Update: minor changes

app.py

@@ -1,13 +1,16 @@
+import os
 import gradio as gr
 from src.brain import generate_answers
 from huggingface_hub import login
-import os
 from dotenv import load_dotenv
-load_dotenv() 
-HUGGINGFACE_TOKEN = os.environ.get("HUGGINGFACE_TOKEN")
-login(token=HUGGINGFACE_TOKEN)
+
+load_dotenv()
+token = os.environ.get("TOKEN")
+login(token=token)
 
 processing = False
+
+
 def response(query, history):
     global processing
     processing = True
@@ -26,9 +29,11 @@ with open("src/content.html", "r") as file:
     title_html = parts[0]
     bts_html = parts[1] if len(parts) > 1 else ""
 
+
 def loading():
     return "Loading ..."
 
+
 with gr.Blocks(css=css) as app:
     with gr.Column(elem_id="column_container"):
         gr.HTML(title_html)

src/brain.py

@@ -4,6 +4,7 @@ model_name = "google/gemma-2b"
 tokenizer = AutoTokenizer.from_pretrained(model_name)
 model = AutoModelForCausalLM.from_pretrained(model_name)
 
+
 def generate_answers(query):
     input_ids = tokenizer(query, return_tensors="pt")
     output = model.generate(**input_ids)