Update app.py

app.py

@@ -3,8 +3,11 @@ import time
 from transformers import AutoModelForCausalLM, AutoTokenizer
 import torch
 
-model = AutoModelForCausalLM.from_pretrained("facebook/opt-13b", torch_dtype=torch.float16).cuda()
-tokenizer = AutoTokenizer.from_pretrained("facebook/opt-13b", use_fast=False)
+@st.cache(allow_output_mutation=True)
+def define_model():
+    model = AutoModelForCausalLM.from_pretrained("facebook/opt-13b", torch_dtype=torch.float16).cuda()
+    tokenizer = AutoTokenizer.from_pretrained("facebook/opt-13b", use_fast=False)
+    return model, tokenizer
 
 @st.cache(allow_output_mutation=True)
 def opt_model(prompt, model, tokenizer, num_sequences = 1, max_length = 50):  
@@ -13,9 +16,12 @@ def opt_model(prompt, model, tokenizer, num_sequences = 1, max_length = 50):
     answer = tokenizer.batch_decode(generated_ids, skip_special_tokens=True)
     return answer
 
+
+model, tokenizer = define_model()
+
 prompt= st.text_area('Your prompt here',
  '''Hello, I'm am conscious and''')
-
+ 
 answer = opt_model(prompt, model, tokenizer,)
 #lst = ['ciao come stai sjfsbd dfhsdf  fuahfuf  feuhfu wefwu ']
 lst = ' '.join(answer)