本文整理汇总了Java中hex.deeplearning.DeepLearningModel类的典型用法代码示例。如果您正苦于以下问题：Java DeepLearningModel类的具体用法？Java DeepLearningModel怎么用？Java DeepLearningModel使用的例子？那么恭喜您, 这里精选的类代码示例或许可以为您提供帮助。
DeepLearningModel类属于hex.deeplearning包，在下文中一共展示了DeepLearningModel类的27个代码示例，这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞，您的评价将有助于我们的系统推荐出更棒的Java代码示例。
示例1: defaultSearchGLM
点赞 3
import hex.deeplearning.DeepLearningModel; //导入依赖的package包/类
public Job<Grid> defaultSearchGLM() {
///////////////////////////////////////////////////////////
// do a random hyperparameter search with GLM
///////////////////////////////////////////////////////////
// TODO: convert to using the REST API
Key<Grid> gridKey = Key.make("GLM_grid_default_" + this._key.toString());
HyperSpaceSearchCriteria.RandomDiscreteValueSearchCriteria searchCriteria = buildSpec.build_control.stopping_criteria;
// TODO: put this into a Provider, which can return multiple searches
GLMModel.GLMParameters glmParameters = new GLMModel.GLMParameters();
setCommonModelBuilderParams(glmParameters);
glmParameters._lambda_search = true;
glmParameters._family = getResponseColumn().isBinary() && !(getResponseColumn().isNumeric()) ? GLMModel.GLMParameters.Family.binomial :
getResponseColumn().isCategorical() ? GLMModel.GLMParameters.Family.multinomial :
GLMModel.GLMParameters.Family.gaussian; // TODO: other continuous distributions!
Map<String, Object[]> searchParams = new HashMap<>();
glmParameters._alpha = new double[] {0.0, 0.2, 0.4, 0.6, 0.8, 1.0}; // Note: standard GLM parameter is an array; don't use searchParams!
// NOTE: removed MissingValuesHandling.Skip for now because it's crashing. See https://0xdata.atlassian.net/browse/PUBDEV-4974
searchParams.put("_missing_values_handling", new DeepLearningModel.DeepLearningParameters.MissingValuesHandling[] {DeepLearningModel.DeepLearningParameters.MissingValuesHandling.MeanImputation /* , DeepLearningModel.DeepLearningParameters.MissingValuesHandling.Skip */});
Job<Grid>glmJob = hyperparameterSearch("GLM", glmParameters, searchParams);
return glmJob;
}
开发者ID:h2oai，
项目名称:h2o-3，
代码行数:27，
代码来源:AutoML.java
示例2: execImpl
点赞 3
import hex.deeplearning.DeepLearningModel; //导入依赖的package包/类
@Override
protected final void execImpl() {
if (dl_model == null) throw new IllegalArgumentException("Deep Learning Model must be specified.");
DeepLearningModel dlm = UKV.get(dl_model);
if (dlm == null) throw new IllegalArgumentException("Deep Learning Model not found.");
StringBuilder sb = new StringBuilder();
if (layer < -1 || layer > dlm.get_params().hidden.length-1) throw new IllegalArgumentException("Layer must be either -1 or between 0 and " + (dlm.get_params().hidden.length-1));
if (layer == -1) layer = dlm.get_params().hidden.length-1;
int features = dlm.get_params().hidden[layer];
sb.append("\nTransforming frame '" + source._key.toString() + "' with " + source.numCols() + " into " + features + " features with model '" + dl_model + "'\n");
Frame df = dlm.scoreDeepFeatures(source, layer);
sb.append("Storing the new features under: " + dest() + ".\n");
Frame output = new Frame(dest(), df.names(), df.vecs());
output.delete_and_lock(null);
output.unlock(null);
}
开发者ID:h2oai，
项目名称:h2o-2，
代码行数:19，
代码来源:DeepFeatures.java
示例3: readExternal
点赞 2
import hex.deeplearning.DeepLearningModel; //导入依赖的package包/类
@SuppressWarnings("unchecked")
@Override
public void readExternal(ObjectInput in) throws IOException, ClassNotFoundException {
storageLocation = (String) in.readObject();
labelToH2OEnumMap = (HashMap<Double, Double>) in.readObject();
@SuppressWarnings("rawtypes")
List<Key> keys = new ObjectTreeBinarySerializer().load(FileUtils.getURI(storageLocation));
this.dlModel = (DeepLearningModel) keys.get(0).get();
}
开发者ID:wso2-attic，
项目名称:carbon-ml，
代码行数:10，
代码来源:MLDeeplearningModel.java
示例4: testSparse
点赞 2
import hex.deeplearning.DeepLearningModel; //导入依赖的package包/类
@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();
}
}
开发者ID:h2oai，
项目名称:h2o-3，
代码行数:29，
代码来源:GLMBasicTestRegression.java
示例5: if
点赞 2
import hex.deeplearning.DeepLearningModel; //导入依赖的package包/类
Job<DeepLearningModel>defaultDeepLearning() {
if (exceededSearchLimits("DeepLearning")) return null;
DeepLearningModel.DeepLearningParameters deepLearningParameters = new DeepLearningModel.DeepLearningParameters();
setCommonModelBuilderParams(deepLearningParameters);
deepLearningParameters._stopping_tolerance = this.buildSpec.build_control.stopping_criteria.stopping_tolerance();
deepLearningParameters._hidden = new int[]{ 10, 10, 10 };
Job deepLearningJob = trainModel(null, "deeplearning", deepLearningParameters);
return deepLearningJob;
}
开发者ID:h2oai，
项目名称:h2o-3，
代码行数:13，
代码来源:AutoML.java
示例6: defaultSearchDL1
点赞 2
import hex.deeplearning.DeepLearningModel; //导入依赖的package包/类
public Job<Grid> defaultSearchDL1(Key<Grid> gridKey) {
///////////////////////////////////////////////////////////
// do a random hyperparameter search with DL
///////////////////////////////////////////////////////////
HyperSpaceSearchCriteria.RandomDiscreteValueSearchCriteria searchCriteria = buildSpec.build_control.stopping_criteria;
// TODO: put this into a Provider, which can return multiple searches
DeepLearningModel.DeepLearningParameters dlParameters = new DeepLearningModel.DeepLearningParameters();
setCommonModelBuilderParams(dlParameters);
dlParameters._epochs = 10000; // early stopping takes care of epochs - no need to tune!
dlParameters._adaptive_rate = true;
dlParameters._activation = RectifierWithDropout;
Map<String, Object[]> searchParams = new HashMap<>();
// common:
searchParams.put("_rho", new Double[] { 0.9, 0.95, 0.99 });
searchParams.put("_epsilon", new Double[] { 1e-6, 1e-7, 1e-8, 1e-9 });
searchParams.put("_input_dropout_ratio", new Double[] { 0.0, 0.05, 0.1, 0.15, 0.2 });
// unique:
searchParams.put("_hidden", new Integer[][] { {50}, {200}, {500} });
searchParams.put("_hidden_dropout_ratios", new Double[][] { { 0.0 }, { 0.1 }, { 0.2 }, { 0.3 }, { 0.4 }, { 0.5 } });
Job<Grid>dlJob = hyperparameterSearch(gridKey, "DL", dlParameters, searchParams);
return dlJob;
}
开发者ID:h2oai，
项目名称:h2o-3，
代码行数:29，
代码来源:AutoML.java
示例7: defaultSearchDL2
点赞 2
import hex.deeplearning.DeepLearningModel; //导入依赖的package包/类
public Job<Grid> defaultSearchDL2(Key<Grid> gridKey) {
///////////////////////////////////////////////////////////
// do a random hyperparameter search with DL
///////////////////////////////////////////////////////////
HyperSpaceSearchCriteria.RandomDiscreteValueSearchCriteria searchCriteria = buildSpec.build_control.stopping_criteria;
// TODO: put this into a Provider, which can return multiple searches
DeepLearningModel.DeepLearningParameters dlParameters = new DeepLearningModel.DeepLearningParameters();
setCommonModelBuilderParams(dlParameters);
dlParameters._epochs = 10000; // early stopping takes care of epochs - no need to tune!
dlParameters._adaptive_rate = true;
dlParameters._activation = RectifierWithDropout;
Map<String, Object[]> searchParams = new HashMap<>();
// common:
searchParams.put("_rho", new Double[] { 0.9, 0.95, 0.99 });
searchParams.put("_epsilon", new Double[] { 1e-6, 1e-7, 1e-8, 1e-9 });
searchParams.put("_input_dropout_ratio", new Double[] { 0.0, 0.05, 0.1, 0.15, 0.2 });
// unique:
searchParams.put("_hidden", new Integer[][] { {50, 50}, {200, 200}, {500, 500} });
searchParams.put("_hidden_dropout_ratios", new Double[][] { { 0.0, 0.0 }, { 0.1, 0.1 }, { 0.2, 0.2 }, { 0.3, 0.3 }, { 0.4, 0.4 }, { 0.5, 0.5 } });
Job<Grid>dlJob = hyperparameterSearch(gridKey, "DL", dlParameters, searchParams);
return dlJob;
}
开发者ID:h2oai，
项目名称:h2o-3，
代码行数:29，
代码来源:AutoML.java
示例8: defaultSearchDL3
点赞 2
import hex.deeplearning.DeepLearningModel; //导入依赖的package包/类
public Job<Grid> defaultSearchDL3(Key<Grid> gridKey) {
///////////////////////////////////////////////////////////
// do a random hyperparameter search with DL
///////////////////////////////////////////////////////////
HyperSpaceSearchCriteria.RandomDiscreteValueSearchCriteria searchCriteria = buildSpec.build_control.stopping_criteria;
// TODO: put this into a Provider, which can return multiple searches
DeepLearningModel.DeepLearningParameters dlParameters = new DeepLearningModel.DeepLearningParameters();
setCommonModelBuilderParams(dlParameters);
dlParameters._epochs = 10000; // early stopping takes care of epochs - no need to tune!
dlParameters._adaptive_rate = true;
dlParameters._activation = RectifierWithDropout;
Map<String, Object[]> searchParams = new HashMap<>();
// common:
searchParams.put("_rho", new Double[] { 0.9, 0.95, 0.99 });
searchParams.put("_epsilon", new Double[] { 1e-6, 1e-7, 1e-8, 1e-9 });
searchParams.put("_input_dropout_ratio", new Double[] { 0.0, 0.05, 0.1, 0.15, 0.2 });
// unique:
searchParams.put("_hidden", new Integer[][] { {50, 50, 50}, {200, 200, 200}, {500, 500, 500} });
searchParams.put("_hidden_dropout_ratios", new Double[][] { { 0.0, 0.0, 0.0 }, { 0.1, 0.1, 0.1 }, { 0.2, 0.2, 0.2 }, { 0.3, 0.3, 0.3 }, { 0.4, 0.4, 0.4 }, { 0.5, 0.5, 0.5 } });
Job<Grid>dlJob = hyperparameterSearch(gridKey, "DL", dlParameters, searchParams);
return dlJob;
}
开发者ID:h2oai，
项目名称:h2o-3，
代码行数:29，
代码来源:AutoML.java
示例9: toHTML
点赞 2
import hex.deeplearning.DeepLearningModel; //导入依赖的package包/类
@Override public boolean toHTML( StringBuilder sb ) {
Job jjob = Job.findJob(job_key);
if (jjob ==null) return true;
DeepLearningModel m = UKV.get(jjob.dest());
if (m!=null) m.generateHTML("Deep Learning Model", sb);
else DocGen.HTML.paragraph(sb, "Pending...");
return true;
}
开发者ID:h2oai，
项目名称:h2o-2，
代码行数:9，
代码来源:DeepLearningProgressPage.java
示例10: visualize
点赞 2
import hex.deeplearning.DeepLearningModel; //导入依赖的package包/类
static public void visualize(final DeepLearningModel dlm) {
Neurons[] neurons = DeepLearningTask.makeNeuronsForTesting(dlm.model_info());
frame.dispose();
frame.setDefaultCloseOperation(WindowConstants.DISPOSE_ON_CLOSE);
DeepLearningVisualization canvas = new DeepLearningVisualization(neurons);
frame.setContentPane(canvas.init());
frame.pack();
frame.setLocationRelativeTo(null);
frame.setVisible(true);
}
开发者ID:h2oai，
项目名称:h2o-2，
代码行数:11，
代码来源:DeepLearningVisualization.java
示例11: MLDeeplearningModel
点赞 2
import hex.deeplearning.DeepLearningModel; //导入依赖的package包/类
public MLDeeplearningModel(DeepLearningModel dlModel) {
this.dlModel = dlModel;
}
开发者ID:wso2-attic，
项目名称:carbon-ml，
代码行数:4，
代码来源:MLDeeplearningModel.java
示例12: getDlModel
点赞 2
import hex.deeplearning.DeepLearningModel; //导入依赖的package包/类
public DeepLearningModel getDlModel() {
return dlModel;
}
开发者ID:wso2-attic，
项目名称:carbon-ml，
代码行数:4，
代码来源:MLDeeplearningModel.java
示例13: buildStackedAutoencodersModel
点赞 2
import hex.deeplearning.DeepLearningModel; //导入依赖的package包/类
/**
* Build the stacked autoencoder model
*
* @param sparkContext
* @param modelID model ID
* @param trainingData training data to train the classifier
* @param testingData testing data to test the classifier and get metrics
* @param workflow workflow
* @param mlModel MLModel to be updated with calcualted values
* @param includedFeatures Included features
* @return
* @throws MLModelBuilderException
*/
private ModelSummary buildStackedAutoencodersModel(JavaSparkContext sparkContext, long modelID,
JavaRDD<LabeledPoint> trainingData, JavaRDD<LabeledPoint> testingData, Workflow workflow, MLModel mlModel,
SortedMap<Integer, String> includedFeatures) throws MLModelBuilderException {
try {
StackedAutoencodersClassifier saeClassifier = new StackedAutoencodersClassifier();
Map<String, String> hyperParameters = workflow.getHyperParameters();
// train the stacked autoencoder
DeepLearningModel deeplearningModel = saeClassifier.train(trainingData,
Integer.parseInt(hyperParameters.get(MLConstants.BATCH_SIZE)),
stringArrToIntArr(hyperParameters.get(MLConstants.LAYER_SIZES)),
hyperParameters.get(MLConstants.ACTIVATION_TYPE),
Integer.parseInt(hyperParameters.get(MLConstants.EPOCHS)),
Integer.parseInt(hyperParameters.get(MLConstants.SEED)),
workflow.getResponseVariable(),
getContext().getModel().getName(), mlModel, modelID);
if (deeplearningModel == null) {
throw new MLModelBuilderException("DeeplearningModel is Null.");
}
// remove from cache
trainingData.unpersist();
// add to cache
testingData.cache();
// make predictions with the trained model
JavaPairRDD<Double, Double> predictionsAndLabels = saeClassifier
.test(sparkContext, deeplearningModel, testingData, mlModel).cache();
// get model summary
DeeplearningModelSummary deeplearningModelSummary = DeeplearningModelUtils
.getDeeplearningModelSummary(sparkContext, testingData, predictionsAndLabels);
// remove from cache
testingData.unpersist();
mlModel.setModel(new MLDeeplearningModel(deeplearningModel));
deeplearningModelSummary.setFeatures(includedFeatures.values().toArray(new String[0]));
deeplearningModelSummary.setAlgorithm(MLConstants.DEEPLEARNING_ALGORITHM.STACKED_AUTOENCODERS.toString());
// set accuracy values
MulticlassMetrics multiclassMetrics = getMulticlassMetrics(sparkContext, predictionsAndLabels);
// remove from cache
predictionsAndLabels.unpersist();
deeplearningModelSummary
.setMulticlassConfusionMatrix(getMulticlassConfusionMatrix(multiclassMetrics, mlModel));
Double modelAccuracy = getModelAccuracy(multiclassMetrics);
deeplearningModelSummary.setModelAccuracy(modelAccuracy);
deeplearningModelSummary.setDatasetVersion(workflow.getDatasetVersion());
return deeplearningModelSummary;
} catch (Exception e) {
throw new MLModelBuilderException(
"An error occurred while building stacked autoencoders model: " + e.getMessage(), e);
}
}
开发者ID:wso2-attic，
项目名称:carbon-ml，
代码行数:76，
代码来源:DeeplearningModelBuilder.java
示例14: persistModel
点赞 2
import hex.deeplearning.DeepLearningModel; //导入依赖的package包/类
private void persistModel(long modelId, String modelName, MLModel model) throws MLModelBuilderException {
try {
MLStorage storage = valueHolder.getDatabaseService().getModelStorage(modelId);
if (storage == null) {
throw new MLModelBuilderException("Invalid model ID: " + modelId);
}
if (!MLUtils.isValidName(modelName)) {
throw new MLModelBuilderException("Model name: " + modelName + " contains restricted path elements.");
}
String storageType = storage.getType();
String storageLocation = storage.getLocation();
String outPath = storageLocation + File.separator + modelName;
// if this is a deeplearning model, need to set the storage location for writing
// then the sparkdeeplearning model will use ObjectTreeBinarySerializer to write it to the given directory
// the DeeplearningModel will be saved as a .bin file
if (MLConstants.DEEPLEARNING.equalsIgnoreCase(model.getAlgorithmClass())) {
MLDeeplearningModel mlDeeplearningModel = (MLDeeplearningModel) model.getModel();
mlDeeplearningModel.setStorageLocation(storageLocation);
model.setModel(mlDeeplearningModel);
// Write POJO if it is a Deep Learning model
// convert model name
String dlModelName = modelName.replace('.', '_').replace('-', '_');
File file = new File(storageLocation + "/" + dlModelName + "_dl" + ".java");
FileOutputStream fileOutputStream = new FileOutputStream(file);
DeepLearningModel deepLearningModel = mlDeeplearningModel.getDlModel();
deepLearningModel.toJava(fileOutputStream, false, false);
fileOutputStream.close();
MLModel dlModel = new MLModel();
dlModel.setAlgorithmClass(model.getAlgorithmClass());
dlModel.setAlgorithmName(model.getAlgorithmName());
dlModel.setEncodings(model.getEncodings());
dlModel.setFeatures(model.getFeatures());
dlModel.setResponseIndex(model.getResponseIndex());
dlModel.setResponseVariable(model.getResponseVariable());
dlModel.setNewToOldIndicesList(model.getNewToOldIndicesList());
// Writing the DL model without Deep Learning logic
// For prediction with POJO
MLIOFactory ioFactoryDl = new MLIOFactory(valueHolder.getMlProperties());
MLOutputAdapter outputAdapterDl = ioFactoryDl.getOutputAdapter(storageType + MLConstants.OUT_SUFFIX);
ByteArrayOutputStream baosDl = new ByteArrayOutputStream();
ObjectOutputStream oosDl = new ObjectOutputStream(baosDl);
oosDl.writeObject(dlModel);
oosDl.flush();
oosDl.close();
InputStream isDl = new ByteArrayInputStream(baosDl.toByteArray());
// adapter will write the model and close the stream.
outputAdapterDl.write(outPath + "_dl", isDl);
}
MLIOFactory ioFactory = new MLIOFactory(valueHolder.getMlProperties());
MLOutputAdapter outputAdapter = ioFactory.getOutputAdapter(storageType + MLConstants.OUT_SUFFIX);
ByteArrayOutputStream baos = new ByteArrayOutputStream();
ObjectOutputStream oos = new ObjectOutputStream(baos);
oos.writeObject(model);
oos.flush();
oos.close();
InputStream is = new ByteArrayInputStream(baos.toByteArray());
// adapter will write the model and close the stream.
outputAdapter.write(outPath, is);
valueHolder.getDatabaseService().updateModelStorage(modelId, storageType, outPath);
log.info(String.format("Successfully persisted the model [id] %s", modelId));
} catch (Exception e) {
throw new MLModelBuilderException("Failed to persist the model [id] " + modelId + ". " + e.getMessage(), e);
}
}
开发者ID:wso2-attic，
项目名称:carbon-ml，
代码行数:70，
代码来源:MLModelHandler.java
示例15: createImpl
点赞 2
import hex.deeplearning.DeepLearningModel; //导入依赖的package包/类
@Override public DeepLearningModel createImpl() {
DeepLearningParameters parms = parameters.createImpl();
return new DeepLearningModel(Key.make() /*dest*/, parms, new DeepLearningModel.DeepLearningModelOutput(null), null, null, 0);
}
开发者ID:kyoren，
项目名称:https-github.com-h2oai-h2o-3，
代码行数:5，
代码来源:DeepLearningModelV3.java
示例16: computeDeepLearningMetalearner
点赞 2
import hex.deeplearning.DeepLearningModel; //导入依赖的package包/类
void computeDeepLearningMetalearner(){
//DeepLearning Metalearner
DeepLearning metaDeepLearningBuilder;
metaDeepLearningBuilder = ModelBuilder.make("DeepLearning", _metalearnerJob, _metalearnerKey);
DeepLearningV3.DeepLearningParametersV3 params = new DeepLearningV3.DeepLearningParametersV3();
params.init_meta();
params.fillFromImpl(metaDeepLearningBuilder._parms); // Defaults for this builder into schema
//Metalearner parameters
if (_hasMetalearnerParams) {
Properties p = new Properties();
HashMap<String, String[]> map = new Gson().fromJson(_metalearner_params, new TypeToken<HashMap<String, String[]>>() {
}.getType());
for (Map.Entry<String, String[]> param : map.entrySet()) {
String[] paramVal = param.getValue();
if (paramVal.length == 1) {
p.setProperty(param.getKey(), paramVal[0]);
} else {
p.setProperty(param.getKey(), Arrays.toString(paramVal));
}
params.fillFromParms(p, true);
}
DeepLearningModel.DeepLearningParameters dlParams = params.createAndFillImpl();
metaDeepLearningBuilder._parms = dlParams;
}
metaDeepLearningBuilder._parms._train = _levelOneTrainingFrame._key;
metaDeepLearningBuilder._parms._valid = (_levelOneValidationFrame == null ? null : _levelOneValidationFrame._key);
metaDeepLearningBuilder._parms._response_column = _model.responseColumn;
metaDeepLearningBuilder._parms._nfolds = _model._parms._metalearner_nfolds; //cross-validation of the metalearner
if (_model._parms._metalearner_fold_column == null) {
metaDeepLearningBuilder._parms._nfolds = _model._parms._metalearner_nfolds; //cross-validation of the metalearner
if (_model._parms._metalearner_nfolds > 1) {
if (_model._parms._metalearner_fold_assignment == null) {
metaDeepLearningBuilder._parms._fold_assignment = Model.Parameters.FoldAssignmentScheme.AUTO;
} else {
metaDeepLearningBuilder._parms._fold_assignment = _model._parms._metalearner_fold_assignment; //cross-validation of the metalearner
}
}
} else {
metaDeepLearningBuilder._parms._fold_column = _model._parms._metalearner_fold_column; //cross-validation of the metalearner
}
metaDeepLearningBuilder.init(false);
Job<DeepLearningModel> j = metaDeepLearningBuilder.trainModel();
while (j.isRunning()) {
try {
_job.update(j._work, "training metalearner(" + _model._parms._metalearner_algorithm + ")");
Thread.sleep(100);
} catch (InterruptedException e) {
}
}
Log.info("Finished training metalearner model(" + _model._parms._metalearner_algorithm + ").");
_model._output._metalearner = metaDeepLearningBuilder.get();
_model.doScoreOrCopyMetrics(_job);
if (_parms._keep_levelone_frame) {
_model._output._levelone_frame_id = _levelOneTrainingFrame; //Keep Level One Training Frame in Stacked Ensemble model object
} else {
DKV.remove(_levelOneTrainingFrame._key); //Remove Level One Training Frame from DKV
}
if (null != _levelOneValidationFrame) {
DKV.remove(_levelOneValidationFrame._key); //Remove Level One Validation Frame from DKV
}
_model.update(_job);
_model.unlock(_job);
}
开发者ID:h2oai，
项目名称:h2o-3，
代码行数:70，
代码来源:Metalearner.java
示例17: createImpl
点赞 2
import hex.deeplearning.DeepLearningModel; //导入依赖的package包/类
@Override public DeepLearningModel createImpl() {
DeepLearningModel.DeepLearningParameters parms = parameters.createImpl();
return new DeepLearningModel(Key.make() /*dest*/, parms, new DeepLearningModel.DeepLearningModelOutput(null), null, null, 0);
}
开发者ID:h2oai，
项目名称:h2o-3，
代码行数:5，
代码来源:DeepLearningModelV3.java
示例18: testPValuesTweedie
点赞 2
import hex.deeplearning.DeepLearningModel; //导入依赖的package包/类
@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();
}
}
开发者ID:h2oai，
项目名称:h2o-3，
代码行数:74，
代码来源:GLMBasicTestRegression.java
示例19: testPValuesPoisson
点赞 2
import hex.deeplearning.DeepLearningModel; //导入依赖的package包/类
@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();
}
}
开发者ID:h2oai，
项目名称:h2o-3，
代码行数:69，
代码来源:GLMBasicTestRegression.java
示例20: testCovtypeNoIntercept
点赞 2
import hex.deeplearning.DeepLearningModel; //导入依赖的package包/类
@Test
public void testCovtypeNoIntercept(){
GLMParameters params = new GLMParameters(Family.multinomial);
GLMModel model = null;
Frame preds = null;
Vec weights = _covtype.anyVec().makeCon(1);
Key k = Key.<Frame>make("cov_with_weights");
Frame f = new Frame(k,_covtype.names(),_covtype.vecs());
f.add("weights",weights);
DKV.put(f);
try {
params._response_column = "C55";
params._train = k;
params._valid = _covtype._key;
params._objective_epsilon = 1e-6;
params._beta_epsilon = 1e-4;
params._weights_column = "weights";
params._missing_values_handling = DeepLearningModel.DeepLearningParameters.MissingValuesHandling.Skip;
params._intercept = false;
double[] alpha = new double[]{0,.5,.1};
Solver s = Solver.L_BFGS;
System.out.println("solver = " + s);
params._solver = s;
params._max_iterations = 5000;
for (int i = 0; i < alpha.length; ++i) {
params._alpha = new double[]{alpha[i]};
// params._lambda[0] = lambda[i];
model = new GLM(params).trainModel().get();
System.out.println(model.coefficients());
// Assert.assertEquals(0,model.coefficients().get("Intercept"),0);
double [][] bs = model._output.getNormBetaMultinomial();
for(double [] b:bs)
Assert.assertEquals(0,b[b.length-1],0);
System.out.println(model._output._model_summary);
System.out.println(model._output._training_metrics);
System.out.println(model._output._validation_metrics);
preds = model.score(_covtype);
ModelMetricsMultinomialGLM mmTrain = (ModelMetricsMultinomialGLM) hex.ModelMetricsMultinomial.getFromDKV(model, _covtype);
assertTrue(model._output._training_metrics.equals(mmTrain));
model.delete();
model = null;
preds.delete();
preds = null;
}
} finally{
weights.remove();
DKV.remove(k);
if(model != null)model.delete();
if(preds != null)preds.delete();
}
}
开发者ID:h2oai，
项目名称:h2o-3，
代码行数:52，
代码来源:GLMBasicTestMultinomial.java
示例21: testCovtypeBasic
点赞 2
import hex.deeplearning.DeepLearningModel; //导入依赖的package包/类
@Test
public void testCovtypeBasic(){
GLMParameters params = new GLMParameters(Family.multinomial);
GLMModel model = null;
Frame preds = null;
Vec weights = _covtype.anyVec().makeCon(1);
Key k = Key.<Frame>make("cov_with_weights");
Frame f = new Frame(k,_covtype.names(),_covtype.vecs());
f.add("weights",weights);
DKV.put(f);
try {
params._response_column = "C55";
params._train = k;
params._valid = _covtype._key;
params._lambda = new double[]{4.881e-05};
params._alpha = new double[]{1};
params._objective_epsilon = 1e-6;
params._beta_epsilon = 1e-4;
params._weights_column = "weights";
params._missing_values_handling = DeepLearningModel.DeepLearningParameters.MissingValuesHandling.Skip;
double[] alpha = new double[]{1};
double[] expected_deviance = new double[]{25499.76};
double[] lambda = new double[]{2.544750e-05};
for (Solver s : new Solver[]{Solver.IRLSM, Solver.COORDINATE_DESCENT, Solver.L_BFGS}) {
System.out.println("solver = " + s);
params._solver = s;
params._max_iterations = params._solver == Solver.L_BFGS?300:10;
for (int i = 0; i < alpha.length; ++i) {
params._alpha[0] = alpha[i];
params._lambda[0] = lambda[i];
model = new GLM(params).trainModel().get();
System.out.println(model._output._model_summary);
System.out.println(model._output._training_metrics);
System.out.println(model._output._validation_metrics);
assertTrue(model._output._training_metrics.equals(model._output._validation_metrics));
assertTrue(((ModelMetricsMultinomialGLM) model._output._training_metrics)._resDev <= expected_deviance[i] * 1.1);
preds = model.score(_covtype);
ModelMetricsMultinomialGLM mmTrain = (ModelMetricsMultinomialGLM) hex.ModelMetricsMultinomial.getFromDKV(model, _covtype);
assertTrue(model._output._training_metrics.equals(mmTrain));
model.delete();
model = null;
preds.delete();
preds = null;
}
}
} finally{
weights.remove();
DKV.remove(k);
if(model != null)model.delete();
if(preds != null)preds.delete();
}
}
开发者ID:h2oai，
项目名称:h2o-3，
代码行数:53，
代码来源:GLMBasicTestMultinomial.java
示例22: testCovtypeNAs
点赞 2
import hex.deeplearning.DeepLearningModel; //导入依赖的package包/类
@Test
public void testCovtypeNAs(){
GLMParameters params = new GLMParameters(Family.multinomial);
GLMModel model = null;
Frame preds = null;
Frame covtype_subset = null, covtype_copy = null;
try {
double expected_deviance = 26000;
covtype_copy = _covtype.deepCopy("covtype_copy");
DKV.put(covtype_copy);
Vec.Writer w = covtype_copy.vec(54).open();
w.setNA(10);
w.setNA(20);
w.setNA(30);
w.close();
covtype_subset = new Frame(Key.<Frame>make("covtype_subset"),new String[]{"C51","C52","C53","C54","C55"},covtype_copy.vecs(new int[]{50,51,52,53,54}));
DKV.put(covtype_subset);
// params._nlambdas = 3;
params._response_column = "C55";
params._train = covtype_copy._key;
params._valid = covtype_copy._key;
params._alpha = new double[]{.99};
params._objective_epsilon = 1e-6;
params._beta_epsilon = 1e-4;
params._max_active_predictors = 50;
params._max_iterations = 500;
params._solver = Solver.L_BFGS;
params._missing_values_handling = DeepLearningModel.DeepLearningParameters.MissingValuesHandling.Skip;
// params._lambda_search = true;
model = new GLM(params).trainModel().get();
assertEquals(covtype_copy.numRows()-3-1,model._nullDOF);
System.out.println(model._output._training_metrics);
System.out.println(model._output._validation_metrics);
assertTrue(model._output._training_metrics.equals(model._output._validation_metrics));
preds = model.score(covtype_copy);
ModelMetricsMultinomialGLM mmTrain = (ModelMetricsMultinomialGLM) hex.ModelMetricsMultinomial.getFromDKV(model, covtype_copy);
assertTrue(model._output._training_metrics.equals(mmTrain));
assertTrue(((ModelMetricsMultinomialGLM) model._output._training_metrics)._resDev <= expected_deviance);
System.out.println(model._output._model_summary);
model.delete();
model = null;
preds.delete();
preds = null;
// now run the same on the subset
params._train = covtype_subset._key;
model = new GLM(params).trainModel().get();
assertEquals(covtype_copy.numRows()-3-1,model._nullDOF);
System.out.println(model._output._training_metrics);
System.out.println(model._output._validation_metrics);
assertTrue(model._output._training_metrics.equals(model._output._validation_metrics));
preds = model.score(_covtype);
System.out.println(model._output._model_summary);
assertTrue(((ModelMetricsMultinomialGLM) model._output._training_metrics)._resDev <= 66000);
model.delete();
model = null;
preds.delete();
preds = null;
} finally{
if(covtype_subset != null) covtype_subset.delete();
if(covtype_copy != null)covtype_copy.delete();
if(model != null)model.delete();
if(preds != null)preds.delete();
}
}
开发者ID:h2oai，
项目名称:h2o-3，
代码行数:66，
代码来源:GLMBasicTestMultinomial.java
示例23: testXValPredictions
点赞 2
import hex.deeplearning.DeepLearningModel; //导入依赖的package包/类
@Test public void testXValPredictions() {
final int nfolds = 3;
Frame tfr = null;
try {
// Load data, hack frames
tfr = parse_test_file("smalldata/iris/iris_wheader.csv");
Frame foldId = new Frame(new String[]{"foldId"}, new Vec[]{AstKFold.kfoldColumn(tfr.vec("class").makeZero(), nfolds, 543216789)});
tfr.add(foldId);
DKV.put(tfr);
// GBM
GBMModel.GBMParameters parms = new GBMModel.GBMParameters();
parms._train = tfr._key;
parms._response_column = "class";
parms._ntrees = 1;
parms._max_depth = 1;
parms._fold_column = "foldId";
parms._distribution = DistributionFamily.multinomial;
parms._keep_cross_validation_predictions=true;
GBM job = new GBM(parms);
GBMModel gbm = job.trainModel().get();
checkModel(gbm, foldId.anyVec(),3);
// DRF
DRFModel.DRFParameters parmsDRF = new DRFModel.DRFParameters();
parmsDRF._train = tfr._key;
parmsDRF._response_column = "class";
parmsDRF._ntrees = 1;
parmsDRF._max_depth = 1;
parmsDRF._fold_column = "foldId";
parmsDRF._distribution = DistributionFamily.multinomial;
parmsDRF._keep_cross_validation_predictions=true;
DRF drfJob = new DRF(parmsDRF);
DRFModel drf = drfJob.trainModel().get();
checkModel(drf, foldId.anyVec(),3);
// GLM
GLMModel.GLMParameters parmsGLM = new GLMModel.GLMParameters();
parmsGLM._train = tfr._key;
parmsGLM._response_column = "sepal_len";
parmsGLM._fold_column = "foldId";
parmsGLM._keep_cross_validation_predictions=true;
GLM glmJob = new GLM(parmsGLM);
GLMModel glm = glmJob.trainModel().get();
checkModel(glm, foldId.anyVec(),1);
// DL
DeepLearningModel.DeepLearningParameters parmsDL = new DeepLearningModel.DeepLearningParameters();
parmsDL._train = tfr._key;
parmsDL._response_column = "class";
parmsDL._hidden = new int[]{1};
parmsDL._epochs = 1;
parmsDL._fold_column = "foldId";
parmsDL._keep_cross_validation_predictions=true;
DeepLearning dlJob = new DeepLearning(parmsDL);
DeepLearningModel dl = dlJob.trainModel().get();
checkModel(dl, foldId.anyVec(),3);
} finally {
if (tfr != null) tfr.remove();
}
}
开发者ID:h2oai，
项目名称:h2o-3，
代码行数:63，
代码来源:XValPredictionsCheck.java
示例24: run
点赞 2
import hex.deeplearning.DeepLearningModel; //导入依赖的package包/类
@Test
public void run() {
long seed = 0xDECAF;
Key file_train = NFSFileVec.make(find_test_file(PATH));
Frame train = ParseDataset2.parse(Key.make(), new Key[]{file_train});
DeepLearning p = new DeepLearning();
p.source = train;
p.autoencoder = true;
p.response = train.lastVec();
p.seed = seed;
p.hidden = new int[]{100, 50, 20};
// p.ignored_cols = new int[]{0,1,2,3,6,7,8,10}; //Optional: ignore all categoricals
// p.ignored_cols = new int[]{4,5,9}; //Optional: ignore all numericals
p.adaptive_rate = true;
p.l1 = 1e-4;
p.activation = DeepLearning.Activation.Tanh;
p.train_samples_per_iteration = -1;
p.loss = DeepLearning.Loss.MeanSquare;
p.epochs = 2;
// p.shuffle_training_data = true;
p.force_load_balance = true;
p.score_training_samples = 0;
p.score_validation_samples = 0;
// p.reproducible = true;
p.invoke();
// Verification of results
StringBuilder sb = new StringBuilder();
sb.append("Verifying results.\n");
DeepLearningModel mymodel = UKV.get(p.dest());
sb.append("Reported mean reconstruction error: " + mymodel.mse() + "\n");
// Training data
// Reconstruct data using the same helper functions and verify that self-reported MSE agrees
final Frame l2 = mymodel.scoreAutoEncoder(train);
final Vec l2vec = l2.anyVec();
sb.append("Actual mean reconstruction error: " + l2vec.mean() + "\n");
// print stats and potential outliers
double quantile = 1 - 5. / train.numRows();
sb.append("The following training points are reconstructed with an error above the "
+ quantile * 100 + "-th percentile - potential \"outliers\" in testing data.\n");
double thresh = mymodel.calcOutlierThreshold(l2vec, quantile);
for (long i = 0; i < l2vec.length(); i++) {
if (l2vec.at(i) > thresh) {
sb.append(String.format("row %d : l2vec error = %5f\n", i, l2vec.at(i)));
}
}
Log.info(sb.toString());
Assert.assertEquals(mymodel.mse(), l2vec.mean(), 1e-8);
// Create reconstruction
Log.info("Creating full reconstruction.");
final Frame recon_train = mymodel.score(train);
// cleanup
recon_train.delete();
train.delete();
p.delete();
mymodel.delete();
l2.delete();
}
开发者ID:h2oai，
项目名称:h2o-2，
代码行数:67，
代码来源:DeepLearningAutoEncoderCategoricalTest.java
示例25: run
点赞 2
import hex.deeplearning.DeepLearningModel; //导入依赖的package包/类
@Test public void run() {
Key file = NFSFileVec.make(find_test_file("smalldata/neural/two_spiral.data"));
Frame frame = ParseDataset2.parse(Key.make(), new Key[]{file});
Key dest = Key.make("spirals2");
for (boolean sparse : new boolean[]{true,false}) {
for (boolean col_major : new boolean[]{false}) {
if (!sparse && col_major) continue;
// build the model
{
DeepLearning p = new DeepLearning();
p.seed = 0xbabe;
p.epochs = 10000;
p.hidden = new int[]{100};
p.sparse = sparse;
p.col_major = col_major;
p.activation = DeepLearning.Activation.Tanh;
p.max_w2 = Float.POSITIVE_INFINITY;
p.l1 = 0;
p.l2 = 0;
p.initial_weight_distribution = DeepLearning.InitialWeightDistribution.Normal;
p.initial_weight_scale = 2.5;
p.loss = DeepLearning.Loss.CrossEntropy;
p.source = frame;
p.response = frame.lastVec();
p.validation = null;
p.score_interval = 2;
p.ignored_cols = null;
p.train_samples_per_iteration = 0; //sync once per period
p.quiet_mode = true;
p.fast_mode = true;
p.ignore_const_cols = true;
p.nesterov_accelerated_gradient = true;
p.classification = true;
p.diagnostics = true;
p.expert_mode = true;
p.score_training_samples = 1000;
p.score_validation_samples = 10000;
p.shuffle_training_data = false;
p.force_load_balance = false;
p.replicate_training_data = false;
p.destination_key = dest;
p.adaptive_rate = true;
p.reproducible = true;
p.rho = 0.99;
p.epsilon = 5e-3;
p.invoke();
}
// score and check result
{
DeepLearningModel mymodel = UKV.get(dest);
double error = mymodel.error();
if (error >= 0.025) {
Assert.fail("Classification error is not less than 0.025, but " + error + ".");
}
mymodel.delete();
mymodel.delete_best_model();
}
}
}
frame.delete();
}
开发者ID:h2oai，
项目名称:h2o-2，
代码行数:67，
代码来源:DeepLearningSpiralsTest.java
示例26: getDeepLearningModel
点赞 1
import hex.deeplearning.DeepLearningModel; //导入依赖的package包/类
/**
* Returns the model
*
* @return model
*/
public DeepLearningModel getDeepLearningModel() {
return this.dlModel;
}
开发者ID:wso2-attic，
项目名称:carbon-ml，
代码行数:9，
代码来源:MLDeeplearningModel.java
示例27: setDeepLearningModel
点赞 1
import hex.deeplearning.DeepLearningModel; //导入依赖的package包/类
/**
* Set the model
*
* @param model model
*/
public void setDeepLearningModel(DeepLearningModel model) {
this.dlModel = model;
}
开发者ID:wso2-attic，
项目名称:carbon-ml，
代码行数:9，
代码来源:MLDeeplearningModel.java
版权声明：本文转自网络文章，转载此文章仅为分享知识，如有侵权，请联系管理员进行删除。
