This document highlights possible improvements and additional implementations
driving this library towards next release (version 0.2.0).
You can check whether someone started work on this task in issues under the label
roadmap and specific improvement or implementation.
Any help with tasks described below will be highly appreciated, thanks!
Following tasks might be considered improvements over what is already done.
Please use label roadmap and improvement and describe the task you are willing to take
(or describe new task you would consider an improvement).
- Check and test support for multiprocessing and PyTorch's workers
- Fix documentation and make it more readable
- Testing with Hypothesis
- Tutorials creation
- Conda release
- General
modifiers,cachersandmapsif any deemed missing. - Cloud-oriented
cacher, which would allow for easier incorporation oftorchdatain more challenging tasks (possible as another repository) - Less I/O intesive disk oriented
cachers(e.g. save 100 samples as one file)