Secrets
+    # We use st.secrets for deployment on Hugging Face
+    try:
+        # For local development, you can set an environment variable
+        # For deployment, set this in Hugging Face Spaces "Secrets"
+        api_key = os.environ.get("GEMINI_API_KEY") or st.secrets["GEMINI_API_KEY"]
+    except KeyError:
+        st.error("GEMINI_API_KEY not found. Please set it in your Hugging Face Spaces secrets.")
+        return
+
+    if not api_key:
+        st.warning("Please add your Gemini API Key to the Hugging Face Space Secrets to use the app.")
+        return
+
+    # File Uploader
+    uploaded_file = st.file_uploader("Upload your CSV file", type=["csv"])
+
+    if uploaded_file is not None:
+        try:
+            # Read the CSV file into a Pandas DataFrame
+            df = pd.read_csv(uploaded_file)
+            st.dataframe(df.head())
+
+            # Initialize the Gemini LLM
+            llm = ChatGoogleGenerativeAI(
+                model="gemini-pro",
+                google_api_key=api_key,
+                temperature=0
+            )
+
+            # Create the Pandas DataFrame Agent
+            # allow_dangerous_code=True is required for the agent to execute Python code
+            agent = create_pandas_dataframe_agent(
+                llm,
+                df,
+                verbose=True,
+                allow_dangerous_code=True 
+            )
+
+            # User question input
+            user_question = st.text_input("Ask a question about your data:")
+
+            if user_question:
+                with st.spinner("Analyzing..."):
+                    try:
+                        # The agent.invoke method runs the agent and returns a dictionary
+                        # The actual answer is in the "output" key
+                        response = agent.invoke(user_question)
+                        st.write("### Answer")
+                        st.success(response["output"])
+                    except Exception as e:
+                        st.error(f"Error analyzing data: {e}")
+
+        except Exception as e:
+            st.error(f"Error reading file: {e}")
+
+if __name__ == "__main__":
+    main()