my blogs

import shapshap.initjs()explainer = shap.TreeExplainer(model)shap_values = explainer.shap_values(X_train)shap.summary_plot(shap_values, X_train,max_display=80)

In the resampling setting, there is a common pitfall that corresponds to resample the entire dataset before splitting it into a train and a test partitions. Note that it would be equivalent to resample the train and test partitions as well.

pip install scikit-optimizepip install hyperoptpip install -i https://pypi.douban.com/simple bayesian-optimization

一、run.py执行过程:1、执行tf.app.run()2、tf.app.run()会调用本类中的的main函数,然后依次执行3、首先获取已经处理好的图片4、然后定义tensorflow占位符，convolution层和pool层，全连接层和逻辑输出层5、最后定义损失函数，反向传播过程和模型评价的一些指标6、获取session，并初始化所有变量，sess.run()启动模型训练的过程7、模型训练结束后，读取模型存放的参数和图，并预测新的数据8、sess.close()整..

from sklearn.ensemble import RandomForestClassifierfrom sklearn.datasets import make_classificationfrom sklearn.model_selection import train_test_splitimport numpy as nprng = np.random.RandomState(0)X, y = make_classification(random_state=rng)rf = .

selectmax(col),percentile(col,0.75),percentile(col,0.5),percentile(col,0.25),min(col),avg(col),stddev_samp(col) from database.hello;

from tempfile import mkdtempfrom shutil import rmtreecachedir = mkdtemp()rmtree(cachedir)

1.安装anaconda2.创建虚拟环境conda create --name datamining python=3.63.安装jupyterlabconda install -c conda-forge jupyterlab4.创建新的核心(jupyter lab自带核心ipykernel，不用安装)python -m ipykernel install --user --name=datamining --display-name datamining5.安装插件con

channels: - defaultsshow_channel_urls: truechannel_alias: https://mirrors.tuna.tsinghua.edu.cn/anacondadefault_channels: - https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main - https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free - https:...

python 3.6sklearn 0.23.2numpy 1.19.1pandas 1.1.3matplotlib 2.2.3 注意:matplotlib出现过兼容问题，不能使用3.3.3版本seaborn 0.9.0