Intro-to-Machine-Learning

Introduction to Machine Learning

There are basically 2 types of Learning process in Machine Learnings

1. Supervised Learning: Data is labled
2. Un-supervised Learning: Data is not labled

Examples of Supervised Learning ss

1. Classification
2. Regression

Road map so far

1. Linear Regression
2. Polynomial Regression

More to come!

---

Steps to follow to solve a problem using Machine Learning

1. Get the right Data Set: Need a right dataset to develop a model that can predict.
2. Understand the Dataset: Understand each feature / column and it's importance.
3. Choose the Depedent variable to be predected
4. Perform Basic cleaning: 
	One of the important steps in ML. Remember Garbage IN Garbage OUT
	Handle different data types in the dataset
	Handle Missing values in the dataset
5. Explore the data in depth:
    Detect Outliners
    Plot and check for distrubances in the data
6. Feature Engineering:
    Next important step in ML.
    Choosing the right features for training your model. More is not always better in ML
    Training the Model with Less necessary features is always better than 
    Training a model with a huge amount of unecessary features.
    See how the features are is co-related between each other.
7. Dimensionality reduction.
8. Choose your algorithm
9. Split the data into Train and Test datasets. 80:20 usually prefered
10. Evaluate Model's performace.