Skip to content
Switch branches/tags
Go to file


Failed to load latest commit information.
Latest commit message
Commit time

Reference implementation of Recursive Cortical Network (RCN)

Reference implementation of a two-level RCN model on MNIST classification. See the Science article "A generative vision model that trains with high data efficiency and breaks text-based CAPTCHAs" and Vicarious Blog for details.

Note: this is an unoptimized reference implementation and is not intended for production.


Note: Python 2.7 is supported. The code was tested on OSX 10.11. It may work on other system platforms but not guaranteed.

Before starting please make sure gcc is installed (brew install gcc) and up to date in order to compile the various dependencies (particularly numpy).

Clone the repository:

git clone

Simple Install:

cd science_rcn

Manual Install (setting up a virtual environment beforehand is recommended):

cd science_rcn
python install


If you installed via make you need to activate the virtual environment:

source venv/bin/activate

To run a small unit test that trains and tests on 20 MNIST images using one CPU (takes ~2 minutes, accuracy is ~60%):

python science_rcn/

To run a slightly more interesting experiment that trains on 100 images and tests on 20 MNIST images using multiple CPUs (takes <1 min using 7 CPUs, accuracy is ~90%):

python science_rcn/ --train_size 100 --test_size 20 --parallel

To test on the full 10k MNIST test set, training on 1000 examples (could take hours depending on the number of available CPUs, average accuracy is ~97.7+%):

python science_rcn/ --full_test_set --train_size 1000 --parallel --pool_shape 25 --perturb_factor 2.0

Blog post

Check out our related blog post.


We used the following datasets for the Science paper:

CAPTCHA datasets

MNIST datasets

MNIST licensing

Yann LeCun (Courant Institute, NYU) and Corinna Cortes (Google Labs, New York) hold the copyright of MNIST dataset, which is a derivative work from original NIST datasets. MNIST dataset is made available under the terms of the Creative Commons Attribution-Share Alike 3.0 license.


Reference implementation of a two-level RCN model




No releases published


No packages published