x-hacker / machine_learning

Basic Machine Learning

Introduction

This project implement classic machine learning algorithms(ML). Motivations for this project includes:

• Helping machine learning freshman to have a better and deeper understanding of the basic algorithms and model in this field.
• Providing the real-life and commercial executing methods in ML filed.
• Keeping my Mathematics Theory and Coding ability fresh due to such cases.

Overview

1.FM

1.1 fast_fm

Show how to use the package of 'fast_fm' to classify the dataset.

1.2 fm_rewrite

Follow the theory of FM , we write the python script by ourselves.

1.3 fm_easy_run

Used by : pip install fm_easy_run.

2.Xgboost

2.1 xgboost

Show how to use the package of 'xgboost' to classify the dataset.

2.2 gridsearch

Show how to use the package of 'gridsearch' to select the best params of the 'xgboost' algorithm.

3.N-gram

An interview problem solved by n-gram instead of Naive Bayes.

4.Svd

@bolg：SVD及扩展的矩阵分解方法

4.1 Matrix decomposition in linalg

4.2 Matrix decomposition with RSVD

5.Collaborative Filtering Recommendation System

@bolg：能够快速实现的协同推荐

5.1 Base of Item

5.2 Base of User

6.Semantic recognition

@bolg：基于自然语言识别下的流失用户预警

6.1 Jieba Process

6.2 Tf-Idf

6.3 Bp Neural Network

6.4 SVM process

6.5 Naive Bayes

6.6 RandomForest

7.Gradient_descent

8.Smote

8.1 Mean of the weight

8.2 Random scale in connected Vector

@bolg：SMOTE算法

9.fast_risk_control

@bolg：风控用户识别方法

Requirements

Python Environment. More details getting from single project requirement.

More

If you find some incorrect content, i'm so sorry about that. PLS contact me by the following way:

• WeChat:sharalion
• E-mail:[email protected]
• Message Board in my bolg