tv.named_transformers_["high_cardinality"].get_feature_names_out()
array(['maintenance, facilities, finance', 'station, state, estate',
- 'behavioral, health, school', 'gaithersburg, the, clarksburg',
- 'procurement, protective, adult', 'warehouse, delivery, liquor',
- 'patrol, 5th, 4th', 'supports, support, network',
- 'traffic, safety, alcohol', 'spring, silver, monitoring',
- 'family, pedophile, crimes', 'rockville, twinbrook, downtown',
- 'administration, battalion, registration',
- 'highway, welfare, child', 'technology, inmate, systems',
- 'management, equipment, budget',
- 'communications, communication, division',
- 'nicholson, transit, taxicab',
- 'investigative, investigations, criminal', 'security, mc311, mccf',
- 'custody, mcdc, quality', 'inspections, inspection, collection',
- 'eligibility, assistance, disability', 'services, service, animal',
- 'programs, projects, program', 'accounts, receivable, members',
- 'planning, training, recruit', 'district, squad, 3rd',
- 'emergency, commuter, duplicating', 'director, kingsview, officer',
- 'firefighter, recruit, rescuer',
- 'paralegal, psychiatrist, employee',
- 'income, assistance, compliance', 'librarian, associate, library',
- 'candidate, police, sergeant', 'manager, projects, project',
+array(['station, state, estate', 'district, patrol, 3rd',
+ 'silver, spring, urban', 'welfare, children, childhood',
+ 'services, highway, service', 'traffic, safety, alcohol',
+ 'gaithersburg, clarksburg, the', 'supports, downtown, support',
+ 'twinbrook, warehouse, rockville',
+ 'investigative, criminal, investigations',
+ 'equipment, automotive, fleet', 'assessment, ombudsman, adult',
+ 'programs, program, commercial', 'development, delivery, cloverly',
+ 'nicholson, transit, taxicab', 'abandoned, division, employee',
+ 'behavioral, health, school', 'maintenance, facilities, finance',
+ 'technology, inmate, systems',
+ 'administrative, administration, battalion',
+ 'toddlers, custody, members', 'building, structural, construction',
+ 'eligibility, assistance, medical',
+ 'communications, communication, immunization',
+ 'training, director, recruit', 'mangement, management, engagement',
+ 'collection, operations, special',
+ 'regulatory, environmental, centers', 'assignment, squad, team',
+ 'security, mc311, mccf', 'representative, legislative, customer',
+ 'manager, projects, project', 'accountant, assistant, library',
+ 'candidate, officer, office', 'operator, bus, operations',
+ 'school, health, room', 'lieutenant, captain, chief',
+ 'firefighter, rescuer, recruit', 'iii, police, of',
+ 'program, programs, graphic',
+ 'information, technology, technologist',
+ 'sergeant, attendant, police',
+ 'correctional, correction, corporal',
+ 'crossing, purchasing, engineer', 'community, nurse, unit',
+ 'coordinator, services, service', 'assistance, income, client',
'enforcement, permitting, inspector',
- 'program, programs, resource', 'operator, bus, operations',
- 'administrative, principal, executive', 'captain, rescue, chief',
- 'technician, mechanic, supply', 'school, room, behavioral',
+ 'technician, mechanic, supply',
+ 'administrative, principal, executive',
+ 'warehouse, welfare, caseworker', 'transit, public, telephone',
+ 'sheriff, deputy, therapist', 'specialist, recreation, special',
+ 'supervisor, supervisory, records', 'master, meter, registered',
'communications, telecommunications, safety',
- 'community, health, nurse', 'correctional, correction, corporal',
- 'liquor, clerk, store', 'services, president, resident',
- 'specialist, special, quality', 'coordinator, coordinating, depot',
- 'officer, office, iii', 'master, registered, meter',
- 'craftsworker, supervisor, advisor', 'sheriff, deputy, autobody',
- 'information, recreation, technology',
- 'warehouse, welfare, caseworker', 'crossing, purchasing, engineer',
- 'lieutenant, shift, records', 'accountant, assistant, county',
- 'equipment, investigator, apprentice'], dtype=object)
+ 'equipment, investment, investigator',
+ 'environmental, budget, analyst', 'liquor, clerk, store'],
+ dtype=object)
-
@@ -864,7 +868,7 @@
A simple prediction pipelineprint(f"R2 score: mean: {np.mean(scores):.3f}; std: {np.std(scores):.3f}\n")
R2 score: mean: 0.922; std: 0.010
+R2 score: mean: 0.921; std: 0.015
The simple pipeline applied on this complex dataset gave us very good results.
@@ -933,7 +937,7 @@ ConclusionTotal running time of the script: (1 minutes 12.266 seconds)
+Total running time of the script: (1 minutes 15.855 seconds)
diff --git a/0.1/auto_examples/02_feature_interpretation_with_gapencoder.html b/0.1/auto_examples/02_feature_interpretation_with_gapencoder.html
index b1220292..7e2ae7c6 100644
--- a/0.1/auto_examples/02_feature_interpretation_with_gapencoder.html
+++ b/0.1/auto_examples/02_feature_interpretation_with_gapencoder.html
@@ -850,7 +850,7 @@ ConclusionGapEncoder, and how it can be used to summarize categorical variables
as a set of latent topics.
-
Total running time of the script: (0 minutes 1.752 seconds)
+Total running time of the script: (0 minutes 1.841 seconds)
diff --git a/0.1/auto_examples/03_datetime_encoder.html b/0.1/auto_examples/03_datetime_encoder.html
index 821f64cb..95f7722a 100644
--- a/0.1/auto_examples/03_datetime_encoder.html
+++ b/0.1/auto_examples/03_datetime_encoder.html
@@ -850,7 +850,7 @@ ConclusionTableVectorizer, which automatically recognizes
and transforms datetime columns by default.
-
Total running time of the script: (0 minutes 4.439 seconds)
+Total running time of the script: (0 minutes 4.448 seconds)
diff --git a/0.1/auto_examples/04_fuzzy_joining.html b/0.1/auto_examples/04_fuzzy_joining.html
index a7f91db7..45905e35 100644
--- a/0.1/auto_examples/04_fuzzy_joining.html
+++ b/0.1/auto_examples/04_fuzzy_joining.html
@@ -1781,7 +1781,7 @@ Using the Best parameters: {'joiner-1__max_dist': 0.1, 'joiner-2__max_dist': 0.9, 'joiner-3__max_dist': 0.1}
-Total running time of the script: (0 minutes 22.247 seconds)
+Total running time of the script: (0 minutes 20.264 seconds)
diff --git a/0.1/auto_examples/05_deduplication.html b/0.1/auto_examples/05_deduplication.html
index 67e670b3..ff47e8c0 100644
--- a/0.1/auto_examples/05_deduplication.html
+++ b/0.1/auto_examples/05_deduplication.html
@@ -665,7 +665,7 @@ ConclusionGapEncoder
or MinHashEncoder.
-
Total running time of the script: (0 minutes 4.893 seconds)
+Total running time of the script: (0 minutes 5.130 seconds)
diff --git a/0.1/auto_examples/06_ken_embeddings.html b/0.1/auto_examples/06_ken_embeddings.html
index 48fc7bc5..f6035b15 100644
--- a/0.1/auto_examples/06_ken_embeddings.html
+++ b/0.1/auto_examples/06_ken_embeddings.html
@@ -672,7 +672,7 @@ Extracting entity embeddingsfetch_ken_table_aliases()
-{'all_entities', 'albums', 'games', 'companies', 'schools', 'movies'}
+{'albums', 'companies', 'games', 'all_entities', 'movies', 'schools'}
The games table is the most relevant to our case.
@@ -968,7 +968,7 @@
Plotting the results
It helped significantly improve the prediction score.
-Total running time of the script: (11 minutes 7.052 seconds)
+Total running time of the script: (10 minutes 42.099 seconds)
diff --git a/0.1/auto_examples/07_multiple_key_join.html b/0.1/auto_examples/07_multiple_key_join.html
index 8b276c21..963f94cd 100644
--- a/0.1/auto_examples/07_multiple_key_join.html
+++ b/0.1/auto_examples/07_multiple_key_join.html
@@ -1514,7 +1514,7 @@ Joining: feature augmentation across tablesJoiner() transformer.
Our final cross-validated accuracy score is 0.58.
-Total running time of the script: (11 minutes 58.676 seconds)
+Total running time of the script: (10 minutes 58.884 seconds)
diff --git a/0.1/auto_examples/08_join_aggregation.html b/0.1/auto_examples/08_join_aggregation.html
index 7770247c..65f2488e 100644
--- a/0.1/auto_examples/08_join_aggregation.html
+++ b/0.1/auto_examples/08_join_aggregation.html
@@ -1096,75 +1096,75 @@ Hyper-parameters tuning and cross validation"value_counts" yields an
exhaustive histogram over all the possible values of ratings
(here 10 different values, from 0.5 to 5).
-Total running time of the script: (0 minutes 23.746 seconds)
+Total running time of the script: (0 minutes 17.626 seconds)
diff --git a/0.1/auto_examples/09_interpolation_join.html b/0.1/auto_examples/09_interpolation_join.html
index 24295bb9..07c4ff1b 100644
--- a/0.1/auto_examples/09_interpolation_join.html
+++ b/0.1/auto_examples/09_interpolation_join.html
@@ -756,9 +756,9 @@ Joining the tablesstate
-AK -3.195786
-MT -0.569793
-WA 0.008274
-ND 1.144235
-MN 1.436454
+AK -2.373219
+MT -0.047876
+WA 0.498786
+MN 1.236130
+ND 1.247732
Name: TMAX, dtype: float64
@@ -1003,11 +1003,11 @@ Sanity checksstate
-LA 21.634984
-FL 24.810720
-HI 26.988138
-VI 30.824549
-PR 31.345255
+LA 21.805534
+FL 24.660732
+HI 26.866774
+VI 31.384267
+PR 31.769181
Name: TMAX, dtype: float64
@@ -1036,7 +1036,7 @@ ConclusionInterpolationJoiner transformer: we give it a table (the weather data) and a set of matching columns (here date, latitude, longitude) and it learns to predict the other columns’ values (such as the max daily temperature).
Then, it transforms tables by predicting values that a matching row would contain, rather than by searching for an actual match.
It is a generalization of the fuzzy_join(), as fuzzy_join() is the same thing as an InterpolationJoiner where the estimators are 1-nearest-neighbor estimators.
-
Total running time of the script: (0 minutes 8.311 seconds)
+Total running time of the script: (0 minutes 5.717 seconds)
diff --git a/0.1/auto_examples/sg_execution_times.html b/0.1/auto_examples/sg_execution_times.html
index 40b2b5d9..a068a939 100644
--- a/0.1/auto_examples/sg_execution_times.html
+++ b/0.1/auto_examples/sg_execution_times.html
@@ -11,10 +11,10 @@
-
+
-
+
Computation times — skrub
@@ -471,7 +471,7 @@
Computation times#
-25:23.383 total execution time for 9 files from auto_examples:
+23:51.862 total execution time for 9 files from auto_examples: