Analysis

Details and available models

See the API documentation for further details on available analysis tools:

etna.analysis.feature_selection.mrmr_selection.AggregationMode(value)

Enum for different aggregation modes.

etna.analysis.feature_relevance.relevance.ModelRelevanceTable()

ModelRelevanceTable builds feature relevance table using feature relevance values obtained from model.

etna.analysis.feature_relevance.relevance.RelevanceTable(...)

Abstract class for relevance table computation.

etna.analysis.feature_relevance.relevance.StatisticsRelevanceTable()

StatisticsRelevanceTable builds feature relevance table with tsfresh statistics.

etna.analysis.outliers.density_outliers.absolute_difference_distance(x, y)

Calculate distance for get_anomalies_density() function by taking absolute value of difference.

etna.analysis.eda.plots.acf_plot(ts[, ...])

Autocorrelation and partial autocorrelation plot for multiple timeseries.

etna.analysis.eda.plots.cross_corr_plot(ts)

Cross-correlation plot between multiple timeseries.

etna.analysis.eda.plots.distribution_plot(ts)

Distribution of z-values grouped by segments and time frequency.

etna.analysis.decomposition.search.find_change_points(ts, ...)

Find trend change points using ruptures models.

etna.analysis.outliers.density_outliers.get_anomalies_density(ts)

Compute outliers according to density rule.

etna.analysis.outliers.hist_outliers.get_anomalies_hist(ts)

Get point outliers in time series using histogram model.

etna.analysis.outliers.median_outliers.get_anomalies_median(ts)

Get point outliers in time series using median model (estimation model-based method).

etna.analysis.outliers.prediction_interval_outliers.get_anomalies_prediction_interval(ts, ...)

Get point outliers in time series using prediction intervals (estimation model-based method).

etna.analysis.eda.utils.get_correlation_matrix(ts)

Compute pairwise correlation of timeseries for selected segments.

etna.analysis.feature_relevance.relevance_table.get_model_relevance_table(df, ...)

Calculate relevance table with feature importance from model.

etna.analysis.forecast.utils.get_residuals(...)

Get residuals for further analysis.

etna.analysis.feature_relevance.relevance_table.get_statistics_relevance_table(df, ...)

Calculate relevance table with p-values from tsfresh.

etna.analysis.forecast.plots.metric_per_segment_distribution_plot(...)

Plot per-segment metrics distribution.

etna.analysis.outliers.plots.plot_anomalies(ts, ...)

Plot a time series with indicated anomalies.

etna.analysis.outliers.plots.plot_anomalies_interactive(ts, ...)

Plot a time series with indicated anomalies.

etna.analysis.forecast.plots.plot_backtest(...)

Plot targets and forecast for backtest pipeline.

etna.analysis.forecast.plots.plot_backtest_interactive(...)

Plot targets and forecast for backtest pipeline using plotly.

etna.analysis.decomposition.plots.plot_change_points_interactive(ts, ...)

Plot a time series with indicated change points.

etna.analysis.eda.plots.plot_clusters(ts, ...)

Plot clusters [with centroids].

etna.analysis.eda.plots.plot_correlation_matrix(ts)

Plot pairwise correlation heatmap for selected segments.

etna.analysis.feature_relevance.plots.plot_feature_relevance(ts, ...)

Plot relevance of the features.

etna.analysis.forecast.plots.plot_forecast(...)

Plot of prediction for forecast pipeline.

etna.analysis.forecast.plots.plot_forecast_decomposition(...)

Plot of prediction and its components.

etna.analysis.eda.plots.plot_holidays(ts, ...)

Plot holidays for segments.

etna.analysis.eda.plots.plot_imputation(ts, ...)

Plot the result of imputation by a given imputer.

etna.analysis.forecast.plots.plot_metric_per_segment(...)

Plot barplot with per-segment metrics.

etna.analysis.eda.plots.plot_periodogram(ts, ...)

Plot the periodogram using scipy.signal.periodogram().

etna.analysis.forecast.plots.plot_residuals(...)

Plot residuals for predictions from backtest against some feature.

etna.analysis.decomposition.plots.plot_time_series_with_change_points(ts, ...)

Plot segments with their trend change points.

etna.analysis.decomposition.plots.plot_trend(ts, ...)

Plot series and trend from trend transform for this series.

etna.analysis.forecast.plots.prediction_actual_scatter_plot(...)

Plot scatter plot with forecasted/actual values for segments.

etna.analysis.forecast.plots.qq_plot(...[, ...])

Plot Q-Q plots for segments.

etna.analysis.decomposition.plots.seasonal_plot(ts)

Plot each season on one canvas for each segment.

etna.analysis.decomposition.plots.stl_plot(ts, ...)

Plot STL decomposition for segments.