Update app.py

app.py

@@ -8,10 +8,11 @@ from transformers import (
 )
 
 MODEL_NAME = "MaxLSB/LeCarnet-8M"
+hf_token = os.environ["HUGGINGFACEHUB_API_TOKEN"]
 
 # Load tokenizer & model locally
-tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
-model = AutoModelForCausalLM.from_pretrained(MODEL_NAME)
+tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME, token=hf_token)
+model = AutoModelForCausalLM.from_pretrained(MODEL_NAME, token=hf_token)
 model.eval()
 
 def respond(
@@ -23,7 +24,6 @@ def respond(
 ):
     inputs = tokenizer(prompt, return_tensors="pt")
 
-    # Text streamer to get one token at a time
     streamer = TextIteratorStreamer(
         tokenizer,
         skip_prompt=True,
@@ -39,11 +39,9 @@ def respond(
         top_p=top_p,
     )
 
-    # Kick off generation in background
     thread = threading.Thread(target=model.generate, kwargs=generate_kwargs)
     thread.start()
 
-    # Stream out partial completions
     accumulated = ""
     for new_text in streamer:
         accumulated += new_text