ds/dx - a data science & ml engineering blog

serverless gcp cloud function cloud run cloud scheduler python spotify spotipy terraform cloud infrastructure

airflow mwaa terraform gcp serverless workflows pipeline dag apache github actions cloud

model containerized machine learning gcp cloud run cloud storage pub-sub terraform prediction batch-prediction serverless cloud

model deployment containerized machine learning api aws lambda api gateway terraform prediction inference serverless s3 cloud

linear regression modeling model diagnostics pitfalls multicollinearity anova t-test residuals outliers log-model log-log-model vif purrr simulation glm glmnet

sklearn speedup tuning performance service endpoint predictions pipeline transformers optimization

tree travel-times geospatial haversine-distance azimuth coordinates taxi bigquery gradientboosting lightgbm xgboost bigquery seaborn numpy pyproj

terraform aws jupyterhub s3 cloud ec2 tljh

bigquery timeseries deduplication bigdata sql island gaps intervals grid events

neuralnetwork embeddings sklearn onehotencoding regression highcardinality

sklearn transformer columntransformer dataframes features scaling pipeline pandas numpy

keras sklearn gridsearch pipeline preprocessing

trees sklearn extrapolation estimator regression ensembling

caching memoization memoise aws.s3

docker travis end-to-end test machine learning s3 data lake random forest pytest ci

forecasting timeseries parallelization ensembleforecasting purrr

sf leaflet shapefile maps