Sergey Kubarev

Ex-Financial Analyst, now a Data Scientist with a Master's from @BolognaBSchool. Passionate about #DataScience, #ML, #Python, #Kaggle

[Sergey Kubarev @SergeyKubarev]

Transitioning from point forecasts to probabilistic forecasts can enrich decision-making. Consider the shift to 'forecasting hitting time': predicting when a specific event will occur. #DataScience #Forecasting

https://t.co/8O38Hhj85F

[Sergei Guriev @sguriev]

Many thanks, Madame Merkel, for coming to Sciences Po and, most importantly, for being a great role model for our students - a 21st century politician with values and emphathy

#ScPoMerkel #ScPoHonorisCausa

https://t.co/A34Ot5Mf26

[Charly Wargnier ♨️ @DataChaz]

#DataLife

[Sergey Kubarev @SergeyKubarev]

Exploring the remarkable capabilities of #ChatGPT4 by @OpenAI. It's inspiring to see how AI has progressed - creating conversational models that understand and generate human-like text. This is a game-changer for #AI, #MachineLearning, and #DataScience.

[François Chollet @fchollet]

Many good ideas will not work well on MNIST (e.g. batch norm). Inversely many bad ideas may work on MNIST and no transfer to real CV.

[Prof. Feynman @ProfFeynman]

If you cannot explain something in simple terms, you don't understand it. The best way to learn is to teach.

[Sergey Kubarev @SergeyKubarev]

MNIST digits dataset with Convolutional Neural Network. Accuracy 99,578%.
#DataScience #MachineLearning #DeepLearning #Kaggle #keras #NeuralNetwork #NeuralNetworks

https://t.co/iMdUoQ8NHa

[Sergey Kubarev @SergeyKubarev]

This is really great!

[The Flemish Scrollers @FlemishScroller]

Every meeting of the flemish government in Belgium is live streamed on a YouTube channel.
Software is searching for phones and tries to identify a distracted politician. This with the help of AI. The results are then posted to social with the politician tagged. By @driesdepoorter

[Sergey Kubarev @SergeyKubarev]

Created a new notebook for the CommonLit Readability competition on Kaggle with simple preprocessing and basic models to start with.

https://t.co/tIROF8RMpT

#Kaggle #DataScience #MachineLearning

[Jeroen Bartelse @JeroenBartelse]

A backpack that flies… This is an all-electric personal flying machine

via @DigitalTrends
#UAM #mobility #drone #aviation #EV
@SpirosMargaris @mvollmer1 @Nicochan33 @FrRonconi @PawlowskiMario @Shi4Tech @Hana_ElSayyed @DrJDrooghaag @labordeolivier

[Sergey Kubarev @SergeyKubarev]

I like this one best - We proved a thing that's been known empirically for 5 years 😀

[Sergey Kubarev @SergeyKubarev]

Correlation is a single statistic (data point), while regression is the entire equation with all of the data points that are represented with a line. Regression explains the relation between X's and Y. Useful article.
#Statistics #DataScience #ML

https://t.co/jEB9NOSpa3.

[Sergey Kubarev @SergeyKubarev]

@sergio_marinho Also, correlation doesn't mean causation, we need always remember this.

https://t.co/jEB9NOSpa3.

[Sergey Kubarev @SergeyKubarev]

@sergio_marinho Sorry, for been late.

Correlation is a single statistic, whereas regression is the entire equation with all of the data points that are represented with a line. Correlation is the relationship between the two variables, regression explains how one affects the other.

[KDnuggets @kdnuggets]

How to Deal with Missing Values in Your Dataset - KDnuggets #KDN https://t.co/5cskwgM2lJ

[Sergey Kubarev @SergeyKubarev]

Today the fastest language model (GPT-3) has up to 175 billion connections. NVIDIA predicts that by 2023, models will have 100 trillion or more connections. Seems we will witness an absolutely different level of science.
#DataScience #MachineLearning #NLP

https://t.co/fRWGhQslsv

[Sergey Kubarev @SergeyKubarev]

There are a lot of tools to detect fake news, most of the algorithms are based on Machine Learning. But still, the final decision is up to the human, the journalist with domain knowledge and fact-checking experience.
#MachineLearning #DataScience
https://t.co/BFosiIBCR1

[Sergey Kubarev @SergeyKubarev]

@WhiteheartVic It's okay, we ended up collecting enough observations. Thank you!

[Sergey Kubarev @SergeyKubarev]

Dear friends,
I need to do a little research for a statistics course. I would be very grateful if you could find the time to take this survey. It will take only about 4 minutes!
Thank you very much!
#Statistics #Travel #smartphone #Apps #data

https://t.co/qrn0gGvjyH

[Sergey Kubarev @SergeyKubarev]

Different country - different name!
https://t.co/gungrKY5YO