Wht is in PyData 2020?

pydata_2020.txt

https://global.pydata.org/

1. Технический трек - от простого к сложному и качественному
----------------------------------------------------------

Датафреймы (базовая табличная структура данных):

    An introduction to DataFrames.jl for pandas users, by Bogumił Kamiński
    Skinny Pandas Riding on a Rocket, by Ian Ozsvald (PyDataLondon) - путанное название 

Параллелизм и ускорение вычислений:

    Parallel processing in Python: The current landscape, by Aaron Richter
    Speed Up Your Data Processing: Parallel and Asynchronous Programming in Data Science, by Chin Hwee Ong
    Hosting Dask: Challenges and Opportunities, by Matthew Rocklin
    Supercharge Scientific Computing in Python with Numba, by Ankit Mahato

Дашборды:

    Quickly deploying explainable AI dashboards, by Oege Dijk
    Streamlit: The Fastest Way to build Data Apps, by Steven Kolawole
    Scalable cross-filtering dashboards with Panel, HoloViews and hvPlot, by Philipp Rudiger and James A. Bednar

"Запаковать в продакшн" (как прототип модели становится работающим блоком в ИТ-системе):

    DevOps for science: using continuous integration for rigorous and reproducible analysis, by Elle O'Brien
    How to guarantee your machine learning model will fail on first contact with the real world., by Jesper Dramsch
    Growing Machine Learning Platforms in the Enterprise, by Hussain Sultan
    Transformation from Research Oriented Code into Machine Learning APIs with Python, by Tetsuya Jesse Hirata
    Monitoring machine learning models in production, by Arnaud Van Looveren
    Meditations on First Deployment: A Practical Guide to Responsible Data Science & Engineering, by Alejandro Saucedo
    Feature drift monitoring as a service for machine learning models at scale, by Keira Zhou and Noriaki Tatsumi

Качество кода ("саентисты" хронически пишут посредственный код)

    Rethinking Software Testing for Data Science, by Eduardo Blancas
    Better Code for Data Science, by Alexander CS Hendorf
    How to review a model, by Andy R. Terrel
    Separation of ~concerns~ scales in software, by Thomas A Caswell


2. А хорошо ли это все работает и что в результате дает
------------------------------------------------------

Этика, explainability и справедливость моделей: 

    Tangible Steps Towards Algorithmic Accountability, by Ayodele Odubela (keynote)
    Building fairer models for finance, by Andrew Weeks
    Responsible ML in Production, by Catherine Nelson and Hannes Hapke
    Open Source Fairness, by Aileen Nielsen 
    Opening the Black Box, by Ben Fowler and Chelsey Kate Meise
    Safe, Fair and Ethical AI - A Practical Framework, by Tariq Rashid

"Small" data:

    The Big Benefits of Small Data, by Christopher Lozinski
    Taking a Close Look in the Mirror: Data Literacy for Data Experts, by Laura J Ludwig
    Data processing pipelines for Small Big Data, by Esteban J. G. Gabancho and Anthony Franklin, PhD
    Dirty Data science: machine-learning on non-curated data, by Gaël Varoquaux

Наука и код:

    Is Coding Science? An interview with Wolfgang Kerzendorf, by Wolfgang Kerzendorf (keynote)
    Computational Social Science with Python, and how Open Source transforms Academia and Research, by Bhargav Srinivasa Desikan


3. Статистические и математические методы 
-----------------------------------------

Отдельные методы - статистика:

    Bayesian Decision Science: A framework for making data informed decisions under uncertainty, by Ravin Kumar
    When features go missing, Bayes’ comes to the rescue, by Narendra Mukherjee
    Modelling the extreme using quantile regression, by Massimiliano Ungheretti
    Geometric and statistical methods in systems biology: the case of metabolic networks, by Haris Zafeiropoulos and Apostolos Chalkis
    Accelerating Differential Equations in R and Python using Julia's SciML Ecosystem, by Chris Rackauckas

Отдельные методы - временные ряды:

    ML-Based Time Series Regression: 10 concepts we learned from Demand Forecasting, by Felix Wick
    Modern Time Series Analysis with STUMPY, by Sean Law
    TimeSeries Forecasting with ML Algorithms and there comparisons, by Sonam Pankaj

Применили методы:

    Multi-Label Classification with Human Rights Data, by Megan Price, PhD and Maria Gargiulo (keynote)
    Climate Change: analyzing remote sensing data with Python, by Luis Lopez
    Leveraging python and open-source for data-science on the buy-side., by James Munro
    Games, Algorithms, and Social Good, by Manojit Nandi


4. Нейросети и навороченный ML
------------------------------

    Building Large-Scale Multilingual Fuzzy Matching Framework, by Abdulrahman Althobaiti
    Inventing Curriculum using Python and spaCy, by Gajendra Deshpande
    Is a neural network better than Ash at detecting Team Rocket? If so, how?, by Juan De Dios Santos
    Snap ML: Accelerated, Accurate, Efficient Machine Learning, by Haris Pozidis and Thomas Parnell
    Taking Care of Parameters So You Don’t Have to with ParamTools, by Hank Doupe
    Thrifty Machine Learning, by Rebecca Bilbro
    Using EOLearn to build a machine learning pipeline to detect plastics in the ocean., by Stuart Lynn
    Why I didn’t use deep learning for my image recognition problem, by Liucija Latanauskaite
    Cardinal: A metrics based Active Learning framework, by Alexandre Abraham
    Complex Network Analysis with NetworkX, by K. Jarrod Millman
    Ensemble-X: Your personal strataGEM to build Ensembled Deep Learning Models for Medical Imaging, by Dipam Paul and Alankrita Tewari
    Ordinary viDeogame Equations: Winning games with PyMC3, sundials and numba, by Adrian Seyboldt
    Visions: An Open-Source Library for Semantic Data, by Ian Eaves and Simon Brugman
    Visual data: abundant, relevant, labelled, cheap. Pick two?, by Irina Vidal Migallon
    What Lies in Word Embeddings, by Vincent D. Warmerdam
    Autonomous Vehicles See More With Thermal Imaging: Multi-modal thin cross section Object Detection, by Laisha Wadhwa
    Basic Pitfalls in Waveform Analysis, by Yukio Okuda
    Building one (multi-task) model to rule them all!, by Nicole Carlson and Michael Sugimura
    Entity matching at scale, by Lorraine D'almeida
    Uncertainty Quantification in Neural Networks with Keras, by Matias Valdenegro-Toro
    Using Algorithm X to re-analyse the last UK general election, by Alex Glaser
    FlyBrainLab: An Interactive Open Computing Platform for Exploring the Drosophila Brain, by Mehmet Kerem Turkcan, Aurel A. Lazar and Yiyin Zhou

5. Разное
---------

Ковид

    COVID-19 Visualizations, the Good, the Bad and the Malicious, by Rongpeng Li 
    What cyber security can teach us about COVID-19 testing, by Hagit Grushka - Cohen

Разные "ништяки" вокруг pandas

    ipywidgets for Education! Using Jupyter tools to make Math Visualization applets for the classroom, by Chiin-Rui Tan
    pandas.(to/from)_sql is simple but not fast, by Uwe Korn
    What's new in pandas?, by Joris Van den Bossche and Tom Augspurger
    pyodide: scientific Python compiled to WebAssembly, by Roman Yurchak

6. Туториалы и короткие выступления (некоторые):
------------------------------------------------

    A Gentle Introduction to Multi-Objective Optimisation, by Eyal Kazin
    Exploratory Data Analysis with Pandas and Matplotlib, by Allen Downey
    Creating a data-driven culture: a social perspective, by Jordi Contestí
    Data Visualization & Storytelling, by Jose Berengueres
    Learning from your (model’s) mistakes, by Simona Maggio
    Rapidly emulating professional visualizations from New York Times in Python using Altair, by Shantam Raj
    Ten Ways to Fizz Buzz, by Joel Grus
    Turn your notebook into a LaTeX-article with TexBook, by Valerio Maggio
    UBI Center: A think tank built on GitHub, Python, and Jupyter, by Max Ghenis
    nbreproduce: Jupyter notebooks in reproducible environments, by Mridul Seth
    Building a Successful Data Science Team, by Justin J. Nguyen (основная программа)