Tensorboard

This is a wrapper class for the Tensorflow's Tensorboard application. It is still in early development, and should be used with a grain of salt. If you want to help be sure to contact me or fork this repository and work on our ToDo list from the development branch. Tensorboard.py is the Wrapper that contains all functions needed to interact, more documentation should come in the future, but as for now, the requirements for this project are:

PyTorch

The project does not install PyTorch so it doesn't force a different version from your environment, so before using the wrapper, be sure to install:

torchvision
torch

Development dependencies

setuptools
wheel
twine
numpy
tensorboard==1.15
torch>=1.6.0 
torchvision>=0.7.0 

install

You can now install this project by using:

pip install tensorboard-wrapper

You can run the tests by executing:

python -m unittest test.py

And the list of the functions already implemented:

• tensorboard.writer.SummaryWritter
• add_scalar
• add_scalar*
• add_histogram*
• add_image
• add_images*
• add_figure
• add_video
• add_audio
• add_text
• add_graph
• add_embedding
• add_pr_curve
• add_custom_scalars
• add_mesh
• add_hparams
• flush*
• close*

*Altered in some shape or form. All changes will be in the documentation in the future.