- Introduction to Python machine learning
- The first chapter is to let the computer learn from the data
- Turn data into knowledge
- Three kinds of machine learning algorithms
- Chapter II Training machine learning classification algorithm
- A glimpse into the history of early machine learning through artificial neurons
- Using Python to implement the perceptual machine algorithm
- Training perceptual machine model based on IRIS data set
- Adaptive linear neurons and convergence problems
- Python implements adaptive linear neurons
- Large-scale machine learning and random gradient descent
- Chapter III tour using Scikit-learn for classifiers
- How to choose the right classifier algorithm
- Scikit-learn Tour
- Logistic regression modeling of class probabilities
- Using regularization to resolve overfitting
- Support Vector Machine
- Using relaxation variables to solve non-linear sub-conditions
- Using kernel SVM to solve nonlinear problems
- Decision Tree Learning
- Maximum information gain
- Building a decision Tree
- Random Forest
- K Nearest neighbor--an algorithm of lazy learning
- Summarize
- The fourth chapter constructs a good training set---data preprocessing
- Handling Missing values
- Eliminate features or samples with missing values
- Overwrite missing values
- Understanding the Estimator API in Sklearn
- Working with categorical data
- Splitting a dataset into training and test sets
- Uniform feature Value Range
- Select a feature that is meaningful
- Using random forest to assess feature importance
- Summarize
- The fifth chapter compresses data by reducing dimension
- PCA for unsupervised dimensionality reduction
- Have a chat about variance
- Feature conversions
- LDA to supervise data compression
- The original data is mapped to the new feature space
- Using kernel PCA for nonlinear mapping
- Implementing kernel PCA with Python
- Mapping a new data point
- Nuclear PCA in the Sklearn
- Summarize
- Sixth. Model Evaluation and parameter adjustment
- Create a workflow from a pipeline
- K-fold cross-validation evaluates model performance
- Debugging algorithms using learning curves and validation curves
- Search Parameters by grid
- Selecting algorithms by nested cross-validation
- Different performance evaluation indicators
- Seventh Chapter Integration Learning
- Integrated learning
- Vote with different classification algorithms
- Eighth chapter Pytorch of Deep Learning
Python Machine learning Chinese version
