amitgayar / bert_hr Goto Github PK

apt-get install default-jre

First Install Grobid by following instructions from "https://grobid.readthedocs.io/en/latest/Install-Grobid/" and "https://grobid.readthedocs.io/en/latest/Grobid-service/"

1. To Install : cd grobid && ./gradlew clean install
2. Run virtual env : in ~/bert_hr/
3. To Run : ./gradlew run to run the server for grobid _parser_pdf.py

Run : script "~/bert_hr/grobid/grobid_parser_pdf.py" with all to-be-processed-pdf-files stored in "../data_pdf/" directory Results would be stored in "../data_xml/" directory

Run : “~/bert_hr/grobid/xml_parser.py” to process data in “../data_xml/” and saving in “../data_input/” directory for BERT prediction.

‘berthr.py’ would take csv files from “bert_hr/data_input/” to do the predictions while saving data in “bert_hr/data_output/” directory.

Run : ‘~/bert_hr/flask_app/berthr.py’ for predictions whose results would be saved in "data_output" directory.

"flask_app/templates/" directory contains the required UI files.

"data" directory contains the checkpoints of the custom trained BERT model (note: files can be found in the shared Drive folder )