Update app.py

app.py

@@ -1,27 +1,24 @@
 import streamlit as st
 import pandas as pd
-import numpy as np
-import matplotlib.pyplot as plt
+import google.generativeai as genai
 import os
-import requests
-from dotenv import load_dotenv
+from io import StringIO
 
-# Load environment variables from .env file
-load_dotenv()
+# Initialize Gemini
+genai.configure(api_key=os.getenv("GEMINI_API_KEY"))
+model = genai.GenerativeModel("gemini-1.5-flash")
 
-# Get the Gemini API key from the environment
-GEMINI_API_KEY = os.getenv("GEMINI_API_KEY")
+# Load solar data
+@st.cache_data
+def load_data():
+    df = pd.read_csv('https://huggingface.co/spaces/MLDeveloper/AI_based_Solar_Project_Estimation_Tool/resolve/main/solar_data_india_2024.csv')
+    return df
 
-# Gemini API URL
-GEMINI_API_URL = "https://generativelanguage.googleapis.com/v1beta/models/gemini-1.5-flash:generateContent"
+df = load_data()
 
-# Load the solar data CSV file
-df = pd.read_csv('https://huggingface.co/spaces/MLDeveloper/AI_based_Solar_Project_Estimation_Tool/resolve/main/solar_data_india_2024.csv')
-
-# Streamlit UI
+# UI - Form for user input
 st.set_page_config(page_title="AI-based Solar Project Estimation Tool", layout="centered")
 st.title("AI-based Solar Project Estimation Tool")
-
 st.write("### Enter Your Details Below:")
 
 with st.form("solar_form"):
@@ -33,6 +30,25 @@ with st.form("solar_form"):
     
     submitted = st.form_submit_button("Get Estimate")
 
+# Build the prompt for Gemini
+def build_prompt(location, roof_size, electricity_bill, ghi, solar_cost_per_kw):
+    prompt = f"""
+    Estimate the solar system for the location '{location}' based on the following details:
+    - Roof size: {roof_size} sq meters
+    - Monthly electricity bill: ₹{electricity_bill}
+    - Average GHI (solar radiation) for {location}: {ghi} kWh/m²/day
+    - Solar system cost per kW in {location}: ₹{solar_cost_per_kw}
+
+    Provide:
+    1. Estimated solar system size in kW
+    2. Estimated daily solar output in kWh
+    3. Total system cost in ₹
+    4. Monthly savings in ₹
+    5. Payback period in years
+    """
+    return prompt
+
+# Generate the solar project estimate via Gemini
 if submitted and location and roof_size > 0 and electricity_bill >= 0:
     state_data = df[df['State'].str.contains(location, case=False)].iloc[0]
     
@@ -40,50 +56,46 @@ if submitted and location and roof_size > 0 and electricity_bill >= 0:
         ghi = state_data['Avg_GHI (kWh/m²/day)']
         solar_cost_per_kw = state_data['Solar_Cost_per_kW (₹)']
         
-        prompt_text = f"""
-        Estimate the solar system for the location '{location}' based on the following details:
-        - Roof size: {roof_size} sq meters
-        - Monthly electricity bill: ₹{electricity_bill}
-        - Average GHI (solar radiation) for {location}: {ghi} kWh/m²/day
-        - Solar system cost per kW in {location}: ₹{solar_cost_per_kw}
-
-        Provide:
-        1. Estimated solar system size in kW
-        2. Estimated daily solar output in kWh
-        3. Total system cost in ₹
-        4. Monthly savings in ₹
-        5. Payback period in years
-        """
-
-        # Define the headers for the API request
-        headers = {
-            "Authorization": f"Bearer {GEMINI_API_KEY}",
-            "Content-Type": "application/json"
-        }
-
-        # Define the request payload
-        payload = {
-            "model": "gemini-1.5-flash",
-            "messages": [{"role": "system", "content": "You are a helpful assistant."},
-                         {"role": "user", "content": prompt_text}],
-            "temperature": 0.7
-        }
-
-        # Make the API request
-        response = requests.post(GEMINI_API_URL, json=payload, headers=headers)
-
-        # Check the response from the API
-        if response.status_code == 200:
-            result = response.json()
-            generated_content = result['choices'][0]['message']['content']
-            
-            # Display the generated content (solar estimates)
+        prompt_text = build_prompt(location, roof_size, electricity_bill, ghi, solar_cost_per_kw)
+        
+        if prompt_text:
+            with st.spinner("Generating solar estimate with Gemini..."):
+                response = model.generate_content(prompt_text)
             st.subheader("Solar Project Estimate")
-            st.write(generated_content)
+            st.text_area("Generated Estimate", response.text, height=200)
         else:
-            st.error(f"Error: {response.status_code} - {response.text}")
-        
+            st.warning("Please check the inputs again.")
     else:
         st.error("Sorry, the location entered does not match any available data.")
 else:
     st.warning("Please fill out all fields to see your solar project estimate.")
+
+# Batch CSV Export
+st.markdown("### Export Solar Estimates")
+batch = st.number_input("How many estimates to generate?", min_value=1, max_value=100, value=5)
+
+if st.button("Generate Batch & Download CSV"):
+    if location and roof_size > 0 and electricity_bill >= 0:
+        csv_buffer = StringIO()
+        writer = csv.writer(csv_buffer)
+        writer.writerow(["Sequence_no", "Solar Estimate"])
+        
+        with st.spinner("Generating estimates..."):
+            for i in range(1, batch + 1):
+                state_data = df[df['State'].str.contains(location, case=False)].iloc[0]
+                ghi = state_data['Avg_GHI (kWh/m²/day)']
+                solar_cost_per_kw = state_data['Solar_Cost_per_kW (₹)']
+                prompt_text = build_prompt(location, roof_size, electricity_bill, ghi, solar_cost_per_kw)
+                
+                # Generate the estimate using Gemini
+                response = model.generate_content(prompt_text)
+                writer.writerow([i, response.text.strip()])
+        
+        st.download_button(
+            label="Download Solar Estimates CSV",
+            data=csv_buffer.getvalue(),
+            file_name="solar_estimates.csv",
+            mime="text/csv"
+        )
+    else:
+        st.warning("Please fill out all fields to generate batch estimates.")