Update app.py
Browse files
app.py
CHANGED
@@ -71,7 +71,7 @@ def proc_submission(
|
|
71 |
st = time.perf_counter()
|
72 |
history = {}
|
73 |
clean_text = clean(input_text, extra_spaces=True, lowercase=True, reg="\b(?!(?:Although|Also)\b)(?:[A-Z][A-Za-z'`-]+)(?:,? (?:(?:and |& )?(?:[A-Z][A-Za-z'`-]+)|(?:et al.?)))*(?:, *(?:19|20)[0-9][0-9](?:, p\.? [0-9]+)?| *\((?:19|20)[0-9][0-9](?:, p\.? [0-9]+)?\))", reg_replace="")
|
74 |
-
max_input_length = 2048 if model_size == "tldr" else max_input_length
|
75 |
processed = truncate_word_count(clean_text, max_input_length)
|
76 |
|
77 |
if processed["was_truncated"]:
|
@@ -167,7 +167,7 @@ if __name__ == "__main__":
|
|
167 |
|
168 |
gr.Markdown("# Automatic summarization of biomedical research papers with neural abstractive methods into a long and comprehensive synopsis or extreme TLDR summary version")
|
169 |
gr.Markdown(
|
170 |
-
"A demo developed for my Master Thesis project using ad-hoc fine-tuned abstractive summarization models to summarize long biomedical articles
|
171 |
)
|
172 |
with gr.Column():
|
173 |
|
@@ -185,7 +185,7 @@ if __name__ == "__main__":
|
|
185 |
value=2,
|
186 |
)
|
187 |
gr.Markdown(
|
188 |
-
"_For optimal results use a GPU as the hosted CPU inference is lacking at times and hinders the
|
189 |
)
|
190 |
with gr.Row():
|
191 |
length_penalty = gr.inputs.Slider(
|
@@ -211,7 +211,7 @@ if __name__ == "__main__":
|
|
211 |
input_text = gr.Textbox(
|
212 |
lines=6,
|
213 |
label="Input Text (for summarization)",
|
214 |
-
placeholder="Enter any scientific text to be condensed into a
|
215 |
)
|
216 |
gr.Markdown("Upload your own file:")
|
217 |
with gr.Row():
|
|
|
71 |
st = time.perf_counter()
|
72 |
history = {}
|
73 |
clean_text = clean(input_text, extra_spaces=True, lowercase=True, reg="\b(?!(?:Although|Also)\b)(?:[A-Z][A-Za-z'`-]+)(?:,? (?:(?:and |& )?(?:[A-Z][A-Za-z'`-]+)|(?:et al.?)))*(?:, *(?:19|20)[0-9][0-9](?:, p\.? [0-9]+)?| *\((?:19|20)[0-9][0-9](?:, p\.? [0-9]+)?\))", reg_replace="")
|
74 |
+
#max_input_length = 2048 if model_size == "tldr" else max_input_length
|
75 |
processed = truncate_word_count(clean_text, max_input_length)
|
76 |
|
77 |
if processed["was_truncated"]:
|
|
|
167 |
|
168 |
gr.Markdown("# Automatic summarization of biomedical research papers with neural abstractive methods into a long and comprehensive synopsis or extreme TLDR summary version")
|
169 |
gr.Markdown(
|
170 |
+
"A demo developed for my Master Thesis project using ad-hoc fine-tuned abstractive summarization models to summarize long biomedical articles into a detailed, explanatory synopsis or extreme TLDR summary."
|
171 |
)
|
172 |
with gr.Column():
|
173 |
|
|
|
185 |
value=2,
|
186 |
)
|
187 |
gr.Markdown(
|
188 |
+
"_For optimal results use a GPU as the hosted CPU inference is lacking at times and hinders the output summary quality as well as forcing to divide the input text into batches._"
|
189 |
)
|
190 |
with gr.Row():
|
191 |
length_penalty = gr.inputs.Slider(
|
|
|
211 |
input_text = gr.Textbox(
|
212 |
lines=6,
|
213 |
label="Input Text (for summarization)",
|
214 |
+
placeholder="Enter any scientific text to be condensed into a detailed, explanatory synopsis or TLDR summary version. The input text is divided into batches of the selected token lengths to fit within the memory constraints, pre-processed and fed into the model of choice. The models were trained to handle long scientific papers but generalize reasonably well also to shorter text documents like scientific abstracts. Might take a while to produce long summaries :)",
|
215 |
)
|
216 |
gr.Markdown("Upload your own file:")
|
217 |
with gr.Row():
|