CSOC ML Week2
Hello and welcome to the week 2 of CSOC ML. Hope you have gone through and practiced all the materials given last week. This week we will learn about two more important libraries matplotlib and pandas. Along with this, starting this week, we will delve right into the core Machine Learning. Also the first assignment has been released. You have time till next Wednesday(29-05-2019 23:59) to submit the assignment. Do not forget to fill the google form after completing the assignment. Now coming to the resources of this week.
pandas
Machine Learning deals with data, loads and loads of them. Understanding, analysing and manuplating this data is essential before applying Machine Learning algorithms. Pandas is a powerful data science library with provides you powerful functions to manipulate and analyze the data. Hence it is very essential to have a strong grasp over some basic functions in this library. Here are some of the good tutorial which you can refer.
Matplotlib
When working with data, we usually want to visualize it to understand the underlying patterns and get insights about the data. Matplotlib is another powerful Data Science library. As the name suggests it is a plotting library which provides powerful functions to plot the data points. A very interesting example demonstrating the importance of plotting is the Anscombe’s quartet. Here are some excellent resources on Matplotlib.
- Official documentation of matplotlib
- Amazing Tutorial
- First seven Videos of this tutorial series by Sentdex
Machine Learning
Machine Learning has become one of the hottest topics of discussion and research today. Most of the big firms and industries are trying to integrate Machine Learning in their technology. But what is this Machine Learning? Machine Learning is an idea to learn from examples and experience, without being explicitly programmed. Instead of writing code, you feed data to the generic algorithm, and it builds logic based on the data given. Machine Learning has several applications. Face detection, Email Filtering, Chatbots, Weather prediction, medical diagnosis are some of them. There are three kinds of Machine Learning Algorithms. They are:
- Supervised Learning
- Unsupervised Learning
- Reinforcement Learning
Refer this amazing blog to get a good insights of these algorithms. In the coming two weeks, we will cover the basics of Supervised Learning. We will be mainly referring to following two courses:
If you have ever asked anyone as to how to start with Machine Learning you must have heard about this course. Going through the course will give you a firm understanding of Machine Learning concepts. Another intresting course is the ud-120, Intro to Machine Learning by Udacity. This course is a bit more hands-on than the former.
In the next two weeks, you need to cover:
- Week 1 - Introduction, Linear Regression with One Variable, Linear Algebra Review,
- Week 2 - Linear Regression with Multiple Variables,
- Week 3 - Logistic Regression, Regularization
- Week 7 - Support Vector Machines part of Machine Learning course by Andrew Ng
- The first seven chapters of Intro to Machine Learning course on Udacity
Be sure to take the the quizzes in Andrew Ng’s course althogh the assignments are optional. In Udacity’s course make sure you do the mini projects at the end of chapters sincerely. Start with the order given above. Doing this will give you a good intution on the core Machine Learning algorithms.
There is a lot of material to cover So I suggest you start early. The assignment has also been kept short for the same reason. Dont feel down if you are feeling a bit rusty will the libraries. As you practise and implement them more and more your grasp on them will become stronger. Next week an assignment on Matplotlib and pandas will be released. Untill then All the best and happy learning!!