Since we launched Teachable Machine in 2017, we’ve noticed people using it for different reasons, from teachers exploring machine learning concepts with students to hands-on workshops for industry professionals.
We’ve also been receiving requests for features that would make Teachable Machine a better tool. So, we’ve been designing an updated version of Teachable Machine with some new features.
Like the original experiment, the new Teachable Machine is all built on Tensorflow.js and works locally in the browser. Our hope is that the new version of Teachable Machine will be a super easy way for anyone to train their own machine learning models and use them in their own projects, wherever tensorflow.js models can be run.
Some of the features we’re working on:
- Train a model on image data
- Train a model on audio data
- Train a model on pose data, with Posenet
- Upload your own datasets to train models
- Train more than 3 classes per model
- Disable classes
- Save your tensorflow.js model
- Download your model
- Deploy your model to use in your own project
We’ve been collaborating with people in different industries with different needs for the tool, like Steve Saling, to test and shape the new Teachable Machine.
If you have a particular use case the updated version would be useful for, and would be willing to give feedback on our beta builds, please reach out to: firstname.lastname@example.org.
Stay tuned for its release later this year!