From 641e16db2c532064faf7501b82c7b0e28fe796d8 Mon Sep 17 00:00:00 2001 From: fliklab Date: Sat, 19 Oct 2019 16:54:22 +0900 Subject: [PATCH 1/7] Update visualization.md MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit # Apply reviewer's comment 인자(argument)의 설명에서 값을 설정한다는 의미의 control 번역시, '조정' 대신 '설정'으로 사용함. review 내용과 다른 문서의 번역내용을 참고하여 수정하였습니다. --- sources/visualization.md | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/sources/visualization.md b/sources/visualization.md index 5aab599b..968106d5 100644 --- a/sources/visualization.md +++ b/sources/visualization.md @@ -10,10 +10,10 @@ plot_model(model, to_file='model.png') `plot_model`은 네 가지 인자를 전달받습니다: -- `show_shapes`(기본값은 `False`)는 결과의 형태을 그래프에 나타낼 것인지 조정합니다. -- `show_layer_names`(기본값은 `True`)는 층의 이름을 그래프에 나타낼 것인지 조정합니다. -- `expand_nested`(기본값은 `False`)는 중첩된 모델을 그래프상에서 클러스터로 확장할 것인지 조정합니다. -- `dpi`(기본값은 96)는 이미지 dpi를 조정합니다. +- `show_shapes`(기본값은 `False`)는 결과의 형태을 그래프에 나타낼지 여부를 설정합니다. +- `show_layer_names`(기본값은 `True`)는 층의 이름을 그래프에 나타낼 것인지 여부를 설정합니다. +- `expand_nested`(기본값은 `False`)는 중첩된 모델을 그래프상에서 클러스터로 확장할 것인지 여부를 설정합니다. +- `dpi`(기본값은 96)는 이미지 dpi를 설정합니다. 또한 직접 `pydot.Graph`오브젝트를 만들어 사용할 수도 있습니다, 예를들어 ipython notebook에 나타내자면 : From 3f5eed8aacac5648cb27b0fb75a355cc9df7fe90 Mon Sep 17 00:00:00 2001 From: fliklab Date: Sat, 19 Oct 2019 16:55:39 +0900 Subject: [PATCH 2/7] Update visualization.md --- sources/visualization.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sources/visualization.md b/sources/visualization.md index 968106d5..123ad67b 100644 --- a/sources/visualization.md +++ b/sources/visualization.md @@ -10,7 +10,7 @@ plot_model(model, to_file='model.png') `plot_model`은 네 가지 인자를 전달받습니다: -- `show_shapes`(기본값은 `False`)는 결과의 형태을 그래프에 나타낼지 여부를 설정합니다. +- `show_shapes`(기본값은 `False`)는 결과의 형태을 그래프에 나타낼 것인지 여부를 설정합니다. - `show_layer_names`(기본값은 `True`)는 층의 이름을 그래프에 나타낼 것인지 여부를 설정합니다. - `expand_nested`(기본값은 `False`)는 중첩된 모델을 그래프상에서 클러스터로 확장할 것인지 여부를 설정합니다. - `dpi`(기본값은 96)는 이미지 dpi를 설정합니다. From edeabac93ca533f1280933a270f04ddbd39721c4 Mon Sep 17 00:00:00 2001 From: fliklab Date: Sat, 19 Oct 2019 17:11:17 +0900 Subject: [PATCH 3/7] Update visualization.md MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit 콜론을 마침표로 변경 --- sources/visualization.md | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/sources/visualization.md b/sources/visualization.md index 123ad67b..6385574e 100644 --- a/sources/visualization.md +++ b/sources/visualization.md @@ -2,21 +2,21 @@ 케라스는(`graphviz`를 사용해서) 케라스 모델을 그래프로 그리기 위한 유틸리티 함수를 제공합니다. -아래 예시는 모델의 그래프를 그려주고 그 결과를 파일로 저장합니다: +아래 예시는 모델의 그래프를 그려주고 그 결과를 파일로 저장합니다. ```python from keras.utils import plot_model plot_model(model, to_file='model.png') ``` -`plot_model`은 네 가지 인자를 전달받습니다: +`plot_model`은 네 가지 인자를 전달받습니다. - `show_shapes`(기본값은 `False`)는 결과의 형태을 그래프에 나타낼 것인지 여부를 설정합니다. - `show_layer_names`(기본값은 `True`)는 층의 이름을 그래프에 나타낼 것인지 여부를 설정합니다. - `expand_nested`(기본값은 `False`)는 중첩된 모델을 그래프상에서 클러스터로 확장할 것인지 여부를 설정합니다. - `dpi`(기본값은 96)는 이미지 dpi를 설정합니다. -또한 직접 `pydot.Graph`오브젝트를 만들어 사용할 수도 있습니다, -예를들어 ipython notebook에 나타내자면 : +또한 직접 `pydot.Graph`오브젝트를 만들어 사용할 수도 있습니다. +아래는 ipython notebook에 나타낸 예시입니다. ```python from IPython.display import SVG from keras.utils import model_to_dot From e70c57c8b7b06e41ddaaf42628d2f18789288405 Mon Sep 17 00:00:00 2001 From: fliklab Date: Sat, 19 Oct 2019 17:14:33 +0900 Subject: [PATCH 4/7] Update visualization.md MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit ipython notebook의 표기를 IPython Notebook으로 수정 (참고 : https://ipython.org/) --- sources/visualization.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sources/visualization.md b/sources/visualization.md index 6385574e..ca3a0341 100644 --- a/sources/visualization.md +++ b/sources/visualization.md @@ -16,7 +16,7 @@ plot_model(model, to_file='model.png') - `dpi`(기본값은 96)는 이미지 dpi를 설정합니다. 또한 직접 `pydot.Graph`오브젝트를 만들어 사용할 수도 있습니다. -아래는 ipython notebook에 나타낸 예시입니다. +아래는 IPython Notebook에 나타낸 예시입니다. ```python from IPython.display import SVG from keras.utils import model_to_dot From 31a54c07794c52ee7990cba9df0ff3a092ee63e8 Mon Sep 17 00:00:00 2001 From: fliklab Date: Sat, 19 Oct 2019 17:24:04 +0900 Subject: [PATCH 5/7] Update visualization.md --- sources/visualization.md | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/sources/visualization.md b/sources/visualization.md index ca3a0341..e37ccbca 100644 --- a/sources/visualization.md +++ b/sources/visualization.md @@ -1,8 +1,8 @@ ## 모델 시각화 -케라스는(`graphviz`를 사용해서) 케라스 모델을 그래프로 그리기 위한 유틸리티 함수를 제공합니다. +케라스는(`graphviz`를 사용해서) 케라스 모델을 그래프로 그리기 위한 유틸리티 함수를 제공합니다. -아래 예시는 모델의 그래프를 그려주고 그 결과를 파일로 저장합니다. +아래 예시는 모델의 그래프를 그려주고 그 결과를 파일로 저장합니다. ```python from keras.utils import plot_model plot_model(model, to_file='model.png') @@ -15,8 +15,8 @@ plot_model(model, to_file='model.png') - `expand_nested`(기본값은 `False`)는 중첩된 모델을 그래프상에서 클러스터로 확장할 것인지 여부를 설정합니다. - `dpi`(기본값은 96)는 이미지 dpi를 설정합니다. -또한 직접 `pydot.Graph`오브젝트를 만들어 사용할 수도 있습니다. -아래는 IPython Notebook에 나타낸 예시입니다. +또한 직접 `pydot.Graph`오브젝트를 만들어 사용할 수도 있습니다. +아래는 IPython Notebook에서 나타낸 예시입니다. ```python from IPython.display import SVG from keras.utils import model_to_dot From d6a6ac1a983ffe4076d51f7faab992efb4ed518f Mon Sep 17 00:00:00 2001 From: fliklab Date: Sat, 19 Oct 2019 17:27:51 +0900 Subject: [PATCH 6/7] Update visualization.md delete spaces at the end of the lines --- sources/visualization.md | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/sources/visualization.md b/sources/visualization.md index e37ccbca..f2684e1c 100644 --- a/sources/visualization.md +++ b/sources/visualization.md @@ -1,8 +1,8 @@ ## 모델 시각화 -케라스는(`graphviz`를 사용해서) 케라스 모델을 그래프로 그리기 위한 유틸리티 함수를 제공합니다. +케라스는(`graphviz`를 사용해서) 케라스 모델을 그래프로 그리기 위한 유틸리티 함수를 제공합니다. -아래 예시는 모델의 그래프를 그려주고 그 결과를 파일로 저장합니다. +아래 예시는 모델의 그래프를 그려주고 그 결과를 파일로 저장합니다. ```python from keras.utils import plot_model plot_model(model, to_file='model.png') @@ -15,7 +15,7 @@ plot_model(model, to_file='model.png') - `expand_nested`(기본값은 `False`)는 중첩된 모델을 그래프상에서 클러스터로 확장할 것인지 여부를 설정합니다. - `dpi`(기본값은 96)는 이미지 dpi를 설정합니다. -또한 직접 `pydot.Graph`오브젝트를 만들어 사용할 수도 있습니다. +또한 직접 `pydot.Graph`오브젝트를 만들어 사용할 수도 있습니다. 아래는 IPython Notebook에서 나타낸 예시입니다. ```python from IPython.display import SVG From 0d4dc5ec2701cb658c5116ddf3cc2dc7c0ac5747 Mon Sep 17 00:00:00 2001 From: fliklab Date: Sun, 20 Oct 2019 19:13:51 +0900 Subject: [PATCH 7/7] Apply Naruu's reivew MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit argument의 설명에서 '여부' 단어 삭제 - ~나타낼 것인지 여부를 설정합니다 -> ~표시할지를 설정합니다. - ~여부를 설정합니다 -> ~것인지를 설정합니다. 예시에 대한 설명 수정 - '아래는 ~~하는 예시입니다.'형식으로 수정 기타 수정 - 형태을 -> 형태를 - 오브젝트 -> 객체 - 그래프로 그리기 위한 유틸리티 함수-> 그래프로 그려주는 유틸리티 함수 - 케라스는(graphviz를 사용해서)-> 케라스는 graphviz를 통해 --- sources/visualization.md | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-) diff --git a/sources/visualization.md b/sources/visualization.md index f2684e1c..6a77071b 100644 --- a/sources/visualization.md +++ b/sources/visualization.md @@ -1,8 +1,8 @@ ## 모델 시각화 -케라스는(`graphviz`를 사용해서) 케라스 모델을 그래프로 그리기 위한 유틸리티 함수를 제공합니다. +케라스는 `graphviz`를 통해 케라스 모델을 그래프로 그려주는 유틸리티 함수를 제공합니다. -아래 예시는 모델의 그래프를 그려주고 그 결과를 파일로 저장합니다. +아래는 모델의 그래프를 그리고 파일로 저장하는 예시입니다. ```python from keras.utils import plot_model plot_model(model, to_file='model.png') @@ -10,12 +10,12 @@ plot_model(model, to_file='model.png') `plot_model`은 네 가지 인자를 전달받습니다. -- `show_shapes`(기본값은 `False`)는 결과의 형태을 그래프에 나타낼 것인지 여부를 설정합니다. -- `show_layer_names`(기본값은 `True`)는 층의 이름을 그래프에 나타낼 것인지 여부를 설정합니다. -- `expand_nested`(기본값은 `False`)는 중첩된 모델을 그래프상에서 클러스터로 확장할 것인지 여부를 설정합니다. +- `show_shapes`(기본값은 `False`)는 결과의 형태를 그래프에 표시할지를 설정합니다. +- `show_layer_names`(기본값은 `True`)는 층의 이름을 그래프에 표시할지를 설정합니다. +- `expand_nested`(기본값은 `False`)는 중첩된 모델을 그래프상에서 클러스터로 확장할 것인지를 설정합니다. - `dpi`(기본값은 96)는 이미지 dpi를 설정합니다. -또한 직접 `pydot.Graph`오브젝트를 만들어 사용할 수도 있습니다. +또한 직접 `pydot.Graph`객체를 만들어 사용할 수도 있습니다. 아래는 IPython Notebook에서 나타낸 예시입니다. ```python from IPython.display import SVG @@ -26,7 +26,7 @@ SVG(model_to_dot(model).create(prog='dot', format='svg')) ## 학습 히스토리 시각화 -케라스 `Model`의 `fit()` 메소드는 `History` 오브젝트를 반환합니다. `History.history` 속성attribute은 각 에폭마다 계산된 학습 손실 및 평가 지표가 순서대로 기록된 딕셔너리입니다. 검증 데이터를 적용한 경우에는 해당 손실 및 지표도 함께 기록됩니다. 아래는 `matplotlib`을 사용하여 학습 및 검증의 손실과 정확도 그래프를 그리는 예시입니다. +케라스 `Model`의 `fit()` 메소드는 `History` 객체를 반환합니다. `History.history` 속성attribute은 각 에폭마다 계산된 학습 손실 및 평가 지표가 순서대로 기록된 딕셔너리입니다. 검증 데이터를 적용한 경우에는 해당 손실 및 지표도 함께 기록됩니다. 아래는 `matplotlib`을 사용하여 학습 및 검증의 손실과 정확도 그래프를 그리는 예시입니다. ```python import matplotlib.pyplot as plt