Update app.py

app.py

@@ -11,8 +11,8 @@ CHECKPOINT_PATH= 'scratch_2-nodes_tokenizer_latbert-original_packing_fcocchi/mod
 CHECKPOINT_PATH= 'itserr/latin_llm_alpha'
 
 print(f"Loading model from: {CHECKPOINT_PATH}")
-#tokenizer = AutoTokenizer.from_pretrained(CHECKPOINT_PATH, token=st.secrets["HF_TOKEN"])
-#model = AutoModelForCausalLM.from_pretrained(CHECKPOINT_PATH, token=st.secrets["HF_TOKEN"])
+tokenizer = AutoTokenizer.from_pretrained(CHECKPOINT_PATH, token=os.environ['HF_TOKEN'])
+model = AutoModelForCausalLM.from_pretrained(CHECKPOINT_PATH, token=os.environ['HF_TOKEN'])
 
 description="""
 This is a Latin Language Model (LLM) based on GPT-2 and it was trained on a large corpus of Latin texts and can generate text in Latin. 
@@ -28,13 +28,11 @@ def generate_text(prompt):
     else: 
         device = torch.device("cpu")
         print("No GPU available")
-    out_text= st.secrets["HF_TOKEN"]
-    return out_text
     
-    #print("***** Generate *****")
-    #text_generator = pipeline("text-generation", model=model, tokenizer=tokenizer, device=device)
-    #generated_text = text_generator(prompt, max_length=50, do_sample=True, temperature=1.0, repetition_penalty=2.0, truncation=True)
-    #return generated_text[0]['generated_text']
+    print("***** Generate *****")
+    text_generator = pipeline("text-generation", model=model, tokenizer=tokenizer, device=device)
+    generated_text = text_generator(prompt, max_length=50, do_sample=True, temperature=1.0, repetition_penalty=2.0, truncation=True)
+    return generated_text[0]['generated_text']
 
 custom_css = """
 #logo {