1
1
"""
2
- fit(model, verbosity, data...) -> fitresult, cache, report
2
+ MLJModelInterface. fit(model, verbosity, data...) -> fitresult, cache, report
3
3
4
4
All models must implement a `fit` method. Here `data` is the
5
5
output of `reformat` on user-provided data, or some some resampling
@@ -16,7 +16,7 @@ fit(::Static, ::Integer, data...) = (nothing, nothing, nothing)
16
16
fit (m:: Supervised , verbosity, X, y, w) = fit (m, verbosity, X, y)
17
17
18
18
"""
19
- update(model, verbosity, fitresult, cache, data...)
19
+ MLJModelInterface. update(model, verbosity, fitresult, cache, data...)
20
20
21
21
Models may optionally implement an `update` method. The fallback calls
22
22
`fit`.
@@ -26,10 +26,10 @@ update(m::Model, verbosity, fitresult, cache, data...) =
26
26
fit (m, verbosity, data... )
27
27
28
28
"""
29
- training_losses(model::M, fitresult, cache)
29
+ MLJModelInterface. training_losses(model::M, fitresult, cache)
30
30
31
31
If `M` is an iterative model type which calculates training losses,
32
- implement this method to return the an `AbstractVector` of the losses
32
+ implement this method to return an `AbstractVector` of the losses
33
33
in historical order. If the model calculates scores instead, then the
34
34
sign of the scores should be reversed.
35
35
@@ -64,7 +64,7 @@ manual](https://alan-turing-institute.github.io/MLJ.jl/dev/adding_models_for_gen
64
64
reformat (model:: Model , args... ) = args
65
65
66
66
"""
67
- selectrows(::Model, I, data...) -> sampled_data
67
+ MLJModelInterface. selectrows(::Model, I, data...) -> sampled_data
68
68
69
69
A model overloads `selectrows` whenever it buys into the optional
70
70
`reformat` front-end for data preprocessing. See [`reformat`](@ref)
0 commit comments