phzpan / 6115_Project

6115_project

pe_pyContextNLP_pe.ipynb

PE NLP pipeline for training and testing. it contains functions including:

• Save annotation dataframes to annotation files
• Read ann file of selected document
• Convert annotations to dataframe
• Convert dataframe to Annotation instance
• Mention level evaluation

dimer_pipe.ipynb

D-dimer NLP pipeline for training and testing. it contains similar functions as PE pipeline.

depolyment.ipynb

Combined PE and d-dimer pipeline, and save the annotation results to output.csv.

csv2modifier_rules.ipynb

Covert csv file format to PyContextNLP modifier_rules format.

DocumentClassifier_pe.py

Modified DocumentClassifier.py.

pynlp_pipe_pe.py

A wrapper of PyContextNLP using PyRush as the sentence segmenter. It return entity annotations, relationships and document classification.

pynlp_valid.py

This module is used for evaluation.

pe_Simple_Classification_System.ipynb

New PE NLP pipeline for training, testing and deployment. Output file: out_table.csv.

dimer_pipe_v0002.ipynb

New d-dimer pipeline for training, testing and deployment. Output file: out_table_dimer.csv.