Remove persistent caching

app.py

@@ -28,7 +28,7 @@ def load_model(ckpt):
 
 tokenizer = MarianTokenizer.from_pretrained("Helsinki-NLP/opus-mt-en-es")
 
-@st.cache(persist=True)
+@st.cache
 def generate_sequence(pixel_values, num_beams, temperature, top_p):
     output_ids = state.model.generate(input_ids=pixel_values, max_length=64, num_beams=num_beams, temperature=temperature, top_p = top_p)
     print(output_ids)