📒
📒
📒
📒
Machine & Deep Learning Compendium
Search…
The Machine & Deep Learning Compendium
Types Of Machine Learning
Data Science
Data Science Tools
Management
Data Science Management
Calculus
Probability & Statistics
Probability
Feature Types
Features
Calibration
Multi Label Classification
Distribution
Distribution Transformation
Information Theory
Game Theory
Datasets
Dataset Confidence
Normalization & Scaling
Regularization
Validation
Meta Learning
Evaluation Metrics
Benchmarking
Hyper Parameter Optimization
Multi CPU Processing
Algorithms 101
Training Strategies
Classic Machine Learning
Label Algorithms
Clustering Algorithms
Anomaly Detection
Decision Trees
Active Learning Algorithms
Linear Separator Algorithms
Ensembles
Reinforcement Learning
Incremental Learning
Dimensionality Reduction Methods
Genetic Algorithms & Genetic Programming
Learning Classifier Systems
Recommender Systems
Timeseries
Fourier Transform
Digital Signal Processing (DSP)
Propensity Score Matching
Natural Language Processing
Graphs
Deep Learning
Experimental Design
Product
Business Domains For Data Science
MLOPs Engineering
Model Formats
Data Engineering
Powered By GitBook
Normalization & Scaling
  • data has varying scales
  • Normalize between range 0 to 1.
    • When the algorithm you are using does not make assumptions about the distribution of your data, such as k-nearest neighbors and artificial neural networks.
  • Standardize, mean of 0 and a std of 1:
    • When the algorithm assumes a gaussian dist, such as linear regression, logistic regression and linear discriminant analysis. LR, LogR, LDA
**Generally, it is a good idea to standardize data that has a Gaussian (bell curve) distribution and normalize otherwise.4. In general terms, we should test 0,1 or -1,1 empirically and possibly match the range to the NN gates/activation function etc.
Previous
Dataset Confidence
Next
Regularization
Last modified 1yr ago
Copy link