Update app.py

app.py

@@ -1,11 +1,55 @@
-from transformers import pipeline
+import os
+import random
+import string
+from datetime import datetime
+from supabase import create_client, Client
+from transformers import AutoModelForSequenceClassification, AutoTokenizer
+from huggingface_hub import login
 import gradio as gr
 
-model = pipeline("text-generation", model="mistralai/Mistral")
+# Load environment variables (for Hugging Face and Supabase integration)
+from dotenv import load_dotenv
+load_dotenv()
 
+# Hugging Face Authentication (for private models)
+huggingface_token = os.getenv("HUGGINGFACE_TOKEN")
+login(token=huggingface_token)  # Authenticate with Hugging Face
+
+# Hugging Face model setup
+model_name = "mistralai/Mistral"  # Replace with the correct model path
+model = AutoModelForSequenceClassification.from_pretrained(model_name, use_auth_token=True)
+tokenizer = AutoTokenizer.from_pretrained(model_name, use_auth_token=True)
+
+# Supabase setup
+SUPABASE_URL = os.getenv("SUPABASE_URL")
+SUPABASE_KEY = os.getenv("SUPABASE_KEY")
+supabase: Client = create_client(SUPABASE_URL, SUPABASE_KEY)
+
+# Function to process the CV and Job Description
 def rate_jd(jd, cv):
-    response = model(f"JD: {jd}\nCV: {cv}", max_length=3000, num_return_sequences=1)
-    return response[0]["generated_text"]
+    # Tokenize the inputs (job description and CV)
+    inputs = tokenizer(jd + " " + cv, return_tensors="pt", truncation=True, padding=True, max_length=512)
+    
+    # Run the model to get the match score
+    outputs = model(**inputs)
+    score = outputs.logits.detach().numpy().tolist()[0][0]  # Adjust based on your model's output
+
+    # Log the action to Supabase
+    email = "guest"  # Placeholder, can be updated to actual user if authentication is added
+    supabase.table('audit_logs').insert({
+        "user_email": email,
+        "action": "rate_jd",
+        "timestamp": datetime.utcnow().isoformat(),
+        "details": {"job_description": jd, "cv": cv, "score": score}
+    }).execute()
+
+    return f"Match Score: {score}"
+
+# Gradio interface to interact with the model
+iface = gr.Interface(fn=rate_jd,
+                     inputs=["text", "text"],  # Job Description and CV
+                     outputs="text",           # Display the match score
+                     live=True)
 
-iface = gr.Interface(fn=rate_jd, inputs=["text", "text"], outputs="text")
+# Launch the Gradio interface
 iface.launch()