Michael (Misha) Smirnov – Medium

Michael (Misha) Smirnov

Michael (Misha) Smirnov

AI, Director’s Cut: You’re the Star in the Movie of Tomorrow

My mother-in-law has a peculiar fondness for JibJab. The eccentric yet entertaining platform takes public domain music videos, superimposes…

Jun 24, 2023

AI, Director’s Cut: You’re the Star in the Movie of Tomorrow

Jun 24, 2023

Michael (Misha) Smirnov

The Creator’s Cut: Advocating for Paid Training Data in AI

Let’s say I’m a data scientist, a futurist, a technophile. Let’s say I even get a little too enthusiastic when it comes to AI and…

May 12, 2023

DALL-E: a robot handing cash to a human, photograph

May 12, 2023

Michael (Misha) Smirnov

Maximizing Accuracy for Rare Occurrences: A Zero-Inflated MSE Approach to Model Training

Training a model when most predictions should be zero.

Apr 4, 2023

lots of zeros with a few small larger single-digit numbers

Apr 4, 2023

Michael (Misha) Smirnov

Failing Upwards: my journey from a PhD in Neuroscience to a Data Scientist at Amazon

Embracing uncertainty and finding a new career path in data science

Mar 18, 2023

Failing Upwards: my journey from a PhD in Neuroscience to a Data Scientist at Amazon

Mar 18, 2023

Michael (Misha) Smirnov

Use BeautifulSoup and WordCloud to visualize the most frequent words in Firefly

I love firefly. Does it hold up? I don’t know. But its script is published online, so we’ll use it as an example to display textual data in…

Feb 22, 2022

Use BeautifulSoup and WordCloud to visualize the most frequent words in Firefly

Feb 22, 2022

Michael (Misha) Smirnov

Common metrics for evaluating natural language processing (NLP) models

You can’t train a good model if you don’t have the right evaluation metric, and you can’t explain your model if you don’t understand the…

Feb 18, 2022

Common metrics for evaluating natural language processing (NLP) models

Feb 18, 2022

Michael (Misha) Smirnov

Explain your text classification model with Lime

The more complex a model, the more of a black box it can be — that is, it can get pretty tricky figuring out just why your predictions…

Feb 4, 2022

Explain your text classification model with Lime

Feb 4, 2022

Michael (Misha) Smirnov

Techniques to represent text for NLP

We can’t just take a paragraph and feed it directly into a mathematical algorithm; the characters, words, or phrases need to be converted…

Jan 12, 2022

Techniques to represent text for NLP

Jan 12, 2022

Michael (Misha) Smirnov
in
Nerd For Tech

Create an automatically expanding TextEditor with a Placeholder in SwiftUI

I’ve banged my head against the desk for a while about this, and there are quite a few buggy hacks about this floating around the web, but…

Nov 27, 2021

Create an automatically expanding TextEditor with a Placeholder in SwiftUI

Nov 27, 2021

Michael (Misha) Smirnov

Create placeholder data with GeoPoint and Date values, and upload it to Firebase Firestore using…

At some point during the development process, you may find yourself wishing you could preview what your app would look like if it had a lot…

Jul 26, 2021

Create placeholder data with GeoPoint and Date values, and upload it to Firebase Firestore using…

Jul 26, 2021

Michael (Misha) Smirnov

Michael (Misha) Smirnov

Data Scientist at Amazon. PhD in Neuroscience. Coder, creator, woodworker, all around cool guy who likes high fives.

Following

Help
Status
About
Careers
Press
Blog
Privacy
Terms
Text to speech
Teams