Quick Start

  Last updated: May 2nd, 2020

Download and Use

Download GroundUpAI

Download python scripts above and unzip scripts.zip. You should find the following files in the scripts directory. Simply refer to the code documentation for info on individual scripts and import the functions and classes you wish to use.

files

utils.py
parameter.py
data_block.py
data_bunch.py
augmentation.py
initialization.py
loss.py
model.py
linear.py
convolution.py
pooling.py
batch_norm.py
other_layers.py
sub_model.py
optimizer.py
training.py
callback.py
learner.py
stats_logging.py
param_scheduling.py
early_stopping.py
progress_bar.py
lr_search.py
stateless_optim.py
resnet.py
lstm.py
gru.py
bleu_score.py
transformer.py
bert.py

Making Contribution


The GitHub stuff

Here is the GitHub repository of GroundUpAI. Please follow this thorough guide from forking to PR reviews.

Development steps

  1. Modify files in the notebooks directory and and make sure to add #export at the top of the notebook cells you want to add to the script
  2. Run python3 generate_scripts.py to reflect your changes in the scripts
  3. Make sure the tests in other notebooks are successful following your change

Below are the mapping from notebook to scripts

notebooks/00_utils.ipynb            -> scripts/utils.py
notebooks/01_parameter.ipynb        -> scripts/parameter.py
notebooks/02_data_block.ipynb       -> scripts/data_block.py
notebooks/03_data_bunch.ipynb       -> scripts/data_bunch.py
notebooks/04_augmentation.ipynb     -> scripts/augmentation.py
notebooks/05_initialization.ipynb   -> scripts/initialization.py
notebooks/06_loss.ipynb             -> scripts/loss.py
notebooks/07_model.ipynb            -> scripts/model.py
notebooks/08_linear.ipynb           -> scripts/linear.py
notebooks/09_convolution.ipynb      -> scripts/convolution.py
notebooks/10_pooling.ipynb          -> scripts/pooling.py
notebooks/11_batch_norm.ipynb       -> scripts/batch_norm.py
notebooks/12_other_layers.ipynb     -> scripts/other_layers.py
notebooks/13_sub_model.ipynb        -> scripts/sub_model.py
notebooks/14_optimizer.ipynb        -> scripts/optimizer.py
notebooks/15_training.ipynb         -> scripts/training.py
notebooks/16_callback.ipynb         -> scripts/callback.py
notebooks/17_learner.ipynb          -> scripts/learner.py
notebooks/18_stats_logging.ipynb    -> scripts/stats_logging.py
notebooks/19_param_scheduling.ipynb -> scripts/param_scheduling.py
notebooks/20_early_stopping.ipynb   -> scripts/early_stopping.py
notebooks/21_progress_bar.ipynb     -> scripts/progress_bar.py
notebooks/22_lr_search.ipynb        -> scripts/lr_search.py
notebooks/23_stateless_optim.ipynb  -> scripts/stateless_optim.py
notebooks/25_resnet.ipynb           -> scripts/resnet.py
notebooks/26_lstm.ipynb             -> scripts/lstm.py
notebooks/27_gru.ipynb              -> scripts/gru.py
notebooks/28_bleu_score.ipynb       -> scripts/bleu_score.py
notebooks/29_cycle_gan.ipynb        -> scripts/cycle_gan.py
notebooks/30_transformer.ipynb      -> scripts/transformer.py
notebooks/31_bert.ipynb             -> scripts/bert.py