ck37
Psychiatry faculty; biostatistics PhD. {Targeted, deep, machine} learning, NLP, IRT, computer vision, exposure mixtures, EHRs.
Harvard Medical School, Mass General HospitalBoston, MA
Pinned Repositories
atlantic-causal-2017
Targeted Learning entry in the Atlantic Causal Inference Conference's 2017 competition
ck37r
R functions for project setup, data cleaning, machine learning, SuperLearner, parallelization, and targeted learning.
coral-ordinal
Tensorflow Keras implementation of ordinal regression using consistent rank logits (CORAL) by Cao et al. (2019)
Predictive-Modeling-in-R
Workshop (2-6 hours): cleaning, missing value imputation, EDA, ensemble learning, calibration, variable importance ranking, accumulated local effect plots. WIP.
superlearner-guide
SuperLearner guide: fitting models, ensembling, prediction, hyperparameters, parallelization, timing, feature selection, etc.
varimpact
Variable importance through targeted causal inference, with Alan Hubbard
Machine-Learning-in-R
Workshop (6 hours): preprocessing, cross-validation, lasso, decision trees, random forest, xgboost, superlearner ensembles
R-Deep-Learning
Workshop (6 hours): Deep learning in R using Keras. Building & training deep nets, image classification, transfer learning, text analysis, visualization
Unsupervised-Learning-in-R
Workshop (6 hours): Clustering (Hdbscan, LCA, Hopach), dimension reduction (UMAP, GLRM), and anomaly detection (isolation forests).
SuperLearner
Current version of the SuperLearner R package
ck37's Repositories
ck37/coral-ordinal
Tensorflow Keras implementation of ordinal regression using consistent rank logits (CORAL) by Cao et al. (2019)
ck37/varimpact
Variable importance through targeted causal inference, with Alan Hubbard
ck37/superlearner-guide
SuperLearner guide: fitting models, ensembling, prediction, hyperparameters, parallelization, timing, feature selection, etc.
ck37/ck37r
R functions for project setup, data cleaning, machine learning, SuperLearner, parallelization, and targeted learning.
ck37/Predictive-Modeling-in-R
Workshop (2-6 hours): cleaning, missing value imputation, EDA, ensemble learning, calibration, variable importance ranking, accumulated local effect plots. WIP.
ck37/featurerank
Ensemble feature ranking for SuperLearner variable selection
ck37/hpc-savio-xsede
Multicore and multi-node parallel R computation via SLURM on the Savio cluster at UC Berkeley, plus XSEDE
ck37/clinsent
Estimate sentiment in clinical notes via keywords or deep learning models
ck37/mimic-clinical-sentiment
Manuscript under review
ck37/toxic-comments-fastai-pytorch
Analyzing Jigsaw's toxic comments Kaggle challenge using fastai + pytorch
ck37/Machine-Learning-in-R
Workshop (6 hours): preprocessing, cross-validation, lasso, decision trees, random forest, xgboost, superlearner ensembles
ck37/SuperLearner
SuperLearner R package: prediction model ensembling method
ck37/tlmixture
Data-adaptive creation of exposure (treatment) mixtures using targeted learning
ck37/causalquantile
R package for estimating quantile treatment effects
ck37/chest-pain-risk-prediction
Article under review
ck37/ck37
ck37/darwin-py
Library and commandline tool for managing datasets on darwin.v7labs.com
ck37/dcurves
Decision Curve Analysis
ck37/embarcadero
🌲🌉 BART species distribution models 🌉 🌲
ck37/iml
iml: interpretable machine learning R package
ck37/mlr3extralearners
Extra learners for use in mlr3.
ck37/mlr3summary
ck37/mlr3viz
Visualizations for mlr3
ck37/Phase2.1DataRPackage
ck37/Phase2.1DockerAnalysis
ck37/Phase2.1SeverityAugRPackage
ck37/quicksectx
ck37/sl3
💪🤓 Modern Super Learning using Pipelines
ck37/treeshap
Compute SHAP values for your tree-based models using the TreeSHAP algorithm
ck37/tuneRanger
Automatic tuning of random forests