Description:

Many people are claiming to learn machine learning using standard libraries while not knowing the math behind it.

My objective is clear to implement and give a intuition of linear regression model while at the same time telling what steps makes a model good fit for training sets.

It includes:-

A. Getting comfortable with libraries by actual implementation

1. Introduction to numpy, pandas and matplotlib

2. Exploring data using pandas

3. Exploring relation between various variables using matplotlib.

4. Knowing what are the problems are for a bad model.

B.Exploratory Data Analysis :-

1. Classifying features as continuous or categorical.

2. Handling missing data.

3. Feature Extraction and Selection.

4. Correlation and causation.

5. Dummy Variables

6. Visualizing Data

C. Implementation of Model

1. Cost function

2. Gradient Descent

3. Normal Equations

Prerequisites:

1. Basic knowledge of python like defining function, declaring variables.
2. Knowledge of Matrix
3. Basic Mathematics.

Content URLs:

https://github.com/RushikeshJachak

https://github.com/Heisenberg0203

Speaker Info:

I am Rushikesh Jachak, Currently pursuing computer science and engineering in government college of engineering, Aurangabad. I moved towards python from last two months due to my interest in data science field especially machine learning. I am complete novice in python environment, i do not know the hooks and crux of python but i do believe the more you share more you learn.So i would definitely like to share my journey till know and and knowledge of maths and intuition behind the most common algorithm of ML. I also have a bit knowledge of Big-data technologies such as Hadoop hive, and poses a keen interest in field of Data Science.

Speaker Links:

https://github.com/Heisenberg0203/Kaggle

https://github.com/Heisenberg0203/MachineLearning/tree/master/Week1

https://www.linkedin.com/in/rushikesh-jachak-44b723135/