Time Series Forecasting Results
When a model finishes training, click on the model to see the results.
Forecast charts
The model report contains a visualization of the time series forecast vs. the ground truth of the target variable. If quantiles were specified, this graph also contains the forecast intervals.
If K-Fold cross-test is used for evaluation, the forecast and forecast intervals are shown for every fold.
For multiple time series datasets, one visualization per time series is provided.
Performance: Metrics
For multiple time series datasets, metrics are aggregated over all time series.
If at least one time series has an undefined metric, then the aggregated metric is also undefined.
If K-Fold cross-test is used for evaluation, these aggregated metrics are then averaged over all folds, ignoring folds that yield undefined metric values.
Performance: Per time series metrics
For multiple time series datasets, DSS also shows the metrics of each indivudal time series.
If K-Fold cross-test is used for evaluation, per time series metrics are aggregated over each fold for each time series, ignoring folds that yield undefined metric values.
Model Information: Algorithm
For multiple time series datasets, some models train one algorithm per time series under the hood (mainly ARIMA and Seasonal LOESS). The resulting per times series hyperparameters are shown in this tab, if any.