Update app.py

app.py

@@ -1,64 +1,80 @@
+from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
 import gradio as gr
-from huggingface_hub import InferenceClient
+import pandas as pd
+import re
 
-"""
-For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-"""
-client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
+# Sample DataFrame setup
+df_clean = pd.read_csv("cleaned_data.csv")
+df = df_clean[['text', 'price', 'location']].copy()
 
+# Clean price column to ensure numeric
+df['price'] = pd.to_numeric(df['price'], errors='coerce')
 
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
+# Load Arabic language transformer (for future use or fine-tuning)
+tokenizer = AutoTokenizer.from_pretrained("UBC-NLP/AraT5-base")
+model = AutoModelForSeq2SeqLM.from_pretrained("UBC-NLP/AraT5-base")
 
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
+# Darija to MSA dictionary
+darija_map = {
+    "bghit": "أريد", "karhba": "سيارة", "auto": "سيارة", "voiture": "سيارة", "f": "في",
+    "drahem": "مال", "kra": "إيجار", "souma": "سعر", "ta7t": "أقل من", "taht": "أقل من",
+    "ghalya": "غالية", "rkhisa": "رخيصة", "wahran": "وهران", "oran": "وهران", "alger": "الجزائر",
+    "dzayer": "الجزائر", "kayn": "يوجد", "chouf": "شاهد", "dir": "قم", "men": "من", "fi": "في",
+    "3andi": "عندي", "nheb": "أريد", "tsoum": "تكلفة", "chwiya": "قليل", "bzaaf": "كثير",
+    "mlih": "جيد", "marque": "ماركة", "modele": "موديل", "nif": "جديدة", "qdima": "قديمة"
+}
 
-    messages.append({"role": "user", "content": message})
+# Translate Darija to Modern Standard Arabic (MSA)
+def translate_darija_to_standard(text):
+    words = text.lower().split()
+    translated = [darija_map.get(word.strip(), word) for word in words]
+    return " ".join(translated)
 
-    response = ""
+# Simple rule-based query filter
+def search_cars(query):
+    df_filtered = df.copy()
 
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
+    # Location filters
+    if "وهران" in query:
+        df_filtered = df_filtered[df_filtered['location'].str.contains("وهران", na=False)]
+    if "الجزائر" in query:
+        df_filtered = df_filtered[df_filtered['location'].str.contains("الجزائر", na=False)]
 
-        response += token
-        yield response
+    # Price filtering (regex for extracting numbers)
+    match = re.search(r"أقل من (\d+)", query)
+    if match:
+        try:
+            price_limit = int(match.group(1)) * 1000  # assuming input in million
+            df_filtered = df_filtered[df_filtered['price'] < price_limit]
+        except:
+            pass
 
+    return df_filtered.head(3).to_dict("records")
 
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
-demo = gr.ChatInterface(
-    respond,
-    additional_inputs=[
-        gr.Textbox(value="You are a friendly Chatbot.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
-    ],
-)
+# Response Generator
+def generate_response(user_input):
+    std_query = translate_darija_to_standard(user_input)
+    results = search_cars(std_query)
 
+    if len(results) > 0:
+        response = f"🚗 Kayn {len(results)} 3roud li ykhdmouk:\n\n"
+        for i, car in enumerate(results, 1):
+            title = car['text']
+            price = int(car['price'])
+            city = car['location']
+            response += f"{i}. 🚙 {title}\n   💰 {price:,} DZD\n   📍 {city}\n"
+        response += "\n📩 T7eb nwarik kter?"
+    else:
+        response = "🫤 Ma l9it walou f had condition... Jرب بكلمات خرا wala souma okhra."
 
-if __name__ == "__main__":
-    demo.launch()
+    return response
+
+# Gradio UI
+gr.Interface(
+    fn=generate_response,
+    inputs="text",
+    outputs="text",
+    title="🚗 Chatbot Dziria - Car Market DZ",
+    description="S2l 3la sayarat b darija, ou chatbot ywarik l3roud 📊",
+    theme="default"
+).launch()