module Stats:sig
..end
type
t = {
|
n_samples : |
(* |
Number of samples used for training
| *) |
|
target_variance : |
(* |
Variance of targets
| *) |
|
sse : |
(* |
Sum of squared errors
| *) |
|
mse : |
(* |
Mean sum of squared errors
| *) |
|
rmse : |
(* |
Root mean sum of squared errors
| *) |
|
smse : |
(* |
Standardized mean squared error
| *) |
|
msll : |
(* |
Mean standardized log loss
| *) |
|
mad : |
(* |
Mean absolute deviation
| *) |
|
maxad : |
(* |
Maximum absolute deviation
| *) |
val calc_n_samples : Gpr_interfaces.Sigs.Eval.Trained.t -> int
calc_n_samples trained
trained
.val calc_target_variance : Gpr_interfaces.Sigs.Eval.Trained.t -> float
calc_target_variance trained
trained
.val calc_sse : Gpr_interfaces.Sigs.Eval.Trained.t -> float
calc_sse trained
trained
model.val calc_mse : Gpr_interfaces.Sigs.Eval.Trained.t -> float
calc_mse trained
trained
model.val calc_rmse : Gpr_interfaces.Sigs.Eval.Trained.t -> float
calc_sse trained
trained
model.val calc_smse : Gpr_interfaces.Sigs.Eval.Trained.t -> float
calc_smse trained
trained
model. This is equivalent to the mean squared error divided
by the target variance.val calc_msll : Gpr_interfaces.Sigs.Eval.Trained.t -> float
calc_msll trained
val calc_mad : Gpr_interfaces.Sigs.Eval.Trained.t -> float
calc_mad trained
trained
model.val calc_maxad : Gpr_interfaces.Sigs.Eval.Trained.t -> float
calc_mad trained
trained
model.val calc : Gpr_interfaces.Sigs.Eval.Trained.t -> t
calc trained
trained
model.