From 742abf5b5913bfa0e4bd5f79a61287c227ebc404 Mon Sep 17 00:00:00 2001 From: Ravi Kumar Thakur Date: Sun, 2 Jun 2019 20:04:08 +0530 Subject: [PATCH 1/2] Update deep_cnn.md 1. Changed 'chapter' to 'sections' while referring to AlexNet paper 2. Added the word 'enhancements'(image processing) for brightness and contrast --- site/en/tutorials/images/deep_cnn.md | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-) diff --git a/site/en/tutorials/images/deep_cnn.md b/site/en/tutorials/images/deep_cnn.md index 827bae96332..823fa310200 100644 --- a/site/en/tutorials/images/deep_cnn.md +++ b/site/en/tutorials/images/deep_cnn.md @@ -39,7 +39,7 @@ designing larger and more sophisticated models in TensorFlow: `tf.nn.max_pool` ([wiki](https://en.wikipedia.org/wiki/Convolutional_neural_network#Pooling_layer)) and `tf.nn.local_response_normalization` -(Chapter 3.3 in +(section 3.3 in [AlexNet paper](https://papers.nips.cc/paper/4824-imagenet-classification-with-deep-convolutional-neural-networks.pdf)). * [Visualization](../../guide/summaries_and_tensorboard.md) of network activities during training, including input images, @@ -125,12 +125,12 @@ The images are processed as follows: * They are `tf.image.per_image_standardization` to make the model insensitive to dynamic range. -For training, we additionally apply a series of random distortions to +For training, we additionally apply a series of random enhancements and distortions to artificially increase the data set size: * `tf.image.random_flip_left_right` the image from left to right. -* Randomly distort the `tf.image.random_brightness`. -* Randomly distort the `tf.image.random_contrast`. +* Randomly enhance the image`tf.image.random_brightness`. +* Randomly enhance the image`tf.image.random_contrast`. Please see the [Images](../../api_guides/python/image.md) page for the list of available distortions. We also attach an From 8d9b3946c910dbb2957233a1afc37c3d8fe6c8d2 Mon Sep 17 00:00:00 2001 From: Billy Lamberta Date: Mon, 3 Jun 2019 10:06:25 -0700 Subject: [PATCH 2/2] wording --- site/en/tutorials/images/deep_cnn.md | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) diff --git a/site/en/tutorials/images/deep_cnn.md b/site/en/tutorials/images/deep_cnn.md index 823fa310200..9a00a31190e 100644 --- a/site/en/tutorials/images/deep_cnn.md +++ b/site/en/tutorials/images/deep_cnn.md @@ -125,12 +125,12 @@ The images are processed as follows: * They are `tf.image.per_image_standardization` to make the model insensitive to dynamic range. -For training, we additionally apply a series of random enhancements and distortions to +For training, we additionally apply a series of random distortions to artificially increase the data set size: * `tf.image.random_flip_left_right` the image from left to right. -* Randomly enhance the image`tf.image.random_brightness`. -* Randomly enhance the image`tf.image.random_contrast`. +* Randomly distort the image with `tf.image.random_brightness`. +* Randomly distort the image with `tf.image.random_contrast`. Please see the [Images](../../api_guides/python/image.md) page for the list of available distortions. We also attach an