sklearn
Machine Learning Toolbox
sklearn function module
Category: identifying an object belongs to which category ------ spam monitoring, image recognition
Regression: Continuous Attributes ------> stock prediction associated with the object
Clustering: The grouping similar objects automatically ------> customer segmentation, grouping results
Dimensionality reduction: reducing the number of the random variable to be considered ------> Visualization
Model Selection: compare, verify, and model selection parameter ------> improve accuracy by adjusting parameters
Pretreatment: feature extraction and normalization ------> converts the input data into data of a machine learning algorithm available
sklearn unified API
sklearn use the map
classification: Category regression: regression clustering: clustering demension reduction: dimensionality reduction
SVC: SVM ---> divided by L-dimensional data set of high dimensional linear boundary (boundary higher dimensional linear dimensionality reduction various curves derived low-dimensional)
KNeighbors: K neighbors
Naive Bayes: Naive Bayes
sklearn learning route
1. Getting Started
sklearn general process: data acquisition, data preprocessing, model training, model evaluation, optimization model
2. Project features
Acquired data, extracting data preprocessing, feature, feature selection
3. The algorithm works
Assessment model training, model, optimization model