Skip to content

Latest commit

 

History

History
45 lines (35 loc) · 3 KB

machine_learning.md

File metadata and controls

45 lines (35 loc) · 3 KB

T o D o

Math

  • Linear Algebra: Fundamental to machine learning and AI. You'll need to understand vectors, matrices, tensor operations, eigenvalues and eigenvectors. They are used in various aspects of ML such as neural networks, support vector machines, dimensionality reduction, etc.
  • Calculus: You should have a good grasp of differential and integral calculus. Understanding concepts such as derivatives and integrals are necessary for optimization problems, understanding learning rates, and in training neural networks (backpropagation).
  • Multivariable Calculus: Since many ML algorithms involve multiple variables, knowing how to perform calculus operations on them is important. Concepts such as partial derivatives and gradients are key.
  • Probability Theory and Statistics: Crucial for understanding and interpreting data. Topics should include probability rules & axioms, Bayes’ theorem, random variables, expectation, variance, distributions (like Gaussian, Binomial, Poisson), central limit theorem, hypothesis testing, and confidence intervals.
  • Optimization: Techniques for finding the best solution to a problem, for instance, gradient descent, are heavily used in ML.
  • Information Theory: Entropy, Information Gain, KL Divergence, etc., are used in various parts of ML including decision trees and in understanding learning in deep networks.

Important links

Coursera machine learning specialization - 3 courses

Machine learning C++ youtube playlist

fast.ai courses - Practical Deep Learning for Coders, From Deep Learning Foundations to Stable Diffusion

Tools/Libraries/Websites/Services

Try scatter plots - https://plotly.com/python/3d-scatter-plots/

Python modules

  • Seaborn - great for ploting values with colors etc.

Key concepts

  • EDA | Understand the data, .shape, .columns, number of values in each columns, Make scatter plots - 2D/3D, understand the axis - labels/scale, Pair-plots, Histograms, univariate, bivariate and multivariate analysis, Probability distribution function (PDF), Cumulative distribution function(CDF), Mean, Variance, Standard deviation,median, percentile, quantile, Interquartile Range (IQR) & Mean Absolute Deviation (MAD), Median Absolute deviation, Interquartile range, Box plot, Whiskers, Violin plots, Multivariate probability density, contour plot,

Misc stuff that helps ML

  • Space & Time complexity

Important key terms

  • EDA - Exploratory data analysis
  • Data point/vector/observation
  • Data Set
  • Feature/Variable/Input variable/Dependent variable
  • Lable/Independent variable/Output var/Class/class lable/ response lable
  • Vectors and Arrays
  • Outlayer
  • Univaraite analysis(PDF, CDF, Boxplot, Voilin plots)
  • Bi-variate analysis (scatter plots, pair-plots)

Math for ML

Linear Algebra

https://www.khanacademy.org/math/linear-algebra/