nupic图像识别程序

The purpose of this repository

• I'll verify the HTM using the task of ImageNet LSVRC-2014.

Index

• Dataset
• Install nupic/pylearn2
• Simple task

Dataset

CIFAR-10

• The CIFAR-10 dataset consists of 60000 32x32 colour images in 10 classes, with 6000 images per class.
• The CIFAR-10 dataset

MNIST

• The MNIST database of handwritten digits.
• THE MNIST DATABASE

ImageNet

• What is ILSVRC2013
  • Large Scale Visual Recognition Challenge 2014
• How to get images from imageNet
  • Get wnid list
    • You need to prepare csv file of wnid.
    • I created the csv file by shaping LSVRC page.(http://image-net.org/challenges/LSVRC/2014/browse-synsets)
    • Here is sample csv file (data/classification_categorys.csv).
  • Run get_image_from_imagenet.py
    • This script to get the image from imagenet.
    • Simply, this script get image url by wnid, and download image file.
    • I have excluded the following file.
      • flicr not found image file
      • cannot open file
      • Too small file

      cd data
      python get_image_from_imagenet.py
      

      • download-API
        • wnids : http://image-net.org/challenges/LSVRC/2014/browse-synsets
        • image : http://imagenet.stanford.edu/api/text/imagenet.synset.geturls?wnid=[wnid]
        • word : http://www.image-net.org/api/text/wordnet.synset.getwords?wnid=[wnid]

Install nupic/pylearn2

Install nupic/

• nupic

Install pylearn2

• install
• tutorial
  • When I execute make_dataset.py, the error has occurred.
  • I edited train_example_path of pylearn2/scripts/tutorials/grbm_smd/make_dataset.py.

  IOError: permission error creating /Library/Python/2.7/site-packages/pylearn2/scripts/tutorials/grbm_smd/cifar10_preprocessed_train.pkl
  

  • environment values

  export PYLEARN2_DATA_PATH=/Users/karino-t/data
  export PYLEARN2_VIEWER_COMMAND="open -Wn"
  

Technique

ImageNet Classification with Deep Convolutional Neural Networks

from https://github.com/kokukuma/nupic_image_recognition

－－－－－－－－－－－

Numenta Platform for Intelligent Computing

Numenta Platform for Intelligent Computing: a brain-inspired machine intelligence platform, and biologically accurate neural network based on cortical learning algorithms. http://numenta.org/

The Numenta Platform for Intelligent Computing (NuPIC) is a machine intelligence platform that implements the HTM learning algorithms. HTM is a detailed computational theory of the neocortex. At the core of HTM are time-based continuous learning algorithms that store and recall spatial and temporal patterns. NuPIC is suited to a variety of problems, particularly anomaly detection and prediction of streaming data sources.

For more information, see numenta.org or the NuPIC wiki.

Installing NuPIC 0.5.4

NuPIC binaries are available for:

• Linux x86 64bit
• OS X 10.9
• OS X 10.10
• Windows 64bit

Dependencies

The following dependencies are required to install NuPIC on all operating systems.

• Python 2.7 & development headers
• pip
• wheel
• numpy
• C++ 11 compiler like gcc (4.8+) or clang

Install OS X

First, you must install Xcode command line tools, which will get you a C++ compiler.

pip install nupic

Install Linux

NOTE: The nupic.bindings binary distribution is not stored on PyPi along with the OS X distribution. NuPIC uses the wheel binary format, and PyPi does not support hosting Linux wheel files. So we are forced to host our own.

pip install https://s3-us-west-2.amazonaws.com/artifacts.numenta.org/numenta/nupic.core/releases/nupic.bindings/nupic.bindings-0.4.4-cp27-none-linux_x86_64.whl
pip install nupic

Install Windows

pip install nupic

For problems installing NuPIC, please see our Installation and Build Wiki.

Test

# From the root of the repo:
py.test tests/unit

Having problems?

• You may need to use the --user flag for the commands above to install in a non-system location (depends on your environment). Alternatively, you can execute the pip commands with sudo (not recommended).
• You may need to add the --use-wheel option if you have an older pip version (wheels are now the default binary package format for pip).

For any other installation issues, please see our FAQ, email the nupic-discuss mailing list, or chat with us on Gitter.

Building NuPIC From Source Code

For details about checking out this repository and building in your local environment, see the Installing and Building NuPIC wiki page.

How to Contribute:

Please see the Contributing to NuPIC wiki page.

• Build:  
• Unit Test Coverage: 
• Regression Tests: 
• To cite this codebase: 

from https://github.com/numenta/nupic

相关帖子：

• 图像识别系统－Huabot Brain