Skip to content
/ antnlp-tproj Public template

a simple and flexible organization of NLP projects

Notifications You must be signed in to change notification settings

AntNLP/antnlp-tproj

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

A Template of Organizing NLP/ML Projects

We provide an example directory of NLP/ML projects filled with data sets, source files, evaluation scripts, and model checkpoints. The hierarchy has been practiced in our previous projects, and it shows reasonable flexibility for the model development process (e.g, tunning hyperparameters, stacking different model components, integrating third-party libraries while keeping the project separated).

Overview

There are several principles we will follow,

Keeping preprocessing and postprocessing steps clear.

Separating data preparation with model source files.

Separating evaluation scripts from model source files.

The subdirectories are

  • data contains datasets, their preprocessing scripts, and intermediate results.
  • modules are basic blocks for building your model. (encoders, decoders, our fancy modification of the Transformer).
  • models are glued modules which accomplish the NLP task.
  • configs hosts configuration files.
  • eval contains evaluation scripts.
  • util are helping tools.
  • ckpts contains model checkpoints.
  • train.py the training script.
  • run.py the deploying script.

About

a simple and flexible organization of NLP projects

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages