Upload folder using huggingface_hub

main.py

@@ -1,7 +1,7 @@
 #uvicorn main:app --reload 
 import os
 os.environ['HF_HOME'] = 'src/cache'
-os.environ['TRANSFORMERS_CACHE'] = 'src/cache'
+# os.environ['TRANSFORMERS_CACHE'] = 'src/cache'
 
 from fastapi import FastAPI, status
 from fastapi.responses import Response, JSONResponse
@@ -10,7 +10,7 @@ from pydantic import BaseModel
 from typing import List
 
 import os
-import json
+# import json
 import time
 
 from src.myNLI import FactChecker

src/myNLI.py

@@ -1,5 +1,5 @@
 import torch
-from transformers import AutoModel, AutoTokenizer, AutoModelForSequenceClassification
+from transformers import AutoModel, AutoTokenizer
 from sentence_transformers import SentenceTransformer, util
 import nltk
 
@@ -26,8 +26,8 @@ class FactChecker:
         self.envir = torch.device('cuda') if torch.cuda.is_available() else torch.device('cpu')
 
         # Load LLM
-        self.tokenizer = AutoTokenizer.from_pretrained("MoritzLaurer/mDeBERTa-v3-base-mnli-xnli", use_auth_token=False)    # LOAD mDEBERTa TOKENIZER
-        self.mDeBertaModel = AutoModel.from_pretrained(f"src/mDeBERTa (ft) V6/mDeBERTa-v3-base-mnli-xnli-{self.INPUT_TYPE}", use_auth_token=False)  # LOAD FINETUNED MODEL
+        self.tokenizer = AutoTokenizer.from_pretrained("MoritzLaurer/mDeBERTa-v3-base-mnli-xnli", token=False)    # LOAD mDEBERTa TOKENIZER
+        self.mDeBertaModel = AutoModel.from_pretrained(f"src/mDeBERTa (ft) V6/mDeBERTa-v3-base-mnli-xnli-{self.INPUT_TYPE}", token=False)  # LOAD FINETUNED MODEL
         # Load classifier model
         self.checkpoints = torch.load(f"src/mDeBERTa (ft) V6/{self.INPUT_TYPE}.pt", map_location=self.envir)