Extending the BiDAF Model with "No Answer"

This is the implementation used in [Zero-Shot Relation Extraction via Reading Comprehension][paper1] (Levy et al., 2017).
It is an extension of the [BiDAF model][paper2] by Seo et al.
This file describes some basic use-cases in the relation-extraction setting. The original implementation's readme file is [BiDAF_README.md][fullreadme].

Requirements

run_prep.sh <run name> calls an internal script (zeroshot2squad.py) that changes our tab-delimited format to SQuAD's JSON format. It then performs any necessary preprocessing for the BiDAF model.
run_train.sh <run name> runs the training procedure.
run_test.sh <run name> runs the testing procedure, and yields an answer file in out/basic/<run name>/test-#####.json
python analyze.py <test set> <answer file> reads the test set and the model's answers, and returns the F1 score broken down by different factors.

[paper1]: [paper2]: https://arxiv.org/abs/1611.01603 [fullreadme]: BiDAF_README.md

Name		Name	Last commit message	Last commit date
Latest commit History 3 Commits
basic		basic
basic_cnn		basic_cnn
my		my
squad		squad
tree		tree
visualization		visualization
zeroshot		zeroshot
.hg_archival.txt		.hg_archival.txt
.hgignore		.hgignore
BiDAF_README.md		BiDAF_README.md
README.md		README.md
analyze.py		analyze.py
run_prep.sh		run_prep.sh
run_test.sh		run_test.sh
run_train.sh		run_train.sh