Dan Oved

`tf.linalg.lstsq` would be great as this would support triangulation of rays or 2d points with known camera parameters into 3d, as is done in [tensorflow graphics ray.triangulate function](https://github.com/tensorflow/graphics/blob/master/tensorflow_graphics/geometry/representation/ray.py)

Suggestions for how to do this, inspired by the structure of [tfjs](https://github.com/tensorflow/tfjs) and [tfjs-models](https://github.com/tensorflow/tfjs-models): Version 1: * Change all models to require `tfjs-core` and `tfjs-converter` for loading converted models, instead...

A nice example of how to switch backends (tfjs browser / tfjs-node (cpu) / tfjs-node-gpu (cuda) is in [tfjs-examples/lstm-text-generation]( https://github.com/tensorflow/tfjs-examples/blob/master/lstm-text-generation/train_node.js#L103) ```javascript const args = parseArgs(); if (args.gpu) { console.log('Using GPU');...

Hey just got back in town. I think trying it for one or two new models makes sense. It should definitely be possible to do with javascript and typescript side...

Yes a standalone thing would be great. A lot of people would find it useful to just have an offline version of PoseNet. I believe this could be just part...

@mncharity how would you imagine the integration of that standalone feature working in your project?

Hi @sklipnoty Yes I will update to 2.0 You can get completely offline by running: `cd client` `yarn distribute` This will create a distributable package for a specific file system...

I would narrow the scope of the try catch block to just be wrapped around the serialization/deserialization part, and if it catches it, callback with the error rather than returning...

You could also do it via dependency injection - I always prefer composition over inheritance: The provider would act as an api, and contain a token repository. This way you...

Working on this https://github.com/oveddan/node-oauth2-provider