Update app.py

app.py

@@ -7,6 +7,23 @@ from huggingface_hub import InferenceClient
 # The client is used for generating the AI's text response.
 client = InferenceClient("microsoft/phi-4")
 
+# Load the clothing database CSV. Make sure 'clothing.csv' is in the same directory.
+try:
+    clothing_df = pd.read_csv("clothing.csv")
+except FileNotFoundError:
+    print("Error: 'clothing.csv' not found. Please ensure the file is in the same directory.")
+    clothing_df = pd.DataFrame(columns=['weather', 'formality', 'category', 'image_path'])
+
+# Simple filtering function to get clothing suggestions based on a query
+import gradio as gr
+import pandas as pd
+from huggingface_hub import InferenceClient
+
+# Load the AI model and clothing database
+# NOTE: Replace 'microsoft/phi-4' with the actual model ID if different
+# The client is used for generating the AI's text response.
+client = InferenceClient("microsoft/phi-4")
+
 # Load the clothing database CSV. Make sure 'clothing.csv' is in the same directory.
 try:
     clothing_df = pd.read_csv("clothing.csv")
@@ -32,8 +49,6 @@ def get_suggestions(query):
         results = results[results['weather'].str.contains('hot', case=False, na=False)]
     elif 'snow' in q:
         results = results[results['weather'].str.contains('snow', case=False, na=False)]
-    else:
-        results = results[results['weather'].str.contains('all', case=False, na=False)]
 
     # Formality-based filtering
     if "formal" in q or "office" in q:
@@ -62,6 +77,57 @@ def get_suggestions(query):
     final_selection = pd.concat(outfit)
     return final_selection
 
+# Chatbot + image output function
+def respond(message, chat_history):
+    
+    # Build conversation messages for the AI model
+    messages = [{"role": "system", "content": "You are a clothing assistant. Suggest suitable clothing items from the database based on the user's query."}]
+    
+    # Iterate through the history tuples and add them to the messages list
+    for user_msg, bot_msg in chat_history:
+        messages.append({"role": "user", "content": user_msg})
+        messages.append({"role": "assistant", "content": bot_msg})
+
+    # Add the current user message to the list for the AI call
+    messages.append({"role": "user", "content": message})
+
+    # Query Hugging Face Inference API
+    ai_response = client.chat_completion(messages=messages, max_tokens=1000)
+
+    # Extract the AI's reasoning or a fallback
+    reasoning = "I'm having trouble generating a response. Please try again."
+    if ai_response.choices and ai_response.choices[0].message:
+        reasoning = ai_response.choices[0].message.content.strip()
+
+    # Get filtered clothing suggestions
+    matches = get_suggestions(message)
+    image_paths = matches['image_path'].tolist()
+
+    # Append the new message and response as a single tuple to the chat history
+    chat_history.append((message, reasoning))
+
+    # Return the updated chat history and the image paths
+    return chat_history, image_paths
+
+# Gradio UI ---
+with gr.Blocks(theme=gr.themes.Soft()) as demo:
+    gr.Markdown("## Fashioneer - your fashion pioneer!")
+    gr.Markdown("Ask me what to wear and I'll suggest clothing with images from the database.")
+
+    # This is the search bar at the top
+    user_input = gr.Textbox(label="Type your outfit requirements here", placeholder="e.g., What should I wear on a rainy day?")
+    
+    # This row will place the chatbot and gallery side-by-side
+    with gr.Row():
+        chatbot = gr.Chatbot(label="Chatbot Conversation")
+        gallery = gr.Gallery(label="Recommended Clothing", columns=2, height='auto')
+
+    # This is the original line that connects the input and output
+    user_input.submit(fn=respond, inputs=[user_input, chatbot], outputs=[chatbot, gallery])
+
+# This starts the Gradio app
+demo.launch()
+
 # Chatbot + image output function
 def respond(message, chat_history):
     """