update code

app.py

@@ -120,12 +120,13 @@ def main():
                     trust_remote_code=True, 
                     cache_dir="model_cache"
                 )
-                base_model = AutoModelForSeq2SeqLM.from_pretrained(
-                    checkpoint, 
-                    torch_dtype=torch.float32, 
-                    trust_remote_code=True, 
-                    cache_dir="model_cache"
-                )
+                base_model = "model_cache/models--ccdv--lsg-bart-base-16384-pubmed/snapshots/4072bc1a7a94e2b4fd860a5fdf1b71d0487dcf15"
+                #base_model = AutoModelForSeq2SeqLM.from_pretrained(
+                #    checkpoint, 
+                #    torch_dtype=torch.float32, 
+                #    trust_remote_code=True, 
+                #    cache_dir="model_cache"
+                #)
             else:  # default Flan T5 small
                 checkpoint = "MBZUAI/LaMini-Flan-T5-77M"
                 tokenizer = AutoTokenizer.from_pretrained(
@@ -135,11 +136,12 @@ def main():
                     model_max_length=1000, 
                     cache_dir="model_cache"
                 )
-                base_model = AutoModelForSeq2SeqLM.from_pretrained(
-                    checkpoint, 
-                    torch_dtype=torch.float32, 
-                    cache_dir="model_cache"
-                )
+                base_model = "model_cache/models--MBZUAI--LaMini-Flan-T5-77M/snapshots/c5b12d50a2616b9670a57189be20055d1357b474"
+                #base_model = AutoModelForSeq2SeqLM.from_pretrained(
+                #    checkpoint, 
+                #    torch_dtype=torch.float32, 
+                #    cache_dir="model_cache"
+                #)
         with col2:
             st.write("Skip any pages?")
             skipfirst = st.checkbox("Skip first page")
@@ -167,8 +169,6 @@ def main():
                 )
                 pdf_viewer = displayPDF(filepath)
             with col2:
-                with st.spinner("Downloading LLM..."):
-                    sleep(5)
                 with st.spinner("Summarizing..."):
                     summary = llm_pipeline(tokenizer, base_model, input_text)
                     postproc_text_length = postproc_count(summary)