{ "id": 436213, "title": "Need Help with KerasCV Starter Notebook : 'keras_cv.layers' has no attribute 'Augmenter'", "url": "/competitions/rsna-2023-abdominal-trauma-detection/discussion/436213", "author_name": "Jude TCHAYE", "post_date": "2023-09-01T11:37:57.994000", "votes": 0, "comment_count": 5, "views": 0, "content": "

Hello Kaggle community,

\n

I've been working with the KerasCV starter notebook recently, and I've encountered an issue.

\n

I'm trying to run the following code snippet \" https://www.kaggle.com/code/tchaye59/attributeerror-augmenter-not-found \":

\n
augmenter = keras_cv.layers.Augmenter(\n        [\n            keras_cv.layers.RandomFlip(mode=),\n            keras_cv.layers.RandomCutout(height_factor=, width_factor=),\n\n        ]\n    )\n
\n

However, I'm getting the following error:

\n

AttributeError: module 'keras_cv.layers' has no attribute 'Augmenter'

\n

I've spent a considerable amount of time trying to troubleshoot this error, but I'm still unable to find a solution. Any help or guidance on how to resolve this issue would be greatly appreciated.

\n

Thank you in advance for your assistance!

", "messages": [ { "id": 2418888, "postDate": "2023-09-01T15:13:07.107Z", "content": "

just replace keras_cv.layers.Augmenter with keras.Sequential
\nmore details here: https://github.com/keras-team/keras-cv/issues/1548
\nThe notebook will be updated too.

", "rawMarkdown": "just replace ```keras_cv.layers.Augmenter``` with ```keras.Sequential```\nmore details here: https://github.com/keras-team/keras-cv/issues/1548\nThe notebook will be updated too.", "votes": 4 }, { "id": 2420818, "postDate": "2023-09-03T00:18:56.573Z", "content": "

Found an alternative way to do it as well

\n

def apply_augmentation(image,labels):
\n image = preprocessing.RandomFlip(mode=\"horizontal_and_vertical\")(image)
\n image = preprocessing.RandomTranslation(height_factor=0.2, width_factor=0.2)(image)
\n return (image, labels)

", "rawMarkdown": "Found an alternative way to do it as well\n\ndef apply_augmentation(image,labels):\n image = preprocessing.RandomFlip(mode=\"horizontal_and_vertical\")(image) \n image = preprocessing.RandomTranslation(height_factor=0.2, width_factor=0.2)(image)\n return (image, labels)\n\n\n", "votes": 2 }, { "id": 2419267, "postDate": "2023-09-01T21:55:26.197Z", "content": "

Thank you, everyone! Here's how I successfully worked around the issue:

\n
 os\n\n\nos.environ[] = \n\n keras_cv\n\n tensorflow  keras\n\n tensorflow.keras  layers\n\naugmenter = keras.Sequential(\n  layers = [\n      tf.keras.layers.Lambda(keras_cv.layers.RandomCutout(height_factor=, \n                                             width_factor=)),\n      tf.keras.layers.Lambda(keras_cv.layers.RandomFlip(mode=)),\n   ]\n)\n
", "rawMarkdown": "Thank you, everyone! Here's how I successfully worked around the issue:\n\n```python\n\nimport os\n# You can use `tensorflow`, `pytorch`, `jax` here\n# KerasCore makes the notebook backend agnostic :)\nos.environ[\"KERAS_BACKEND\"] = \"tensorflow\"\n\nimport keras_cv\n#import keras_core as keras\nfrom tensorflow import keras\n#from keras_core import layers\nfrom tensorflow.keras import layers\n\naugmenter = keras.Sequential(\n layers = [\n tf.keras.layers.Lambda(keras_cv.layers.RandomCutout(height_factor=0.2, \n width_factor=0.2)),\n tf.keras.layers.Lambda(keras_cv.layers.RandomFlip(mode=\"horizontal_and_vertical\")),\n ]\n)\n```", "votes": 2 }, { "id": 2418838, "postDate": "2023-09-01T14:47:56.720Z", "content": "

I get the same error running with either of the two GPU resources on kaggle.

\n

When running on my local machine it does NOT fail at that same point

\n
 -  fails   model.fit cell apparently  a   NVIDIA hell.  Tensorflow  Keras always a bitch   working   system when you want your GPU's  play nicely.  \n
", "rawMarkdown": "I get the same error running with either of the two GPU resources on kaggle.\n\nWhen running on my local machine it does NOT fail at that same point \n\n - but fails in the model.fit cell apparently as a result of NVIDIA hell. Tensorflow and Keras always a bitch to get working on local system when you want your GPU's to play nicely. " }, { "id": 2418835, "postDate": "2023-09-01T14:45:19.117Z", "content": "

I'm facing the same issue and haven't had any luck trying to solve it so far.
\nCould you please help here? @awsaf49 @fchollet @gusthema @ianstenbit @philculliton

", "rawMarkdown": "I'm facing the same issue and haven't had any luck trying to solve it so far.\nCould you please help here? @awsaf49 @fchollet @gusthema @ianstenbit @philculliton " }, { "id": 2418564, "postDate": "2023-09-01T11:37:57.993Z", "content": "

Hello Kaggle community,

\n

I've been working with the KerasCV starter notebook recently, and I've encountered an issue.

\n

I'm trying to run the following code snippet \" https://www.kaggle.com/code/tchaye59/attributeerror-augmenter-not-found \":

\n
augmenter = keras_cv.layers.Augmenter(\n        [\n            keras_cv.layers.RandomFlip(mode=),\n            keras_cv.layers.RandomCutout(height_factor=, width_factor=),\n\n        ]\n    )\n
\n

However, I'm getting the following error:

\n

AttributeError: module 'keras_cv.layers' has no attribute 'Augmenter'

\n

I've spent a considerable amount of time trying to troubleshoot this error, but I'm still unable to find a solution. Any help or guidance on how to resolve this issue would be greatly appreciated.

\n

Thank you in advance for your assistance!

", "rawMarkdown": "Hello Kaggle community,\n\nI've been working with the [KerasCV starter notebook ](https://www.kaggle.com/code/aritrag/kerascv-starter-notebook-train) recently, and I've encountered an issue.\n\nI'm trying to run the following code snippet \" https://www.kaggle.com/code/tchaye59/attributeerror-augmenter-not-found \":\n\n```python\n\naugmenter = keras_cv.layers.Augmenter(\n [\n keras_cv.layers.RandomFlip(mode=\"horizontal_and_vertical\"),\n keras_cv.layers.RandomCutout(height_factor=0.2, width_factor=0.2),\n \n ]\n )\n\n\n```\nHowever, I'm getting the following error:\n\n`AttributeError: module 'keras_cv.layers' has no attribute 'Augmenter'`\n\nI've spent a considerable amount of time trying to troubleshoot this error, but I'm still unable to find a solution. Any help or guidance on how to resolve this issue would be greatly appreciated.\n\nThank you in advance for your assistance!" } ], "comments": [ { "id": 2418888, "author_name": "Martin Görner", "author_url": "", "post_date": "2023-09-01T15:13:07.107000", "content": "

just replace keras_cv.layers.Augmenter with keras.Sequential
\nmore details here: https://github.com/keras-team/keras-cv/issues/1548
\nThe notebook will be updated too.

", "votes": 4, "replies": [] }, { "id": 2420818, "author_name": "Neeraj Karoshi", "author_url": "", "post_date": "2023-09-03T00:18:56.573000", "content": "

Found an alternative way to do it as well

\n

def apply_augmentation(image,labels):
\n image = preprocessing.RandomFlip(mode=\"horizontal_and_vertical\")(image)
\n image = preprocessing.RandomTranslation(height_factor=0.2, width_factor=0.2)(image)
\n return (image, labels)

", "votes": 2, "replies": [] }, { "id": 2419267, "author_name": "Jude TCHAYE", "author_url": "", "post_date": "2023-09-01T21:55:26.197000", "content": "

Thank you, everyone! Here's how I successfully worked around the issue:

\n
 os\n\n\nos.environ[] = \n\n keras_cv\n\n tensorflow  keras\n\n tensorflow.keras  layers\n\naugmenter = keras.Sequential(\n  layers = [\n      tf.keras.layers.Lambda(keras_cv.layers.RandomCutout(height_factor=, \n                                             width_factor=)),\n      tf.keras.layers.Lambda(keras_cv.layers.RandomFlip(mode=)),\n   ]\n)\n
", "votes": 2, "replies": [] }, { "id": 2418838, "author_name": "PC Jimmmy", "author_url": "", "post_date": "2023-09-01T14:47:56.720000", "content": "

I get the same error running with either of the two GPU resources on kaggle.

\n

When running on my local machine it does NOT fail at that same point

\n
 -  fails   model.fit cell apparently  a   NVIDIA hell.  Tensorflow  Keras always a bitch   working   system when you want your GPU's  play nicely.  \n
", "votes": 0, "replies": [] }, { "id": 2418835, "author_name": "the real L", "author_url": "", "post_date": "2023-09-01T14:45:19.117000", "content": "

I'm facing the same issue and haven't had any luck trying to solve it so far.
\nCould you please help here? @awsaf49 @fchollet @gusthema @ianstenbit @philculliton

", "votes": 0, "replies": [] } ], "raw_markdown_by_id": { "2418888": "just replace ```keras_cv.layers.Augmenter``` with ```keras.Sequential```\nmore details here: https://github.com/keras-team/keras-cv/issues/1548\nThe notebook will be updated too.", "2420818": "Found an alternative way to do it as well\n\ndef apply_augmentation(image,labels):\n image = preprocessing.RandomFlip(mode=\"horizontal_and_vertical\")(image) \n image = preprocessing.RandomTranslation(height_factor=0.2, width_factor=0.2)(image)\n return (image, labels)\n\n\n", "2419267": "Thank you, everyone! Here's how I successfully worked around the issue:\n\n```python\n\nimport os\n# You can use `tensorflow`, `pytorch`, `jax` here\n# KerasCore makes the notebook backend agnostic :)\nos.environ[\"KERAS_BACKEND\"] = \"tensorflow\"\n\nimport keras_cv\n#import keras_core as keras\nfrom tensorflow import keras\n#from keras_core import layers\nfrom tensorflow.keras import layers\n\naugmenter = keras.Sequential(\n layers = [\n tf.keras.layers.Lambda(keras_cv.layers.RandomCutout(height_factor=0.2, \n width_factor=0.2)),\n tf.keras.layers.Lambda(keras_cv.layers.RandomFlip(mode=\"horizontal_and_vertical\")),\n ]\n)\n```", "2418838": "I get the same error running with either of the two GPU resources on kaggle.\n\nWhen running on my local machine it does NOT fail at that same point \n\n - but fails in the model.fit cell apparently as a result of NVIDIA hell. Tensorflow and Keras always a bitch to get working on local system when you want your GPU's to play nicely. ", "2418835": "I'm facing the same issue and haven't had any luck trying to solve it so far.\nCould you please help here? @awsaf49 @fchollet @gusthema @ianstenbit @philculliton ", "2418564": "Hello Kaggle community,\n\nI've been working with the [KerasCV starter notebook ](https://www.kaggle.com/code/aritrag/kerascv-starter-notebook-train) recently, and I've encountered an issue.\n\nI'm trying to run the following code snippet \" https://www.kaggle.com/code/tchaye59/attributeerror-augmenter-not-found \":\n\n```python\n\naugmenter = keras_cv.layers.Augmenter(\n [\n keras_cv.layers.RandomFlip(mode=\"horizontal_and_vertical\"),\n keras_cv.layers.RandomCutout(height_factor=0.2, width_factor=0.2),\n \n ]\n )\n\n\n```\nHowever, I'm getting the following error:\n\n`AttributeError: module 'keras_cv.layers' has no attribute 'Augmenter'`\n\nI've spent a considerable amount of time trying to troubleshoot this error, but I'm still unable to find a solution. Any help or guidance on how to resolve this issue would be greatly appreciated.\n\nThank you in advance for your assistance!" } }