Beginner's Guide: Some Useful Machine Learning Websites(Part 2)

In the previous tutorial (Part 1) I've talked about adventures-in-machine-learning.com. Today I will talk about another online resource that comes in handy when one needs to quickly implement neural network models without going through all the hassle of the theories. As a result, I recommend this AFTER you've gone through the articles in  adventures-in-machine-learning.com. So, today we're going to talk about python-programming-dot-net.

When I was doing my thesis, I at first went through the theories and step by step implementation  in adventures-in-machine-learning.com. When I had a better grasp about the topics in neural networks, I started following this website. They have tutorials on various topics. But I'm going to focus on machine learning tutorial which starts from this link.

Here the tutorial for a particular topic( i.e. regression, neural network)  spans several articles. I found the articles fun to read and the code easier to understand(since I've already studied  adventures-in-machine-learning.com ). I recommend you implement the code in your own computer/laptop as well rather than just watching the implementation in the videos. The tutorial topics are laid below in order:

• Regression: Regression is taught in any ML course in the beginning. Many even include assignments to implement this. In my thesis I also had to use regression to figure out the relation among some variables. It also helps one to ease into ML course before moving onto more difficult topics. So definitely check it out. The vlogger notes the important points to implement this in python. So you won't miss much in the case of implementation. But I also recommend studying the theories a bit before watching the tutorial for better grasp.

Linear regression

• The next topic is K-nearest neighbor starting from here. I only had to study the theory for my academic course. But you can check it out.
• After K-nearest neighbor comes SVM. Some of my classmates had to implement this in their thesis. So you may have to as well. SVM is also given as homework/assignment in many varsities. So this series of articles may come to your aid when you are given work to do using SVM.
• Now comes clustering. Like K-nearest neighbor, I studied the theories only so I didn't go through the articles here. I recommend you study these articles, if you have to do specific tasks in academia that requires them. 
• After clustering comes deep learning tutorial in TensorFlow. This part has greatly helped me when I was just beginning to learn the implementation of neural network in Python. It begins with using a multi-layer neural network on MNIST data-set. The explanations of the codes are quite easy to understand and so very suitable for beginners. The second example is positive/negative sentiment analysis of text data. So there's one example with image data-set and one example with text data-set. Most of the data-sets, one deals with while learning machine learning, are either text data or image data. Short tutorials are also provided for Recurrent Neural Network( RNN)  and Convolutional Neural Network( CNN). RNN is used when the data samples are related to one another( sequential data) and CNN is used for image data-sets because of its efficient feature extracting abilities. Their implementations are given in TensorFlow. The implementation can be used for your academic or professional work with ease and understanding. I highly recommend you go through this section if you are interested in deep learning.
• Another important section of this website is the 'Natural Language Processing' tutorials. Many CS students, taking up machine learning courses or theses, may have to study NLP and implement NLP. Here too the website has created simple yet comprehensive tutorial to learn the implementation of NLP in Python. I do recommend some prior theoretical study before diving in. The tutorial is long but it's totally worth it!! It will let you grasp the basic coding skills you need to perform NLP-related tasks with ease. Before copy-pasting the code, do go through what the article says about its code so that you don't get confused while writing down the code.

That's it for today. These are the sources that helped me most to learn about the wonderful world of deep learning. Do let me know what you think in the comment section and follow me on Google+ to get more updates in my blog.