@WinVectorLLC profile picture
WinVector LLC @WinVectorLLC

#DataScience and #R consulting and training by John Mount and Nina Zumel. practicaldatascience.com/ and blog win-vector.com/blog/

Joined April 2013
  • Tweets 3,964
  • Following 2,113
  • Followers 2,011
731 Photos and videos
I would like to share a new article on some of the methods and pitfalls of time series forecasting win-vector.com/2023/05/07/a-… #TimeSeries #rstats #Statistics #DataScience #Analytics

A Time Series Apologia

I would like to share a new article on some of the methods and pitfalls of time series forecasting: “A Time Series Apologia”. In it I work the seemingly simple problem of forecasting a …

win-vector.com
2
5
1,749
Google search????
254
The Sell ∀ ∃ as ∃ ∀ Scam win-vector.com/2023/04/22/th… #ai #machinelearning #deeplearning

The Sell ∀ ∃ as ∃ ∀ Scam

Artificial intelligence, like machine learning before it, is making big money off what I call the “sell ∀ ∃ as ∃ ∀ scam.” The scam works as follows. Build a sy…

win-vector.com
2
6
245
Dr. Nina Zumel just shared a nice Pandas/Polars Rosetta Stone. The common needed data wrangling operations in Pandas and Polars. This can help with the data wrangling in your projects. Please check it out github.com/WinVector/Example… #pydata #pandas #polars #datascience #Python
2
5
754
youtu.be/qMe7qwNMe-s

Partial Pooling

This video is an introduction to the idea of partial pooling: combi...

youtube.com
195
The standard method to estimated an expected value is to use the average from a sample. When we know something about how the data is grouped, you can do better. Please check it out: win-vector.com/2023/04/02/do… #python #data #pydata #datascience #statistics
1
2
310
How much data do you need for a data science project? youtu.be/j-FqGy8clxg

How much data do you need?

This is an introduction to my article on how much data is needed in...

youtube.com
1
148
Bounding Excess Generalization Error for Linear Regression Models win-vector.com/2023/03/16/bo… #statistics #datascience

Bounding Excess Generalization Error for Linear Regression Models

Introduction The goal of this note is to try and characterize excess generalization error: how much worse your model works in production versus how well it appeared to work during training. The cla…

win-vector.com
1
118
Having some fun with a short lecture on using the decomposition of regression problems (predicting numbers) into a function plus error or residuals. youtu.be/UgB6b-E5MqE #statistics #datascience

Regression Lecture: decomposing out the error and residuals

This lecture uses ideas and notation from this article: https://win...

youtube.com
1
111
Thinking About Linear Regression win-vector.com/2023/03/12/th… #statistics #datascience #rstats #pydata

Thinking About Linear Regression

Introduction I want to spend some time thinking out loud about linear regression. As a data science consultant and teacher I spend a lot of time using linear regression and teaching linear regressi…

win-vector.com
1
3
1,429
Data Algebra now fully support Polars for data engineering. win-vector.com/2022/12/22/da… #dataengineering #pydata #Polars #DataScience

Data Algebra over Polars Ready for Production Use

The data algebra is a system for composing data manipulation tasks in Python. In the data algebra, operator pipelines (or even directed acyclic graphs) are the primary objects. Applying operations …

win-vector.com
4
430
The whole purpose of this account has been to try to drive traffic to the Win Vector blog win-vector.com/blog-2/ . I admit I haven't kept that. But if that is no going to be allowed ...

Win Vector LLC

Data science advising, consulting, and training

win-vector.com
1
166
WinVector LLC retweeted
Not Jerome Powell
@alifarhat79
13 Dec 2022
In all fairness your honor, my client said he’ll mint a new coin to repay all FTX investors in full
56
245
2,287
Touching the 3rd Rail of Data Science: “R or Python?” (including my own myopic benchmark) win-vector.com/2022/12/13/to… #rstats #pydata #datascience #datatable #polars

Touching the 3rd Rail of Data Science: “R or Python?”

I’ve been seeing a lot of hot takes on if one should do data science in R or in Python. I’ll comment generally on the topic, and then add my own myopic gear-head micro benchmark. I̵…

win-vector.com
1
Hi, #polars Peeps, would please love some help with aggregating strings in Python Polars. I have an open question here: "How to get max string per group in Polars?" (works for numbers, not for strings) stackoverflow.com/q/74763636…
1
I am starting to experiment with a the Polars data frame library in Python. In fact I have started development on a data algebra to Polars adapter. Please check out my initial notes and experiment here: win-vector.com/2022/12/07/ex… #python #pydata #datascience #polars

Experimenting with Polars for Data in Python

I’ve just started experimenting with the Polars data frame library in Python. I really like the programmable API it exposes. In fact I am starting an experimental adapter from the data algebr…

win-vector.com
2
Couldn't resist trying to reproduce @__mharrison__ 's brilliant Pandas example in the data algebra: github.com/WinVector/data_al… #pandas #pydata
3
WinVector LLC retweeted
Matt Harrison
@__mharrison__
23 Nov 2022
Once you understand the common patterns in Pandas, you piece them together to build up powerful analytics quickly! 🐼🤯
4
33
325
WinVector LLC retweeted
Christopher Bouzy (spoutible.com/cbouzy)
@cbouzy
19 Nov 2022
A former Twitter employee wrote an essay about safety but didn't take responsibility for the targeted harassment directed at women of color that occurred under his watch. Prior to Elon, accounts were created to target VP Harris and Meghan Markle, and Twitter was like, "meh."
158
2,332
13,414
Forget about video games and music. Read this as a tech contractor: medium.com/@mickgordon/my-fu…
Load more