Skip to content
Branch: master
Find file History
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
..
Failed to load latest commit information.
data
mnist_extraction_source
README.md
__init__.py
convertImages.py
create_training_sample.sh
extract_mnist.sh
run_mnist_experiment.py
run_saccades_experiment.pyw
saccade_network.py
saccades_experiment.py

README.md

HTM for MNIST

These examples shows how to use an HTM network to classify images from the MNIST dataset.

Instructions

  1. Download and extract the MNIST data:

    python -m nupic.vision.mnist.data.extract
    

    This will download, extract, and preprocess training and test images. This MNIST dataset contains 60,000 training samples and 10,000 testing samples. By default, images are located in a directory called data in your current working directory. Use the --data option to specify a different data directory.

  2. Train and test an HTM SP-only network:

    python -m nupic.vision.mnist.run_mnist_experiment
    

Results

This example achieves 95.56% accuracy on the 10,000 image training set as of 2015-04-13.

Acknowledgements

Numenta wishes to express its gratitude to Dr. Yann LeCun of the Courant Institute of Mathematical Sciences, New York, NY, for publishing the MNIST data set. Details regarding MNIST and related publications are available at the following URL: http://yann.lecun.com/exdb/mnist

You can’t perform that action at this time.