NEW: Tues 31 March = overhaul of coronavirus trajectories
— John Burn-Murdoch (@jburnmurdoch) March 31, 2020
As the story shifts, the charts shift. We’re now focusing on *new* cases & deaths rather than running totals.
Why? Because key for any country is when today’s new cases are fewer than yesterday’shttps://t.co/JxVd2cG7KI pic.twitter.com/yPMUzlVLQv
Belgian seismologists pick up #covid19 lockdown as less noisy seismometer readings https://t.co/b7t7uuRYz3 On @nature pic.twitter.com/QCC2ZN6skW
— Maarten Lambrechts (@maartenzam) March 31, 2020
Number of pet adoptions have increased a lot since quarantine began. Humans are social beings and desire company. Source: https://t.co/4Y7znKQdIN pic.twitter.com/ij4KdZi6qw
— Simon Kuestenmacher (@simongerman600) March 31, 2020
Difference Attention Based Error Correction LSTM Model for Time Series Prediction
— Thomas (@evolvingstuff) March 31, 2020
"difference-attention LSTM model and error-correction LSTM model are respectively employed and combined in a cascade way"https://t.co/JwyttchDuq pic.twitter.com/U6bEwTvs6t
#pandas groupby silently drops NaNs:
— Daily Python Tip 🐍🐧 (@python_tip) March 31, 2020
import pandas as pd
import numpy as np
birds = pd.Series([390., 350., 30., 20., np.NaN], index=['Falcon', 'Falcon', 'Parrot', 'Parrot', 'Dodo'])
len(birds) # = 5
birds.groupby(birds > 100).count().sum() # = 4 https://t.co/sOlPLaE1dO
“How to be Curious Instead of Contrarian About COVID-19: Eight Data Science Lessons From Coronavirus Perspective” https://t.co/zBXZvnpSvT
— Andrew Gelman (@StatModeling) March 31, 2020
A Metric Learning Reality Check: “[…] state of the art loss functions perform marginally better than, and sometimes on par with, classic methods” 🔥https://t.co/vqVvncuogX pic.twitter.com/DACbVP9Eoi
— Denny Britz (@dennybritz) March 31, 2020
Nicely written, @EricTopol!
— Jeff Dean (@🏡) (@JeffDean) March 31, 2020
I completely agree with the conclusion: "The handling of the COVID-19 pandemic in the United States will go down as the worst public health disaster in the history of the country."
We may have 100k people die who didn't have to (cont) https://t.co/DMXaYyiabK
I wrote about America's worst public health disaster and how our country has compromised clinicians and left the medical community highly #COVID19 vulnerable, along with our patients https://t.co/9uvTQ8YyMt @Medscape pic.twitter.com/Mm0HcqnT5x
— Eric Topol (@EricTopol) March 31, 2020
More cool things built with nbdev 😊 https://t.co/xPMEP9JsN9
— Sylvain Gugger (@GuggerSylvain) March 30, 2020
This is the model from Report 13 of the Imperial College COVID-19 response team. It fits the death data jointly from 11 European countries to estimate the reproduction number and the effect of lockdowns. Such a remarkable piece of @mcmc_stan https://t.co/Hhmbt3KVcQ
— Gilles Louppe (@glouppe) March 30, 2020
This U.S. District Court for D.C. decision is a huge win for data journalists:
— Julia Angwin (@JuliaAngwin) March 30, 2020
"Criminalizing terms-of-
service violations risks turning
each website into its own criminal jurisdiction and each webmaster
into his own legislature.”https://t.co/d4sN8U1JV4
