Metadata-Version: 2.1
Name: dhnn
Version: 0.1.8
Summary: Discrete Hopfield Network (DHNN) implemented with Python
Home-page: https://github.com/Zeroto521/DHNN
Author: yosukekatada,Zeroto521
Author-email: Zeroto521@gmail.com
License: MIT
Download-URL: https://github.com/Zeroto521/DHNN/archive/master.zip
Description: # DHNN
        
        DHNN is a minimalistic and Numpy based implementation of the [Discrete Hopfield Network](http://en.wikipedia.org/wiki/Hopfield_network). DHNN can learn (memorize) patterns and remember (recover) the patterns when the network feeds those with noises.
        
        ## Installation
        
        Just use pip:
        
            pip install dhnn
        
        Or download `dhnn` to a directory which your choice and use `setup` to install script:
        
            python setup.py install
        
        ## Prerequisites
        
        Prior to running this package, please install the following libraries.
        
        -   `numpy`
        
        ## Example (Image Restoration)
        
        ### Step1
        
        Input a neat picture like this(yosukekatada's smile face).
        
        ![train](assets/yosuke.jpg)
        
        ### Step2
        
        Get the network to memorize the pattern, this program will automatically transform RGB Jpg into black-white picture.
        
        ### Step3
        
        After the network memorized it, put the picture with noise like this(yosukekatada's smile face with **sunglasses**) into the network.
        
        ![test](assets/yosuke_test.jpg)
        
        ### Step4
        
        The network can strip off the sunglasses, because the network ready remembers the former picture.
        
        ![recovery](assets/recovery_0.jpg)
        
        ## Authors
        
        | <img src="https://avatars3.githubusercontent.com/u/4463558?v=4" alt="yosukekatada" width="100px" height="100px"/> | <img src="https://avatars1.githubusercontent.com/u/25895405?v=4" alt="Zeroto521" width="100px" height="100px"/> |
        | :---------------------------------------------------------------------------------------------------------------: | :-------------------------------------------------------------------------------------------------------------: |
        |                                  [yosukekatada](https://github.com/yosukekatada)                                  |                                    [Zeroto521](https://github.com/Zeroto521)                                    |
        
        ## TODO
        
        -   [ ] more flag, add 0/1 flag or other flag.
        -   [x] optimize loop, try numba, Cpython or any other ways.
        -   [ ] optimize memory.
        
        ## License
        
        MIT License. [@yosukekatada](https://github.com/yosukekatada), [@Zeroto521](https://github.com/Zeroto521)
        
        ## References
        
        -   http://rishida.hatenablog.com/entry/2014/03/03/174331
        
Keywords: machine learning,neural networks,hopfield,DHNN
Platform: UNKNOWN
Requires: numpy
Requires: numba
Requires-Python: >=3
Description-Content-Type: text/markdown
