Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

JA: Translated site/ja/beta/tutorials/eager/custom_layers.ipynb #830

Conversation

@masa-ita
Copy link
Contributor

commented Jul 18, 2019

No description provided.

@tfdocsbot

This comment has been minimized.

Copy link
Collaborator

commented Jul 18, 2019

Preview and run these notebook edits with Google Colab:

Notebook diffs available on ReviewNB.com.
@tfdocsbot

This comment has been minimized.

Copy link
Collaborator

commented Jul 18, 2019

Reviewers added, please take a look.
@ohtaman, @sfujiwara

When your review is finished, approve the pull request or include "LGTM" in your comment.

@googlebot googlebot added the cla: yes label Jul 18, 2019

@AseiSugiyama
Copy link
Contributor

left a comment

@masa-ita I reviewed your PR, and it is almost perfect. I left some comment about it, please check it.

"layer = tf.keras.layers.Dense(100)\n",
"# 入力の次元数は多くの場合不要となっています。それは、レイヤーが最初に使われる際に\n",
"# 推定可能だからです。ただし、複雑なモデルを構築する場合などでは、\n",
"# 手動で設定するために、引数として渡してやることも可能です。\n",

This comment has been minimized.

Copy link
@AseiSugiyama

AseiSugiyama Sep 7, 2019

Contributor

[suggestion]

# 推定可能だからです。ただし、複雑なモデルを構築する場合などでは、
# 手動で設定するために、引数として渡してやることも可能です。
==>
# 推定可能だからです。ただし、引数として渡すことで手動で指定することも可能です。
# これは複雑なモデルを構築する場合に役に立つでしょう。

This comment has been minimized.

Copy link
@masa-ita

masa-ita Sep 7, 2019

Author Contributor

Thank you. It's better! I fixed it.

"\n",
"独自のレイヤーを実装する最良の方法は、tf.keras.Layer クラスを拡張し、下記のメソッドを実装することです。\n",
" * `__init__` , 入力に依存しないすべての初期化を行う\n",
" * `build`, 入力の形状を知った上で、残りの初期化を行う\n",

This comment has been minimized.

Copy link
@AseiSugiyama

AseiSugiyama Sep 7, 2019

Contributor

[suggestion]
The word shapes used here is a technical term of Keras and Japanese translation of Keras Documentation does not translate the word shape(s).

I guess the below is better for consistency:
入力のshapeを知った上で、残りの初期化を行う

This comment has been minimized.

Copy link
@masa-ita

masa-ita Sep 7, 2019

Author Contributor

Thanks. I've modified it.

"\n",
"機械学習では、多くのレイヤーに類するものが、既存のレイヤーを組み合わせることで実装されています。例えば、ResNetの残差ブロックは、畳込み、バッチ正規化とショートカットの組み合わせです。\n",
"\n",
"他のレイヤーからなるレイヤーに類するものを定義する際の主役は、tf.keras.Model クラスです。このクラスを継承することで実装することができます。"

This comment has been minimized.

Copy link
@AseiSugiyama

AseiSugiyama Sep 7, 2019

Contributor

I guess できます is better than することができます.

This comment has been minimized.

Copy link
@masa-ita

masa-ita Sep 7, 2019

Author Contributor

Thank you. I agree.

"id": "wYfucVw65PMj"
},
"source": [
"しかし、ほとんどの場合には、モデルはレイヤーを次々に呼び出すことで構成されます。tf.keras.Sequential クラスを使うことで、これを最少の行数で行うことができます。"

This comment has been minimized.

Copy link
@AseiSugiyama

AseiSugiyama Sep 7, 2019

Contributor

[suggestion]

これを最少の行数で行うことができます
==>
これをかなり短いコードで実装できます。

This comment has been minimized.

Copy link
@masa-ita

masa-ita Sep 7, 2019

Author Contributor

Thank you. It's better. I fixed it.

masa-ita added 10 commits Sep 7, 2019
rebased and move custom_layers.ipynb
from site/ja/beta/tutorials/eager/ to site/ja/tutorilas/customization/
@masa-ita

This comment has been minimized.

Copy link
Contributor Author

commented Sep 7, 2019

I've rebased and move custom_layer.ipynb from site/ja/beta/tutorials/eager/ to site/ja/tutorials/customization/.

lamberta added 2 commits Sep 9, 2019
@lamberta
Copy link
Member

left a comment

Thanks for the translation and review!

TensorFlow-Docs-Copybara pushed a commit that referenced this pull request Sep 9, 2019
Copybara-Service
Merge pull request #830 from masa-ita:site_ja_beta_tutorials_eager_cu…
…stom_layers

PiperOrigin-RevId: 268091389

@TensorFlow-Docs-Copybara TensorFlow-Docs-Copybara merged commit cf9132a into tensorflow:master Sep 9, 2019

3 checks passed

Ubuntu Sanity Check Internal CI build successful
Details
cla/google All necessary CLAs are signed
import/copybara Change imported to the internal review system
Details
@masa-ita

This comment has been minimized.

Copy link
Contributor Author

commented Sep 10, 2019

@lamberta Thank you. I'm sorry to forget to fix them.

@masa-ita masa-ita deleted the masa-ita:site_ja_beta_tutorials_eager_custom_layers branch Sep 10, 2019

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
7 participants
You can’t perform that action at this time.