Update app.py

app.py

@@ -1,67 +1,36 @@
-import os
-import numpy as np
-import pandas as pd
-import faiss
 from groq import Groq
-from huggingface_hub import InferenceClient  # If you need Hugging Face support
-import radio  # Assuming you meant `radio` as a placeholder for UI or any framework
-
-# Set Groq API Key
-os.environ["GROQ_API_KEY"] = "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"
-client = Groq(api_key=os.environ.get("GROQ_API_KEY"))
-
-# Function to query Groq API
-def query_groq(prompt, model="llama3-8b-8192"):
-    chat_completion = client.chat.completions.create(
-        messages=[
-            {"role": "user", "content": prompt}
-        ],
-        model=model,
-    )
-    return chat_completion.choices[0].message.content
-
-# Simulated Traffic Dataset
-data = {
-    "Traffic_Density": [30, 60, 90, 70],  # Vehicles/km
-    "Signal_Wait_Time": [2, 5, 10, 7],   # in minutes
-    "Weather_Condition": ["Clear", "Rainy", "Foggy", "Clear"],  # categorical
-    "Time_of_Day": ["Morning", "Evening", "Night", "Afternoon"],  # categorical
-    "Avg_Speed": [50, 30, 20, 40],  # km/h
-}
-df = pd.DataFrame(data)
-
-# Encode categorical columns
-df_encoded = pd.get_dummies(df, columns=["Weather_Condition", "Time_of_Day"])
-
-# FAISS index setup
-def build_faiss_index(df_vectorized):
-    data_values = df_vectorized.values.astype('float32')
-    index = faiss.IndexFlatL2(data_values.shape[1])
-    index.add(data_values)
-    return index
-
-index = build_faiss_index(df_encoded.drop(columns=["Avg_Speed"]))
-query_vector = np.array([[60, 5, 1, 0, 0, 0, 0, 1, 0, 0]]).astype('float32')  # Sample
-
-# Nearest Neighbor Search
-D, I = index.search(query_vector, k=2)
-print("Distances:", D)
-print("Indices:", I)
-
-# Generate Recommendations
-def generate_traffic_insights(input_vector):
-    D, I = index.search(np.array([input_vector]).astype('float32'), k=3)
-    similar_conditions = df.iloc[I[0]]
-    print("Similar Conditions:\n", similar_conditions)
-
-    prompt = (
-        f"Given traffic data: {similar_conditions.to_dict(orient='records')}, "
-        "suggest strategies to optimize traffic flow and reduce signal wait times."
-    )
-    response = query_groq(prompt)
-    return response
-
-# Use Case: Predict & Recommend
-input_vector = [60, 5, 1, 0, 0, 0, 0, 1, 0, 0]  # Example input
-insights = generate_traffic_insights(input_vector)
-print("Traffic Optimization Insights:\n", insights)
+import gradio as gr
+
+# Groq API key (hardcoded as requested)
+GROQ_API_KEY = "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"
+
+# Initialize Groq client
+client = Groq(api_key=GROQ_API_KEY)
+
+# Function to get traffic recommendation
+def get_traffic_recommendation(input_situation):
+    try:
+        chat_completion = client.chat.completions.create(
+            messages=[
+                {
+                    "role": "user",
+                    "content": f"Suggest a solution to optimize traffic flow in the following situation:\n\n{input_situation}"
+                }
+            ],
+            model="llama-3-70b-8192"
+        )
+        return chat_completion.choices[0].message.content
+    except Exception as e:
+        return f"Error: {e}"
+
+# Gradio UI
+demo = gr.Interface(
+    fn=get_traffic_recommendation,
+    inputs=gr.Textbox(lines=5, placeholder="Describe the traffic scenario..."),
+    outputs="text",
+    title="Traffic Flow Optimizer",
+    description="Enter a traffic scenario to get suggestions for improving traffic flow using the LLaMA-3 model powered by Groq."
+)
+
+if __name__ == "__main__":
+    demo.launch()