Neal Wu 052e5e8b6e Converted the models repo to TF 1.0 using the upgrade script		8 éve
..
README.md	863f6c0919 Update README.md	8 éve
data_utils.py	052e5e8b6e Converted the models repo to TF 1.0 using the upgrade script	8 éve
model.py	052e5e8b6e Converted the models repo to TF 1.0 using the upgrade script	8 éve
neural_programmer.py	052e5e8b6e Converted the models repo to TF 1.0 using the upgrade script	8 éve
nn_utils.py	052e5e8b6e Converted the models repo to TF 1.0 using the upgrade script	8 éve
parameters.py	052e5e8b6e Converted the models repo to TF 1.0 using the upgrade script	8 éve
wiki_data.py	052e5e8b6e Converted the models repo to TF 1.0 using the upgrade script	8 éve

Neural Programmer

Implementation of the Neural Programmer model described in paper

Download and extract the data from dropbox. Change the data_dir FLAG to the location of the data.

python neural_programmer.py

The models are written to FLAGS.output_dir

python neural_programmer.py --evaluator_job=True

The models are loaded from FLAGS.output_dir. The evaluation is done on development data.

In case of errors because of encoding, add "# -*- coding: utf-8 -*-" as the first line in wiki_data.py

Maintained by Arvind Neelakantan (arvind2505)