Update app.py

app.py

@@ -1,71 +1,87 @@
 import os
-import numpy as np
 import pandas as pd
-import faiss
-from sentence_transformers import SentenceTransformer
+import numpy as np
+import streamlit as st
 from groq import Groq
 
-# Load pre-trained Sentence Transformer model
-model = SentenceTransformer('all-MiniLM-L6-v2')
+GROQ_API_KEY = "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"
 
 # Initialize Groq API client
-GROQ_API_KEY = "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"
 client = Groq(api_key=GROQ_API_KEY)
 
-# Generate embeddings using Sentence Transformers
-def generate_embeddings(text):
-    return model.encode(text)
-
-# Build FAISS index for retrieval
-def build_faiss_index(data):
-    index = faiss.IndexFlatL2(384)  # 384-dimensional embeddings for MiniLM
-    embeddings = [generate_embeddings(row.to_string()) for _, row in data.iterrows()]
-    embeddings = np.array(embeddings).astype("float32")
-    index.add(embeddings)
-    return index, embeddings
+# Function to analyze energy usage
+def analyze_energy_usage(data, household_id=None):
+    if household_id:
+        # Filter data for a specific household
+        household_data = data[data["Household ID"] == household_id]
+    else:
+        household_data = data
 
-# Query FAISS index
-def query_index(query, data, index):
-    query_embedding = generate_embeddings(query).astype("float32")
-    distances, indices = index.search(np.array([query_embedding]), k=5)
-    results = data.iloc[indices[0]]
-    return results
-
-# Generate a detailed report using Groq's generative model
-def generate_report_with_groq(query, results):
-    input_text = f"Based on the query '{query}', the following insights are generated:\n\n{results.to_string(index=False)}"
-    response = client.chat.completions.create(
-        messages=[{"role": "user", "content": input_text}],
-        model="llama3-8b-8192",
-        stream=False
+    # Aggregate data
+    total_usage = household_data["Energy Usage (kWh)"].sum()
+    avg_cost = household_data["Cost"].mean()
+    peak_time_period = (
+        household_data.groupby("Time Period")["Energy Usage (kWh)"]
+        .sum()
+        .idxmax()
     )
-    return response.choices[0].message.content
 
-# Main function to execute the workflow
-if __name__ == "__main__":
-    # Load dataset
-    csv_path = "energy_usage_data.csv"  # Ensure this CSV is uploaded to your working directory
-    data = pd.read_csv(csv_path)
+    # Generate a summary report
+    report_summary = f"""
+    Total Energy Usage: {total_usage} kWh
+    Average Cost: ${avg_cost:.2f}
+    Peak Usage Time Period: {peak_time_period}
+    """
+    return report_summary, household_data
 
-    # Preprocess data (if needed)
-    data.fillna("", inplace=True)
+# Function to generate recommendations using Groq's API
+def generate_recommendations(context):
+    try:
+        response = client.chat.completions.create(
+            messages=[
+                {
+                    "role": "user",
+                    "content": f"Based on the following data:\n{context}\nProvide energy-saving recommendations and insights."
+                }
+            ],
+            model="llama3-8b-8192",
+            stream=False,
+        )
+        return response.choices[0].message.content
+    except Exception as e:
+        return f"An error occurred: {e}"
 
-    # Build FAISS index
-    print("Building FAISS index...")
-    index, embeddings = build_faiss_index(data)
+# Streamlit App
+st.title("Energy Usage Analysis Report Generator")
+
+# Upload Dataset
+uploaded_file = st.file_uploader("Upload your energy usage dataset (CSV)", type="csv")
+if uploaded_file:
+    # Load dataset
+    data = pd.read_csv(uploaded_file)
+    st.write("Dataset Preview:", data.head())
 
-    # User query
-    query = "Show households with high energy usage in the North region"
-    print(f"User Query: {query}")
+    # User Input
+    household_id = st.text_input("Enter Household ID for specific analysis (optional)")
 
-    # Query FAISS index
-    print("Retrieving relevant data...")
-    results = query_index(query, data, index)
+    # Analyze Data
+    if st.button("Analyze Energy Usage"):
+        with st.spinner("Analyzing..."):
+            report_summary, filtered_data = analyze_energy_usage(data, household_id)
+            st.subheader("Energy Usage Summary")
+            st.text(report_summary)
 
-    # Generate report
-    print("Generating report using Groq API...")
-    report = generate_report_with_groq(query, results)
+            # Generate recommendations
+            st.subheader("Recommendations")
+            context = filtered_data.to_string(index=False)
+            recommendations = generate_recommendations(context)
+            st.text(recommendations)
 
-    # Output the report
-    print("Generated Report:\n")
-    print(report)
+# Footer
+st.sidebar.title("About")
+st.sidebar.info(
+    """
+    This app generates energy usage reports and recommendations based on uploaded data.
+    Built with Streamlit and powered by Groq's language model.
+    """
+)