Spaces:
Sleeping
Sleeping
bertugmirasyedi
commited on
Commit
·
3316ef5
1
Parent(s):
1ec7e79
Changed summarization model and added onnxruntime options
Browse files- .DS_Store +0 -0
- __pycache__/app.cpython-310.pyc +0 -0
- app.py +45 -45
.DS_Store
ADDED
Binary file (6.15 kB). View file
|
|
__pycache__/app.cpython-310.pyc
ADDED
Binary file (10.5 kB). View file
|
|
app.py
CHANGED
@@ -21,7 +21,7 @@ def search(
|
|
21 |
classification: bool = True,
|
22 |
summarization: bool = True,
|
23 |
similarity: bool = False,
|
24 |
-
add_chatgpt_results: bool =
|
25 |
n_results: int = 10,
|
26 |
):
|
27 |
import time
|
@@ -316,7 +316,7 @@ def search(
|
|
316 |
|
317 |
return similar_books
|
318 |
|
319 |
-
def summarize(descriptions):
|
320 |
"""
|
321 |
Summarize the descriptions and return the results.
|
322 |
"""
|
@@ -325,10 +325,17 @@ def search(
|
|
325 |
AutoModelForSeq2SeqLM,
|
326 |
pipeline,
|
327 |
)
|
|
|
|
|
328 |
|
329 |
# Define the summarizer model and tokenizer
|
330 |
-
|
331 |
-
|
|
|
|
|
|
|
|
|
|
|
332 |
|
333 |
# Create the summarizer pipeline
|
334 |
summarizer_pipe = pipeline(
|
@@ -349,7 +356,7 @@ def search(
|
|
349 |
|
350 |
return summaries
|
351 |
|
352 |
-
def classify(combined_data,
|
353 |
"""
|
354 |
Create classifier pipeline and return the results.
|
355 |
"""
|
@@ -358,15 +365,25 @@ def search(
|
|
358 |
AutoModelForSequenceClassification,
|
359 |
pipeline,
|
360 |
)
|
|
|
|
|
361 |
|
362 |
-
|
363 |
-
|
364 |
-
|
365 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
366 |
|
367 |
-
model = AutoModelForSequenceClassification.from_pretrained(
|
368 |
-
"sileod/deberta-v3-base-tasksource-nli"
|
369 |
-
)
|
370 |
classifier_pipe = pipeline(
|
371 |
"zero-shot-classification",
|
372 |
model=model,
|
@@ -374,49 +391,30 @@ def search(
|
|
374 |
hypothesis_template="This book is {}.",
|
375 |
batch_size=1,
|
376 |
device=-1,
|
377 |
-
multi_label=
|
378 |
)
|
379 |
|
380 |
# Define the candidate labels
|
381 |
-
|
382 |
"Introductory",
|
383 |
"Advanced",
|
384 |
-
"Academic",
|
385 |
-
"Not Academic",
|
386 |
-
"Manual",
|
387 |
]
|
388 |
|
389 |
-
|
390 |
-
import ray
|
391 |
-
import psutil
|
392 |
-
|
393 |
-
# Define the number of cores to use
|
394 |
-
num_cores = psutil.cpu_count(logical=True)
|
395 |
|
396 |
-
|
397 |
-
|
398 |
-
|
399 |
-
|
400 |
-
|
401 |
-
|
402 |
-
|
403 |
-
classifier = ray.get(classifier_id)
|
404 |
-
return classifier(doc, candidate_labels)
|
405 |
-
|
406 |
-
# Get the predicted labels
|
407 |
-
classes = [
|
408 |
-
classify_parallel.remote(classifier_id, doc, candidate_labels)
|
409 |
-
for doc in combined_data
|
410 |
-
]
|
411 |
-
else:
|
412 |
-
# Get the predicted labels
|
413 |
-
classes = [classifier_pipe(doc, candidate_labels) for doc in combined_data]
|
414 |
|
415 |
return classes
|
416 |
|
417 |
# If true then run the similarity, summarize, and classify functions
|
418 |
if classification:
|
419 |
-
classes = classify(combined_data,
|
420 |
else:
|
421 |
classes = [
|
422 |
{"labels": ["No labels available."], "scores": [0]}
|
@@ -428,7 +426,7 @@ def search(
|
|
428 |
classification_time = int(fourth_checkpoint - third_checkpoint)
|
429 |
|
430 |
if summarization:
|
431 |
-
summaries = summarize(descriptions)
|
432 |
else:
|
433 |
summaries = [
|
434 |
[{"summary_text": description}]
|
@@ -467,8 +465,10 @@ def search(
|
|
467 |
"author": authors[i],
|
468 |
"publisher": publishers[i],
|
469 |
"image_link": images[i],
|
470 |
-
"
|
471 |
-
"
|
|
|
|
|
472 |
"summary": summaries[i][0]["summary_text"],
|
473 |
"similar_books": similar_books[i]["sorted_by_similarity"],
|
474 |
"runtime": {
|
|
|
21 |
classification: bool = True,
|
22 |
summarization: bool = True,
|
23 |
similarity: bool = False,
|
24 |
+
add_chatgpt_results: bool = False,
|
25 |
n_results: int = 10,
|
26 |
):
|
27 |
import time
|
|
|
316 |
|
317 |
return similar_books
|
318 |
|
319 |
+
def summarize(descriptions, runtime="normal"):
|
320 |
"""
|
321 |
Summarize the descriptions and return the results.
|
322 |
"""
|
|
|
325 |
AutoModelForSeq2SeqLM,
|
326 |
pipeline,
|
327 |
)
|
328 |
+
from optimum.onnxruntime import ORTModelForSeq2SeqLM
|
329 |
+
from optimum.bettertransformer import BetterTransformer
|
330 |
|
331 |
# Define the summarizer model and tokenizer
|
332 |
+
if runtime == "normal":
|
333 |
+
tokenizer = AutoTokenizer.from_pretrained("lidiya/bart-base-samsum")
|
334 |
+
model = AutoModelForSeq2SeqLM.from_pretrained("lidiya/bart-base-samsum")
|
335 |
+
model = BetterTransformer.transform(model)
|
336 |
+
elif runtime == "onnxruntime":
|
337 |
+
tokenizer = AutoTokenizer.from_pretrained("optimum/t5-small")
|
338 |
+
model = ORTModelForSeq2SeqLM.from_pretrained("optimum/t5-small")
|
339 |
|
340 |
# Create the summarizer pipeline
|
341 |
summarizer_pipe = pipeline(
|
|
|
356 |
|
357 |
return summaries
|
358 |
|
359 |
+
def classify(combined_data, runtime="normal"):
|
360 |
"""
|
361 |
Create classifier pipeline and return the results.
|
362 |
"""
|
|
|
365 |
AutoModelForSequenceClassification,
|
366 |
pipeline,
|
367 |
)
|
368 |
+
from optimum.onnxruntime import ORTModelForSequenceClassification
|
369 |
+
from optimum.bettertransformer import BetterTransformer
|
370 |
|
371 |
+
if runtime == "normal":
|
372 |
+
# Define the zero-shot classifier
|
373 |
+
tokenizer = AutoTokenizer.from_pretrained(
|
374 |
+
"sileod/deberta-v3-base-tasksource-nli"
|
375 |
+
)
|
376 |
+
model = AutoModelForSequenceClassification.from_pretrained(
|
377 |
+
"sileod/deberta-v3-base-tasksource-nli"
|
378 |
+
)
|
379 |
+
elif runtime == "onnxruntime":
|
380 |
+
tokenizer = AutoTokenizer.from_pretrained(
|
381 |
+
"optimum/distilbert-base-uncased-mnli"
|
382 |
+
)
|
383 |
+
model = ORTModelForSequenceClassification.from_pretrained(
|
384 |
+
"optimum/distilbert-base-uncased-mnli"
|
385 |
+
)
|
386 |
|
|
|
|
|
|
|
387 |
classifier_pipe = pipeline(
|
388 |
"zero-shot-classification",
|
389 |
model=model,
|
|
|
391 |
hypothesis_template="This book is {}.",
|
392 |
batch_size=1,
|
393 |
device=-1,
|
394 |
+
multi_label=False,
|
395 |
)
|
396 |
|
397 |
# Define the candidate labels
|
398 |
+
level = [
|
399 |
"Introductory",
|
400 |
"Advanced",
|
|
|
|
|
|
|
401 |
]
|
402 |
|
403 |
+
audience = ["Academic", "Not Academic", "Manual"]
|
|
|
|
|
|
|
|
|
|
|
404 |
|
405 |
+
classes = [
|
406 |
+
{
|
407 |
+
"audience": classifier_pipe(doc, audience),
|
408 |
+
"level": classifier_pipe(doc, level),
|
409 |
+
}
|
410 |
+
for doc in combined_data
|
411 |
+
]
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
412 |
|
413 |
return classes
|
414 |
|
415 |
# If true then run the similarity, summarize, and classify functions
|
416 |
if classification:
|
417 |
+
classes = classify(combined_data, runtime="normal")
|
418 |
else:
|
419 |
classes = [
|
420 |
{"labels": ["No labels available."], "scores": [0]}
|
|
|
426 |
classification_time = int(fourth_checkpoint - third_checkpoint)
|
427 |
|
428 |
if summarization:
|
429 |
+
summaries = summarize(descriptions, runtime="normal")
|
430 |
else:
|
431 |
summaries = [
|
432 |
[{"summary_text": description}]
|
|
|
465 |
"author": authors[i],
|
466 |
"publisher": publishers[i],
|
467 |
"image_link": images[i],
|
468 |
+
"audience": classes[i]["audience"]["labels"][0],
|
469 |
+
"audience_confidence": classes[i]["audience"]["scores"][0],
|
470 |
+
"level": classes[i]["level"]["labels"][0],
|
471 |
+
"level_confidence": classes[i]["level"]["scores"][0],
|
472 |
"summary": summaries[i][0]["summary_text"],
|
473 |
"similar_books": similar_books[i]["sorted_by_similarity"],
|
474 |
"runtime": {
|