TensorFlow Hub ❤️ Kaggle — The TensorFlow Website

[ad_1]

Posted by Luiz GUStavo Martins, Google AI Developer Advocate

We’re enthusiastic to announce our new integration with Kaggle Styles, a not long ago launched pre-trained model system. All 2,300+ TensorFlow models posted on TFHub by Google, DeepMind, and more are now discoverable and usable on Kaggle Types, with documentation and sample code.

Why Kaggle?

Kaggle is a global community of over 12 million machine learners who exam their knowledge in competitions and share equipment studying assets, which includes above 200,000 public datasets. In excess of the previous 10+ several years, Kaggle’s competitions have come to be a proving ground for what operates very well and what does not across a multitude of ML use situations. This is why Kaggle just lately introduced its open up model hub, Kaggle Types, to greater help the ML community to stress examination and validate styles publicly and at scale.

Web hosting TensorFlow types on Kaggle makes them much more simply available to the broader ML community, democratizing design developing and advancement. We won’t be able to wait around to see what remedies occur from this partnership.

How to Get Began

The Chook Vocalization Classifier product was just open-sourced by the Google Investigate staff on TFHub (and subsequently Kaggle Designs 🙌). It truly is a world wide fowl embedding and classification model that can identify a lot more than 10k chook species’ vocalizations, and also results in embedding vectors that can be utilized for other duties.

To try out the design on Kaggle:

Navigate to the model below.
Simply click the “New Notebook” button, which will open a Kaggle Notebooks editor.
Simply click the “Copy Code” button on the suitable-hand aspect of the editor, which will duplicate sample code that hundreds the design utilizing the TensorFlow Hub library.
Paste the code into the notebook’s mobile, and you’re completely ready to go!
Click the “Add Model” button at the bottom. This will attach the product to your notebook.

Moving image showing the user's experience of the Bird Vocalization Classifier Model on Kaggle

The snippet imports TFHub library and masses the recently posted Bird Vocalization Classifier model. To uncover extra data about this model, you can look at its documentation and even perform with a entire illustration that demonstrates how to use the product in the competition in this article.

import tensorflow_hub as hub

keras_layer = hub.KerasLayer('https://kaggle.com/versions/google/bird-vocalization-classifier/frameworks/TensorFlow2/versions/fowl-vocalization-classifier/versions/1')

For a lot more facts on Kaggle Versions which includes its present element established and upcoming roadmap, check out the official announcement in this article. We glance forward to looking at what you construct as a outcome of this integration!

[ad_2]

Resource link