How to starting with Machine Learning

The following are 5 steps for starting with machine learning:


Step-1: Decide to work with Machine Learning

At first, you have to decide to work with machine learning. As machine learning becoming the most attractive field in computer science, it's also a choice of millions of programmers.

Step-2: Just pick a process- 

Supervised Learning or Unsupervised Learning-based work you would like to do.

Step-3: Choose Tool

It is very important that you choose a tool for machine learning. There are two well-known tools for a machine is MATLAB and Python. Here python is a programming language that acts like natural language. Though MATLAB has reach libraries and resources, Python is also wide-spreading tools for machine learning.

Sample Code for MATLAB:


% Remove unnecessary double quotes from certain attributes
bank = datasetfun(@removequotes,bank,'DatasetOutput',true);

% Convert all the categorical variables into nominal arrays
[nrows, ncols] = size(bank);
category = false(1,ncols);
for i = 1:ncols
    if isa(bank.(names{i}),'cell') || isa(bank.(names{i}),'nominal')
        category(i) = true;
        bank.(names{i}) = nominal(bank.(names{i}));
    end
end
% Logical array keeping track of categorical attributes
catPred = category(1:end-1);
% Set the random number seed to make the results repeatable in this script
rng('default');

Sample Code for Python

 
# Python version
import sys
print('Python: {}'.format(sys.version))
# scipy
import scipy
print('scipy: {}'.format(scipy.__version__))
# numpy
import numpy
print('numpy: {}'.format(numpy.__version__))
# matplotlib
import matplotlib
print('matplotlib: {}'.format(matplotlib.__version__))
# pandas
import pandas
print('pandas: {}'.format(pandas.__version__))
# scikit-learn
import sklearn
print('sklearn: {}'.format(sklearn.__version__))

Step-4: Make a Datasets: Make a dataset on any phenomena that make your confident in machine learning first.

Step-5: Build a portfolio with the datasets. Machine Learning Tutorials for beginners will be upload here very soon...

Comments