pycatcher 0.0.20

This package identifies outlier(s) for a given time-series dataset in simple steps. It supports day, week, month and quarter level time-series data.

Outlier Detection for Time-series Data

This package identifies outlier(s) for a given time-series dataset in simple steps. It supports day, week, month and quarter level time-series data.

DataFrame Arguments:

First column in the dataframe must be a date column ('YYYY-MM-DD') and the last column a count column.

Package Functions:

• detect_outliers(df): Detect outliers in a time-series dataframe using seasonal trend decomposition when there is at least 2 years of data, otherwise we can use Interquartile Range (IQR) for smaller timeframe.
• detect_outliers_today(df) Detect outliers for the current date in a time-series dataframe.
• detect_outliers_latest(df): Detect latest outliers in a time-series dataframe.
• find_outliers_iqr(df): Detect outliers in a time-series dataframe when there's less than 2 years of data.

Diagnostic Plots:

• build_seasonal_plot(df): Build seasonal plot (additive or multiplicative) for a given dataframe.
• build_iqr_plot(df): Build IQR plot for a given dataframe (for less than 2 years of data).
• build_monthwise_plot(df): Build month-wise plot for a given dataframe.
• build_decomposition_results(df): Get seasonal decomposition results for a given dataframe.
• conduct_stationarity_check(df): Conduct stationarity check (trend only) for a feature (dataframe's count column).