Jaegerist🔺

Telling stories with data | Jumping around in Web3 playgrounds | This boy is electric ⚡

[Jaegerist🔺 @timijobs]

I graduated with a Master's in Data Science in December of last year, yet I've never felt more like an imposter career-wise than I do now.

That said, there's only one cure for imposter syndrome, and that is to build in public.

Over the next month, that's what I'll be doing.

[Jaegerist🔺 @timijobs]

I'll be taking a short break from my ML studies to focus on some more pressing priorities.

Hope to be back within a week or so. When I return, I will continue posting my updates as usual.

[Jaegerist🔺 @timijobs]

SVMs are very powerful but can be slow with big datasets. They are used in image recognition, text analysis, biology research and many other areas.
Thanks for reading 🙏🏾

#30DaysOfML #typefully

[Jaegerist🔺 @timijobs]

DAY 7/30
Today, let's look at Support Vector Machine (SVM) algorithm, a powerful technique for supervised tasks like classification and regression.

SVMs are a way to classify data into different groups. They work by finding the best line or plane that separates the groups.

[Jaegerist🔺 @timijobs]

Tuning parameters like the error allowance, kernel type and kernel settings is important to get the best performance from SVMs. Cross-validation can also help find the best settings.

[Jaegerist🔺 @timijobs]

In summary, I believe our understanding of how ML algorithms like Decision Trees work would help us make better decisions and solve problems in machine learning.

Thanks for reading.

#30DaysOfML #typefully

[Jaegerist🔺 @timijobs]

DAY 6/30
Still on Decision Trees. This algo estimates class probabilities by looking at how many data points belong to each class in a leaf.
E.g., if you were to classify apples vs oranges, and there were lots of apples vs a few oranges in a leaf, it'll say, "Probably an apple!"

[Jaegerist🔺 @timijobs]

So, Decision Trees act as really smart guides in a forest of data. They estimate probabilities, keep things clear with impurity metrics, tidy up with regularization, and work hard during training to make predictions possible.

[Jaegerist🔺 @timijobs]

Even though they're simple, Decision Trees are really good at handling tricky problems and figuring out what's important. Understanding how they work helps us make smarter decisions in machine learning.

[Jaegerist🔺 @timijobs]

Day 5/30: Focusing on Decision Trees today! These are like flowcharts that help make decisions based on different things. They're easy to understand and can work with numbers or categories. #30DaysofML #typefully

[Jaegerist🔺 @timijobs]

When making decisions, Decision Trees look at different factors and choose the best one to split the data. This helps them make better predictions.