├── .gitignore
├── Procfile
├── hocr
├── requirements.txt
├── README.md
├── templates
└── upload_form.html
├── app.json
├── app.py
├── cdqa.ipynb
└── pdfCDQA.ipynb
/.gitignore:
--------------------------------------------------------------------------------
1 | .DS_store
2 | *.pyc
--------------------------------------------------------------------------------
/Procfile:
--------------------------------------------------------------------------------
1 | web: gunicorn app:app --log-file -
2 |
--------------------------------------------------------------------------------
/hocr:
--------------------------------------------------------------------------------
1 | tessedit_create_hocr 1
2 | tessedit_pageseg_mode 1
--------------------------------------------------------------------------------
/requirements.txt:
--------------------------------------------------------------------------------
1 | gunicorn==19.5.0
2 | Flask==0.12.3
3 | Werkzeug==0.10.4
4 |
--------------------------------------------------------------------------------
/README.md:
--------------------------------------------------------------------------------
1 | # Closed-Domain-Question-Answering
2 | End-to-End Closed Domain Question Answering on Financial Data extracted from given PDFs using OCR
3 | Based on #@andrelmfarias and #@fmikaelian cdQA (https://github.com/cdqa-suite/cdQA)
4 |
--------------------------------------------------------------------------------
/templates/upload_form.html:
--------------------------------------------------------------------------------
1 |
2 |
| \n", 317 | " | author | \n", 318 | "content | \n", 319 | "date | \n", 320 | "tag | \n", 321 | "title | \n", 322 | "url | \n", 323 | "website | \n", 324 | "
|---|---|---|---|---|---|---|---|
| 0 | \n", 329 | "The Quint | \n", 330 | "Tulsi Gabbard, the first Hindu lawmaker from H... | \n", 331 | "2018-11-13 | \n", 332 | "Tulsi Gabbard | \n", 333 | "Tulsi Gabbard: Hindu War Veteran US Lawmaker M... | \n", 334 | "https://www.thequint.com/news/world/tulsi-gabb... | \n", 335 | "quint | \n", 336 | "
| 1 | \n", 339 | "Pushpesh Pant | \n", 340 | "Swami Vivekananda ranks among the highest in t... | \n", 341 | "2018-07-03 | \n", 342 | "Swami Vivekananda,Hinduism,Tolerance,Politics ... | \n", 343 | "Swami Vivekananda’s 156th Birth Anniversary 20... | \n", 344 | "https://www.thequint.com/voices/opinion/swami-... | \n", 345 | "quint | \n", 346 | "
| 2 | \n", 349 | "Meeran Chadha Borwankar | \n", 350 | "In the news for the wrong reasons, the Central... | \n", 351 | "2018-07-18 | \n", 352 | "Modi Government,Central Bureau of Investigatio... | \n", 353 | "Rift Between CBI Chief & Special Director: Thi... | \n", 354 | "https://www.thequint.com/voices/opinion/cbi-co... | \n", 355 | "quint | \n", 356 | "
| 3 | \n", 359 | "Mayukh Ghosh | \n", 360 | "Reminiscing about the past is a dreadful tool ... | \n", 361 | "2018-05-14 | \n", 362 | "Satyajit Ray,Shyam Benegal,Smita Patil,Politic... | \n", 363 | "The Late Filmmaker Mrinal Sen’s ‘Cinema of Res... | \n", 364 | "https://www.thequint.com/entertainment/indian-... | \n", 365 | "quint | \n", 366 | "
| 4 | \n", 369 | "Akanksha Kumar | \n", 370 | ": Akanksha Kumar : Sandeep Suman Twenty-three ... | \n", 371 | "2018-04-30 | \n", 372 | "Tihar Jail,Supreme Court of India,Life Impriso... | \n", 373 | "Tandoor Murder: Former Delhi Police Joint Comm... | \n", 374 | "https://www.thequint.com/videos/news-videos/ho... | \n", 375 | "quint | \n", 376 | "
| \n", 443 | " | date | \n", 444 | "title | \n", 445 | "category | \n", 446 | "link | \n", 447 | "abstract | \n", 448 | "paragraphs | \n", 449 | "
|---|---|---|---|---|---|---|
| 0 | \n", 454 | "13.05.2019 | \n", 455 | "The banking jobs : Assistant Vice President – ... | \n", 456 | "Careers | \n", 457 | "https://group.bnpparibas/en/news/banking-jobs-... | \n", 458 | "Within the Group’s Corporate and Institutional... | \n", 459 | "[I manage a team in charge of designing and im... | \n", 460 | "
| 1 | \n", 463 | "13.05.2019 | \n", 464 | "BNP Paribas at #VivaTech : discover the progra... | \n", 465 | "Innovation | \n", 466 | "https://group.bnpparibas/en/news/bnp-paribas-v... | \n", 467 | "From Thursday 16 to Saturday 18 May 2019, join... | \n", 468 | "[With François Hollande, Chairman of French fo... | \n", 469 | "
| 2 | \n", 472 | "13.05.2019 | \n", 473 | "\"The bank with an IT budget of more than EUR6 ... | \n", 474 | "Group | \n", 475 | "https://group.bnpparibas/en/news/the-bank-budg... | \n", 476 | "Interview with Jean-Laurent Bonnafé, Director ... | \n", 477 | "[We did the groundwork between 2012 and 2016, ... | \n", 478 | "
| 3 | \n", 481 | "10.05.2019 | \n", 482 | "BNP Paribas at #VivaTech : discover the progra... | \n", 483 | "Innovation | \n", 484 | "https://group.bnpparibas/en/news/bnp-paribas-v... | \n", 485 | "From Thursday 16 to Saturday 18 May 2019, join... | \n", 486 | "[As part of the ‘United Tech of Europe’ theme,... | \n", 487 | "
| 4 | \n", 490 | "10.05.2019 | \n", 491 | "When Artificial Intelligence participates in r... | \n", 492 | "Careers | \n", 493 | "https://group.bnpparibas/en/news/artificial-in... | \n", 494 | "As the competition to attract talent intensifi... | \n", 495 | "[Online recruitment is already the norm. Accor... | \n", 496 | "
| \n", 410 | " | title | \n", 411 | "paragraphs | \n", 412 | "
|---|---|---|
| 0 | \n", 417 | "GLAccount00762980-2.pdf | \n", 418 | "[Report Name: Test Report , 12/09/2017 09:15:... | \n", 419 | "