Skip to main content


Showing posts from September, 2020

Featured post

The Active Ones Take It All

Hey, you! Yes.. you! Are you still delaying that wonderful idea you may have been nursing for a while now? Have you been hesitating on starting that business, journey, career, course, or work you have  to do?

Have you identified a favorable opportunity, but you've not been able to utilize it because you're thinking too much about it? Then this article is for you. I want you to bear this at the back of your mind: "The active ones take it all."

Life offers everything to the ones who are active. Life doesn't care about your intention or what you're thinking of doing. It cares about what you're doing!

Let's say there are two people who intend to start a similar business, let's say it's a small restaurant. One of them has been nursing the idea for a long time and is very passionate about it. He keeps thinking and thinking of how to start up the business and get everything ready but has done nothing yet.

The other one also nurses the idea though he…

Motivation and Machine Learning (Lesson 3) Part 2

Feature Selection:Helps you answer the question: "What are the features that are most useful for a given model?" One of the reasons we need to apply this is that the number of features in your original dataset may be very very high.Benefits: Eliminates irrelevant, redundant and highly correlated features Reduce dimensionality for increased performance. As many ML models do not cope well on data with very large dimensions(many features).We can improve the situation of having too many features through dimensionality reduction.Commonly used techniques are:PCA (Principal Component Analysis)t-SNE (t-Distributed Stochastic Neighboring Entities)Feature embeddingAzure ML prebuilt modules:Filter-based feature selection: identify columns in the input dataset that have the greatest predictive powerPermutation feature importance: determine the best features to use by computing the feature importance scores*********************Data DriftData drift is change in the input data for a model.…