Skip to content

Conversation

@mattdangerw
Copy link
Member

I had to pin nightly a few days book, look like there was some issue with ResNet on the latest version. This also needs a change to Keras that will not yet be on nightly until tonight

I had to pin nightly a few days book, look like there was some
issue with ResNet on the latest version. This also needs a change
to Keras that will not yet be on nightly until tonight
@divyashreepathihalli
Copy link
Collaborator

Hi Matt!! The resizing has been added back and the new presets have been updated. So is the nightly working fine for you?

Copy link
Collaborator

@divyashreepathihalli divyashreepathihalli left a comment

Choose a reason for hiding this comment

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

Thanks Matt! left a few NIT comments.

This is just scratching the surface of what you can do with the KerasHub.
This guide shows a few of the high-level tasks that we ship with the KerasHub library,
but there are many tasks we did not cover here. Try [generating images with Stable
Copy link
Collaborator

Choose a reason for hiding this comment

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

we can maybe link to this page too here to point to the full list of models to try out - https://keras.io/api/keras_hub/models/

keras.mixed_precision.set_global_policy("mixed_float16")
1. Go to the [Gemma 2](https://www.kaggle.com/models/keras/gemma2) model page, and accept
the license at the banner at the top.
2. Generate an Kaggle API key by going to [Kaggle settings](https://www.kaggle.com/settings)
Copy link
Collaborator

Choose a reason for hiding this comment

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

NIT: a Kaggle

and clicking "Create New Token" button under the "API" section.
3. Inside your colab notebook, click on the key icon on the left hand toolbar. Add two
secrets: `KAGGLE_USERNAME` with your username, and `KAGGLE_KEY` with the API key you just
created. Make these secrets visible to the notebook you are running.
Copy link
Collaborator

Choose a reason for hiding this comment

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

NIT: Provide access to these secrets in your notebook when prompted while running the Colab.

KerasHub will use [tf.data](https://www.tensorflow.org/guide/data) as the default API for
running multi-threaded preprocessing on the CPU. `tf.data` is a powerful API for training
input pipelines that can scale up to complex, multi-host training jobs easily. Using it
does not restrict your choice of backend, a `tf.data.Dataset` can be as an iterator of
Copy link
Collaborator

Choose a reason for hiding this comment

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

NIT: can be used as an iterator

@fchollet fchollet merged commit a93f382 into keras-team:master Oct 20, 2024
4 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants