A small example package
Project description
Auto_ML_C 0.0.8
Illustrate:
这是崔连山和小伙伴们的机器学习拓展包,代有浓厚的社会主义开源分享精神,极富创造力和战斗力。在这里让我们为他们鼓掌 :clinking_glasses:
Spend
测试集数据位置:{Example}
如果全流程,源码的速度是 Fast_Example.ipynb是 38s
函数【有感知机版本】 | 耗时 | 函数【无感知机版本】 | 耗时 |
---|---|---|---|
以下数据在i9-9750H | 12核上测试 | ||
binary_ROC(X,Y,5,) | 566.51 | 也就是去掉NuSVC | 52.63 |
auto_model(X,Y,5) | 556.12 | 52.60 |
Request_install
可以参考学习当前目录下的环境备份:Auto_ML_C.yaml
主要是涉及到的软件如下:
Package | 最低版本——待检测 |
---|---|
python=3.8.10 | |
seaborn=0.11.2 | |
pandas=1.3.3 | |
matplotlib=3.4.2 | |
numpy=1.20.3 | |
py-xgboost=1.4.0 | |
Content:
该包是基于Sklearn,imblance等机器学习拓展包之上的Package,共计划分为两个部分,
-
分类任务
-
binary_classfication.py
内部可用函数如下
函数名 功能 返回值 cal_add_1(num1,num2):wave: 简单的欢迎函数 num1,num2 LogisticRegressionCV_mdoel(X, Y,cv) SGDClassifier_model(X,Y,cv) LinearDiscriminantAnalysis_model(X, Y,cv) LinearSVC_model(X, Y,cv) SVC_model(X, Y,cv) NuSVC_model(X,Y,cv) DecisionTreeClassifier_model(X,Y,cv) AdaBoostClassifier_model(X,Y,cv) BaggingClassifier_model(X, Y,cv) GradientBoostingClassifier_model(X, Y,cv) RandomForestClassifier_model(X, Y,cv) xgboost_model(X, Y,cv) KNeighborsClassifier_model(X, Y,cv) NearestNeighbors_model(X, Y,cv) BernoulliNB_model(X, Y,cv) GaussianNB_model(X,Y,cv) 下面是新增函数 binary_ROC(X,Y,k,fig_name) 绘制标量超参数搜索下最佳的ROC fig auto_model(X, Y, k) 模型的标量超参数搜索结果 Auc_data, Acc_data,
Recall_data, Precision_dataestimator_violion(df1,df2,fig_name) 为auto_model结果的Dataframe绘制小提琴图 fig -
binary_classfication_ws.py
这是为了速度考虑,舍弃占用90%时间的NuSVC函数的函数
-
多分类函数
等待
-
数据增强函数
-
waited
-
How to Use
Install
# Method 1
# Create a new environment, here is conda as an example
conda create --name Auto_ML_C python=3.8.10
# Activate the newly created environment
conda activate Auto_ML_C
# Installation package
pip install Auto_ML_C==0.0.7
# Suggest the pipeline of Jupyter notebook [optional, recommended]
conda install jupyter notebook
conda install ipykernel
python -m ipykernel install --user --name Auto_ML_C --display-name "Auto_ML_C"
# Install Sklearn 0.6. this will fixed next version
conda install -c conda-forge sklearn-contrib-lightning
conda install xgboost
[之后完善可以看到,这里依赖项比较多
_py-xgboost-mutex anaconda/pkgs/main/osx-64::_py-xgboost-mutex-2.0-cpu_0
libxgboost anaconda/pkgs/main/osx-64::libxgboost-1.3.3-h23ab428_0
py-xgboost anaconda/pkgs/main/osx-64::py-xgboost-1.3.3-py38hecd8cb5_0
xgboost anaconda/pkgs/main/osx-64::xgboost-1.3.3-py38hecd8cb5_0
]
# Method2
# Use the yaml environment file on the GitHub homepage to directly copy the current environment
conda env create -n Auto_ML_C -f Auto_ML_C.yaml
# Activate the newly created environment
conda activate
# Suggest the pipeline of Jupyter notebook [optional, recommended]
conda install jupyter notebook
conda install ipykernel
python -m ipykernel install --user --name Auto_ML_C --display-name "Auto_ML_C"
Binary Classication
# Here is an example of the function binary_classfication_ws
# 这里以函数binary_classfication_ws举例
# if you use macOS,please use function binary_classfication_ws_for_mac,This function abandons xgboost which is more troublesome for mac
ConTact
VX:Cuizy13390906310_ic
QQ:1776228595
E-mail:1776228595@qq.com
GitHub:地址待填写
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Auto_ML_C-0.0.8.tar.gz
(11.5 kB
view hashes)
Built Distribution
Auto_ML_C-0.0.8-py3-none-any.whl
(26.6 kB
view hashes)
Close
Hashes for Auto_ML_C-0.0.8-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 6ccf7723f0dd01b1b3a7a8891b904d888ddd4f4e68a733169d666d0bd8b89d23 |
|
MD5 | cec5569ef31ea57fe1a6b86e5a7cd4a4 |
|
BLAKE2b-256 | 91f1c26be2a69b7a1d52060909808a2306c28774fc32175102bbe52f940d45d7 |