Update app.py

app.py

@@ -6,7 +6,7 @@ import gradio as gr
 import nltk
 from cleantext import clean
 
-from summarize import load_model_and_tokenizer, summarize_via_tokenbatches
+from summ import load_model_and_tokenizer, summarize_via_tokenbatches
 from utils import load_example_filenames, truncate_word_count
 
 _here = Path(__file__).parent
@@ -155,7 +155,7 @@ if __name__ == "__main__":
 
         gr.Markdown("# Automatic summarization of biomedical research papers with neural abstractive methods into a long and comprehensive synopsis or extreme TLDR summary version")
         gr.Markdown(
-            "A rather simple demo using an ad-hoc fine-tuned LongT5 or LED model to summarize long biomedical articles (or any scientific text related to the biomedical domain) into a detailed or extreme TLDR version."
+            "A rather simple demo (developed for my Master Thesis project) using an ad-hoc fine-tuned LongT5 or LED model to summarize long biomedical articles (or any scientific text related to the biomedical domain) into a detailed, explanatory synopsis or extreme TLDR version."
         )
         with gr.Column():
 
@@ -164,8 +164,11 @@ if __name__ == "__main__":
                 "Enter text below in the text area. The text will be summarized [using the selected text generation parameters](https://huggingface.co/blog/how-to-generate). Optionally load an available example below or upload a file."
             )
             with gr.Row():
-                model_size = gr.Radio(
-                    choices=["tldr", "sumpubmed"], label="Model Variant", value="sumpubmed"
+                summary_type = gr.Radio(
+                    choices=["tldr", "detailed"], label="Summary type", value="detailed"
+                )
+                model_type = gr.Radio(
+                    choices=["LongT5", "LED"], label="Model type", value="LongT5"
                 )
                 num_beams = gr.Radio(
                     choices=[2, 3, 4],
@@ -173,7 +176,7 @@ if __name__ == "__main__":
                     value=2,
                 )
             gr.Markdown(
-                "_The LED model is less performant than the LongT5 model, but is faster and will accept up to 2048 words per input (Large model accepts up to 768)._"
+                "_The LED model is less performant than the LongT5 model, but it's smaller in terms of size and therefore all other parameters being equal allows for a larger _"
             )
             with gr.Row():
                 length_penalty = gr.inputs.Slider(
@@ -245,7 +248,7 @@ if __name__ == "__main__":
                 "The summary scores can be thought of as representing the quality of the summary. less-negative numbers (closer to 0) are better:"
             )
             summary_scores = gr.Textbox(
-                label="Summary Scores", placeholder="Summary scores will appear here"
+                label="Compression rate 🗜", placeholder="🗜 will appear here"
             )
 
             gr.Markdown("---")