Skip to content

JA: add ja/r2/guide/eager.ipynb#752

Closed
chie8842 wants to merge 4 commits into
tensorflow:masterfrom
chie8842:eager_r2
Closed

JA: add ja/r2/guide/eager.ipynb#752
chie8842 wants to merge 4 commits into
tensorflow:masterfrom
chie8842:eager_r2

Conversation

@chie8842
Copy link
Copy Markdown
Contributor

@chie8842 chie8842 commented Jul 5, 2019

No description provided.

@googlebot googlebot added the cla: yes CLA has been signed label Jul 5, 2019
@tfdocsbot
Copy link
Copy Markdown
Collaborator

tfdocsbot commented Jul 5, 2019

Preview and run these notebook edits with Google Colab:

Notebook diffs available on ReviewNB.com.

@tfdocsbot tfdocsbot added ja translation Community-provided translation for certain languages labels Jul 5, 2019
@tfdocsbot
Copy link
Copy Markdown
Collaborator

Reviewers added, please take a look.
@ohtaman, @sfujiwara, @masa-ita

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

@chie8842 chie8842 marked this pull request as ready for review July 5, 2019 11:53
Copy link
Copy Markdown
Member

@lamberta lamberta left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks.
Please move this into the site/ja/beta/guide directory.
Also, update the buttons to match

@lamberta lamberta dismissed their stale review July 10, 2019 22:41

Applied

@lamberta
Copy link
Copy Markdown
Member

Added commits to update path and fix notebook. Switched to GPU to train faster (#780)

Will keep this pull request open to allow time for translation reviews.

Copy link
Copy Markdown
Contributor

@sfujiwara sfujiwara left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@chie8842 Sorry for late review and thank you for your translation!
Your translation is almost perfect, but I found some typo and have suggestions.
Check my comments!

"source": [
"\n",
"\n",
"Tensorflow の Eager Execution は、計算グラフの作成と評価を同時におこなう命令的なプログラミングを行うための環境です:\n",
Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

==> TensorFlow

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

fixed.

"* *より簡単なデバッグ*— ops を直接呼び出すことで、実行中のモデルを調査したり、変更をテストすることができます。 Python 標準のデバッグツールを用いて即座にエラーのレポーティングができます。\n",
"* *自然な制御フロー*— TensorFlow のグラフ制御フローの代わりに Python の制御フローを利用するため、動的なモデルのパラメータ変更をシンプルに行うことができます。\n",
" \n",
"Eager Execution は Tensorflow のほとんどのオペレーションとGPUアクセラレーションをサポートします。\n",
Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

==> TensorFlow

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

fixed.

"\n",
"Eager Execution は研究や実験のための柔軟な機械学習環境として、以下を提供します。\n",
"\n",
"* *直感的なインタフェース*— Python のデータ構造を使用して、コードをナチュラルに記述することができます。スモールなモデルとデータに対してすばやく実験を繰り返すことができます。\n",
Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

[weak suggestion]

ナチュラルに ==> 自然に
スモールな ==> 小規模な

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

fixed.

"\n",
"* *直感的なインタフェース*— Python のデータ構造を使用して、コードをナチュラルに記述することができます。スモールなモデルとデータに対してすばやく実験を繰り返すことができます。\n",
"* *より簡単なデバッグ*— ops を直接呼び出すことで、実行中のモデルを調査したり、変更をテストすることができます。 Python 標準のデバッグツールを用いて即座にエラーのレポーティングができます。\n",
"* *自然な制御フロー*— TensorFlow のグラフ制御フローの代わりに Python の制御フローを利用するため、動的なモデルのパラメータ変更をシンプルに行うことができます。\n",
Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

[discussion]

制御フロー or control flow

Control flow is a technical term for TensorFlow:
https://www.tensorflow.org/api_docs/cc/group/control-flow-ops

So, I think we don't need to translate it.
But 制御フロー is good and natural translation, so I can not tell which is better.
How is your opinion?

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

[suggestion]

動的なモデルのパラメータ変更 ==> 動的なモデルの作成

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

制御フロー or control flow

I'd like "制御フロー" because Chinese documents translate it.
And in the another part of documents such as autograph uses "制御フロー", too..

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

動的なモデルのパラメータ変更 ==> 動的なモデルの作成

specification's direct translation isn't "作成", but it's natural. So I use your suggestion.

"Eager Execution を有効化することで、 TensorFlow の挙動は変わります—TensorFlowは即座に式を評価して結果をPythonに返すようになります。\n",
"`tf.Tensor` オブジェクトは計算グラフのノードへのシンボリックハンドルの代わりに具体的な値を参照します。\n",
"セッションの中で構築して実行する計算グラフが存在しないため、`print()`やデバッガを使って容易に結果を調べることができます。\n",
"勾配計算を終了することなくテンソル値を評価、出力、およびチェックすることができます。\n",
Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

[suggestion]
I think 終了することなく ==> 遮ることなく is more natural translation.

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

fixed.

"source": [
"## Eager Execution の途中でオブジェクトのステータスを使用する\n",
"\n",
"TF 1.x の Graph Execution では、プログラムの状態(変数など)は global collection に格納され、それらの存続期間は `tf.Session` オブジェクトによって管理されます。\n",
Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I commented about Variable above.

Copy link
Copy Markdown
Contributor

@sfujiwara sfujiwara Jul 26, 2019

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Variable and queue are only objects which have "state" (that is, mutabule in session) in TF v1.
They refer to it here.
So, I guess it's better to distinguish Variable (a technical term for TensorFlow) and 変数 (a technical term for general programming or mathematical optimization) explicitly.

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

the original document in English uses (such as the variables).
variables in this case use lowercase, so I doubt if we should translate or use handle as tf.Variable.
I translated it as (Variableなど).(used capital V)

"id": "xEL4yJe5qEhD"
},
"source": [
"## 高度な自動分類トピック\n",
Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

自動分類 ==> 自動微分

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

fixed.

"### カスタム勾配\n",
"\n",
"カスタム勾配は、勾配を上書きする簡単な方法です。 フォワード関数では、\n",
"入力、出力、または中間結果に関する勾配を定義します。たとえば、逆方向パスにおいて勾配のノルムを切り取る簡単な方法は次のとおりです:"
Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

[weak suggestion]
I guess it's more natural to translate clip to 制限する or 固定する (not 切り取る)

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

fixed.

"colab": {}
},
"source": [
"# 購買計算は x = 0 のときはうまくいく。\n",
Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

typo: 購買 ==> 勾配

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

fixed.

"Eager Execution は開発とデバッグをより対話的にしますが、\n",
"TensorFlow 1.x スタイルの Graph Execution は分散学習、パフォーマンスの最適化、そしてプロダクション環境へのデプロイの観点で利点があります。\n",
"\n",
"2つの手法のギャップを埋めるために、 TensorFlow 2.0 は `tf.function` API に似た `function` を紹介します。\n",
Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

[suggestion]

TensorFlow 2.0 は `tf.function` API に似た `function` を紹介します。
==>
TensorFlow 2.0 では `tf.function` という機能を導入しています。

Copy link
Copy Markdown
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

fixed.

@lamberta lamberta added awaiting-contributor-response Indicates a question, draft, or work-in-progress and removed awaiting-translation-review labels Jul 31, 2019
@lamberta
Copy link
Copy Markdown
Member

Hi @chie8842, can you please apply the reviewer suggestions and then I can merge this pull request. Thank you

@lamberta
Copy link
Copy Markdown
Member

Closing this pull requests because of inactivity. Feel free to apply the reviewer suggestions and re-open when ready. Thanks

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

awaiting-contributor-response Indicates a question, draft, or work-in-progress cla: yes CLA has been signed ja translation Community-provided translation for certain languages

Projects

None yet

Development

Successfully merging this pull request may close these issues.

5 participants