No description, website, or topics provided.
Python
Latest commit 8640b18 Oct 21, 2016 @zacapa zacapa committed on GitHub Create nico_nslstm.py
Permalink
Failed to load latest commit information.
README.md modify Oct 12, 2016
maesyori1.py modify Oct 12, 2016
maesyori2.py mod Oct 12, 2016
nico_lstm.py Update nico_lstm.py Oct 12, 2016
nico_nslstm.py Create nico_nslstm.py Oct 21, 2016
play_lstm.py Update play_lstm.py Oct 12, 2016

README.md

chainer-nicolstm

Chainer implementation of estimation next word from current words.
(http://qiita.com/ixixi/items/a3d56b2db6e09249a519)

About

This model trains niconico-douga comments, estimates comment from bag-of-comment. For example,

Input Output
┗(^ ┗(^o^ )┓三
日本語 日本語でおk
/hi /hidden
おっく おっくせんまん!おっくせんまん!
らんら らんらんるー
ξ*・ ξ・ヮ・
わっふ わっふるわっふる
かわい かわいいww

More description is here. ( http://www.monthly-hack.com/ )

Requirements

  • Python 2.7 or 3.4
  • Chainer 1.16.0

Usage

Step1

Downloads data from NII.
ニコニコ動画コメント等データ(http://www.nii.ac.jp/dsc/idr/nico/nicocomm-apply.html)

Step2

Get latest 10 comments from each nico-douga.
It outputs 'last10comments.pkl'.

$ python maesyori1.py /path/to/data/thread/

Step3

Get sample comments that were selected randomly.
It outputs 'sample_texts.pkl' and 'sample_vocab.pkl'.

$ python maesyori2.py

Step4

Train comment-data by lstm.
It outputs '/result/lstm_model.npz'.

$ python nico_lstm.py -g <gpu_id>

Step5

Play generating comments.
Input bag of comments.

$ python play_lstm.py
>> ┗(^
┗(^o^ )┓三
>> 日本語
日本語でおk
>>