Why Python instead of Matlab and R in Machine Learning?

2 min readApr 22, 2020

Machine learning has become an important tool for a vast of scientiﬁc and engineering disciplines and there are a lot of environments and tools to use when applying machine learning solutions, but which one to choose? Python, R, or Matlab? Let’s make a quick comparison between these and discuss the advantages of each one.

Why Python?

Python is becoming an increasingly central tool for data science. This was not always the case, ten years ago everyone was using Matlab instead of Python. However, due to high prices, licensing issues, and the rapid development of Python, scientiﬁc Python started to gain its user community. Nowadays, Python rapidly becoming the default programming language for practical machine learning applications. Python’s strength is in its simple syntax, readability, variability, and huge community. There are 2 versions: Python 2.7 and 3.6. however, version 2.x will not be supported anymore.

Alternatives to Python in Science

“Matlab is made for mathematicians, R for statisticians and Python for
programmers.”

Essential Modules of Python

NumPy: The fundamental package for scientific computing and matrix/numerical analysis
SciPy: Scientiﬁc computing utilities (linear algebra, FFT, signal/image processing…)
scikit-learn: Machine learning
scikit-image: Image processing
OpenCV: Open Source Computer Vision Library
Matplotlib: Plotting and visualization
Seaborn: Statistical data visualization
Pandas: Data analysis and manipulation
Tensorﬂow, Keras, PyTorch: Deep Learning
PyCharm: Editor
Jupyter Notebook: Notebook format editor
Spyder: Scientiﬁc Python Development Environment (Comes with Anaconda)

Things to Come

The editor and the environment: Matlab is slightly better than Python
Linear algebra: Matlab is better than Python
Programming constructs (loops, classes, etc.): Python better than Matlab
Machine Learning: Python is a lot better than Matlab

References

[1] TUT SGN-41007 Pattern Recognition and Machine Learning Course