README.md 703 Bytes
Newer Older
Devansh D's avatar
Devansh D committed
1
# Neural Programmer
Quoc Le's avatar
Quoc Le committed
2

Devansh D's avatar
Devansh D committed
3
Implementation of the Neural Programmer model described in [paper](https://openreview.net/pdf?id=ry2YOrcge)
Quoc Le's avatar
Quoc Le committed
4

Devansh D's avatar
Devansh D committed
5
Download and extract the data from [dropbox](https://www.dropbox.com/s/9tvtcv6lmy51zfw/data.zip?dl=0). Change the ``data_dir FLAG`` to the location of the data.
Quoc Le's avatar
Quoc Le committed
6

Devansh D's avatar
Devansh D committed
7
8
### Training 
``python neural_programmer.py`` 
Quoc Le's avatar
Quoc Le committed
9

Devansh D's avatar
Devansh D committed
10
11
12
13
14
15
16
17
The models are written to FLAGS.output_dir

### Testing 
``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``
Arvind Neekalantan's avatar
Arvind Neekalantan committed
18

Quoc Le's avatar
Quoc Le committed
19
Maintained by Arvind Neelakantan (arvind2505)