package hex.glm;
import hex.GLMMetrics;
import hex.ModelMetricsRegressionGLM;
import hex.deeplearning.DeepLearningModel;
import hex.glm.GLMModel.GLMParameters;
import hex.glm.GLMModel.GLMParameters.Family;
import hex.glm.GLMModel.GLMParameters.Solver;
import org.junit.AfterClass;
import org.junit.Assert;
import org.junit.BeforeClass;
import org.junit.Test;
import water.DKV;
import water.Key;
import water.TestUtil;
import water.exceptions.H2OModelBuilderIllegalArgumentException;
import water.fvec.Frame;
import water.fvec.NFSFileVec;
import water.parser.ParseDataset;
import java.io.File;
import java.io.IOException;
import java.util.HashMap;
import java.util.Map;
import water.fvec.*;
import static water.util.FileUtils.*;
import static org.junit.Assert.assertEquals;
import static org.junit.Assert.assertFalse;
import static org.junit.Assert.assertTrue;
/**
* Created by tomasnykodym on 6/4/15.
*/
public class GLMBasicTestRegression extends TestUtil {
static Frame _canCarTrain;
static Frame _earinf;
static Frame _weighted;
static Frame _upsampled;
static Vec _merit, _class;
static Frame _prostateTrain;
static Frame _airlines;
static Frame _airlinesMM;
@BeforeClass
public static void setup() throws IOException {
stall_till_cloudsize(1);
File f = getFile("smalldata/glm_test/cancar_logIn.csv");
assert f.exists();
NFSFileVec nfs = NFSFileVec.make(f);
Key outputKey = Key.make("prostate_cat_train.hex");
_canCarTrain = ParseDataset.parse(outputKey, nfs._key);
_canCarTrain.add("Merit", (_merit = _canCarTrain.remove("Merit")).toCategoricalVec());
_canCarTrain.add("Class", (_class = _canCarTrain.remove("Class")).toCategoricalVec());
DKV.put(_canCarTrain._key, _canCarTrain);
f = getFile("smalldata/glm_test/earinf.txt");
nfs = NFSFileVec.make(f);
outputKey = Key.make("earinf.hex");
_earinf = ParseDataset.parse(outputKey, nfs._key);
DKV.put(_earinf._key, _earinf);
f = getFile("smalldata/glm_test/weighted.csv");
nfs = NFSFileVec.make(f);
outputKey = Key.make("weighted.hex");
_weighted = ParseDataset.parse(outputKey, nfs._key);
DKV.put(_weighted._key, _weighted);
f = getFile("smalldata/glm_test/upsampled.csv");
nfs = NFSFileVec.make(f);
outputKey = Key.make("upsampled.hex");
_upsampled = ParseDataset.parse(outputKey, nfs._key);
DKV.put(_upsampled._key, _upsampled);
_prostateTrain = parse_test_file("smalldata/glm_test/prostate_cat_train.csv");
_airlines = parse_test_file("smalldata/airlines/AirlinesTrain.csv.zip");
Vec v = _airlines.remove("IsDepDelayed");
Vec v2 = v.makeCopy(null);
_airlines.add("IsDepDelayed",v2);
v.remove();
DKV.put(_airlines._key,_airlines);
// System.out.println("made copy of vec " + v._key + " -> " + v2._key + ", in DKV? src =" + ((DKV.get(v._key) != null)) + ", dst = " + (DKV.get(v2._key) != null));
_airlinesMM = parse_test_file(Key.make("AirlinesMM"), "smalldata/airlines/AirlinesTrainMM.csv.zip");
v = _airlinesMM.remove("IsDepDelayed");
_airlinesMM.add("IsDepDelayed",v.makeCopy(null));
v.remove();
DKV.put(_airlinesMM._key,_airlinesMM);
}
@Test
public void testSingleCatNoIcpt(){
Vec cat = Vec.makeVec(new long[]{1,1,1,0,0},new String[]{"black","red"},Vec.newKey());
Vec res = Vec.makeVec(new double[]{1,1,0,0,0},cat.group().addVec());
Frame fr = new Frame(Key.<Frame>make("fr"),new String[]{"x","y"},new Vec[]{cat,res});
DKV.put(fr);
GLMParameters parms = new GLMParameters();
parms._train = fr._key;
parms._alpha = new double[]{0};
parms._response_column = "y";
parms._intercept = false;
parms._family = Family.binomial;
// just make sure it runs
GLMModel model = new GLM(parms).trainModel().get();
Map<String,Double> coefs = model.coefficients();
System.out.println("coefs = " + coefs);
Assert.assertEquals(coefs.get("Intercept"),0,0);
Assert.assertEquals(4.2744474,((GLMMetrics)model._output._training_metrics).residual_deviance(),1e-4);
System.out.println();
model.delete();
fr.delete();
}
@Test
public void testSparse() {
double [] exp_coefs = new double [] {0.0233151691783671,-0.00543776852277619,-0.0137359312181047,0.00770037200907652,0.0328856331139761,-0.0242845468071283,-0.0101698117745265,0.00868844870137727,0.000349121513384513,-0.0106962199761512,-0.00705001448025939,0.00821574637914086,0.00601015905212279,0.0021278467162546,-0.0233079168835112,0.00535473896013676,-0.00897667301004576,0.00788272228017582,0.00237442711371947,-0.013136425134371,0.00134003869245749,0.0240118046676911,0.000607214787933269,-0.0112908513868027,0.000443119443631777,0.00749330452744921,-0.00558704122833295,0.000533036850835694,0.0130008059852934,-4.40634889376063e-05,-0.00580285872202347,0.0117029111583238,-0.00685480666428133,0.00809526311326634,-0.0088567165389072,-0.0363126456378731,-0.00267237519808936,-0.01669554043682,0.00556943053195684,0.0178196407614288,-0.000903204442155076,-0.0085363297586185,-0.00421147221966977,-0.00828702756129772,0.017027928644479,0.00710126315700672,0.019819043342772,-0.0165232485929677,0.00439570108491533,0.0188325734374437,0.00799712968759025,-0.0100388875424171,-0.0062415137856855,-0.00258013659839137,-6.58516379178382e-05,0.0135032332096949,-0.00776869619293087,-0.00544035128543343,-0.0110626226606883,-0.00768490011210769,-0.00684181016695251,-0.0144627862333649,-0.0262830557415184,-0.0102290180164706,0.00368252955770187,0.015824495748353,0.00383484095683782,0.0151193905626625,-0.00615077094420626,0.0142842231522414,0.00150448184871646,0.0521491615912011,0.0128661232226479,0.00225580439739044,-0.0117476427864401,-0.0059792656068627,0.000787012740598272,0.00255419488737936,0.00406033118385186,0.0102551045653601,0.00423949002681056,-0.0116986428989079,0.00232448128787425,-0.00296198808290357,-0.00793738689381332,-0.000771158906679964,0.00435708760153937,-0.0138922325725763,0.00264561130131037,-0.0156128295187466,-0.0102023187068811,0.0074744189329328,0.0102377749189598,-0.0304739969497646,0.00692556661464647,0.00151065993974025,0.0133704258946895,-0.0167391228441308,0.0111804482435337,-0.0062469732087272,-0.00930165243463748,-0.00418698783410104,0.00190918091726462,0.00632982717347925,-0.00277608255480933,-0.00175463261672652,-0.00267223587651978,-0.00329264073314718,0.000960091877616874,-0.00946014799557438,-0.0112302467393988,-0.00870512647578646,-0.00238582834931644,-0.0100845163232815,-0.00675861103174491,-0.000689229731411459,0.0127651588318169,-0.0062753105816655,-0.00240575758827749,0.00439570108491531,0.00934971690544427,-0.0184380964678117,-0.00474253892124699,0.00522916014066936,-0.0105148336464531,0.0088372219244051,0.0100429095740915,-0.0107657032259033,-0.00512476269437683,-0.00558487620671732,-0.000637298812579742,-0.00118460090105795,-0.00369801350318738,-0.00556276860695209,0.00789011470305446,-0.00248367841256358,0.00677762904717052,-0.00640135771848287,0.00797532960057465,-0.00117508910987595,0.000986931150778778,-0.0148237721063735,0.0053001635341953,-0.0139698571439444,-0.0172255105183439,-0.0177416268392445,-0.0107062660197562,-0.00735448768491512,-0.00418482390542493,0.00933957546887131,-0.00761657876743367,0.0107862806984669,6.99667442150322e-05,-0.00151054027221715,0.00941377216029456,0.0112882845381545,0.0014423575345095,0.00845773223444363,-0.00675939077916714,-0.00329806028742896,0.000276998824889068,0.00206337643122044,-0.00173085772672239,0.00169616445468346,0.00281297187309321,-0.0152343998246272,0.0126261762792184,-0.000224959505615703,-0.00476466349783071,-0.0102541605421868,-0.000561674281900828,0.00367777757696579,-0.000960272764476094,0.00255704179717728,-0.000696266184051808,0.0470920125432207,0.0115016691642458,-0.00287666464467251,-0.00132912286075637,0.00201932482935891,0.00119899092739739,0.00380417340899902,-0.00394363983208331,-0.00294543812868618,-1.77894150438862e-05,-0.00455002740798846,0.000613307426862812,0.00348274063618593,0.00161877234851832,0.0231608701706833,-0.00390062462708628,0.00244047437999614,-0.00143984617445982,-0.00221831741496412,-0.00744853810342609,-0.00575689075773469,-0.00567890661011033,0.00384589889309526,-0.00173241442296732,-0.00526995531653655,-0.00310819786514896,0.00740596461822877,-0.0790037392468225,0.0239744234187787,0.0514310481067108,0.034335426530007,0.0254604884688754,0.0531375235023675,-0.0228335779154641,0.546865402727144};
GLMModel model1 = null;
GLMParameters parms = new GLMParameters(Family.gaussian);
_airlinesMM.add("weights",_airlinesMM.anyVec().makeCon(1.0));
DKV.put(_airlinesMM._key,_airlinesMM);
parms._weights_column = "weights";
parms._train = _airlinesMM._key;
parms._lambda = new double[]{1e-2};
parms._alpha = new double[]{0};
parms._solver = Solver.IRLSM;
parms._ignored_columns = new String[]{"C1"};
// parms._remove_collinear_columns = true;
parms._response_column = "IsDepDelayed";
parms._standardize = true;
parms._objective_epsilon = 0;
parms._gradient_epsilon = 1e-10;
parms._max_iterations = 1000;
parms._missing_values_handling = DeepLearningModel.DeepLearningParameters.MissingValuesHandling.Skip;
try {
model1 = new GLM(parms).trainModel().get();
for(int i = 0; i < model1._output._coefficient_names.length; ++i)
assertEquals(exp_coefs[i],model1._output.getNormBeta()[i],Math.abs(exp_coefs[i])*1e-8);
} finally {
if (model1 != null) model1.delete();
}
}
@Test public void testWeights() {
GLMModel model1 = null, model2 = null;
GLMParameters parms = new GLMParameters(Family.gaussian);
parms._train = _weighted._key;
parms._ignored_columns = new String[]{_weighted.name(0)};
parms._response_column = _weighted.name(1);
parms._standardize = true;
parms._objective_epsilon = 0;
parms._gradient_epsilon = 1e-10;
parms._max_iterations = 1000;
for (Solver s : GLMParameters.Solver.values()) {
// if(s != Solver.IRLSM)continue; //fixme: does not pass for other than IRLSM now
System.out.println("===============================================================");
System.out.println("Solver = " + s);
System.out.println("===============================================================");
try {
parms._lambda = new double[]{1e-5};
parms._alpha = null;
parms._train = _weighted._key;
parms._solver = s;
parms._weights_column = "weights";
model1 = new GLM(parms).trainModel().get();
HashMap<String, Double> coefs1 = model1.coefficients();
System.out.println("coefs1 = " + coefs1);
parms._train = _upsampled._key;
parms._weights_column = null;
parms._lambda = new double[]{1e-5};
parms._alpha = null;
model2 = new GLM(parms).trainModel().get();
HashMap<String, Double> coefs2 = model2.coefficients();
System.out.println("coefs2 = " + coefs2);
System.out.println("mse1 = " + model1._output._training_metrics.mse() + ", mse2 = " + model2._output._training_metrics.mse());
System.out.println( model1._output._training_metrics);
System.out.println( model2._output._training_metrics);
assertEquals(model2._output._training_metrics.mse(), model1._output._training_metrics.mse(),1e-4);
} finally {
if(model1 != null) model1.delete();
if(model2 != null) model2.delete();
}
}
}
@Test public void testOffset() {
GLMModel model1 = null, model2 = null;
GLMParameters parms = new GLMParameters(Family.gaussian);
parms._train = _weighted._key;
parms._ignored_columns = new String[]{_weighted.name(0)};
parms._response_column = _weighted.name(1);
parms._standardize = true;
parms._objective_epsilon = 0;
parms._gradient_epsilon = 1e-10;
parms._max_iterations = 1000;
Solver s = Solver.IRLSM;
try {
parms._lambda = new double[]{0};
parms._alpha = new double []{0};
parms._train = _weighted._key;
parms._solver = s;
parms._offset_column = "C20";
parms._compute_p_values = true;
parms._standardize = false;
model1 = new GLM(parms).trainModel().get();
HashMap<String, Double> coefs1 = model1.coefficients();
System.out.println("coefs1 = " + coefs1);
/**
* Call:
glm(formula = C2 ~ . - C1 - C20, data = data, offset = data$C20)
Deviance Residuals:
Min 1Q Median 3Q Max
-3.444 -0.821 -0.021 0.878 2.801
Coefficients:
Estimate Std. Error t value Pr(>|t|)
(Intercept) -0.026928 0.479281 -0.056 0.9553
C3 -0.064657 0.144517 -0.447 0.6558
C4 -0.076132 0.163746 -0.465 0.6432
C5 0.397962 0.161458 2.465 0.0158 *
C6 0.119644 0.173165 0.691 0.4916
C7 -0.124615 0.151145 -0.824 0.4121
C8 0.142455 0.164912 0.864 0.3902
C9 0.087358 0.158266 0.552 0.5825
C10 -0.012873 0.155429 -0.083 0.9342
C11 0.277392 0.181299 1.530 0.1299
C12 0.004988 0.170290 0.029 0.9767
C13 -0.091400 0.172910 -0.529 0.5985
C14 -0.248876 0.177311 -1.404 0.1643
C15 0.053598 0.167305 0.320 0.7495
C16 0.156302 0.157823 0.990 0.3249
C17 0.296317 0.167453 1.770 0.0806 .
C18 0.013306 0.162185 0.082 0.9348
C19 0.115939 0.160250 0.723 0.4715
weights -0.005771 0.303477 -0.019 0.9849
*
*/
double [] expected_coefs = new double[]{-0.064656782,-0.076131880,0.397962147,0.119644094,-0.124614842,0.142455018,0.087357855,-0.012872522,0.277392182,0.004987961,-0.091400128,-0.248875970
,0.053597896,0.156301780,0.296317472,0.013306398,0.115938809,-0.005771429,-0.026928297};
double [] expected_pvals = new double[]{0.65578062,0.64322317,0.01582348,0.49158786,0.41209217,0.39023637,0.58248959,0.93419972,0.12990598,0.97670462,0.59852911,0.16425679,0.74951951,0.32494727
,0.08056447,0.93481349,0.47146503,0.98487376,0.95533301};
double [] actual_coefs = model1.beta();
double [] actual_pvals = model1._output.pValues();
for(int i = 0; i < expected_coefs.length; ++i) {
assertEquals(expected_coefs[i], actual_coefs[i],1e-4);
assertEquals(expected_pvals[i], actual_pvals[i],1e-4);
}
} finally {
if (model1 != null) model1.delete();
if (model2 != null) model2.delete();
}
}
@Test public void testTweedie() {
GLMModel model = null;
Frame scoreTrain = null;
// -------------------------------------- R examples output ----------------------------------------------------------------
// Call: glm(formula = Infections ~ ., family = tweedie(0), data = D)
//
// Coefficients:
// (Intercept) SwimmerOccas LocationNonBeach Age20-24 Age25-29 SexMale
// 0.8910 0.8221 0.7266 -0.5033 -0.2679 -0.1056
//
// Degrees of Freedom: 286 Total (i.e. Null); 281 Residual
// Null Deviance: 1564
// Residual Deviance: 1469 AIC: NA
// Call: glm(formula = Infections ~ ., family = tweedie(1), data = D)
//
// Coefficients:
// (Intercept) SwimmerOccas LocationNonBeach Age20-24 Age25-29 SexMale
// -0.12261 0.61149 0.53454 -0.37442 -0.18973 -0.08985
//
// Degrees of Freedom: 286 Total (i.e. Null); 281 Residual
// Null Deviance: 824.5
// Residual Deviance: 755.4 AIC: NA
// Call: glm(formula = Infections ~ ., family = tweedie(1.25), data = D)
//
// Coefficients:
// (Intercept) SwimmerOccas LocationNonBeach Age20-24 Age25-29 SexMale
// 1.02964 -0.14079 -0.12200 0.08502 0.04269 0.02105
//
// Degrees of Freedom: 286 Total (i.e. Null); 281 Residual
// Null Deviance: 834.2
// Residual Deviance: 770.8 AIC: NA
// Call: glm(formula = Infections ~ ., family = tweedie(1.5), data = D)
//
// Coefficients:
// (Intercept) SwimmerOccas LocationNonBeach Age20-24 Age25-29 SexMale
// 1.05665 -0.25891 -0.22185 0.15325 0.07624 0.03908
//
// Degrees of Freedom: 286 Total (i.e. Null); 281 Residual
// Null Deviance: 967
// Residual Deviance: 908.9 AIC: NA
// Call: glm(formula = Infections ~ ., family = tweedie(1.75), data = D)
//
// Coefficients:
// (Intercept) SwimmerOccas LocationNonBeach Age20-24 Age25-29 SexMale
// 1.08076 -0.35690 -0.30154 0.20556 0.10122 0.05375
//
// Degrees of Freedom: 286 Total (i.e. Null); 281 Residual
// Null Deviance: 1518
// Residual Deviance: 1465 AIC: NA
// Call: glm(formula = Infections ~ ., family = tweedie(2), data = D)
//
// Coefficients:
// (Intercept) SwimmerOccas LocationNonBeach Age20-24 Age25-29 SexMale
// 1.10230 -0.43751 -0.36337 0.24318 0.11830 0.06467
//
// Degrees of Freedom: 286 Total (i.e. Null); 281 Residual
// Null Deviance: 964.4
// Residual Deviance: 915.7 AIC: NA
// ---------------------------------------------------------------------------------------------------------------------------
String [] cfs1 = new String [] { "Intercept", "Swimmer.Occas", "Location.NonBeach", "Age.20-24", "Age.25-29", "Sex.Male" };
double [][] vals = new double[][] {{ 0.89100, 0.82210, 0.72660, -0.50330, -0.26790, -0.10560 },
{ -0.12261, 0.61149, 0.53454, -0.37442, -0.18973, -0.08985 },
{ 1.02964, -0.14079, -0.12200, 0.08502, 0.04269, 0.02105 },
{ 1.05665, -0.25891, -0.22185, 0.15325, 0.07624, 0.03908 },
{ 1.08076, -0.35690, -0.30154, 0.20556, 0.10122, 0.05375 },
{ 1.10230, -0.43751, -0.36337, 0.24318, 0.11830, 0.06467 },
};
int dof = 286, res_dof = 281;
double [] nullDev = new double[]{1564,824.5,834.2,967.0,1518,964.4};
double [] resDev = new double[]{1469,755.4,770.8,908.9,1465,915.7};
double [] varPow = new double[]{ 0, 1.0, 1.25, 1.5,1.75, 2.0};
GLMParameters parms = new GLMParameters(Family.tweedie);
parms._train = _earinf._key;
parms._ignored_columns = new String[]{};
// "response_column":"Claims","offset_column":"logInsured"
parms._response_column = "Infections";
parms._standardize = false;
parms._lambda = new double[]{0};
parms._alpha = new double[]{0};
parms._gradient_epsilon = 1e-10;
parms._max_iterations = 1000;
parms._objective_epsilon = 0;
parms._beta_epsilon = 1e-6;
for(int x = 0; x < varPow.length; ++x) {
double p = varPow[x];
parms._tweedie_variance_power = p;
parms._tweedie_link_power = 1 - p;
for (Solver s : /*new Solver[]{Solver.IRLSM}*/ GLMParameters.Solver.values()) {
if(s == Solver.COORDINATE_DESCENT_NAIVE) continue; // ignore for now, has trouble with zero columns
try {
parms._solver = s;
model = new GLM(parms).trainModel().get();
HashMap<String, Double> coefs = model.coefficients();
System.out.println("coefs = " + coefs);
for (int i = 0; i < cfs1.length; ++i)
assertEquals(vals[x][i], coefs.get(cfs1[i]), 1e-4);
assertEquals(nullDev[x], (GLMTest.nullDeviance(model)), 5e-4*nullDev[x]);
assertEquals(resDev[x], (GLMTest.residualDeviance(model)), 5e-4*resDev[x]);
assertEquals(dof, GLMTest.nullDOF(model), 0);
assertEquals(res_dof, GLMTest.resDOF(model), 0);
// test scoring
scoreTrain = model.score(_earinf);
assertTrue(model.testJavaScoring(_earinf,scoreTrain,1e-8));
hex.ModelMetricsRegressionGLM mmTrain = (ModelMetricsRegressionGLM) hex.ModelMetricsRegression.getFromDKV(model, _earinf);
assertEquals(model._output._training_metrics._MSE, mmTrain._MSE, 1e-8);
assertEquals(GLMTest.residualDeviance(model), mmTrain._resDev, 1e-8);
assertEquals(GLMTest.nullDeviance(model), mmTrain._nullDev, 1e-8);
} finally {
if (model != null) model.delete();
if (scoreTrain != null) scoreTrain.delete();
}
}
}
}
@Test
public void testPoissonWithOffset(){
GLMModel model = null;
Frame scoreTrain = null;
// Call: glm(formula = formula, family = poisson, data = D)
//
// Coefficients:
// (Intercept) Merit1 Merit2 Merit3 Class2 Class3 Class4 Class5
// -2.0357 -0.1378 -0.2207 -0.4930 0.2998 0.4691 0.5259 0.2156
//
// Degrees of Freedom: 19 Total (i.e. Null); 12 Residual
// Null Deviance: 33850
// Residual Deviance: 579.5 AIC: 805.9
String [] cfs1 = new String [] { "Intercept", "Merit.1", "Merit.2", "Merit.3", "Class.2", "Class.3", "Class.4", "Class.5"};
double [] vals = new double [] { -2.0357, -0.1378, -0.2207, -0.4930, 0.2998, 0.4691, 0.5259, 0.2156};
GLMParameters parms = new GLMParameters(Family.poisson);
parms._train = _canCarTrain._key;
parms._ignored_columns = new String[]{"Insured", "Premium", "Cost"};
// "response_column":"Claims","offset_column":"logInsured"
parms._response_column = "Claims";
parms._offset_column = "logInsured";
parms._standardize = false;
parms._lambda = new double[]{0};
parms._alpha = new double[]{0};
parms._objective_epsilon = 0;
parms._beta_epsilon = 1e-6;
parms._gradient_epsilon = 1e-10;
parms._max_iterations = 1000;
for (Solver s : GLMParameters.Solver.values()) {
if(s == Solver.COORDINATE_DESCENT_NAIVE) continue; // skip for now, does not handle zero columns (introduced by extra missing bucket with no missing in the dataset)
try {
parms._solver = s;
model = new GLM(parms).trainModel().get();
HashMap<String, Double> coefs = model.coefficients();
System.out.println("coefs = " + coefs);
for (int i = 0; i < cfs1.length; ++i)
assertEquals(vals[i], coefs.get(cfs1[i]), 1e-4);
assertEquals(33850, GLMTest.nullDeviance(model), 5);
assertEquals(579.5, GLMTest.residualDeviance(model), 1e-4*579.5);
assertEquals(19, GLMTest.nullDOF(model), 0);
assertEquals(12, GLMTest.resDOF(model), 0);
assertEquals(805.9, GLMTest.aic(model), 1e-4*805.9);
// test scoring
try {
Frame fr = new Frame(_canCarTrain.names(),_canCarTrain.vecs());
fr.remove(parms._offset_column);
scoreTrain = model.score(fr);
assertTrue("shoul've thrown IAE", false);
} catch (IllegalArgumentException iae) {
assertTrue(iae.getMessage().contains("Test/Validation dataset is missing offset column"));
}
scoreTrain = model.score(_canCarTrain);
hex.ModelMetricsRegressionGLM mmTrain = (ModelMetricsRegressionGLM)hex.ModelMetricsRegression.getFromDKV(model, _canCarTrain);
assertEquals(model._output._training_metrics._MSE, mmTrain._MSE, 1e-8);
assertEquals(GLMTest.residualDeviance(model), mmTrain._resDev, 1e-8);
assertEquals(GLMTest.nullDeviance(model), mmTrain._nullDev, 1e-8);
} finally {
if(model != null) model.delete();
if(scoreTrain != null) scoreTrain.delete();
}
}
}
static double [] tweedie_se_fit = new double[]{
0.0613489490790154,0.0613489490790154,0.0613489490790154,0.0613489490790154,0.0613489490790154,0.0613489490790154,0.0613489490790154,0.0613489490790154,0.0613489490790154,0.0613489490790154,0.0613489490790154,0.0613489490790154,0.0613489490790154,0.0613489490790154,0.0613489490790154,0.0613489490790154,0.0613489490790154,0.0613489490790154,0.0613489490790154,0.0613489490790154,0.0613489490790154,0.0613489490790154,0.0613489490790154,0.0613489490790154,0.0613489490790154,0.0613489490790154,0.0613489490790154,0.0613489490790154,0.0613489490790154,0.0613489490790154,0.0613489490790154,0.0925127769415089,0.0925127769415089,0.0925127769415089,0.0925127769415089,0.0987894311775416,0.0987894311775416,0.0987894311775416,0.0987894311775416,0.0987894311775416,0.0987894311775416,0.0987894311775416,0.0987894311775416,0.0987894311775416,0.0987894311775416,0.0987894311775416,0.0987894311775416,0.100083115466028,0.100083115466028,0.100083115466028,0.100083115466028,0.100083115466028,0.100083115466028,0.100083115466028,0.100083115466028,0.100083115466028,0.100083115466028,0.100083115466028,0.100884077314708,0.100884077314708,0.100884077314708,0.100884077314708,0.100884077314708,0.100884077314708,0.100884077314708,0.115835959352225,0.115835959352225,0.115835959352225,0.115835959352225,0.0841383955582187,0.0841383955582187,0.0841383955582187,0.0841383955582187,0.0841383955582187,0.0841383955582187,0.0841383955582187,0.0841383955582187,0.0841383955582187,0.0841383955582187,0.0841383955582187,0.0841383955582187,0.0841383955582187,0.0841383955582187,0.0841383955582187,0.0841383955582187,0.0841383955582187,0.0841383955582187,0.0841383955582187,0.0841383955582187,0.0841383955582187,0.0841383955582187,0.0841383955582187,0.0841383955582187,0.0841383955582187,0.0841383955582187,0.0841383955582187,0.0841383955582187,0.0841383955582187,0.0841383955582187,0.0841383955582187,0.0841383955582187,0.110599707082871,0.110599707082871,0.110599707082871,0.110599707082871,0.111858985562116,0.111858985562116,0.111858985562116,0.111858985562116,0.111858985562116,0.111858985562116,0.111858985562116,0.111858985562116,0.111858985562116,0.111858985562116,0.111858985562116,0.111858985562116,0.111858985562116,0.114576682598884,0.114576682598884,0.114576682598884,0.114576682598884,0.114576682598884,0.114576682598884,0.114576682598884,0.114576682598884,0.114576682598884,0.114576682598884,0.115282471068922,0.115282471068922,0.115282471068922,0.115282471068922,0.115282471068922,0.115282471068922,0.115282471068922,0.115282471068922,0.129955861024206,0.129955861024206,0.129955861024206,0.129955861024206,0.0858288225981346,0.0858288225981346,0.0858288225981346,0.0858288225981346,0.0858288225981346,0.0858288225981346,0.0858288225981346,0.0858288225981346,0.0858288225981346,0.0858288225981346,0.0858288225981346,0.0858288225981346,0.0858288225981346,0.0858288225981346,0.0858288225981346,0.0858288225981346,0.0858288225981346,0.0858288225981346,0.0858288225981346,0.0858288225981346,0.0959405152884533,0.0959405152884533,0.0959405152884533,0.0959405152884533,0.0959405152884533,0.0959405152884533,0.0959405152884533,0.0959405152884533,0.0959405152884533,0.0959405152884533,0.0959405152884533,0.0959405152884533,0.0959405152884533,0.0959405152884533,0.0959405152884533,0.121964163974085,0.121964163974085,0.121964163974085,0.121964163974085,0.121964163974085,0.121964163974085,0.121964163974085,0.110343150778848,0.110343150778848,0.110343150778848,0.110343150778848,0.110343150778848,0.110343150778848,0.110343150778848,0.110343150778848,0.110343150778848,0.108157177224978,0.108157177224978,0.108157177224978,0.108157177224978,0.108157177224978,0.108157177224978,0.108157177224978,0.108157177224978,0.108157177224978,0.108157177224978,0.108157177224978,0.108157177224978,0.108157177224978,0.108157177224978,0.108157177224978,0.108157177224978,0.109459685499218,0.109459685499218,0.109459685499218,0.109459685499218,0.109459685499218,0.109459685499218,0.109459685499218,0.109459685499218,0.100845471768361,0.100845471768361,0.100845471768361,0.100845471768361,0.100845471768361,0.100845471768361,0.100845471768361,0.100845471768361,0.100845471768361,0.100845471768361,0.100845471768361,0.100845471768361,0.100845471768361,0.100845471768361,0.100845471768361,0.100845471768361,0.100845471768361,0.100845471768361,0.100845471768361,0.100845471768361,0.111202113814401,0.111202113814401,0.111202113814401,0.111202113814401,0.111202113814401,0.111202113814401,0.111202113814401,0.111202113814401,0.111202113814401,0.111202113814401,0.111202113814401,0.111202113814401,0.111202113814401,0.111202113814401,0.130828072545958,0.130828072545958,0.130828072545958,0.130828072545958,0.130828072545958,0.130828072545958,0.130828072545958,0.121550168454726,0.121550168454726,0.121550168454726,0.121550168454726,0.121550168454726,0.121550168454726,0.121550168454726,0.121550168454726,0.121550168454726,0.121550168454726,0.119574547454639,0.119574547454639,0.119574547454639,0.119574547454639,0.119574547454639,0.119574547454639,0.119574547454639,0.119574547454639,0.119574547454639,0.119574547454639,0.119574547454639,0.119574547454639,0.119574547454639,0.119574547454639,0.119574547454639,0.122227760425649,0.122227760425649,0.122227760425649,0.122227760425649,0.122227760425649,0.122227760425649
};
@Test
public void testPValuesTweedie() {
// Call:
// glm(formula = Infections ~ ., family = tweedie(var.power = 1.5),
// data = D)
//
// Deviance Residuals:
// Min 1Q Median 3Q Max
// -2.6355 -2.0931 -1.8183 0.5046 4.9458
//
// Coefficients:
// Estimate Std. Error t value Pr(>|t|)
// (Intercept) 1.05665 0.11120 9.502 < 2e-16 ***
// SwimmerOccas -0.25891 0.08455 -3.062 0.00241 **
// LocationNonBeach -0.22185 0.08393 -2.643 0.00867 **
// Age20-24 0.15325 0.10041 1.526 0.12808
// Age25-29 0.07624 0.10099 0.755 0.45096
// SexMale 0.03908 0.08619 0.453 0.65058
// ---
// Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
//
// (Dispersion parameter for Tweedie family taken to be 2.896306)
//
// Null deviance: 967.05 on 286 degrees of freedom
// Residual deviance: 908.86 on 281 degrees of freedom
// AIC: NA
//
// Number of Fisher Scoring iterations: 7
double [] sderr_exp = new double[]{ 0.11120211, 0.08454967, 0.08393315, 0.10041150, 0.10099231, 0.08618960};
double [] zvals_exp = new double[]{ 9.5021062, -3.0622693, -2.6431794, 1.5262357, 0.7548661, 0.4534433};
double [] pvals_exp = new double[]{ 9.508400e-19, 2.409514e-03, 8.674149e-03, 1.280759e-01, 4.509615e-01, 6.505795e-01 };
GLMParameters parms = new GLMParameters(Family.tweedie);
parms._tweedie_variance_power = 1.5;
parms._tweedie_link_power = 1 - parms._tweedie_variance_power;
parms._train = _earinf._key;
parms._standardize = false;
parms._lambda = new double[]{0};
parms._alpha = new double[]{0};
parms._response_column = "Infections";
parms._compute_p_values = true;
parms._objective_epsilon = 0;
parms._missing_values_handling = DeepLearningModel.DeepLearningParameters.MissingValuesHandling.Skip;
GLMModel model = null;
Frame predict = null;
try {
model = new GLM(parms).trainModel().get();
String[] names_expected = new String[]{"Intercept", "Swimmer.Occas", "Location.NonBeach", "Age.20-24", "Age.25-29", "Sex.Male"};
String[] names_actual = model._output.coefficientNames();
HashMap<String, Integer> coefMap = new HashMap<>();
for (int i = 0; i < names_expected.length; ++i)
coefMap.put(names_expected[i], i);
double[] stder_actual = model._output.stdErr();
double[] zvals_actual = model._output.zValues();
double[] pvals_actual = model._output.pValues();
for (int i = 0; i < sderr_exp.length; ++i) {
int id = coefMap.get(names_actual[i]);
assertEquals(sderr_exp[id], stder_actual[i], sderr_exp[id] * 1e-3);
assertEquals(zvals_exp[id], zvals_actual[i], Math.abs(zvals_exp[id]) * 1e-3);
assertEquals(pvals_exp[id], pvals_actual[i], Math.max(1e-8,pvals_exp[id]) * 5e-3);
}
predict = model.score(parms._train.get());
Vec.Reader r = predict.vec("StdErr").new Reader();
for(int i = 0; i < 10; i++)
System.out.println(tweedie_se_fit[i] + " ?=? " + r.at(i));
for(int i = 0; i < tweedie_se_fit.length; ++i)
assertEquals(tweedie_se_fit[i],r.at(i),1e-4);
} finally {
if(model != null) model.delete();
if(predict != null) predict.delete();
}
}
static double [] poisson_se_fit = new double [] {
0.00214595071236062, 0.00743699599697046, 0.00543894401842774, 0.00655714683196705, 0.0110212478876686, 0.0075848798597348, 0.0145966442532301, 0.0119334418854485, 0.0119310044426751, 0.0206323555670128, 0.00651512689814114, 0.0126291877824898, 0.0101512423391255, 0.0125677132679544, 0.0177401092625854, 0.0050285508709862, 0.00984147775616493, 0.0100843481643067, 0.00920309580050661, 0.0135853678325585
};
@Test
public void testPValuesPoisson() {
// Coefficients:
// Estimate Std. Error z value Pr(>|z|)
// (Intercept) -1.279e+00 3.481e-01 -3.673 0.000239 ***
// Merit1 -1.498e-01 2.972e-02 -5.040 4.64e-07 ***
// Merit2 -2.364e-01 3.859e-02 -6.127 8.96e-10 ***
// Merit3 -3.197e-01 5.095e-02 -6.274 3.52e-10 ***
// Class2 6.899e-02 8.006e-02 0.862 0.388785
// Class3 2.892e-01 6.333e-02 4.566 4.97e-06 ***
// Class4 2.708e-01 4.911e-02 5.515 3.49e-08 ***
// Class5 -4.468e-02 1.048e-01 -0.427 0.669732
// Insured 1.617e-06 5.069e-07 3.191 0.001420 **
// Premium -3.630e-05 1.087e-05 -3.339 0.000840 ***
// Cost 2.021e-05 6.869e-06 2.943 0.003252 **
// logInsured 9.390e-01 2.622e-02 35.806 < 2e-16 ***
// ---
// Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
//
// (Dispersion parameter for poisson family taken to be 1)
//
// Null deviance: 961181.685 on 19 degrees of freedom
// Residual deviance: 42.671 on 8 degrees of freedom
// AIC: 277.08
double [] sderr_exp = new double[]{ 3.480733e-01, 2.972063e-02, 3.858825e-02, 5.095260e-02,8.005579e-02, 6.332867e-02, 4.910690e-02, 1.047531e-01, 5.068602e-07, 1.086939e-05, 6.869142e-06 ,2.622370e-02};
double [] zvals_exp = new double[]{ -3.6734577, -5.0404946, -6.1269397, -6.2739848, 0.8618220, 4.5662083 , 5.5148904, -0.4265158 , 3.1906387, -3.3392867, 2.9428291, 35.8061272 };
double [] pvals_exp = new double[]{ 2.392903e-04, 4.643302e-07, 8.958540e-10 , 3.519228e-10, 3.887855e-01 , 4.966252e-06, 3.489974e-08 , 6.697321e-01 , 1.419587e-03, 8.399383e-04, 3.252279e-03, 8.867127e-281};
GLMParameters parms = new GLMParameters(Family.poisson);
parms._train = _canCarTrain._key;
parms._standardize = false;
parms._lambda = new double[]{0};
parms._alpha = new double[]{0};
parms._response_column = "Claims";
parms._compute_p_values = true;
parms._objective_epsilon = 0;
parms._missing_values_handling = DeepLearningModel.DeepLearningParameters.MissingValuesHandling.Skip;
GLMModel model = null;
Frame predict = null;
try {
model = new GLM(parms).trainModel().get();
predict = model.score(parms._train.get());
String[] names_expected = new String[]{"Intercept", "Merit.1", "Merit.2", "Merit.3", "Class.2", "Class.3", "Class.4", "Class.5","Insured","Premium", "Cost", "logInsured" };
String[] names_actual = model._output.coefficientNames();
HashMap<String, Integer> coefMap = new HashMap<>();
for (int i = 0; i < names_expected.length; ++i)
coefMap.put(names_expected[i], i);
double[] stder_actual = model._output.stdErr();
double[] zvals_actual = model._output.zValues();
double[] pvals_actual = model._output.pValues();
for (int i = 0; i < sderr_exp.length; ++i) {
int id = coefMap.get(names_actual[i]);
assertEquals(sderr_exp[id], stder_actual[i], sderr_exp[id] * 1e-4);
assertEquals(zvals_exp[id], zvals_actual[i], Math.abs(zvals_exp[id]) * 1e-4);
assertEquals(pvals_exp[id], pvals_actual[i], Math.max(1e-15,pvals_exp[id] * 1e-3));
}
Vec.Reader r = predict.vec("StdErr").new Reader();
for(int i = 0; i < 10; i++)
System.out.println(fit_se[i] + " ?=? " + r.at(i));
for(int i = 0; i < poisson_se_fit.length; ++i)
assertEquals(poisson_se_fit[i],r.at(i),1e-4);
} finally {
if(model != null) model.delete();
if(predict != null) predict.delete();
}
}
double [] fit_se = new double[]{
0.0696804506329891, 0.101281071858829, 0.10408152157582, 0.111268172403792, 0.0864807746842875, 0.115236360340542, 0.12221727704254, 0.124866677872014, 0.128339261626057, 0.0979129217095552, 0.110152626457565, 0.0990293981783657, 0.0797922311345053, 0.0792463216677848, 0.111426861426166, 0.0738441273359852, 0.120203649843966, 0.121693896359032, 0.0761455129888811, 0.26100084308246, 0.0720456904900178, 0.081097918838208, 0.0789702474051714, 0.0808142534572709, 0.105472082060368, 0.0840482368559722, 0.063378670290797, 0.105606912248152, 0.100466329601425, 0.0612898868494135, 0.109616337415397, 0.0701794746336352, 0.0640797475112887, 0.0673121686325539, 0.0634803009724142, 0.0640736894596905, 0.136481170989144, 0.263714808298936, 0.108661073153972, 0.0679992878660712, 0.0692653698520285, 0.0578825801685848, 0.0692824549011659, 0.158918750046892, 0.105821946814859, 0.062539478239644, 0.0645559360159139, 0.0675850464571084, 0.0747554134125586, 0.0615564429388638, 0.0654697687695094, 0.0917602397221548, 0.0585224587976278, 0.0778560274291999, 0.0680261708103141, 0.0958827924243588, 0.058974124112217, 0.072913090525014, 0.0689795760272738, 0.0713170788962477, 0.065706257508678, 0.128042541188024, 0.0649749667059613, 0.0613806345806654, 0.0750782449165757, 0.100075191507371, 0.0690401878038698, 0.0663993405278943, 0.0722234100213727, 0.114421672443619, 0.110357013874037, 0.0642985002654091, 0.0671725856291289, 0.063523258944993, 0.0715597141096345, 0.0646566408141189, 0.0633033140683379, 0.0670491504275652, 0.0603642211488642, 0.0560144665111521, 0.0671727628266449, 0.0738384805508671, 0.1247199741748, 0.0554223809418321, 0.0650037579647878, 0.0727634600806498, 0.0575637383983063, 0.0616609512372853, 0.0682789218401665, 0.0966026797905161, 0.12463988175174, 0.108735909355295, 0.0640657895777542, 0.0691809892888932, 0.0805455198436419, 0.0723317376403597, 0.0782641712930961, 0.104008893620461, 0.0854140524746924, 0.0495807108524011,
0.0520203427103241, 0.0629693638202253, 0.054824519906118, 0.0664522679377852, 0.0709937504956703, 0.0522528199125061, 0.116792628883851, 0.127959068214529, 0.0588829864765987, 0.0938071273144982, 0.0638448982296692, 0.095474139348608, 0.0636920146973271, 0.102824928294982, 0.0546954905581237, 0.0957477006105716, 0.0516295701222635, 0.0679538008921464, 0.067911254988675, 0.11772719691146, 0.0626934169760874, 0.0755070350639548, 0.0581558616336498, 0.0873377370618371, 0.0654538358047351, 0.0693235931850606, 0.0962317603498954, 0.0552842877956681, 0.077459867942534, 0.0626998557114978, 0.0531665050182605, 0.0495451968026518, 0.0531904147974664, 0.0773863775170239, 0.0570467158542459, 0.0615088358357168, 0.0653655052453002, 0.0958225208725932, 0.0821004080317487, 0.0554118772903184, 0.062705388445474, 0.115252227824609, 0.0930756784532364, 0.0856558971929684, 0.0976473251692103, 0.0710701529636323, 0.050750991917379, 0.0564411256187975, 0.0775449777496427, 0.115494288850098,0.0682381145402218, 0.0515555125627838, 0.0670040023710206, 0.0712685707513018, 0.0532727639007648, 0.0546917068101745, 0.0717446129579534, 0.0801494525268998, 0.0472679272457015, 0.0730855772596969, 0.0656353433724242, 0.0670760966162116, 0.086126622468753, 0.0867455394873098, 0.117762705091036, 0.0552308514888129, 0.0567599016061833, 0.0761215691699384, 0.0699603827190508, 0.0611526828602172, 0.0665649473386548, 0.068400044275874, 0.052851970203728, 0.0947351046167158, 0.075626919466335, 0.0986954326552911, 0.158600667788559, 0.0997971513046435, 0.0558735275034329, 0.055050981781157, 0.0543870270651114, 0.0885427466948035, 0.0912282011735491, 0.0501764251426058, 0.065519936856806, 0.126597731978782, 0.0571871738429555, 0.0601312366784372, 0.134633469314707, 0.0636293392600048, 0.0822701720606341, 0.0998238113866312, 0.264894832552688, 0.0649884289638972, 0.0708677960760423, 0.0806790608308702, 0.101119270743047, 0.0550422649696084, 0.0648419030353994, 0.0558608594291732,
0.0526453344402306, 0.0610806341536575, 0.0609287420297426, 0.092034974197124, 0.0654686061501201, 0.0876869833195622, 0.0632671529428891, 0.0537964056797385, 0.0578936987690603, 0.0610938723761382, 0.0639353712535906, 0.168259497679141, 0.0962447000659448, 0.0651311057869981, 0.057491792983237, 0.0833792244616626, 0.086830040351315, 0.0774225857956364, 0.0664132437698603, 0.0574733178794812, 0.0647095391454638, 0.0608749267574728, 0.0534737593003958, 0.0864207446374423, 0.0630820817777617, 0.07226313326455, 0.0657499714305992, 0.121316445806293, 0.05853768423366, 0.0768645928103155, 0.0561109648914227, 0.0746288344339693, 0.0657484453780335, 0.0969340921119936, 0.0794439588324644, 0.0748828899207881, 0.100497037474176, 0.10675969143369, 0.0684810175839798, 0.0824837244664557, 0.0892814658999665, 0.0573638625958212, 0.0646309493356802, 0.0940910063257154, 0.0673435846353654, 0.0957497261909759, 0.0664402337808255, 0.0781546316899442, 0.0742122328375746, 0.0582089765051909,0.0781545857991108, 0.104152580875285, 0.0711435121130216, 0.0983829670734453, 0.0815684611863238, 0.102263743443002, 0.0936000092997729, 0.128533232616524, 0.0641557720833701, 0.111115887875877, 0.0638681893568514, 0.101074063878806, 0.06424466347809, 0.064441266436105, 0.110618016393452, 0.0712315373586064, 0.0657094575123701, 0.0705967310833688, 0.068439218729386, 0.103666086457174, 0.0787150533390872, 0.107851546439191, 0.142558987347935, 0.0756230725139849, 0.0812011758847381, 0.0710836161067677, 0.0662009215101577, 0.130219300771016, 0.0951028456739751, 0.0774634922652527, 0.100986990070013, 0.0810216431052252, 0.0836836265752558, 0.0897897867952456, 0.174853086617412, 0.0750505478534531, 0.105468755484224, 0.102115887997378, 0.102894682905793, 0.0651020673618454
};
@Test
public void testPValuesGaussian(){
// 1) NON-STANDARDIZED
// summary(m)
//
// Call:
// glm(formula = CAPSULE ~ ., family = gaussian, data = D)
//
// Deviance Residuals:
// Min 1Q Median 3Q Max
// -0.8394 -0.3162 -0.1113 0.3771 0.9447
//
// Coefficients:
// Estimate Std. Error t value Pr(>|t|)
// (Intercept) -0.6870832 0.4035941 -1.702 0.08980 .
// ID 0.0003081 0.0002387 1.291 0.19791
// AGE -0.0006005 0.0040246 -0.149 0.88150
// RACER2 -0.0147733 0.2511007 -0.059 0.95313
// RACER3 -0.1456993 0.2593492 -0.562 0.57471
// DPROSb 0.1462512 0.0657117 2.226 0.02684 *
// DPROSc 0.2297207 0.0713659 3.219 0.00144 **
// DPROSd 0.1144974 0.0937208 1.222 0.22286
// DCAPSb 0.1430945 0.0888124 1.611 0.10827
// PSA 0.0047237 0.0015060 3.137 0.00189 **
// VOL -0.0019401 0.0013920 -1.394 0.16449
// GLEASON 0.1438776 0.0273259 5.265 2.81e-07 ***
// ---
// Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
//
// (Dispersion parameter for gaussian family taken to be 0.1823264)
//
// Null deviance: 69.600 on 289 degrees of freedom
// Residual deviance: 50.687 on 278 degrees of freedom
// AIC: 343.16
//
// Number of Fisher Scoring iterations: 2
GLMParameters params = new GLMParameters(Family.gaussian);
params._response_column = "CAPSULE";
params._standardize = false;
params._train = _prostateTrain._key;
params._compute_p_values = true;
params._lambda = new double[]{0};
params._missing_values_handling = DeepLearningModel.DeepLearningParameters.MissingValuesHandling.Skip;
try {
params._solver = Solver.L_BFGS;
new GLM(params).trainModel().get();
assertFalse("should've thrown, p-values only supported with IRLSM",true);
} catch(H2OModelBuilderIllegalArgumentException t) {
}
try {
params._solver = Solver.COORDINATE_DESCENT_NAIVE;
new GLM(params).trainModel().get();
assertFalse("should've thrown, p-values only supported with IRLSM",true);
} catch(H2OModelBuilderIllegalArgumentException t) {
}
try {
params._solver = Solver.COORDINATE_DESCENT;
new GLM(params).trainModel().get();
assertFalse("should've thrown, p-values only supported with IRLSM",true);
} catch(H2OModelBuilderIllegalArgumentException t) {
}
params._solver = Solver.IRLSM;
GLM glm = new GLM(params);
try {
params._lambda = new double[]{1};
glm.trainModel().get();
assertFalse("should've thrown, p-values only supported with no regularization",true);
} catch(H2OModelBuilderIllegalArgumentException t) {
}
params._lambda = new double[]{0};
try {
params._lambda_search = true;
glm.trainModel().get();
assertFalse("should've thrown, p-values only supported with no regularization (i.e. no lambda search)",true);
} catch(H2OModelBuilderIllegalArgumentException t) {
}
params._lambda_search = false;
GLMModel model = null;
Frame predict = null;
try {
model = new GLM(params).trainModel().get();
String[] names_expected = new String[]{"Intercept", "ID", "AGE", "RACE.R2", "RACE.R3", "DPROS.b", "DPROS.c", "DPROS.d", "DCAPS.b", "PSA", "VOL", "GLEASON"};
double[] stder_expected = new double[]{0.4035941476, 0.0002387281, 0.0040245520, 0.2511007120, 0.2593492335, 0.0657117271, 0.0713659021, 0.0937207659, 0.0888124376, 0.0015060289, 0.0013919737, 0.0273258788};
double[] zvals_expected = new double[]{-1.70241133, 1.29061005, -0.14920829, -0.05883397, -0.56178799, 2.22564893, 3.21891333, 1.22168646, 1.61119882, 3.13650800, -1.39379859, 5.26524961 };
double[] pvals_expected = new double[]{8.979610e-02, 1.979113e-01, 8.814975e-01, 9.531266e-01, 5.747131e-01, 2.683977e-02, 1.439295e-03, 2.228612e-01, 1.082711e-01, 1.893210e-03, 1.644916e-01, 2.805776e-07};
String[] names_actual = model._output.coefficientNames();
HashMap<String, Integer> coefMap = new HashMap<>();
for (int i = 0; i < names_expected.length; ++i)
coefMap.put(names_expected[i], i);
double[] stder_actual = model._output.stdErr();
double[] zvals_actual = model._output.zValues();
double[] pvals_actual = model._output.pValues();
for (int i = 0; i < stder_expected.length; ++i) {
int id = coefMap.get(names_actual[i]);
assertEquals(stder_expected[id], stder_actual[i], stder_expected[id] * 1e-5);
assertEquals(zvals_expected[id], zvals_actual[i], Math.abs(zvals_expected[id]) * 1e-5);
assertEquals(pvals_expected[id], pvals_actual[i], pvals_expected[id] * 1e-3);
}
predict = model.score(params._train.get());
Vec.Reader r = predict.vec("StdErr").new Reader();
for(int i = 0; i < 10; i++)
System.out.println(fit_se[i] + " ?=? " + r.at(i));
for(int i = 0; i < fit_se.length; ++i)
assertEquals(fit_se[i],r.at(i),1e-4);
} finally {
if(model != null) model.delete();
if(predict != null) predict.delete();
}
// 2) STANDARDIZED
// Call:
// glm(formula = CAPSULE ~ ., family = binomial, data = Dstd)
//
// Deviance Residuals:
// Min 1Q Median 3Q Max
// -2.0601 -0.8079 -0.4491 0.8933 2.2877
//
// Coefficients:
// Estimate Std. Error z value Pr(>|z|)
// (Intercept) -1.28045 1.56879 -0.816 0.41438
// ID 0.19054 0.15341 1.242 0.21420
// AGE -0.02118 0.14498 -0.146 0.88384
// RACER2 0.06831 1.54240 0.044 0.96468
// RACER3 -0.74113 1.58272 -0.468 0.63959
// DPROSb 0.88833 0.39509 2.248 0.02455 *
// DPROSc 1.30594 0.41620 3.138 0.00170 **
// DPROSd 0.78440 0.54265 1.446 0.14832
// DCAPSb 0.61237 0.51796 1.182 0.23710
// PSA 0.60917 0.22447 2.714 0.00665 **
// VOL -0.18130 0.16204 -1.119 0.26320
// GLEASON 0.91751 0.19633 4.673 2.96e-06 ***
// ---
// Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
//
// (Dispersion parameter for binomial family taken to be 1)
//
// Null deviance: 390.35 on 289 degrees of freedom
// Residual deviance: 297.65 on 278 degrees of freedom
// AIC: 321.65
//
// Number of Fisher Scoring iterations: 5
// Estimate Std. Error z value Pr(>|z|)
// (Intercept) -1.28045434 1.5687858 -0.81620723 4.143816e-01
// ID 0.19054396 0.1534062 1.24208800 2.142041e-01
// AGE -0.02118315 0.1449847 -0.14610616 8.838376e-01
// RACER2 0.06830776 1.5423974 0.04428674 9.646758e-01
// RACER3 -0.74113331 1.5827190 -0.46826589 6.395945e-01
// DPROSb 0.88832948 0.3950883 2.24843259 2.454862e-02
// DPROSc 1.30594011 0.4161974 3.13779030 1.702266e-03
// DPROSd 0.78440312 0.5426512 1.44550154 1.483171e-01
// DCAPSb 0.61237150 0.5179591 1.18227779 2.370955e-01
// PSA 0.60917093 0.2244733 2.71377864 6.652060e-03
// VOL -0.18129997 0.1620383 -1.11887108 2.631951e-01
// GLEASON 0.91750972 0.1963285 4.67333842 2.963429e-06
params._standardize = true;
try {
model = new GLM(params).trainModel().get();
String[] names_expected = new String[]{"Intercept", "ID", "AGE", "RACE.R2", "RACE.R3", "DPROS.b", "DPROS.c", "DPROS.d", "DCAPS.b", "PSA", "VOL", "GLEASON"};
// do not compare std_err here, depends on the coefficients
// double[] stder_expected = new double[]{1.5687858, 0.1534062, 0.1449847, 1.5423974, 1.5827190, 0.3950883, 0.4161974, 0.5426512, 0.5179591, 0.2244733, 0.1620383, 0.1963285};
// double[] zvals_expected = new double[]{1.14158283, 1.29061005, -0.14920829, -0.05883397, -0.56178799, 2.22564893, 3.21891333, 1.22168646, 1.61119882, 3.13650800, -1.39379859, 5.26524961 };
// double[] pvals_expected = new double[]{2.546098e-01, 1.979113e-01, 8.814975e-01, 9.531266e-01, 5.747131e-01, 2.683977e-02, 1.439295e-03, 2.228612e-01, 1.082711e-01, 1.893210e-03, 1.644916e-01, 2.805776e-07 };
double[] stder_expected = new double[]{0.4035941476, 0.0002387281, 0.0040245520, 0.2511007120, 0.2593492335, 0.0657117271, 0.0713659021, 0.0937207659, 0.0888124376, 0.0015060289, 0.0013919737, 0.0273258788};
double[] zvals_expected = new double[]{-1.70241133, 1.29061005, -0.14920829, -0.05883397, -0.56178799, 2.22564893, 3.21891333, 1.22168646, 1.61119882, 3.13650800, -1.39379859, 5.26524961 };
double[] pvals_expected = new double[]{8.979610e-02, 1.979113e-01, 8.814975e-01, 9.531266e-01, 5.747131e-01, 2.683977e-02, 1.439295e-03, 2.228612e-01, 1.082711e-01, 1.893210e-03, 1.644916e-01, 2.805776e-07};
String[] names_actual = model._output.coefficientNames();
HashMap<String, Integer> coefMap = new HashMap<>();
for (int i = 0; i < names_expected.length; ++i)
coefMap.put(names_expected[i], i);
double[] zvals_actual = model._output.zValues();
double[] pvals_actual = model._output.pValues();
for (int i = 0; i < zvals_expected.length; ++i) {
int id = coefMap.get(names_actual[i]);
assertEquals(zvals_expected[id], zvals_actual[i], Math.abs(zvals_expected[id]) * 1e-5);
assertEquals(pvals_expected[id], pvals_actual[i], pvals_expected[id] * 1e-3);
}
predict = model.score(params._train.get());
Vec.Reader r = predict.vec("StdErr").new Reader();
for(int i = 0; i < 10; i++)
System.out.println(fit_se[i] + " ?=? " + r.at(i));
for(int i = 0; i < fit_se.length; ++i)
assertEquals(fit_se[i],r.at(i),1e-4);
} finally {
if(model != null) model.delete();
if(predict != null) predict.delete();
}
// Airlines (has collinear columns)
params._standardize = true;
params._remove_collinear_columns = true;
params._train = _airlines._key;
params._response_column = "IsDepDelayed";
params._ignored_columns = new String[]{"IsDepDelayed_REC"};
try {
model = new GLM(params).trainModel().get();
String[] names_expected = new String[] {"Intercept","fYearf1988","fYearf1989","fYearf1990","fYearf1991","fYearf1992","fYearf1993","fYearf1994","fYearf1995","fYearf1996","fYearf1997","fYearf1998","fYearf1999","fYearf2000","fDayofMonthf10","fDayofMonthf11","fDayofMonthf12","fDayofMonthf13","fDayofMonthf14","fDayofMonthf15","fDayofMonthf16","fDayofMonthf17","fDayofMonthf18","fDayofMonthf19","fDayofMonthf2","fDayofMonthf20","fDayofMonthf21","fDayofMonthf22","fDayofMonthf23","fDayofMonthf24", "fDayofMonthf25", "fDayofMonthf26", "fDayofMonthf27", "fDayofMonthf28", "fDayofMonthf29" , "fDayofMonthf3" , "fDayofMonthf30" , "fDayofMonthf31", "fDayofMonthf4", "fDayofMonthf5", "fDayofMonthf6", "fDayofMonthf7" , "fDayofMonthf8" , "fDayofMonthf9" , "fDayOfWeekf2" , "fDayOfWeekf3" , "fDayOfWeekf4" , "fDayOfWeekf5", "fDayOfWeekf6" , "fDayOfWeekf7" , "DepTime", "ArrTime", "UniqueCarrierCO", "UniqueCarrierDL", "UniqueCarrierHP", "UniqueCarrierPI", "UniqueCarrierTW" ,"UniqueCarrierUA" ,"UniqueCarrierUS", "UniqueCarrierWN" , "OriginABQ", "OriginACY", "OriginALB", "OriginATL", "OriginAUS", "OriginAVP", "OriginBDL", "OriginBGM", "OriginBHM", "OriginBNA", "OriginBOS", "OriginBTV", "OriginBUF", "OriginBUR", "OriginBWI", "OriginCAE", "OriginCHO", "OriginCHS", "OriginCLE", "OriginCLT", "OriginCMH", "OriginCOS", "OriginCRW", "OriginCVG", "OriginDAY", "OriginDCA", "OriginDEN", "OriginDFW", "OriginDSM", "OriginDTW", "OriginERI", "OriginEWR", "OriginFLL", "OriginGSO", "OriginHNL", "OriginIAD", "OriginIAH", "OriginICT", "OriginIND", "OriginISP", "OriginJAX", "OriginJFK", "OriginLAS", "OriginLAX", "OriginLEX", "OriginLGA", "OriginLIH", "OriginLYH", "OriginMCI", "OriginMCO", "OriginMDT", "OriginMDW", "OriginMFR", "OriginMHT", "OriginMIA", "OriginMKE", "OriginMLB", "OriginMRY", "OriginMSP", "OriginMSY", "OriginMYR", "OriginOAK", "OriginOGG", "OriginOMA", "OriginORD", "OriginORF", "OriginPBI", "OriginPHF", "OriginPHL", "OriginPHX", "OriginPIT", "OriginPSP", "OriginPVD", "OriginPWM", "OriginRDU", "OriginRIC", "OriginRNO", "OriginROA", "OriginROC", "OriginRSW", "OriginSAN", "OriginSBN", "OriginSCK", "OriginSDF", "OriginSEA", "OriginSFO", "OriginSJC", "OriginSJU", "OriginSLC", "OriginSMF", "OriginSNA", "OriginSRQ", "OriginSTL", "OriginSTX", "OriginSWF", "OriginSYR", "OriginTLH", "OriginTPA", "OriginTRI", "OriginTUS", "OriginTYS", "OriginUCA", "DestABQ", "DestACY", "DestALB", "DestATL", "DestAVP", "DestBDL", "DestBGM", "DestBNA", "DestBOS", "DestBTV", "DestBUF", "DestBUR", "DestBWI", "DestCAE", "DestCAK", "DestCHA", "DestCHS", "DestCLE", "DestCLT", "DestCMH", "DestDAY", "DestDCA", "DestDEN", "DestDFW", "DestDTW", "DestELM", "DestERI", "DestEWR", "DestFAT", "DestFAY", "DestFLL", "DestFNT", "DestGEG", "DestGRR", "DestGSO", "DestGSP", "DestHNL", "DestHTS", "DestIAD", "DestIAH", "DestICT", "DestIND", "DestISP", "DestJAX", "DestJFK", "DestKOA", "DestLAS", "DestLAX", "DestLEX", "DestLGA", "DestLIH", "DestLYH", "DestMCI", "DestMCO", "DestMDT", "DestMDW", "DestMHT", "DestMIA", "DestMRY", "DestMSY", "DestOAJ", "DestOAK", "DestOGG", "DestOMA", "DestORD", "DestORF", "DestORH", "DestPBI", "DestPDX", "DestPHF", "DestPHL", "DestPHX", "DestPIT", "DestPSP", "DestPVD", "DestRDU", "DestRIC", "DestRNO", "DestROA", "DestROC", "DestRSW", "DestSAN", "DestSCK", "DestSDF", "DestSEA", "DestSFO", "DestSJC", "DestSMF", "DestSNA", "DestSTL", "DestSWF", "DestSYR", "DestTOL", "DestTPA", "DestTUS", "DestUCA", "Distance"};
double[] exp_coefs = new double[] {3.383044e-01,-1.168214e-01,-4.405621e-01,-3.365341e-01,-4.925256e-01,-5.374542e-01,-4.149143e-01,-2.694969e-01,-2.991095e-01,-2.776553e-01,-2.921466e-01,-4.336252e-01
,-3.597812e-01,-3.812643e-01,1.024025e-02,2.549787e-02,3.877628e-02,1.650942e-02,-2.981043e-02,-1.167855e-02,1.025499e-02,-4.574083e-03,-2.502898e-02,-5.803535e-02
,7.679039e-02,-5.247306e-02,-5.918685e-02,-3.339667e-02,-2.885718e-02,-4.225694e-02,-7.500997e-02,-5.145179e-02,-7.093373e-02,-5.634115e-02,-3.643811e-02,1.284665e-01
,-8.150175e-02,-4.724434e-02,1.511024e-01,5.498057e-02,4.411630e-02,1.278961e-02,7.276038e-03,4.672048e-02,-2.128594e-02,1.629933e-02,3.721499e-02,5.933446e-02
,-2.303705e-02,1.141451e-02,1.258241e-04,1.271866e-05,7.155502e-02,1.444990e-01,-8.685535e-02,-2.602512e-02,4.227022e-01,2.639493e-01,2.600565e-01,5.409442e-02
,5.106308e-02,-1.993041e-01,5.663324e-01,2.524168e-01,-8.032071e-02,1.959854e-02,3.110741e-01,2.711911e-01,-1.480432e-01,2.711969e-02,1.298365e-01,3.051547e-01
,1.747017e-01,-6.282101e-03,1.542743e-01,-3.037726e-01,3.808392e-01,1.829607e-01,4.841763e-02,9.353007e-02,2.154611e-01,6.469679e-02,-1.950998e-01,7.957484e-02
,2.430247e-01,1.942201e-02,5.701321e-02,2.770389e-01,1.497383e-01,4.943089e-02,2.598871e-01,5.930680e-02,3.748394e-01,4.204685e-02,-3.574776e-01,2.153817e-02
,-1.719974e-01,4.806820e-01,2.678204e-01,4.266956e-02,6.340217e-02,-1.536324e-02,-1.294344e-02,1.985872e-01,4.831069e-01,2.726364e-01,-4.813763e-01,4.199029e-01
,3.054954e-01,1.784330e-01,-2.500409e-02,2.978489e-03,-9.356699e-02,1.246280e-01,2.858306e-01,-6.533971e-02,-1.403327e-01,-3.924693e-01,5.947271e-02,-7.903152e-03
,-2.135489e-01,-1.454085e-01,-2.049959e-01,1.704250e-01,1.826566e-01,1.896976e-01,2.541375e-01,-9.746707e-02,1.990703e-01,9.068512e-02,2.848977e-01,3.409567e-01
,8.689141e-02,-6.294297e-02,2.402344e-02,9.583028e-02,4.207585e-01,2.096370e-01,2.184863e-01,1.316822e-01,4.863172e-02,4.918303e-01,-7.990361e-02,-4.499847e-02
,6.140887e-02,7.329919e-02,-1.658663e-01,1.850334e-01,-2.165094e-01,-1.054388e-01,8.943775e-02,3.809166e-01,-9.766444e-02,2.645371e-01,-5.147078e-02,2.323637e-01
,-3.746418e-01,1.841517e-01,-2.121584e-01,-1.888144e-02,-8.009574e-02,1.801828e-01,1.216036e-01,4.123190e-03,-4.747419e-02,-1.001471e-01,3.611426e-02,1.427218e-01
,-1.154052e-01,-2.388724e-01,-8.097489e-03,-3.321890e-02,-8.470654e-02,8.609431e-03,2.278746e-02,2.959335e-01,-8.363623e-02,-1.736324e-01,2.140292e-01,-1.252043e-01
,2.086573e-02,7.549936e-02,-2.339204e-01,1.009014e-01,1.396302e-01,-2.180753e-01,-1.118935e-02,-3.345582e-01,-1.490167e-01,-5.455654e-03,-2.884281e-02,-7.778542e-02
,1.481921e-01,-9.387787e-02,2.894362e-01,-2.599589e-01,1.210906e-01,1.721670e-02,6.271491e-02,-5.077020e-01,2.524418e-01,-1.146321e-01,-3.418030e-01,-7.056448e-03
,-1.948121e-01,-1.716377e-01,-5.915873e-02,3.465761e-01,-3.964155e-02,9.297146e-02,6.840982e-02,-2.694979e-02,3.489802e-01,4.473631e-01,9.045849e-02,1.195621e-01
,8.137467e-04,-8.754947e-02,2.089706e-02,2.676953e-03,-1.381342e-01,5.200934e-02,2.208028e-01,-1.096369e-01,4.753661e-01,2.876296e-02,2.256874e-02,-9.231270e-02
,2.507403e-02,1.529442e-01,-2.173190e-02,-1.180872e-01,-3.305849e-02,1.091687e-01,9.174085e-02,-6.172636e-02,5.983764e-02,1.094581e-01,1.537772e-01,1.117601e-01
,-9.674298e-02,3.111324e-02,1.404767e-01,-4.243193e-03,9.218955e-02,2.554272e-01,-4.434348e-02,1.222306e-01,1.960349e-02,1.308767e-01,-2.830042e-03,-3.212863e-02
,-1.035897e-01,-2.828326e-02,-2.452788e-01,5.876054e-02,6.094385e-02,-6.242541e-02,5.535717e-05};
double[] stder_expected = new double[]{8.262325e-02,1.960654e-02,5.784259e-02,5.211346e-02,5.351436e-02,5.364119e-02,5.377681e-02,5.361611e-02,5.480210e-02,5.916530e-02,5.924352e-02,5.947477e-02,5.684859e-02
,6.015367e-02,2.359873e-02,2.364261e-02,2.366028e-02,2.346965e-02,2.331776e-02,2.348358e-02,2.366537e-02,2.371736e-02,2.353753e-02,2.345702e-02,2.360676e-02,2.353096e-02
,2.352809e-02,2.354292e-02,2.381824e-02,2.360087e-02,2.357901e-02,2.352439e-02,2.333820e-02,2.348150e-02,2.349408e-02,2.388143e-02,2.363605e-02,2.369714e-02,2.384589e-02
,2.360301e-02,2.346261e-02,2.365805e-02,2.377684e-02,2.374369e-02,1.093338e-02,1.091722e-02,1.094858e-02,1.089616e-02,1.127837e-02,1.099223e-02,1.243150e-05,1.193431e-05
,6.185154e-02,5.842257e-02,4.797840e-02,4.082146e-02,6.764477e-02,4.904281e-02,4.661126e-02,4.949252e-02,7.194630e-02,1.080608e-01,1.000542e-01,7.206225e-02,6.866783e-02
,9.183712e-02,8.937756e-02,9.509039e-02,1.101394e-01,7.333840e-02,6.976195e-02,1.139758e-01,7.902871e-02,6.688118e-02,6.842836e-02,1.228471e-01,1.290408e-01,8.980176e-02
,6.808851e-02,7.095243e-02,6.932701e-02,7.036599e-02,1.021726e-01,7.566290e-02,7.743516e-02,7.012655e-02,6.722331e-02,7.756484e-02,2.146603e-01,8.390956e-02,1.138773e-01
,6.896196e-02,8.394126e-02,7.983643e-02,8.101956e-02,8.960544e-02,8.278554e-02,2.417453e-01,6.988129e-02,1.085592e-01,9.274580e-02,1.206031e-01,7.400875e-02,6.750358e-02
,1.107047e-01,6.957462e-02,1.139873e-01,1.340117e-01,7.976223e-02,6.979235e-02,7.837532e-02,1.285433e-01,1.334371e-01,1.198966e-01,8.332708e-02,1.229658e-01,1.149044e-01
,1.130423e-01,1.090638e-01,8.406530e-02,9.600642e-02,7.247142e-02,1.140837e-01,9.506082e-02,6.926602e-02,7.590418e-02,7.459985e-02,1.287070e-01,6.815592e-02,7.411458e-02
,6.592406e-02,9.179115e-02,7.223151e-02,7.670526e-02,7.764917e-02,7.343286e-02,1.999711e-01,1.175572e-01,7.108214e-02,7.409246e-02,6.847739e-02,2.476394e-01,1.080218e-01
,1.120317e-01,8.137946e-02,6.754660e-02,7.897969e-02,7.867300e-02,1.044366e-01,8.260141e-02,7.542126e-02,1.116638e-01,7.481728e-02,1.126226e-01,1.286945e-01,7.009628e-02
,1.346972e-01,6.941736e-02,1.228611e-01,7.884636e-02,1.089254e-01,1.178960e-01,6.487494e-02,1.141428e-01,6.337383e-02,1.044082e-01,9.881149e-02,6.748862e-02,7.802332e-02
,7.989152e-02,4.877654e-02,8.606809e-02,6.446482e-02,5.276630e-02,5.072148e-02,1.073048e-01,1.054882e-01,2.695275e-01,8.023848e-02,5.665850e-02,5.273383e-02,6.096450e-02
,7.907020e-02,5.261070e-02,5.180430e-02,1.142093e-01,5.580208e-02,2.354317e-01,2.681434e-01,5.047968e-02,1.029695e-01,7.947606e-02,6.167620e-02,1.260100e-01,1.094464e-01
,1.044411e-01,6.861138e-02,1.122694e-01,6.168966e-02,1.033369e-01,9.571271e-02,5.958640e-02,1.168745e-01,4.831583e-02,7.683862e-02,7.909215e-02,8.397850e-02,1.069573e-01
,5.494288e-02,4.744649e-02,2.133179e-01,5.407477e-02,1.070343e-01,1.207816e-01,5.898603e-02,5.647888e-02,1.076070e-01,7.977657e-02,2.690687e-01,1.077435e-01,3.279724e-01
,1.140342e-01,1.154527e-01,5.419787e-02,1.098867e-01,1.049436e-01,5.082173e-02,6.118521e-02,2.107675e-01,7.758130e-02,7.001571e-02,1.073186e-01,4.963340e-02,5.394587e-02
,4.612111e-02,7.909675e-02,7.081853e-02,7.685204e-02,1.132175e-01,6.811432e-02,1.231347e-01,7.004574e-02,1.089064e-01,5.191893e-02,2.689951e-01,3.267575e-01,1.008663e-01
,4.802894e-02,6.230837e-02,1.109208e-01,6.627911e-02,8.130255e-02,1.094653e-01,5.568541e-02,9.874917e-02,5.701293e-02,7.421695e-02,1.393040e-01,8.828166e-06};
double[] zvals_expected = new double[]{4.094542787,-5.958287216,-7.616568859,-6.457719779,-9.203616729,-10.019431514,-7.715486715,-5.026416071,-5.457993778,-4.692873330,-4.931283164,-7.290910329
,-6.328761834,-6.338172537,0.433932435,1.078470756,1.638876645,0.703437006,-1.278442927,-0.497307097,0.433333243,-0.192857977,-1.063364622,-2.474114538
,3.252898021,-2.229957876,-2.515581926,-1.418544300,-1.211558333,-1.790482149,-3.181217772,-2.187168299,-3.039383181,-2.399384279,-1.550948532,5.379347430
,-3.448197372,-1.993672779,6.336619935,2.329387922,1.880280495,0.540602946,0.306013647,1.967700936,-1.946876164,1.492992904,3.399070000,5.445447763
,-2.042586155,1.038415698,10.121391544,1.065722732,1.156883305,2.473342141,-1.810301011,-0.637535148,6.248852336,5.382017137,5.579262483,1.092981554
,0.709738700,-1.844369286,5.660253498,3.502760590,-1.169699226,0.213405474,3.480449353,2.851929432,-1.344144285,0.369788430,1.861136098,2.677364258
,2.210610981,-0.093929275,2.254537133,-2.472769164,2.951309213,2.037384116,0.711098412,1.318208131,3.107896103,0.919432700,-1.909512146,1.051702242
,3.138428148,0.276956565,0.848116635,3.571707066,0.697559521,0.589097196,2.282167858,0.859992866,4.465496733,0.526662485,-4.412237900,0.240366792
,-2.077626172,1.988382134,3.832504877,0.393053401,0.683612236,-0.127386732,-0.174890682,2.941876192,4.363924639,3.918618005,-4.223069230,3.133329974
,3.830075787,2.556626718,-0.319030122,0.023171093,-0.701206599,1.039462023,3.430224574,-0.531365033,-1.221299087,-3.471879279,0.545302258,-0.094012061
,-2.224318698,-2.006425249,-1.796891224,1.792799985,2.637030567,2.499172484,3.406675572,-0.757278429,2.920806636,1.223580021,4.321604030,3.714483644
,1.202957243,-0.820582100,0.309384395,1.305005410,2.104096487,1.783276250,3.073716393,1.777268003,0.710186572,1.986074319,-0.739698581,-0.401658318
,0.754599117,1.085164836,-2.100113836,2.351929605,-2.073117895,-1.276477554,1.185842715,3.411279361,-1.305372693,2.348881190,-0.399945460,3.314921843
,-2.781362443,2.652819094,-1.726814586,-0.239471355,-0.735326303,1.528320564,1.874431642,0.036123072,-0.749113504,-0.959187862,0.365486433,2.114753848
,-1.479111228,-2.989960091,-0.166011968,-0.385960754,-1.313996380,0.163161548,0.449266356,2.757878211,-0.792848866,-0.644210444,2.667413046,-2.209806237
,0.395680070,1.238415063,-2.958388604,1.917886602,2.695339429,-1.909435881,-0.200518466,-1.421041172,-0.555735022,-0.108076244,-0.280110165,-0.978727707
,2.402743720,-0.745003303,2.644547495,-2.489047558,1.764876031,0.153351678,1.016619439,-4.913074631,2.637495147,-1.923795851,-2.924531170,-0.146048350
,-2.535340379,-2.170097247,-0.704450917,3.240322900,-0.721504747,1.959501429,0.320694226,-0.498380144,3.260451771,3.703901572,1.533557842,2.116934588
,0.007562213,-1.097433427,0.077664399,0.024845610,-0.421176265,0.456085590,1.912495640,-2.022900342,4.325966158,0.274080250,0.444076651,-1.508741999
,0.118965357,1.971405595,-0.310386063,-1.100341487,-0.666053319,2.023671705,1.989128987,-0.780390564,0.844943233,1.424270602,1.358245515,1.640771855
,-0.785668134,0.444184574,1.289884997,-0.081727285,0.342718282,0.781702452,-0.439626096,2.544935912,0.314620525,1.179911368,-0.042698859,-0.395173744
,-0.946324367,-0.507911593,-2.483856451,1.030652857,0.821158118,-0.448123713,6.270517743
};
double[] pvals_expected = new double[]{4.243779e-05,2.584251e-09,2.700448e-14,1.083124e-10,3.733573e-20,1.392306e-23,1.251677e-14,5.032991e-07,4.862783e-08,2.708701e-06,8.223295e-07,3.173337e-13,2.514741e-10
,2.366114e-10,6.643414e-01,2.808345e-01,1.012520e-01,4.817902e-01,2.011056e-01,6.189770e-01,6.647766e-01,8.470718e-01,2.876273e-01,1.336350e-02,1.143912e-03,2.575939e-02
,1.189004e-02,1.560448e-01,2.256933e-01,7.338895e-02,1.468427e-03,2.873979e-02,2.373166e-03,1.643019e-02,1.209271e-01,7.545084e-08,5.653029e-04,4.619905e-02,2.390030e-10
,1.984672e-02,6.008188e-02,5.887863e-01,7.595969e-01,4.911388e-02,5.156115e-02,1.354521e-01,6.772461e-04,5.217869e-08,4.110425e-02,2.990870e-01,4.958814e-24,2.865596e-01
,2.473315e-01,1.339242e-02,7.026154e-02,5.237824e-01,4.203831e-10,7.434137e-08,2.441273e-08,2.744128e-01,4.778730e-01,6.514157e-02,1.528620e-08,4.612949e-04,2.421336e-01
,8.310125e-01,5.014564e-04,4.349160e-03,1.789144e-01,7.115434e-01,6.273710e-02,7.425404e-03,2.707212e-02,9.251661e-01,2.417131e-02,1.341390e-02,3.167338e-03,4.162244e-02
,4.770301e-01,1.874465e-01,1.886429e-03,3.578785e-01,5.620789e-02,2.929467e-01,1.700608e-03,7.818158e-01,3.963814e-01,3.553510e-04,4.854594e-01,5.558016e-01,2.248808e-02
,3.898015e-01,8.024485e-06,5.984328e-01,1.027500e-05,8.100480e-01,3.775434e-02,4.678071e-02,1.271662e-04,6.942835e-01,4.942266e-01,8.986354e-01,8.611670e-01,3.265400e-03
,1.282792e-05,8.930334e-05,2.418793e-05,1.730416e-03,1.284274e-04,1.057531e-02,7.497064e-01,9.815140e-01,4.831808e-01,2.986003e-01,6.040895e-04,5.951707e-01,2.219847e-01
,5.177320e-04,5.855507e-01,9.251004e-01,2.613621e-02,4.482202e-02,7.236537e-02,7.301742e-02,8.368881e-03,1.245495e-02,6.586634e-04,4.488905e-01,3.494493e-03,2.211226e-01
,1.555177e-05,2.040773e-04,2.290047e-01,4.118924e-01,7.570318e-01,1.919034e-01,3.538034e-02,7.455389e-02,2.116458e-03,7.553673e-02,4.775953e-01,4.703635e-02,4.594901e-01
,6.879391e-01,4.504969e-01,2.778595e-01,3.572917e-02,1.868429e-02,3.817188e-02,2.017990e-01,2.356961e-01,6.476464e-04,1.917784e-01,1.883792e-02,6.892002e-01,9.180365e-04
,5.417320e-03,7.987482e-03,8.421375e-02,8.107421e-01,4.621479e-01,1.264461e-01,6.088301e-02,9.711845e-01,4.537961e-01,3.374737e-01,7.147515e-01,3.446114e-02,1.391236e-01
,2.792949e-03,8.681489e-01,6.995291e-01,1.888599e-01,8.703926e-01,6.532436e-01,5.822158e-03,4.278737e-01,5.194451e-01,7.648862e-03,2.712795e-02,6.923446e-01,2.155742e-01
,3.095516e-03,5.513717e-02,7.036561e-03,5.621772e-02,8.410768e-01,1.553177e-01,5.783972e-01,9.139361e-01,7.793954e-01,3.277243e-01,1.628008e-02,4.562770e-01,8.185310e-03
,1.281526e-02,7.759723e-02,8.781222e-01,3.093447e-01,9.024527e-07,8.357430e-03,5.439191e-02,3.452960e-03,8.838844e-01,1.124006e-02,3.000919e-02,4.811588e-01,1.195559e-03
,4.706060e-01,5.006557e-02,7.484449e-01,6.182207e-01,1.113888e-03,2.127814e-04,1.251516e-01,3.427559e-02,9.939663e-01,2.724629e-01,9.380957e-01,9.801783e-01,6.736301e-01
,6.483325e-01,5.582446e-02,4.309441e-02,1.524737e-05,7.840253e-01,6.569911e-01,1.313778e-01,9.053038e-01,4.868889e-02,7.562701e-01,2.711943e-01,5.053834e-01,4.301493e-02
,4.669822e-02,4.351687e-01,3.981509e-01,1.543811e-01,1.743985e-01,1.008578e-01,4.320696e-01,6.569131e-01,1.971029e-01,9.348643e-01,7.318134e-01,4.343971e-01,6.602119e-01
,1.093594e-02,7.530525e-01,2.380471e-01,9.659419e-01,6.927182e-01,3.439926e-01,6.115200e-01,1.300354e-02,3.027140e-01,4.115643e-01,6.540679e-01,3.659400e-10};
double[] stder_actual = model._output.stdErr();
double[] zvals_actual = model._output.zValues();
double[] pvals_actual = model._output.pValues();
String[] names_actual = model._output.coefficientNames();
HashMap<String, Integer> coefMap = new HashMap<>();
for (int i = 0; i < names_expected.length; ++i)
coefMap.put(names_expected[i], i);
double[] coefs_actual = model._output._global_beta;
for (int i = 0; i < exp_coefs.length; ++i) {
String s = removeDot(names_actual[i]);
if(!coefMap.containsKey(s)) { // removed col, check we removed it too
assertTrue(coefs_actual[i] == 0 && Double.isNaN(zvals_actual[i]));
System.out.println("found removed col " + s);
} else {
int id = coefMap.get(s);
assertEquals(exp_coefs[id], coefs_actual[i], 1e-4);
assertEquals(stder_expected[id], stder_actual[i],Math.abs(stder_expected[id]*1e-4));
assertEquals(zvals_expected[id], zvals_actual[i],Math.abs(zvals_expected[id]*1e-4));
assertEquals(pvals_expected[id], pvals_actual[i],pvals_expected[id]*1e-4);
}
}
predict = model.score(_airlines);
} finally {
if(model != null) model.delete();
if(predict != null) predict.delete();
}
}
private static String removeDot(String s) {
int id = s.indexOf(".");
if(id ==-1) return s;
return s.substring(0,id) + s.substring(id+1);
}
@AfterClass
public static void cleanUp() {
if(_canCarTrain != null)
_canCarTrain.delete();
if(_merit != null)
_merit.remove();
if(_class != null)
_class.remove();
if(_earinf != null)
_earinf.delete();
if(_weighted != null)
_weighted.delete();
if(_upsampled != null)
_upsampled.delete();
if(_prostateTrain != null)
_prostateTrain.delete();
if(_airlines != null)
_airlines.delete();
if(_airlinesMM != null)
_airlinesMM.delete();
}
}