/
regressors.jl
568 lines (385 loc) · 17 KB
/
regressors.jl
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
#= ======================
LINEAR REGRESSOR (OLS)
====================== =#
"""
$(doc_header(LinearRegressor))
This model provides standard linear regression with objective function
``|Xθ - y|₂²/2``
$DOC_SOLVERS
# Training data
In MLJ or MLJBase, bind an instance `model` to data with
mach = machine(model, X, y)
where:
- `X` is any table of input features (eg, a `DataFrame`) whose columns
have `Continuous` scitype; check column scitypes with `schema(X)`
- `y` is the target, which can be any `AbstractVector` whose element scitype is
`Continuous`; check the scitype with `scitype(y)`
Train the machine using `fit!(mach, rows=...)`.
# Hyperparameters
$TYPEDFIELDS
$(example_docstring("LinearRegressor"))
"""
@with_kw_noshow mutable struct LinearRegressor <: MMI.Deterministic
"whether to fit the intercept or not."
fit_intercept::Bool = true
""""any instance of `MLJLinearModels.Analytical`. Use `Analytical()`
for Cholesky and `CG()=Analytical(iterative=true)` for conjugate-gradient.
If `solver = nothing` (default) then `Analytical()` is used. """
solver::Option{Solver} = nothing
end
glr(m::LinearRegressor) = LinearRegression(fit_intercept=m.fit_intercept)
#= ===============
RIDGE REGRESSOR
=============== =#
"""
$(doc_header(RidgeRegressor))
Ridge regression is a linear model with objective function
``|Xθ - y|₂²/2 + n⋅λ|θ|₂²/2``
where ``n`` is the number of observations.
If `scale_penalty_with_samples = false` then the objective function is instead
``|Xθ - y|₂²/2 + λ|θ|₂²/2``.
$DOC_SOLVERS
# Training data
In MLJ or MLJBase, bind an instance `model` to data with
mach = machine(model, X, y)
where:
- `X` is any table of input features (eg, a `DataFrame`) whose columns
have `Continuous` scitype; check column scitypes with `schema(X)`
- `y` is the target, which can be any `AbstractVector` whose element scitype is
`Continuous`; check the scitype with `scitype(y)`
Train the machine using `fit!(mach, rows=...)`.
# Hyperparameters
$TYPEDFIELDS
$(example_docstring("RidgeRegressor"))
See also [`ElasticNetRegressor`](@ref).
"""
@with_kw_noshow mutable struct RidgeRegressor <: MMI.Deterministic
"strength of the L2 regularization."
lambda::Real = 1.0
"whether to fit the intercept or not."
fit_intercept::Bool = true
"whether to penalize the intercept."
penalize_intercept::Bool = false
"whether to scale the penalty with the number of observations."
scale_penalty_with_samples::Bool = true
"""any instance of `MLJLinearModels.Analytical`. Use `Analytical()` for
Cholesky and `CG()=Analytical(iterative=true)` for conjugate-gradient.
If `solver = nothing` (default) then `Analytical()` is used. """
solver::Option{Solver} = nothing
end
glr(m::RidgeRegressor) =
RidgeRegression(m.lambda,
fit_intercept=m.fit_intercept,
penalize_intercept=m.penalize_intercept,
scale_penalty_with_samples=m.scale_penalty_with_samples)
#= ===============
LASSO REGRESSOR
=============== =#
"""
$(doc_header(LassoRegressor))
Lasso regression is a linear model with objective function
``|Xθ - y|₂²/2 + n⋅λ|θ|₁``
where ``n`` is the number of observations.
If `scale_penalty_with_samples = false` the objective function is
``|Xθ - y|₂²/2 + λ|θ|₁``.
$DOC_SOLVERS
# Training data
In MLJ or MLJBase, bind an instance `model` to data with
mach = machine(model, X, y)
where:
- `X` is any table of input features (eg, a `DataFrame`) whose columns
have `Continuous` scitype; check column scitypes with `schema(X)`
- `y` is the target, which can be any `AbstractVector` whose element scitype is
`Continuous`; check the scitype with `scitype(y)`
Train the machine using `fit!(mach, rows=...)`.
# Hyperparameters
$TYPEDFIELDS
$(example_docstring("LassoRegressor"))
See also [`ElasticNetRegressor`](@ref).
"""
@with_kw_noshow mutable struct LassoRegressor <: MMI.Deterministic
"strength of the L1 regularization."
lambda::Real = 1.0
"whether to fit the intercept or not."
fit_intercept::Bool = true
"whether to penalize the intercept."
penalize_intercept::Bool = false
"whether to scale the penalty with the number of observations."
scale_penalty_with_samples::Bool = true
"""any instance of `MLJLinearModels.ProxGrad`.
If `solver=nothing` (default) then `ProxGrad(accel=true)` (FISTA) is used.
Solver aliases: `FISTA(; kwargs...) = ProxGrad(accel=true, kwargs...)`,
`ISTA(; kwargs...) = ProxGrad(accel=false, kwargs...)`. """
solver::Option{Solver} = nothing
end
glr(m::LassoRegressor) =
LassoRegression(m.lambda,
fit_intercept=m.fit_intercept,
penalize_intercept=m.penalize_intercept,
scale_penalty_with_samples=m.scale_penalty_with_samples)
#= =====================
ELASTIC NET REGRESSOR
===================== =#
"""
$(doc_header(ElasticNetRegressor))
Elastic net is a linear model with objective function
``|Xθ - y|₂²/2 + n⋅λ|θ|₂²/2 + n⋅γ|θ|₁``
where ``n`` is the number of observations.
If `scale_penalty_with_samples = false` the objective function is instead
``|Xθ - y|₂²/2 + λ|θ|₂²/2 + γ|θ|₁``.
$DOC_SOLVERS
# Training data
In MLJ or MLJBase, bind an instance `model` to data with
mach = machine(model, X, y)
where:
- `X` is any table of input features (eg, a `DataFrame`) whose columns
have `Continuous` scitype; check column scitypes with `schema(X)`
- `y` is the target, which can be any `AbstractVector` whose element scitype is
`Continuous`; check the scitype with `scitype(y)`
Train the machine using `fit!(mach, rows=...)`.
# Hyperparameters
$TYPEDFIELDS
$(example_docstring("ElasticNetRegressor"))
See also [`LassoRegressor`](@ref).
"""
@with_kw_noshow mutable struct ElasticNetRegressor <: MMI.Deterministic
"strength of the L2 regularization."
lambda::Real = 1.0
"strength of the L1 regularization."
gamma::Real = 0.0
"whether to fit the intercept or not."
fit_intercept::Bool = true
"whether to penalize the intercept."
penalize_intercept::Bool = false
"whether to scale the penalty with the number of observations."
scale_penalty_with_samples::Bool = true
"""any instance of `MLJLinearModels.ProxGrad`.
If `solver=nothing` (default) then `ProxGrad(accel=true)` (FISTA) is used.
Solver aliases: `FISTA(; kwargs...) = ProxGrad(accel=true, kwargs...)`,
`ISTA(; kwargs...) = ProxGrad(accel=false, kwargs...)`. """
solver::Option{Solver} = nothing
end
glr(m::ElasticNetRegressor) =
ElasticNetRegression(m.lambda, m.gamma,
fit_intercept=m.fit_intercept,
penalize_intercept=m.penalize_intercept,
scale_penalty_with_samples=m.scale_penalty_with_samples)
#= ==========================
ROBUST REGRESSOR (General)
========================== =#
"""
$(doc_header(RobustRegressor))
Robust regression is a linear model with objective function
``∑ρ(Xθ - y) + n⋅λ|θ|₂² + n⋅γ|θ|₁``
where ``ρ`` is a robust loss function (e.g. the Huber function) and
``n`` is the number of observations.
If `scale_penalty_with_samples = false` the objective function is instead
``∑ρ(Xθ - y) + λ|θ|₂² + γ|θ|₁``.
$DOC_SOLVERS
# Training data
In MLJ or MLJBase, bind an instance `model` to data with
mach = machine(model, X, y)
where:
- `X` is any table of input features (eg, a `DataFrame`) whose columns
have `Continuous` scitype; check column scitypes with `schema(X)`
- `y` is the target, which can be any `AbstractVector` whose element scitype is
`Continuous`; check the scitype with `scitype(y)`
Train the machine using `fit!(mach, rows=...)`.
# Hyperparameters
$TYPEDFIELDS
$(example_docstring("RobustRegressor"))
See also [`HuberRegressor`](@ref), [`QuantileRegressor`](@ref).
"""
@with_kw_noshow mutable struct RobustRegressor <: MMI.Deterministic
"the type of robust loss, which can be any instance of
`MLJLinearModels.L` where `L` is one of: `AndrewsRho`,
`BisquareRho`, `FairRho`, `HuberRho`, `LogisticRho`,
`QuantileRho`, `TalwarRho`, `HuberRho`, `TalwarRho`. "
rho::RobustRho = HuberRho(0.1)
"strength of the regularizer if `penalty` is `:l2` or `:l1`.
Strength of the L2 regularizer if `penalty` is `:en`."
lambda::Real = 1.0
"strength of the L1 regularizer if `penalty` is `:en`."
gamma::Real = 0.0
"the penalty to use, either `:l2`, `:l1`, `:en` (elastic net) or `:none`."
penalty::SymStr = :l2
"whether to fit the intercept or not."
fit_intercept::Bool = true
"whether to penalize the intercept."
penalize_intercept::Bool = false
"whether to scale the penalty with the number of observations."
scale_penalty_with_samples::Bool = true
"""some instance of `MLJLinearModels.S` where `S` is one of: `LBFGS`, `IWLSCG`,
`Newton`, `NewtonCG`, if `penalty = :l2`, and `ProxGrad` otherwise.
If `solver = nothing` (default) then `LBFGS()` is used, if `penalty = :l2`, and
otherwise `ProxGrad(accel=true)` (FISTA) is used.
Solver aliases: `FISTA(; kwargs...) = ProxGrad(accel=true, kwargs...)`,
`ISTA(; kwargs...) = ProxGrad(accel=false, kwargs...)`"""
solver::Option{Solver} = nothing
end
glr(m::RobustRegressor) =
RobustRegression(m.rho, m.lambda, m.gamma;
penalty=Symbol(m.penalty),
fit_intercept=m.fit_intercept,
penalize_intercept=m.penalize_intercept,
scale_penalty_with_samples=m.scale_penalty_with_samples)
#= ===============
HUBER REGRESSOR
=============== =#
"""
$(doc_header(HuberRegressor))
This model coincides with [`RobustRegressor`](@ref), with the exception that the robust
loss, `rho`, is fixed to `HuberRho(delta)`, where `delta` is a new hyperparameter.
$DOC_SOLVERS
# Training data
In MLJ or MLJBase, bind an instance `model` to data with
mach = machine(model, X, y)
where:
- `X` is any table of input features (eg, a `DataFrame`) whose columns
have `Continuous` scitype; check column scitypes with `schema(X)`
- `y` is the target, which can be any `AbstractVector` whose element scitype is
`Continuous`; check the scitype with `scitype(y)`
Train the machine using `fit!(mach, rows=...)`.
# Hyperparameters
$TYPEDFIELDS
$(example_docstring("HuberRegressor"))
See also [`RobustRegressor`](@ref), [`QuantileRegressor`](@ref).
"""
@with_kw_noshow mutable struct HuberRegressor <: MMI.Deterministic
"parameterizes the `HuberRho` function (radius of the ball within which the loss
is a quadratic loss)"
delta::Real = 0.5
"strength of the regularizer if `penalty` is `:l2` or `:l1`.
Strength of the L2 regularizer if `penalty` is `:en`."
lambda::Real = 1.0
"strength of the L1 regularizer if `penalty` is `:en`."
gamma::Real = 0.0
"the penalty to use, either `:l2`, `:l1`, `:en` (elastic net) or `:none`."
penalty::SymStr = :l2
"whether to fit the intercept or not."
fit_intercept::Bool = true
"whether to penalize the intercept."
penalize_intercept::Bool = false
"whether to scale the penalty with the number of observations."
scale_penalty_with_samples::Bool = true
"""some instance of `MLJLinearModels.S` where `S` is one of: `LBFGS`, `IWLSCG`,
`Newton`, `NewtonCG`, if `penalty = :l2`, and `ProxGrad` otherwise.
If `solver = nothing` (default) then `LBFGS()` is used, if `penalty = :l2`, and
otherwise `ProxGrad(accel=true)` (FISTA) is used.
Solver aliases: `FISTA(; kwargs...) = ProxGrad(accel=true, kwargs...)`,
`ISTA(; kwargs...) = ProxGrad(accel=false, kwargs...)`"""
solver::Option{Solver} = nothing
end
glr(m::HuberRegressor) =
HuberRegression(m.delta, m.lambda, m.gamma;
penalty=Symbol(m.penalty),
fit_intercept=m.fit_intercept,
penalize_intercept=m.penalize_intercept,
scale_penalty_with_samples=m.scale_penalty_with_samples)
#= ==================
QUANTILE REGRESSOR
================== =#
"""
$(doc_header(QuantileRegressor))
This model coincides with [`RobustRegressor`](@ref), with the exception that the robust
loss, `rho`, is fixed to `QuantileRho(delta)`, where `delta` is a new hyperparameter.
$DOC_SOLVERS
# Training data
In MLJ or MLJBase, bind an instance `model` to data with
mach = machine(model, X, y)
where:
- `X` is any table of input features (eg, a `DataFrame`) whose columns
have `Continuous` scitype; check column scitypes with `schema(X)`
- `y` is the target, which can be any `AbstractVector` whose element scitype is
`Continuous`; check the scitype with `scitype(y)`
Train the machine using `fit!(mach, rows=...)`.
# Hyperparameters
$TYPEDFIELDS
$(example_docstring("QuantileRegressor"))
See also [`RobustRegressor`](@ref), [`HuberRegressor`](@ref).
"""
@with_kw_noshow mutable struct QuantileRegressor <: MMI.Deterministic
"parameterizes the `QuantileRho` function (indicating the quantile to use
with default `0.5` for the median regression)"
delta::Real = 0.5
"strength of the regularizer if `penalty` is `:l2` or `:l1`.
Strength of the L2 regularizer if `penalty` is `:en`."
lambda::Real = 1.0
"strength of the L1 regularizer if `penalty` is `:en`."
gamma::Real = 0.0
"the penalty to use, either `:l2`, `:l1`, `:en` (elastic net) or `:none`."
penalty::SymStr = :l2
"whether to fit the intercept or not."
fit_intercept::Bool = true
"whether to penalize the intercept."
penalize_intercept::Bool = false
"whether to scale the penalty with the number of observations."
scale_penalty_with_samples::Bool = true
"""some instance of `MLJLinearModels.S` where `S` is one of: `LBFGS`, `IWLSCG`,
if `penalty = :l2`, and `ProxGrad` otherwise.
If `solver = nothing` (default) then `LBFGS()` is used, if `penalty = :l2`, and
otherwise `ProxGrad(accel=true)` (FISTA) is used.
Solver aliases: `FISTA(; kwargs...) = ProxGrad(accel=true, kwargs...)`,
`ISTA(; kwargs...) = ProxGrad(accel=false, kwargs...)`"""
solver::Option{Solver} = nothing
end
glr(m::QuantileRegressor) =
QuantileRegression(m.delta, m.lambda, m.gamma;
penalty=Symbol(m.penalty),
fit_intercept=m.fit_intercept,
penalize_intercept=m.penalize_intercept,
scale_penalty_with_samples=m.scale_penalty_with_samples)
#= ==================================
LEAST ABSOLUTE DEVIATION REGRESSOR
================================== =#
"""
$(doc_header(LADRegressor))
Least absolute deviation regression is a linear model with objective function
``∑ρ(Xθ - y) + n⋅λ|θ|₂² + n⋅γ|θ|₁``
where ``ρ`` is the absolute loss and ``n`` is the number of observations.
If `scale_penalty_with_samples = false` the objective function is instead
``∑ρ(Xθ - y) + λ|θ|₂² + γ|θ|₁``.
$DOC_SOLVERS
# Training data
In MLJ or MLJBase, bind an instance `model` to data with
mach = machine(model, X, y)
where:
- `X` is any table of input features (eg, a `DataFrame`) whose columns
have `Continuous` scitype; check column scitypes with `schema(X)`
- `y` is the target, which can be any `AbstractVector` whose element scitype is
`Continuous`; check the scitype with `scitype(y)`
Train the machine using `fit!(mach, rows=...)`.
# Hyperparameters
See also `RobustRegressor`.
## Parameters
$TYPEDFIELDS
$(example_docstring("LADRegressor"))
"""
@with_kw_noshow mutable struct LADRegressor <: MMI.Deterministic
"strength of the regularizer if `penalty` is `:l2` or `:l1`.
Strength of the L2 regularizer if `penalty` is `:en`."
lambda::Real = 1.0
"strength of the L1 regularizer if `penalty` is `:en`."
gamma::Real = 0.0
"the penalty to use, either `:l2`, `:l1`, `:en` (elastic net) or `:none`."
penalty::SymStr = :l2
"whether to fit the intercept or not."
fit_intercept::Bool = true
"whether to penalize the intercept."
penalize_intercept::Bool = false
"whether to scale the penalty with the number of observations."
scale_penalty_with_samples::Bool = true
"""some instance of `MLJLinearModels.S` where `S` is one of: `LBFGS`, `IWLSCG`,
if `penalty = :l2`, and `ProxGrad` otherwise.
If `solver = nothing` (default) then `LBFGS()` is used, if `penalty = :l2`, and
otherwise `ProxGrad(accel=true)` (FISTA) is used.
Solver aliases: `FISTA(; kwargs...) = ProxGrad(accel=true, kwargs...)`,
`ISTA(; kwargs...) = ProxGrad(accel=false, kwargs...)`"""
solver::Option{Solver} = nothing
end
glr(m::LADRegressor) =
LADRegression(m.lambda, m.gamma;
penalty=Symbol(m.penalty),
fit_intercept=m.fit_intercept,
penalize_intercept=m.penalize_intercept,
scale_penalty_with_samples=m.scale_penalty_with_samples)