grid_search_cv¶

In [ ]:

grid_search_cv(estimator,
               param_grid: dict,
               input_relation: (str, vDataFrame),
               X: list,
               y: str,
               metric: str = "auto",
               cv: int = 3,
               pos_label: (int, float, str) = None,
               cutoff: float = -1,
               training_score: bool = True,
               skip_error: bool = False,
               print_info: bool = False,)

Computes the k-fold grid search of an estimator.

Parameters¶

Name	Type	Optional	Description
estimator	object	❌	Vertica estimator having a fit method.
param_grid	dict	❌	Dictionary of the parameters to test.
input_relation	str / vDataFrame	❌	Input Relation.
X	list	❌	List of the predictor columns.
y	str	❌	Response Column.
metric	str / list	✓	Metric used to do the model evaluation. auto : logloss for classification & rmse for regression. For Classification: accuracy : Accuracy auc : Area Under the Curve (ROC) bm : Informedness = tpr + tnr - 1 csi : Critical Success Index = tp / (tp + fn + fp) f1 : F1 Score logloss : Log Loss mcc : Matthews Correlation Coefficient mk : Markedness = ppv + npv - 1 npv : Negative Predictive Value = tn / (tn + fn) prc_auc : Area Under the Curve (PRC) precision : Precision = tp / (tp + fp) recall : Recall = tp / (tp + fn) specificity : Specificity = tn / (tn + fp) For Regression: max : Max Error mae : Mean Absolute Error median : Median Absolute Error mse : Mean Squared Error msle : Mean Squared Log Error r2 : R-squared coefficient r2a : R2 adjusted rmse : Root Mean Squared Error var : Explained Variance
cv	int	✓	Number of folds.
pos_label	int / float / str	✓	The main class to be considered as positive (classification only).
cutoff	float	✓	The model cutoff (classification only).
training_score	bool	✓	If set to True, the training score will be computed with the validation score.
skip_error	bool	✓	If set to True and an error occurs, it will be displayed and not raised.
print_info	bool	✓	If set to True, prints the model information at each step.

Returns¶

tablesample : An object containing the result. For more information, see utilities.tablesample.

Example¶

In [1]:

from verticapy.learn.linear_model import LogisticRegression
model = LogisticRegression(name = "public.LR_titanic",
                           tol = 1e-4,
                           max_iter = 100, 
                           solver = 'Newton')

from verticapy.learn.model_selection import grid_search_cv
grid_search_cv(model,
               {"tol": [1e-2, 1e-4, 1e-6],
                "max_iter": [3, 10, 100],
                "solver": ["Newton", "BFGS"]},
               input_relation = "public.titanic_clean", 
               X = ["age", "fare", "parch", "sex", "boat"], 
               y = "survived", 
               cv = 3)

Out[1]:

	avg_score	avg_train_score	avg_time	score_std	score_train_std
1	0.0338729774949778	0.042819954059593664	0.8094018300374349	0.00565326273572761	0.0026011100736774693
2	0.03638378295427573	0.037116710300958866	1.1810380617777507	0.0092360291409996	0.004284315151296585
3	0.03663900999960346	0.03667922580142363	1.180067777633667	0.007031505944186805	0.00335596962457923
4	0.038035896828295036	0.03684032107512557	8.305554628372192	0.006716665605499628	0.0032610693751197366
5	0.0393751924931145	0.036514722765467	1.1204415957132976	0.01636541409818188	0.006214431324554634
6	0.04093988426819627	0.038781269785237264	0.7672536373138428	0.003416065161440895	0.0019369056238435882
7	0.0430579535520046	0.03374866715715193	8.017044067382812	0.004822256222439908	0.002044611191173046
8	0.04371481817399013	0.033395073738859866	1.580355962117513	0.00976870983603849	0.004421770934265993
9	0.044695520727177834	0.036775604889886696	0.835712750752767	0.005776140288910381	0.003256567627819591
10	0.04476292747143957	0.0330470223193894	1.0389548937479656	0.010057665153580024	0.004098239604327156
11	0.051121787729512536	0.031876769277613536	1.5491706530253093	0.012440634094598314	0.0023881436104915476
12	0.17661663941675715	0.19457229430494133	2.1853217283884683	0.09901133178214414	0.04113450964911643
13	0.25053464912229034	0.18591090857426532	2.400744676589966	0.017529837063691368	0.029291844865067175
14	0.28198058097338635	0.29342552714167036	0.7699193159739176	0.0261824607043102	0.01430022477430164
15	0.284285535090221	0.28928903809336864	0.8779073556264242	0.018441039125622852	0.012384382206053984
16	0.284903816694282	0.286138110508654	0.9634076754252116	0.015394463866562272	0.015584296153305211
17	0.7392297287561823	0.7068537220271557	1.8459896246592205	0.033825111006344916	0.12850141468168103
18	0.778475661462934	0.7197495919965173	0.9956333637237549	0.12113196473090262	0.05978336554898876

Rows: 1-18 | Columns: 6

In [2]:

# Adding the train result

grid_search_cv(model, 
               {"tol": [1e-2, 1e-4, 1e-6],
                "max_iter": [3, 10, 100],
                "solver": ["Newton", "BFGS"]},
               input_relation = "public.titanic_clean", 
               X = ["age", "fare", "parch", "sex", "boat"], 
               y = "survived", 
               cv = 3,
               metric = "auc",
               training_score = True)

Out[2]:

	avg_score	avg_train_score	avg_time	score_std	score_train_std
1	0.9957047916764754	0.989765011285069	8.342910210291544	0.0020086505397561255	0.0015396929438997338
2	0.9955958259447818	0.989867333597278	0.9211256504058838	0.00227082384081601	0.0007777915676018286
3	0.9931314320527086	0.9914753451010898	0.6805299917856852	0.0014536457654249085	0.0010435965770400793
4	0.9928137362195392	0.991425675404001	6.990610996882121	0.0010721664799684674	0.0008190828012935012
5	0.9919795398611593	0.9917641194860017	1.3745268980662029	0.005338571855018673	0.0026096842645156894
6	0.9908019645200694	0.9925100639312938	1.3107425371805828	0.0020762380526588493	0.00038865894434134764
7	0.9899609751192194	0.993060487653058	1.1924467086791992	0.003295238577443329	0.001779423267114273
8	0.9899369477340081	0.9931409935408261	0.9819730122884115	0.00289535523223408	0.0016788104414725243
9	0.9897228732999466	0.9927205298186418	1.5839755535125732	0.0025619260903407114	0.0010945148894646786
10	0.9888940008999041	0.9932717476147086	1.0834623177846272	0.003611677816088391	0.0012912109439143444
11	0.9884583799543014	0.9936549979019944	0.7054110368092855	0.002195185492339849	0.0005424783641344345
12	0.9791950721291849	0.9797737789373318	2.452951749165853	0.004643277275395455	0.005688250823444079
13	0.9419300765344258	0.9475056591575056	2.2448383967081704	0.025058948912847295	0.029946625243526214
14	0.7610859578519552	0.7662901015442332	1.0182945728302002	0.1412487074952268	0.14779754216228183
15	0.7522799785536499	0.7732061458954418	1.145751714706421	0.16640051565784883	0.14445890706301523
16	0.6592479750060258	0.6390384033281947	0.8590865929921468	0.03206423830359214	0.0548401353926451
17	0.6537411096959576	0.678152866103498	0.7217996915181478	0.013673479053865704	0.03781222947910539
18	0.6469267462702876	0.6285409464804869	0.8040386835734049	0.008179088926644757	0.026820845014610482

Rows: 1-18 | Columns: 6

In [3]:

# Printing the results

grid_search_cv(model,
               {"tol": [1e-2, 1e-4, 1e-6],
                "max_iter": [3, 10, 100],
                "solver": ["Newton", "BFGS"]},
               input_relation = "public.titanic_clean",
               X = ["age", "fare", "parch", "sex", "boat"],
               y = "survived",
               cv = 3,
               print_info = True)

Model: LogisticRegression; Parameters: {'tol': 0.01, 'max_iter': 3, 'solver': 'Newton'}; Test_score: 0.0416958046346368; Train_score: 0.038424415884418965; Time: 0.5732765197753906;
Model: LogisticRegression; Parameters: {'tol': 0.01, 'max_iter': 3, 'solver': 'BFGS'}; Test_score: 0.29398654547193737; Train_score: 0.2988679312647527; Time: 0.5154391924540201;
Model: LogisticRegression; Parameters: {'tol': 0.01, 'max_iter': 10, 'solver': 'Newton'}; Test_score: 0.052628245146305465; Train_score: 0.0301507442632964; Time: 0.832193930943807;
Model: LogisticRegression; Parameters: {'tol': 0.01, 'max_iter': 10, 'solver': 'BFGS'}; Test_score: 0.29820891278644834; Train_score: 0.301466737798373; Time: 0.4071214993794759;
Model: LogisticRegression; Parameters: {'tol': 0.01, 'max_iter': 100, 'solver': 'Newton'}; Test_score: 0.047525308310065496; Train_score: 0.03363348375788787; Time: 0.90839155515035;
Model: LogisticRegression; Parameters: {'tol': 0.01, 'max_iter': 100, 'solver': 'BFGS'}; Test_score: 0.28703377137989067; Train_score: 0.286418012223701; Time: 0.6809303760528564;
Model: LogisticRegression; Parameters: {'tol': 0.0001, 'max_iter': 3, 'solver': 'Newton'}; Test_score: 0.03951581142934674; Train_score: 0.0396534039992437; Time: 0.6992436250050863;
Model: LogisticRegression; Parameters: {'tol': 0.0001, 'max_iter': 3, 'solver': 'BFGS'}; Test_score: 0.7642205218056733; Train_score: 0.6709010814005953; Time: 0.8295307159423828;
Model: LogisticRegression; Parameters: {'tol': 0.0001, 'max_iter': 10, 'solver': 'Newton'}; Test_score: 0.0683423306806069; Train_score: 0.0264460511577688; Time: 1.2579163710276287;
Model: LogisticRegression; Parameters: {'tol': 0.0001, 'max_iter': 10, 'solver': 'BFGS'}; Test_score: 0.22774776768970934; Train_score: 0.197865157052058; Time: 2.21067746480306;
Model: LogisticRegression; Parameters: {'tol': 0.0001, 'max_iter': 100, 'solver': 'Newton'}; Test_score: 0.03324017062712357; Train_score: 0.038765923578225; Time: 1.2479968865712483;
Model: LogisticRegression; Parameters: {'tol': 0.0001, 'max_iter': 100, 'solver': 'BFGS'}; Test_score: 0.036668136667377164; Train_score: 0.036630915731166905; Time: 6.029535373051961;
Model: LogisticRegression; Parameters: {'tol': 1e-06, 'max_iter': 3, 'solver': 'Newton'}; Test_score: 0.043027230095132804; Train_score: 0.03786501867226883; Time: 0.6162236531575521;
Model: LogisticRegression; Parameters: {'tol': 1e-06, 'max_iter': 3, 'solver': 'BFGS'}; Test_score: 0.6341258178264964; Train_score: 0.576540349383636; Time: 0.9407570362091064;
Model: LogisticRegression; Parameters: {'tol': 1e-06, 'max_iter': 10, 'solver': 'Newton'}; Test_score: 0.0307525613810747; Train_score: 0.039218161825095264; Time: 1.073122262954712;
Model: LogisticRegression; Parameters: {'tol': 1e-06, 'max_iter': 10, 'solver': 'BFGS'}; Test_score: 0.19364803033332262; Train_score: 0.235090021592102; Time: 1.710917075475057;
Model: LogisticRegression; Parameters: {'tol': 1e-06, 'max_iter': 100, 'solver': 'Newton'}; Test_score: 0.039198022867401934; Train_score: 0.035456319389767364; Time: 1.282687822977702;
Model: LogisticRegression; Parameters: {'tol': 1e-06, 'max_iter': 100, 'solver': 'BFGS'}; Test_score: 0.0462506713535187; Train_score: 0.0327620077852528; Time: 12.775736331939697;

Model: LogisticRegression; Best_Parameters: {'solver': 'Newton', 'penalty': 'none', 'max_iter': 10, 'tol': 1e-06}; Best_Test_score: 0.0307525613810747; Train_score: 0.039218161825095264; Time: 1.073122262954712;

Out[3]:

	avg_score	avg_train_score	avg_time	score_std	score_train_std
1	0.0307525613810747	0.039218161825095264	1.073122262954712	0.007100662672899838	0.0035947942452672996
2	0.03324017062712357	0.038765923578225	1.2479968865712483	0.009133179807192505	0.004709129784307957
3	0.036668136667377164	0.036630915731166905	6.029535373051961	0.0040768836647644795	0.0014954728251516528
4	0.039198022867401934	0.035456319389767364	1.282687822977702	0.011278317867966861	0.005932776297560999
5	0.03951581142934674	0.0396534039992437	0.6992436250050863	0.005776079765172715	0.003074111974969664
6	0.0416958046346368	0.038424415884418965	0.5732765197753906	0.006409462434274896	0.0037795543778024313
7	0.043027230095132804	0.03786501867226883	0.6162236531575521	0.003050647757996853	0.0014852789809571398
8	0.0462506713535187	0.0327620077852528	12.775736331939697	0.012283730621080905	0.0057697535215526065
9	0.047525308310065496	0.03363348375788787	0.90839155515035	0.027200156222674646	0.009522105722295802
10	0.052628245146305465	0.0301507442632964	0.832193930943807	0.00997109580023	0.0028717054009224905
11	0.0683423306806069	0.0264460511577688	1.2579163710276287	0.0262843614565434	0.008707878886529262
12	0.19364803033332262	0.235090021592102	1.710917075475057	0.08782225931619451	0.021348701782719817
13	0.22774776768970934	0.197865157052058	2.21067746480306	0.028296871355096332	0.040256901002149334
14	0.28703377137989067	0.286418012223701	0.6809303760528564	0.005838639127727358	0.020291135406136366
15	0.29398654547193737	0.2988679312647527	0.5154391924540201	0.02014352370680819	0.0046708678871667944
16	0.29820891278644834	0.301466737798373	0.4071214993794759	0.007766157523915822	0.0022740280025160234
17	0.6341258178264964	0.576540349383636	0.9407570362091064	0.33343479658575037	0.26146235649895955
18	0.7642205218056733	0.6709010814005953	0.8295307159423828	0.42411732666221186	0.36763046489498064

Rows: 1-18 | Columns: 6