Where to begin?

                    This is the first question that pops into our heads whenever we decide to learn something. Machine learning is also no different. The biggest challenge I faced was not that I did not have enough resources but the fact that there were too many resources that I could not fix on a roadmap for learning.

                      The path I took lead me to make a base on machine learning. You could apply this while learning anything for that matter.

• Set a milestone: Make a milestone for each week and give deadlines to yourself. Select some courses beforehand and complete them one after the other. 
• Focus on a single course: There are quite a lot of courses available. Try to select and complete a single course. Jumping from one to another will make you lose interest in the process.
• Repeating is good: If you find that you are reading things again and again in different courses or resources, it’s a good sign. It means you are getting familiar with the terms of the subject.
• Join support groups: Join groups interested in Machine Learning both online and offline. I am not endorsing any groups but the School of AI was useful for me. Follow hashtags related to machine learning in Facebook, LinkedIn, etc.
• Make notes: Even though the resources are always available, making notes of your own will always be useful.

Machine learning terms

               I would just give you a list of terms and topics which you could familiarise with. Once you have done that you would be in a position to start on a mini-project.

• What is the role of data in Machine Learning?
• What is a Model
• Types of Machine Learning
  • Supervised
  • Unsupervised
  • Reinforcement
• Clustering
• Classification
• Linear Regression
• Python
• Jupyter notebook
• Kaggle
• K-nearest algorithm.

Machine Learning languages

               There are quite a few languages available.

• Python: I am biased to Python as there are a lot of libraries available for python which will help you focus on the problem at hand. 
  • NumPy
  • SciPy
  • Matplotlib
  • Pandas
  • Tensorflow
  • Also, the support and resources available for python are better compared to python. 
  • Another advantage is it’s incredibly simple and easy to understand for beginners.
  • It’s very adaptable and easy to plug and play
• R – programming: If statistics and data visualization are prime concerns you can prefer this language as this exactly does the job. You can also use it to create machine learning algorithms. It’s free and thus make it a better choice than its counterparts like SAS and MATLAB. It also has an open-source community.
• Javascript: Tensorflow.js is the main reason people would opt for javascript but it still makes it an option for machine learning.
• Then there are some others which I do not have much information on like C, C#, Java, Unix Shell Scripting, Julia, Typescript, Scala, etc

     What next?

          So now that we have learned the basics and related terms, let’s get our hands dirty. Select a project and try to complete it, that’s the phase where you will get your doubts and deeper learning starts. You have a lot of projects in GitHub for this homework. 

      Sharing knowledge is always part of learning. So I will try and keep sharing what I learn. Happy learning.