Represents Machine Learning and Deep Learning Core Library functionality
Install Experiment Structure
# installs experiment structure in $HOME/dev-platform/experiments
./scripts/experiment_snapshot_installer.sh template_name # create experiment structureActivate Environment
# update anaconda
conda update conda
# create a mck anaconda environment first based on python 3.x
cd env
conda env create -f <template>.v3_x.yml # create python nlp environment package dependencies from existing
conda env list # list environments
source activate ml.base.v37 # select environment (make sure executed anytime python scripts for nlp depdendencies are executed)
# conda create --name deeplearning.core.v37 --clone ml.base.v37
# for spacy nlp specifics
python -m spacy download en # download a smaller model
python -m spacy download en_core_web_md # download a larger model for a larger vocabularyInstall Libraries and Dependencies
# installs the library in 'edit' mode for development into prior installed environment
# show be installed as: `hunter-workflows` in listing the packages (conda list |less )
cd hunter-core
pip install -e . Create symbolic links to execute within Experiment Structure if applicable
cd $HOME/experiments/snapshots/{template_name}
rm -rf apps
ln -s $HOME/Projects/repositories/{template_name}/apps/ apps
rm -rf data/conf
ln -s $HOME/Projects/repositories/{template_name}/resources/data/conf/ data/confIf you reference in your research, please cite:
@article{Hunter2020,
author = {Kamlani, Ari},
title = {{Spinning up core Machine Leanring and Deep Learning Frameworks}},
year = {2020}
}