* -1->None (#823)

* unify paddle 1.6 api in understand_sentiment (#824)


* unify paddle 1.6 api in understand_sentiment

* Upgrade w2v & srl's api (#828)

* Upgrade w2v & srl's api

* Upgrade label semantic roles api

* Rewrite 08.machine_translation using Paddle-1.6 apis. (#826)

* Rewrite 08.machine_translation using Paddle-1.6 apis.

* Delete the old train.py in 08.machine_translation

* Update train.py to seq2seq.py in README_cn 08.machine_translation.

* Fix the print content of seq2seq.py.

* Update code format in README_cn of 08.machine_translation.

* add 1.6 requirement (#830)

* fix bugs in Readme.md of 04 and 07 (#831) (#832)

* Polish optimizer name in 04.word2vec (#841) (#842)

* change opt name in word2vec

* update index.cn.html

* Cherry pick for paddle 1.6 (#843)
Co-authored-by: Nceci3 <ceci3@users.noreply.github.com>
Co-authored-by: NLi Fuchen <lfchener@outlook.com>
Co-authored-by: NYibing Liu <liuyibing01@baidu.com>
Co-authored-by: NGuo Sheng <whucsgs@163.com>
Co-authored-by: Nruri <shipeng1108@163.com>
Co-authored-by: NChen Weihang <sunny_cwh@163.com>

* change opt name in word2vec

* update index.cn.html

fix : http://newicafe.baidu.com:80/issue/DLTP-1929/show?from=page， fix book code is not compatible with python3

use numpy.asarray as data initializer

* Fixes anchor issues, except for ch1.

* Remove old files.

Compressing the model dramatically slows down model read and write
with a little benefit: model compression ratio is very low.

* Seperate convert_jinja2_into_html into different repo.
  * It seems a bug of pre-commit tools