Model configuration
Common Model Configuration
The following fields are configurable on all models:
Training data range
How far back in time to look for training data. Default 90 days. Ideally this range should not be greater than the retention period of the underlying data source, as no data would be returned beyond it. Setting this to a higher value will not lead to errors, but could limit your training data interval options.
Training data interval
The resolution of the data being trained on and of the generated predictions. Default 5 minutes.
Note that at most 50,000 samples per series can be used to train a model, so training data range / training data interval
must result in less than 50,000. For reference, 90 days of training data at a 5 minute resolution is 25,920 samples.
Prophet Hyperparameters
Changepoint Prior Scale
- Default: 0.05
- Minimum: 0.001
- Maximum: 10.0
Determines the flexibility of the trend and in particular how much the trend changes at the trend changepoints. If it is too small the trend will be underfit and variance that should have been modeled with trend changes will instead end up being handled with the noise term. If it is too large the trend will overfit and in the most extreme case you can end up with the trend capturing yearly seasonality. A range of [0.001, 0.5] would likely be about right.
Changepoint Range
- Default: 0.8
- Minimum: 0.01
- Maximum: 1.0
Controls what portion of the training data to check for changepoints. For the default, 0.8, the first 80% of the data is checked for changepoints, and any changepoints in the last 20% will be ignored.
Seasonality Prior Scale
- Default: 10.0
- Minimum: 0.01
- Maximum: 10.0
Controls the flexibility to fit the seasonality. A large value allows the seasonality to fit large fluctuations, whereas a small value shrinks the magnitude of the seasonality
Uncertainty Interval Width
- Default: 0.95
- Minimum: 0.01
- Maximum: 1
Uncertainty intervals (yhat_upper, yhat_lower) are computed as quantiles of the predicted value to use. The default of 0.95 provides an 95% confidence interval. 95% of future data should be expected to be between (yhat_upper, yhat_lower).
Seasonality Mode
- Default: additive
- Options: additive, multiplicative
Set to “multiplicative” if it appears that the magnitude of seasonal fluctuations grows with the magnitude of the time series.
Growth
- Default: linear
- Options: flat, linear, logistic
The type of model used for the growth trend component. Set to ‘flat’ only if you are certain that the trend is constant and that the time series mostly exhibits seasonality patterns, rather than trend changes. A flat growth model can reduce the width of uncertainty intervals in such cases. Choose ’linear’ if modelling something that always increases linearly, for example, disk space consumed by a constant rate of write operations, Choose ’logistic’ if attempting to model growth with a known maximum value. This requires specifying a maximum achievable point, or carrying capacity, that the forecast should saturate at. An optional floor may also be specified.
Logistic Cap
Expected maximum achievable value (required if and only if ‘growth’ is ’logistic’)
Logistic Floor
Expected minimum achievable value (optional)
Weekly Fourier Order
- Default: 3
- Minimum: 0 (disabled)
- Maximum: 25
How many Fourier orders to use when calculating weekly seasonality. Leaving this value empty will use the default of 3 when the training data range is larger than two weeks, and disable seasonality on smaller training data ranges. Setting the Fourier order to 0 will disable weekly seasonality. If there is a strong weekday vs weekend component to your data it is recommended to use a Fourier order between 10 and 25.
Separate weekly seasonality for holiday
Optionally, a holiday for which weekly seasonality should be modelled separately. This allows the model to capture the fact that the holiday may have a different day-of-week effects to non-holidays. For example, public holidays may not show the same weekday effects as non-holidays.
Daily Fourier Order
- Default: 4
- Minimum: 0 (disabled)
- Maximum: 25
How many Fourier orders to use when calculating daily seasonality. Leaving this value empty will use the default of 4 when the training data range is larger than two days, and disable seasonality on smaller training data ranges. Setting the Fourier order to 0 will disable daily seasonality.
Separate daily seasonality for holiday
Optionally, a holiday for which daily seasonality should be modelled separately. This allows the model to capture the fact that the holiday may have a different hour-of-day effects to non-holidays. For example, public holidays may not show the same spike during work hours as non-holidays.
Holidays Prior Scale
- Default: 10.0
- Minimum: 0.01
- Maximum: 10.0
Controls the flexibility to fit holidays. A large value allows fitting to large fluctuations on holidays, whereas a small value shrinks the impact of those fluctuations on the fit.
Was this page helpful?
Related resources from Grafana Labs
![Getting started with the Grafana LGTM Stack](/static/assets/featured/getting-started-with-grafana-lgtm-stack-amer_featured.png?w=200 200w, /static/assets/featured/getting-started-with-grafana-lgtm-stack-amer_featured.png?w=400 400w)
![Intro to Kubernetes monitoring in Grafana Cloud](/media/webinar/Screenshot-2024-02-12-at-4.09.30-PM.png?w=200 200w, /media/webinar/Screenshot-2024-02-12-at-4.09.30-PM.png?w=400 400w)
![Building advanced Grafana dashboards](/media/webinar/AMER_buildadv_notxt.png?w=200 200w, /media/webinar/AMER_buildadv_notxt.png?w=400 400w)