Thread by @pvergadia

👋Day 2 #31DayofML

Let's explore #MachineLearning terms for supervised learning:
🔸Labels - the thing we're predicting
🔸Features - an input variable
🔸Examples - particular instance of data (Labeled/Unlabeled)
🔸Models - defines the relationship between features & label.

A 🧵

🔸Labels - the thing we're predicting

Eg: The y variable in simple linear regression. The label could be the future price of wheat, the kind of animal shown in a picture, the meaning of an audio clip, or just about anything.

#31DayofML

🔸Features - an input variable x in linear regression.

A simple ML project might use 1 feature X, while a sophisticated project could use millions of features X1, X2, ..... Xn

Eg: In the spam detector, features could be:
📌Words in email
📌Sender's address
📌Time
📌Some phrase

🔸Examples - particular instance of data (X) ➡️is a vector

📌Labeled - includes both feature(s) & labels
📌Unlabeled - contains features but not labels

Model is trained on labeled examples & is used to predict the label on unlabeled examples.

#31DayofML

🔸Models - defines the relationship between features & label.

There are 2 phases of a model's life:
1️⃣ Training - creating/learning the model from labeled examples
2️⃣ Inference - applying the trained model to unlabeled examples.

#31DayofML