29 March 2019, Charlottesville: Women in Data Science Conference

[Daniel-Mietchen]

nothing public yet, but see #355 for the 2018 one.

[Daniel-Mietchen]

Website: https://datascience.virginia.edu/2019-charlottesville-women-data-science-conference

[Daniel-Mietchen]

Agenda: https://datascience.virginia.edu/pages/2019-women-data-science-conference-agenda

[Daniel-Mietchen]

hashtag: WiDS2019Cville

[Daniel-Mietchen]

• xerox brother dominic commercial introducing the idea of a copy machine (1977)
• 2017 remake involving artificial intelligence

[Daniel-Mietchen]

livestream: https://livestream.com/tavco/womenindatascience?mc_cid=651d7369ee&mc_eid=feefa3b4bc

[Daniel-Mietchen]

parent conference: https://www.widsconference.org/

[Daniel-Mietchen]

• Declare your biases

[Daniel-Mietchen]

Book: Humanities Data in R — seen in this Twitter bio via this tweet

[Daniel-Mietchen]

biases in police traffic stops:

• overview
• data
• link to legalization of marihuana

[Daniel-Mietchen]

more on the members of the discussion panel: https://datascience.virginia.edu/pages/panel-discussion-women-founders-executives

[Daniel-Mietchen]

One of the panelists — Tatenda Ndambakuwa — founded https://shirifly.org/ , "an agricultural practice that uses software and mobile applications as agricultural service solutions for farmers while creating a vibrant ecosystem of agricultural alliances"

[Daniel-Mietchen]

Interesting sideline of the panel discussion was on imposter syndrome and on admitting that even leaders and otherwise successful people can (and do) have self-doubts. There are cultural factors (including gender) that modulate how people and even teams deal with that.

Also: "failure is not actually failure. it is just a path to learning" (Kim Scott)

[Daniel-Mietchen]

Going to the skills session "Data visualization with Python and Jupyter notebooks" now.

Session lead is from https://www.thorn.org/ : Erin Braswell, trained as an astronomer, with a focus on education and at the Center for Open Science until two weeks ago..

[Daniel-Mietchen]

Plug for Python Charlottesville meetups: https://www.meetup.com/PythonCharlottesville/events/past/ .

[Daniel-Mietchen]

Install environments:

• Anaconda
• virtualenv / virtualenvwrapper
• Pipenv / Pyenv

[Daniel-Mietchen]

visualization modules:

• http://matplotlib.org/ (rich functionality for basic 2D plots)
• http://pandas.pydata.org/ (inspired by R's data frames)
• https://plot.ly/ (proprietary; need to request API key)
• http://bokeh.pydata.org/ (for interactive visualizations)

[Daniel-Mietchen]

Now Jupyter

[Daniel-Mietchen]

Potential data sources:

• http://opendata.charlottesville.org/
• https://www.ncdc.noaa.gov/cag/
• https://www.data.gov/
• http://opendata.cern.ch/
• Google Dataset Search https://github.com/Daniel-Mietchen/ideas/issues/942
• https://osf.io
• https://opendatanetwork.com/

[Daniel-Mietchen]

Python development environments:

• Spyder
• PyCharm

[Daniel-Mietchen]

Back to Jupyter:

• https://github.com/jupyter/jupyter/
  • https://github.com/jupyter/jupyter/wiki
  • https://github.com/jupyter/jupyter/wiki/A-gallery-of-interesting-Jupyter-Notebooks

[Daniel-Mietchen]

Rendering Jupyter notebooks on the web

• https://nbviewer.jupyter.org/
• GitHub and OSF render Jupyter notebooks as well
• https://mybinder.org/
• http://colab.research.google.com/
• https://jupyter.org/try

[Daniel-Mietchen]

Now to hands-on part:

• https://github.com/erinspace/wids_dataviz , which links (via the README) to https://colab.research.google.com/drive/1ZXHpkthMUWhjeRithtDGa8j3KlvzFngc , of which I took a "copy to [Google] drive": https://colab.research.google.com/drive/1PGFqSFsQhdpjA4lx5ztFnWXCVVZr8cN1 .

[Daniel-Mietchen]

Matplotlib Basemap

[Daniel-Mietchen]

Matplotlib folium

[Daniel-Mietchen]

Amazon has Boto3 for Python.

[Daniel-Mietchen]

Closing keynote: Meredith Broussard on "Artificial Unintelligence", based on her book of the same title.

[Daniel-Mietchen]

Titanic dataset: http://biostat.mc.vanderbilt.edu/wiki/pub/Main/DataSets/titanic.html

[Daniel-Mietchen]

Tweet: "If you have ever had a problem grasping the importance of diversity in tech and its impact on society, watch this video"

Book: Algorithms of oppression

She speaks of "Technochauvinism", as it discriminates by default based on (often unconscious) bias

[Daniel-Mietchen]

Several other books mentioned - captured in this tweet

• Automating Inequality
• Technically Wrong

This is followed by a shout-out for grassroots orgs like Black in AI, Data & Society and AI Now! or Source.

[Daniel-Mietchen]

A tab I still had open from the event:

• https://source.opennews.org/articles/hacking-our-hiring-part-1/