Skip to content

Commit fbf8975

Browse files
SvenKlaassenSvenKlaassen
committed
update plr results
1 parent 14dfe0d commit fbf8975

File tree

5 files changed

+100
-36
lines changed

5 files changed

+100
-36
lines changed

‎monte-cover/src/montecover/plm/plr_ate.py‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -49,9 +49,9 @@ def _convert_ml_string_to_object(self, ml_string):
4949
if ml_string == "Lasso":
5050
learner = LassoCV()
5151
elif ml_string == "Random Forest":
52-
learner = RandomForestRegressor(n_estimators=100, max_features=20, max_depth=5, min_samples_leaf=2)
52+
learner = RandomForestRegressor(n_estimators=200, max_features=10, max_depth=5, min_samples_leaf=20)
5353
elif ml_string == "LGBM":
54-
learner = LGBMRegressor(n_estimators=100, learning_rate=0.05, verbose=-1, n_jobs=1)
54+
learner = LGBMRegressor(n_estimators=500, learning_rate=0.01, verbose=-1, n_jobs=1)
5555
else:
5656
raise ValueError(f"Unknown learner type: {ml_string}")
5757

‎results/plm/plr_ate_config.yml‎

Lines changed: 70 additions & 16 deletions
Original file line numberDiff line numberDiff line change
@@ -83,15 +83,15 @@ dml_parameters:
8383
- ccp_alpha
8484
- monotonic_cst
8585
max_depth: 5
86-
max_features: 20
86+
max_features: 10
8787
max_leaf_nodes: null
8888
max_samples: null
8989
min_impurity_decrease: 0.0
90-
min_samples_leaf: 2
90+
min_samples_leaf: 20
9191
min_samples_split: 2
9292
min_weight_fraction_leaf: 0.0
9393
monotonic_cst: null
94-
n_estimators: 100
94+
n_estimators: 200
9595
n_jobs: null
9696
oob_score: false
9797
random_state: null
@@ -122,15 +122,15 @@ dml_parameters:
122122
splitter: best
123123
estimator_params: *id001
124124
max_depth: 5
125-
max_features: 20
125+
max_features: 10
126126
max_leaf_nodes: null
127127
max_samples: null
128128
min_impurity_decrease: 0.0
129-
min_samples_leaf: 2
129+
min_samples_leaf: 20
130130
min_samples_split: 2
131131
min_weight_fraction_leaf: 0.0
132132
monotonic_cst: null
133-
n_estimators: 100
133+
n_estimators: 200
134134
n_jobs: null
135135
oob_score: false
136136
random_state: null
@@ -179,15 +179,15 @@ dml_parameters:
179179
splitter: best
180180
estimator_params: *id001
181181
max_depth: 5
182-
max_features: 20
182+
max_features: 10
183183
max_leaf_nodes: null
184184
max_samples: null
185185
min_impurity_decrease: 0.0
186-
min_samples_leaf: 2
186+
min_samples_leaf: 20
187187
min_samples_split: 2
188188
min_weight_fraction_leaf: 0.0
189189
monotonic_cst: null
190-
n_estimators: 100
190+
n_estimators: 200
191191
n_jobs: null
192192
oob_score: false
193193
random_state: null
@@ -218,15 +218,15 @@ dml_parameters:
218218
splitter: best
219219
estimator_params: *id001
220220
max_depth: 5
221-
max_features: 20
221+
max_features: 10
222222
max_leaf_nodes: null
223223
max_samples: null
224224
min_impurity_decrease: 0.0
225-
min_samples_leaf: 2
225+
min_samples_leaf: 20
226226
min_samples_split: 2
227227
min_weight_fraction_leaf: 0.0
228228
monotonic_cst: null
229-
n_estimators: 100
229+
n_estimators: 200
230230
n_jobs: null
231231
oob_score: false
232232
random_state: null
@@ -270,12 +270,12 @@ dml_parameters:
270270
class_weight: null
271271
colsample_bytree: 1.0
272272
importance_type: split
273-
learning_rate: 0.05
273+
learning_rate: 0.01
274274
max_depth: -1
275275
min_child_samples: 20
276276
min_child_weight: 0.001
277277
min_split_gain: 0.0
278-
n_estimators: 100
278+
n_estimators: 500
279279
n_jobs: 1
280280
num_leaves: 31
281281
objective: null
@@ -306,12 +306,12 @@ dml_parameters:
306306
class_weight: null
307307
colsample_bytree: 1.0
308308
importance_type: split
309-
learning_rate: 0.05
309+
learning_rate: 0.01
310310
max_depth: -1
311311
min_child_samples: 20
312312
min_child_weight: 0.001
313313
min_split_gain: 0.0
314-
n_estimators: 100
314+
n_estimators: 500
315315
n_jobs: 1
316316
num_leaves: 31
317317
objective: null
@@ -322,6 +322,60 @@ dml_parameters:
322322
subsample_for_bin: 200000
323323
subsample_freq: 0
324324
verbose: -1
325+
- ml_g: !!python/tuple
326+
- LGBM
327+
- !!python/object:lightgbm.sklearn.LGBMRegressor
328+
_Booster: null
329+
_best_iteration: -1
330+
_best_score: {}
331+
_class_map: null
332+
_class_weight: null
333+
_classes: null
334+
_evals_result: {}
335+
_n_classes: -1
336+
_n_features: -1
337+
_n_features_in: -1
338+
_objective: null
339+
_other_params:
340+
verbose: -1
341+
boosting_type: gbdt
342+
class_weight: null
343+
colsample_bytree: 1.0
344+
importance_type: split
345+
learning_rate: 0.01
346+
max_depth: -1
347+
min_child_samples: 20
348+
min_child_weight: 0.001
349+
min_split_gain: 0.0
350+
n_estimators: 500
351+
n_jobs: 1
352+
num_leaves: 31
353+
objective: null
354+
random_state: null
355+
reg_alpha: 0.0
356+
reg_lambda: 0.0
357+
subsample: 1.0
358+
subsample_for_bin: 200000
359+
subsample_freq: 0
360+
verbose: -1
361+
ml_m: !!python/tuple
362+
- Lasso
363+
- !!python/object:sklearn.linear_model._coordinate_descent.LassoCV
364+
_sklearn_version: 1.5.2
365+
alphas: null
366+
copy_X: true
367+
cv: null
368+
eps: 0.001
369+
fit_intercept: true
370+
max_iter: 1000
371+
n_alphas: 100
372+
n_jobs: null
373+
positive: false
374+
precompute: auto
375+
random_state: null
376+
selection: cyclic
377+
tol: 0.0001
378+
verbose: false
325379
score:
326380
- partialling out
327381
- IV-type

‎results/plm/plr_ate_coverage.csv‎

Lines changed: 25 additions & 17 deletions
Original file line numberDiff line numberDiff line change
@@ -1,17 +1,25 @@
1-
Learner g,Learner m,score,level,Coverage,CI Length,Bias,repetition
2-
Lasso,Lasso,IV-type,0.9,0.881,0.1393979255576113,0.0352891099128789,1000
3-
Lasso,Lasso,IV-type,0.95,0.945,0.16610287331091153,0.0352891099128789,1000
4-
Lasso,Lasso,partialling out,0.9,0.908,0.14646362984437974,0.034686755904342816,1000
5-
Lasso,Lasso,partialling out,0.95,0.956,0.17452217926042807,0.034686755904342816,1000
6-
Lasso,Random Forest,IV-type,0.9,0.898,0.14672410325315619,0.036150217819390894,1000
7-
Lasso,Random Forest,IV-type,0.95,0.954,0.1748325524704008,0.036150217819390894,1000
8-
Lasso,Random Forest,partialling out,0.9,0.817,0.143330638432391,0.04191262539042403,1000
9-
Lasso,Random Forest,partialling out,0.95,0.886,0.17078898973477283,0.04191262539042403,1000
10-
Random Forest,Lasso,IV-type,0.9,0.877,0.141921407652651,0.035804568203505666,1000
11-
Random Forest,Lasso,IV-type,0.95,0.948,0.16910978768971624,0.035804568203505666,1000
12-
Random Forest,Lasso,partialling out,0.9,0.903,0.15198241550336503,0.03618790663815028,1000
13-
Random Forest,Lasso,partialling out,0.95,0.954,0.1810982179746172,0.03618790663815028,1000
14-
Random Forest,Random Forest,IV-type,0.9,0.907,0.1493755431098695,0.03649624902429271,1000
15-
Random Forest,Random Forest,IV-type,0.95,0.949,0.1779919379264572,0.03649624902429271,1000
16-
Random Forest,Random Forest,partialling out,0.9,0.884,0.14638253610723148,0.036967300943126565,1000
17-
Random Forest,Random Forest,partialling out,0.95,0.942,0.17442555011265587,0.036967300943126565,1000
1+
Learner g,Learner m,Score,level,Coverage,CI Length,Bias,repetition
2+
LGBM,LGBM,IV-type,0.9,0.873,0.1600979705787658,0.041736453670531866,1000
3+
LGBM,LGBM,IV-type,0.95,0.937,0.19076849829726017,0.041736453670531866,1000
4+
LGBM,LGBM,partialling out,0.9,0.807,0.14701046434404016,0.043407336971130486,1000
5+
LGBM,LGBM,partialling out,0.95,0.887,0.17517377275621215,0.043407336971130486,1000
6+
LGBM,Lasso,IV-type,0.9,0.878,0.14908437061246727,0.03862236976140604,1000
7+
LGBM,Lasso,IV-type,0.95,0.926,0.17764498449616647,0.03862236976140604,1000
8+
LGBM,Lasso,partialling out,0.9,0.892,0.15981114790450912,0.039529461666836616,1000
9+
LGBM,Lasso,partialling out,0.95,0.945,0.19042672800093632,0.039529461666836616,1000
10+
Lasso,Lasso,IV-type,0.9,0.877,0.1404646476549414,0.03601132554422347,1000
11+
Lasso,Lasso,IV-type,0.95,0.928,0.16737395108830316,0.03601132554422347,1000
12+
Lasso,Lasso,partialling out,0.9,0.9,0.14736731076791165,0.03607425067346851,1000
13+
Lasso,Lasso,partialling out,0.95,0.938,0.17559898149657663,0.03607425067346851,1000
14+
Lasso,Random Forest,IV-type,0.9,0.839,0.13067912086907693,0.03663040619086667,1000
15+
Lasso,Random Forest,IV-type,0.95,0.901,0.1557137767385692,0.03663040619086667,1000
16+
Lasso,Random Forest,partialling out,0.9,0.753,0.14303542962976423,0.04845002977336067,1000
17+
Lasso,Random Forest,partialling out,0.95,0.854,0.17043722675016001,0.04845002977336067,1000
18+
Random Forest,Lasso,IV-type,0.9,0.883,0.14162830043619654,0.03632778967870182,1000
19+
Random Forest,Lasso,IV-type,0.95,0.933,0.16876052889948315,0.03632778967870182,1000
20+
Random Forest,Lasso,partialling out,0.9,0.888,0.1512642580778129,0.037711493804103186,1000
21+
Random Forest,Lasso,partialling out,0.95,0.948,0.18024248062130577,0.037711493804103186,1000
22+
Random Forest,Random Forest,IV-type,0.9,0.852,0.13195953745670919,0.036614704598155204,1000
23+
Random Forest,Random Forest,IV-type,0.95,0.911,0.15723948720656877,0.036614704598155204,1000
24+
Random Forest,Random Forest,partialling out,0.9,0.896,0.14282075762578683,0.03605962387417277,1000
25+
Random Forest,Random Forest,partialling out,0.95,0.94,0.17018142928016597,0.03605962387417277,1000

‎results/plm/plr_ate_metadata.csv‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -1,2 +1,2 @@
11
DoubleML Version,Script,Date,Total Runtime (minutes),Python Version,Config File
2-
0.10.dev0,PLRATECoverageSimulation,2025-03-21 17:43,12.395700534184774,3.12.9,scripts/plm/plr_ate_config.yml
2+
0.10.dev0,PLRATECoverageSimulation,2025-04-28 10:01,17.98821387688319,3.12.9,scripts/plm/plr_ate_config.yml

‎scripts/plm/plr_ate_config.yml‎

Lines changed: 2 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -24,6 +24,8 @@ dml_parameters:
2424
ml_m: ["Lasso"]
2525
- ml_g: ["LGBM"]
2626
ml_m: ["LGBM"]
27+
- ml_g: ["LGBM"]
28+
ml_m: ["Lasso"]
2729

2830
score: ["partialling out", "IV-type"]
2931

0 commit comments

Comments
 (0)