kiyukuta
/ autoencoder.py
Last active
January 23, 2020 06:16
Minimum implementation of denoising autoencoder.Error function is cross-entropy of reconstruction.Optimizing by SGD with mini-batch.Dataset is available at http://deeplearning.net/data/mnist/mnist.pkl.gz
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| #coding: utf8 | |
| """ | |
| 1. Download this gist. | |
| 2. Get the MNIST data. | |
| wget http://deeplearning.net/data/mnist/mnist.pkl.gz | |
| 3. Run this code. | |
| python autoencoder.py 100 -e 1 -b 20 -v | |
| """ | |
| import numpy | |
| import argparse |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| # -*- coding: utf-8 -*- | |
| # 句読点のない文字列を文単位に区切る | |
| # | |
| # 基本的にはルールベースでルールが適用できない場合は機械学習器で判定する。 | |
| # これは青空文庫のデータ( http://www.udp.jp/misc/aozora-data/ )から学習した。 | |
| # * ruby 1.8の場合は $KCODE='u'としないと動かない | |
| =begin | |
| s = <<TEXT | |
| 昼飯のスパゲティナポリタンを眺めながら |