update

chatbot.py

@@ -5,31 +5,29 @@ from llama_index.embeddings.huggingface import HuggingFaceEmbedding
 from llama_index.llms.groq import Groq
 from llama_index.vector_stores.pinecone import PineconeVectorStore
 from pinecone import Pinecone
-from transformers import AutoModelForSeq2SeqLM, AutoTokenizer
+from transformers import AutoModelForSeq2SeqLM, AutoTokenizer, AutoModel
 from PyPDF2 import PdfReader
 from flask_cors import CORS
 from functools import wraps
 from dotenv import load_dotenv
 from huggingface_hub import InferenceClient
-import re, torch, jwt, os, json, gc
+import re, jwt, os, json
 
 load_dotenv()
 
 SECRET_KEY = os.getenv("SECRET_KEY")
 
 # Initialize Hugging Face Inference Client for embeddings
-client = InferenceClient(
-    provider="hf-inference",
-    api_key=os.getenv("HF_API_KEY")  # Add your Hugging Face API key to .env
-)
+client = InferenceClient(token=os.getenv("HF_API_KEY"))
 
 # Load summarization model and tokenizer
 model_path = "Jurisight/legal_led"
-model = AutoModelForSeq2SeqLM.from_pretrained(model_path,use_auth_token=os.getenv("HF_API_KEY"))
-tokenizer = AutoTokenizer.from_pretrained(model_path,use_auth_token=os.getenv("HF_API_KEY"))
+model = AutoModelForSeq2SeqLM.from_pretrained(model_path,token=os.getenv("HF_API_KEY"))
+tokenizer = AutoTokenizer.from_pretrained(model_path,token=os.getenv("HF_API_KEY"))
 
+embed_model = "BAAI/bge-base-en-v1.5"
 # Configure LlamaIndex settings
-Settings.embed_model = HuggingFaceEmbedding(model_name="BAAI/bge-base-en-v1.5")
+Settings.embed_model = HuggingFaceEmbedding(model_name=embed_model)
 Settings.llm = Groq(model="llama3-8b-8192", api_key=os.getenv("GROQ_API_KEY"))
 
 # Initialize Pinecone
@@ -208,12 +206,7 @@ def summarize(user_id):
 @authenticate_user
 def retrieve_cases(user_id):
     def generate_embedding(text):
-        # Use Hugging Face Inference API for embeddings
-        result = client.feature_extraction(
-            model="BAAI/bge-base-en-v1.5",
-            inputs=text,
-            provider="hf-inference",
-        )
+        result = client.feature_extraction(model=embed_model, text=text)
         return result
 
     def query_pinecone(query_text, top_k=10):