Add tensorboardX recipe

[mdraw]

Packages https://github.com/lanpa/tensorboard-pytorch, a library for direct tensorboard logging of NumPy arrays, PyTorch tensors etc. without TensorFlow or TensorBoard requirements.

Notes:

• The repository is still called tensorboard-pytorch, but the project name has been changed
• As far as I see, the log format of tensorboardX is not compatible with recent versions of the official TensorBoard server https://github.com/tensorflow/tensorboard, but only with https://github.com/dmlc/tensorboard, which has a name conflict with Google's Tensorflow (see The PyPI package is no longer maintain due to some reasons dmlc/tensorboard#50) and is not packaged on conda-forge. I am not sure how we should deal with this issue. If we decided to also package dmlc/tensorboard, how should we name it?

[conda-forge-linter]

Hi! This is the friendly automated conda-forge-linting service.

I just wanted to let you know that I linted all conda-recipes in your PR (recipes/tensorboardx) and found it was in an excellent condition.

[CJ-Wright]

Please add more tests. (more imports are fine)

[mdraw]

All submodules are now imported at test time. Is it okay now?

[CJ-Wright]

This is noarch, and thus only needs to pass on circle.

[mdraw]

Thanks for reviewing and merging, @CJ-Wright.

[lanpa]

@mdraw I always test with tensorboardX with tensorflow-tensorboard. Can you provide more information about the compatibility? thanks

[mdraw]

I tried it again on a new Anaconda installation with tensorflow-tensorboard and it works now 👍. There was probably an issue with my previous install that I can't reproduce.