Skip to content

A skeleton pytorch codebase commonly used across my projects

License

Notifications You must be signed in to change notification settings

nileshkulkarni/project_skeleton

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

5 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Project Skeleton Utilities

This repository provides a skeleton pytorch codebase commonly used across my projects. Some examples projects using this are factored3d, CMR, 3D-relnet, and CSM.

Feel free to modify and/or build on this.

Usage Instructions

Any experiment using this code base typically creates a class that inherits from the 'Trainer' class in nnutils/train_utils.py. See this file in CMR code as an example. To define this 'Trainer', one has to simply instantiate the functions to define a dataset, model, register scalars to plot, and create visualizations. Once these are done, the training, logging, and saving models proceeds using the defined functions in the base class. Please see the flags in nnutils/train_utils.py to see the default parameters.

About

A skeleton pytorch codebase commonly used across my projects

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%