get_accuracy | R Documentation |
This is a function to get the predictions for a specific output simulated by SCOPE and calculate the accuracy.
get_accuracy(
obs_vec,
predictions,
metric_function,
Filter = FALSE,
timestamp,
month_start = 1,
month_end = 12,
period = c("day", "dawn_dusk", "night"),
nu_interations,
interations = 101,
dryhours_vec,
dry_hours = 0,
neg_vec,
neg_null = 0,
pred_neg = 0,
neg_values = NA,
lat = 52.46,
lon = 13.32
)
obs_vec |
for get_accuracy a vector of observed values in the same timestamp as the predictions. |
predictions |
the name of the file result of the get_predictions function. |
metric_function |
for get_accuracy is possible to select the metric, |
The result is a table with the accuracy metrics for all simulation starting with that name.
Examples of uses of the get_predictions function
###############
library(tidyverse)
possible metrics: yardstick::rmse for Root mean squared error,
yardstick::mae for Mean absolute error,
yardstick::msd for Mean signed deviation,
yardstick::ccc for Concordance correlation coefficient,
yardstick::mase for Mean absolute scaled error - order by time
yardstick::rsq for R squared - correlation
yardstick::rsq_trad for R squared - traditional
yardstick::rpiq for Ratio of performance to inter-quartile
yardstick::rpd for Ratio of performance to deviation
yardstick::iic for Index of ideality of correlation
yardstick::mpe for Mean percentage error
yardstick::mape for Mean absolute percent error
yardstick::smape for Symmetric mean absolute percentage error
Get model accuracy for ET corrected
Predictions_metrics_1169 <- get_accuracy(obs_vec = EC_ROTH$ET_clean0,
predictions = Predictions_pixel_1169,
metric_function = yardstick::metric_set(yardstick::rsq,
yardstick::rmse,
yardstick::mae))
Predictions_metrics_1169
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.