Spaces:
Sleeping
Sleeping
Update src/streamlit_app.py
Browse files- src/streamlit_app.py +76 -47
src/streamlit_app.py
CHANGED
|
@@ -205,6 +205,41 @@ if 'df_ner' in st.session_state and not st.session_state.df_ner.empty:
|
|
| 205 |
fig_treemap.update_layout(margin=dict(t=50, l=25, r=25, b=25), paper_bgcolor='#F5FFFA', plot_bgcolor='#F5FFFA')
|
| 206 |
st.plotly_chart(fig_treemap)
|
| 207 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 208 |
# --- Question Answering Section ---
|
| 209 |
@st.cache_resource
|
| 210 |
def load_gliner_model():
|
|
@@ -280,56 +315,50 @@ if st.button("Extract Answers"):
|
|
| 280 |
st.session_state.df_qa = df_qa # Store QA results in session state
|
| 281 |
st.subheader("Extracted Answers", divider="green")
|
| 282 |
st.dataframe(df_qa, use_container_width=True)
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 283 |
else:
|
| 284 |
st.warning("No answers were found for the provided questions.")
|
| 285 |
if 'df_qa' in st.session_state:
|
| 286 |
del st.session_state.df_qa
|
| 287 |
except Exception as e:
|
| 288 |
st.error(f"An error occurred during answer extraction: {e}")
|
| 289 |
-
|
| 290 |
-
|
| 291 |
-
|
| 292 |
-
# --- Download Button Section ---
|
| 293 |
-
def create_zip_file_and_get_bytes():
|
| 294 |
-
"""Generates a zip file in memory with all available dataframes."""
|
| 295 |
-
# Define the glossary DataFrame here to ensure it's always available
|
| 296 |
-
dfa = pd.DataFrame(
|
| 297 |
-
data={
|
| 298 |
-
'Column Name': ['text', 'label', 'score', 'start', 'end', 'category'],
|
| 299 |
-
'Description': [
|
| 300 |
-
'entity extracted from your text data',
|
| 301 |
-
'label (tag) assigned to a given extracted entity',
|
| 302 |
-
'accuracy score; how accurately a tag has been assigned to a given entity',
|
| 303 |
-
'index of the start of the corresponding entity',
|
| 304 |
-
'index of the end of the corresponding entity',
|
| 305 |
-
'the broader category the entity belongs to',
|
| 306 |
-
]
|
| 307 |
-
}
|
| 308 |
-
)
|
| 309 |
-
if 'df_ner' not in st.session_state and 'df_qa' not in st.session_state:
|
| 310 |
-
return None, None
|
| 311 |
-
buf = io.BytesIO()
|
| 312 |
-
with zipfile.ZipFile(buf, "w") as myzip:
|
| 313 |
-
if 'df_ner' in st.session_state and not st.session_state.df_ner.empty:
|
| 314 |
-
myzip.writestr("Extracted_Entities.csv", st.session_state.df_ner.to_csv(index=False))
|
| 315 |
-
if 'df_qa' in st.session_state and not st.session_state.df_qa.empty:
|
| 316 |
-
myzip.writestr("Extracted_Answers.csv", st.session_state.df_qa.to_csv(index=False))
|
| 317 |
-
myzip.writestr("Glossary_of_tags.csv", dfa.to_csv(index=False))
|
| 318 |
-
return buf.getvalue(), "nlpblogs_results.zip"
|
| 319 |
-
|
| 320 |
-
st.divider()
|
| 321 |
-
|
| 322 |
-
if ('df_ner' in st.session_state and not st.session_state.df_ner.empty) or \
|
| 323 |
-
('df_qa' in st.session_state and not st.session_state.df_qa.empty):
|
| 324 |
-
zip_data, file_name = create_zip_file_and_get_bytes()
|
| 325 |
-
if zip_data:
|
| 326 |
-
with stylable_container(
|
| 327 |
-
key="download_button",
|
| 328 |
-
css_styles="""button { background-color: red; border: 1px solid black; padding: 5px; color: white; }""",
|
| 329 |
-
):
|
| 330 |
-
st.download_button(
|
| 331 |
-
label="Download results and glossary (zip)",
|
| 332 |
-
data=zip_data,
|
| 333 |
-
file_name=file_name,
|
| 334 |
-
mime="application/zip",
|
| 335 |
-
)
|
|
|
|
| 205 |
fig_treemap.update_layout(margin=dict(t=50, l=25, r=25, b=25), paper_bgcolor='#F5FFFA', plot_bgcolor='#F5FFFA')
|
| 206 |
st.plotly_chart(fig_treemap)
|
| 207 |
|
| 208 |
+
|
| 209 |
+
|
| 210 |
+
dfa = pd.DataFrame(
|
| 211 |
+
data={
|
| 212 |
+
'Column Name': ['text', 'label', 'score', 'start', 'end', 'category'],
|
| 213 |
+
'Description': [
|
| 214 |
+
'entity extracted from your text data',
|
| 215 |
+
'label (tag) assigned to a given extracted entity',
|
| 216 |
+
'accuracy score; how accurately a tag has been assigned to a given entity',
|
| 217 |
+
'index of the start of the corresponding entity',
|
| 218 |
+
'index of the end of the corresponding entity',
|
| 219 |
+
'the broader category the entity belongs to',]}
|
| 220 |
+
)
|
| 221 |
+
buf = io.BytesIO()
|
| 222 |
+
with zipfile.ZipFile(buf, "w") as myzip:
|
| 223 |
+
myzip.writestr("Summary of the results.csv", df.to_csv(index=False))
|
| 224 |
+
myzip.writestr("Glossary of tags.csv", dfa.to_csv(index=False))
|
| 225 |
+
|
| 226 |
+
with stylable_container(
|
| 227 |
+
key="download_button",
|
| 228 |
+
css_styles="""button { background-color: red; border: 1px solid black; padding: 5px; color: white; }""",
|
| 229 |
+
):
|
| 230 |
+
st.download_button(
|
| 231 |
+
label="Download results and glossary (zip)",
|
| 232 |
+
data=buf.getvalue(),
|
| 233 |
+
file_name="nlpblogs_results.zip",
|
| 234 |
+
mime="application/zip",)
|
| 235 |
+
|
| 236 |
+
if comet_initialized:
|
| 237 |
+
experiment.log_figure(figure=fig_treemap, figure_name="entity_treemap_categories")
|
| 238 |
+
experiment.end()
|
| 239 |
+
else:
|
| 240 |
+
st.warning("No entities were found in the provided text.")
|
| 241 |
+
|
| 242 |
+
|
| 243 |
# --- Question Answering Section ---
|
| 244 |
@st.cache_resource
|
| 245 |
def load_gliner_model():
|
|
|
|
| 315 |
st.session_state.df_qa = df_qa # Store QA results in session state
|
| 316 |
st.subheader("Extracted Answers", divider="green")
|
| 317 |
st.dataframe(df_qa, use_container_width=True)
|
| 318 |
+
csv_data = df_qa.to_csv(index=False).encode('utf-8')
|
| 319 |
+
with stylable_container(
|
| 320 |
+
key="download_button",
|
| 321 |
+
css_styles="""button { background-color: red; border: 1px solid black; padding: 5px; color: white; }""",
|
| 322 |
+
):
|
| 323 |
+
st.download_button(
|
| 324 |
+
label="Download CSV",
|
| 325 |
+
data=csv_data,
|
| 326 |
+
file_name="nlpblogs_extracted_answers.csv",
|
| 327 |
+
mime="text/csv",
|
| 328 |
+
)
|
| 329 |
+
|
| 330 |
+
if comet_initialized:
|
| 331 |
+
experiment.log_metric("processing_time_seconds", elapsed_time)
|
| 332 |
+
experiment.log_table("predicted_entities", df)
|
| 333 |
+
experiment.log_figure(figure=fig_treemap, figure_name="entity_treemap")
|
| 334 |
+
|
| 335 |
+
experiment.end()
|
| 336 |
+
else:
|
| 337 |
+
st.info("No answers were found in the text with the defined questions.")
|
| 338 |
+
if comet_initialized:
|
| 339 |
+
experiment.end()
|
| 340 |
+
except Exception as e:
|
| 341 |
+
st.error(f"An error occurred during processing: {e}")
|
| 342 |
+
st.write(f"Error details: {e}")
|
| 343 |
+
if comet_initialized:
|
| 344 |
+
experiment.log_text(f"Error: {e}")
|
| 345 |
+
experiment.end()
|
| 346 |
+
|
| 347 |
+
|
| 348 |
+
|
| 349 |
+
|
| 350 |
+
|
| 351 |
+
|
| 352 |
+
|
| 353 |
+
|
| 354 |
+
|
| 355 |
+
|
| 356 |
+
|
| 357 |
else:
|
| 358 |
st.warning("No answers were found for the provided questions.")
|
| 359 |
if 'df_qa' in st.session_state:
|
| 360 |
del st.session_state.df_qa
|
| 361 |
except Exception as e:
|
| 362 |
st.error(f"An error occurred during answer extraction: {e}")
|
| 363 |
+
|
| 364 |
+
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|