TSErrors:timeseries数据的各种错误
TSErrors的Python项目详细描述
此存储库的目的是收集各种性能指标或错误 在一个地方计算时间序列/序列/1D数据。目前只支持1d数据。在
如何安装
使用pip
pip install TSErrors
使用github链接获取最新代码
^{pr2}$使用安装文件，转到下载repo的文件夹
python setup.py install
错误
当前正在计算以下错误。在
| Name | Name in this repository |
|---|---|
| Absolute Percent Bias | ^{ |
| Agreement Index | ^{ |
| Aitchison Distance | ^{ |
| Alpha decomposition of the NSE | ^{ |
| Anomaly correction coefficient | ^{ |
| Bias | ^{ |
| Beta decomposition of NSE | ^{ |
| Bounded NSE | ^{ |
| Bounded KGE | ^{ |
| Brier Score | ^{ |
| Correlation Coefficient | ^{ |
| Coefficient of Determination | ^{ |
| Centered Root Mean Square Deviation | ^{ |
| Covariances | ^{ |
| Decomposed Mean Square Error | ^{ |
| Explained variance score | ^{ |
| Euclid Distance | ^{ |
| Geometric Mean Difference | ^{ |
| Geometric Mean Absolute Error | ^{ |
| Geometric Mean Relative Absolute Error | ^{ |
| Inertial Root Squared Error | ^{ |
| Integral Normalized Root Squared Error | ^{ |
| Inter-percentile Normalized Root Mean Squared Error | ^{ |
| Jensen-shannon divergence | ^{ |
| Kling-Gupta Efficiency | ^{ |
| Legate-McCabe Efficiency Index | ^{ |
| Logrithmic Nash Sutcliff Efficiency | ^{ |
| Logrithmic probability distribution | ^{ |
| maximum error | ^{ |
| Mean Absolute Error | ^{ |
| Mean Absolute Percentage Deviation | ^{ |
| Mean Absolute Percentage Error | ^{ |
| Mean Absolute Relative Error | ^{ |
| Mean Absolute Scaled Error | ^{ |
| Mean Arctangle Absolute Percentage Error | ^{ |
| Mean Bias Error | ^{ |
| Mean Bounded relative Absolute Error | ^{ |
| Mean Errors | ^{ |
| Mean Gamma Deviances | ^{ |
| Mean Log Error | ^{ |
| Mean Normalized Root Mean Square Error | ^{ |
| Mean Percentage Error | ^{ |
| Mean Poisson Deviance | ^{ |
| Mean Relative Absolute Error | ^{ |
| Mean Square Error | ^{ |
| Mean Square Logrithmic Errors | ^{ |
| Mean Variance | ^{ |
| Median Absolute Error | ^{ |
| Median Absolute Percentage Error | ^{ |
| Median Dictionary Accuracy | |
| Median Error | ^{ |
| Median Relative Absolute Error | ^{ |
| Median Squared Error | ^{ |
| Mielke-Berry R | ^{ |
| Modified Agreement of Index | ^{ |
| Modified Kling-Gupta Efficiency | ^{ |
| Modified Nash-Sutcliff Efficiency | ^{ |
| Nash-Sutcliff Efficiency | ^{ |
| Non parametric Kling-Gupta Efficiency | ^{ |
| Normalized Absolute Error | ^{ |
| Normalized Absolute Percentage Error | ^{ |
| Normalized Euclid Distance | ^{ |
| Normalized Root Mean Square Error | ^{ |
| Peak flow bias of the flow duration curve | ^{ |
| Pearson correlation coefficient | ^{ |
| Percent Bias | ^{ |
| Range Normalized root mean square | ^{ |
| Refined Agreement of Index | ^{ |
| Relative Agreement of Index | ^{ |
| Relative Absolute Error | ^{ |
| Relative Root Mean Squared Error | ^{ |
| Relative Nash-Sutcliff Efficiency | ^{ |
| Root Mean Square Errors | ^{ |
| Root Mean Square Log Error | ^{ |
| Root Mean Square Percentage Error | ^{ |
| Root Mean Squared Scaled Error | ^{ |
| Root Median Squared Scaled Error | ^{ |
| Root Relative Squared Error | ^{ |
| RSR | ^{ |
| Separmann correlation coefficient | ^{ |
| Skill Score of Murphy | ^{ |
| Spectral Angle | ^{ |
| Spectral Correlation | ^{ |
| Spectral Gradient Angle | ^{ |
| Spectral Information Divergence | ^{ |
| Symmetric kullback-leibler divergence | ^{ |
| Symmetric Mean Absolute Percentage Error | ^{ |
| Symmetric Median Absolute Percentage Error | ^{ |
| sum of squared errors | ^{ |
| Volume Errors | ^{ |
| Volumetric Efficiency | ^{ |
| Unscaled Mean Bounded Relative Absolute Error | ^{ |
| Watterson's M | ^{ |
| Weighted Mean Absolute Percent Errors | ^{ |
| Weighted Absolute Percentage Error | ^{ |
如何使用
importnumpyasnpfromTSErrorsimportFindErrorstrue=np.random.random((20,1))pred=np.random.random((20,1))er=FindErrors(true,pred)forminer.all_methods:print("{:20}".format(m))# get names of all availabe methodser.nse()# calculate Nash Sutcliff efficiencyer.calculate_all(verbose=True)# or calculate errors using all available methodser.stats(verbose=True)# get some important stats about true and predicted arrays
相关
- 项目
标签：
欢迎加入QQ群-->： 979659372
