data-science.md

data science

list of useful tools

• https://landscape.lfai.foundation/

references to further similar lists

• https://github.com/r0f1/datascience

basics

• https://missing.csail.mit.edu/

exploratory data analysis (EDA)

• automated EDA
  • python
    • https://sci-analysis.readthedocs.io/en/latest/
    • https://github.com/pandas-profiling/pandas-profiling
    • https://amueller.github.io/dabl/dev/
    • https://www.visidata.org/
    • https://github.com/microsoft/vscode-data-wrangler
  • R
    • https://boxuancui.github.io/DataExplorer/

anomaly detection

• https://pyod.readthedocs.io/en/latest/index.html

visualization

• https://charticulator.com/
• https://www.datawrapper.de/
• https://flourish.studio/
• https://rawgraphs.io/
• high dimensional data https://github.com/facebookresearch/hiplot

scaling pandas

• https://github.com/ibis-project/ibis
• elasticsearch: https://github.com/elastic/eland
• dask
• ray/modin
• RAPIDS
• vaex
• https://ibis-project.org/
• https://github.com/pola-rs/polars

experiment design

• http://docs.pyro.ai/en/stable/contrib.oed.html

experiment tracking

MLops overview: https://github.com/visenger/awesome-mlops

• mlFlow
• https://docs.bentoml.org
• https://neptune.ai/blog/experiment-management
• https://neptune.ai/blog/mlops-what-it-is-why-it-matters-and-how-to-implement-it-from-a-data-scientist-perspective
• https://towardsdatascience.com/machine-learning-experiment-tracking-93b796e501b0
• https://neptune.ai/blog/machine-learning-model-management

bayesian analyses

• frameworks
  • stand
  • pymc
  • pyro https://arviz-devs.github.io/arviz/
• https://botorch.org/
• https://arviz-devs.github.io/arviz/
• https://bayesiancomputationbook.com/welcome.html

deep learning

• books

  • https://github.com/fastai/fastbook

• a really useful link collection on my blog: https://georgheiler.com/2020/12/03/intersting-links-about-deep-learning/

• automated/easy DL

  • https://github.com/geekjr/quickai
  • https://pythonawesome.com/a-framework-that-implements-automl-algorithms-for-model-architecture-search-at-scale/

• tabular deeplearning:

  • https://github.com/dreamquark-ai/tabnet

• https://lets-unify.ai/

neighbor search

• https://ai.googleblog.com/2020/07/announcing-scann-efficient-vector.html
• https://github.com/spotify/annoy

subgroup discovery

• https://github.com/flemmerich/pysubgroup
• https://github.com/imoscovitz/wittgenstein
• https://github.com/scikit-learn-contrib/skope-rules

NLP

• https://pythonawesome.com/attention-based-grapheme-to-phoneme-with-python/

rules

• https://github.com/adaamko/POTATO

reprodrucibility

population drift monitoring

• https://github.com/ing-bank/popmon

exploration of models

• exploration in a nice HTML UI by the end user to get a feel for the model or feedback: https://github.com/gradio-app/gradio

fairness

• https://www.youtube.com/watch?v=WfcCXlxkBb0, https://github.com/koaning/scikit-lego
• https://koaning.io/posts/the-future-is-past/
• https://github.com/linkedin/LiFT

explainable AI

• https://github.com/interpretml/interpret

version control

• improve jupyter git integration http://timstaley.co.uk/posts/making-git-and-jupyter-notebooks-play-nice/
• https://elyra.readthedocs.io/en/latest/

streaming data science

libraries

• https://github.com/alibaba/alink
• https://github.com/scikit-multiflow/scikit-multiflow
• https://riverml.xyz/latest/

data

audio

• feature calculation
  • https://github.com/jcvasquezc/DisVoice

streams

• jupyter
  • bokeh on streams in jupyter https://medium.com/swlh/streaming-time-series-with-jupyter-and-influxdb-fec33803c460

feature store

• hopsworks
• Behaviors compose better than states
  • perhaps a simple python library with some transformations & DBT & Dagster will do the job very well and https://metriql.com/ for metrics
• https://feast.dev/

forecasting (time series)

• https://www.youtube.com/watch?v=jo12CWZ00Lo
• https://github.com/linkedin/greykite
• https://facebook.github.io/prophet/
• https://github.com/winedarksea/AutoTS
• https://github.com/timeseriesAI/tsai

generic timeseries operations

• https://khiva.readthedocs.io/en/latest/
• cleaning https://towardsdatascience.com/a-common-mistake-to-avoid-when-working-with-time-series-data-eedf60a8b4c1

website tracking

google analytics

• schema overview http://storage.googleapis.com/e-nor/visualizations/bigquery/ga360-schema.html

CDM/CDP customer data management & customer data platform

basically SQL-based sync & incremental updates of data to many systems and reports without sending a lot of CSV files around

• https://www.hightouch.io/
• https://www.getcensus.com/

deployment

• https://github.com/whylabs/whylogs
• clean machine learning code https://leanpub.com/cleanmachinelearningcode
  • code refactoring example https://github.com/moutai/tdml-example/tree/master/with_tests/tests

cloud instance handling

• cheap training using spot instances https://www.spotml.io/

use cases

marketing

• https://facebookexperimental.github.io/Robyn/

text processing

ocr

• https://github.com/VikParuchuri/surya