/**
* Train a Deep Learning neural net model
*
* @param model Input model (e.g., from initModel(), or from a previous training run)
* @return Trained model
*/
public final DeepLearningModel trainModel(DeepLearningModel model) {
Frame validScoreFrame = null;
Frame train, trainScoreFrame;
try {
// if (checkpoint == null && !quiet_mode) logStart(); //if checkpoint is given, some
// Job's params might be uninitialized (but the restarted model's parameters are correct)
if (model == null) {
model = DKV.get(dest()).get();
}
Log.info(
"Model category: "
+ (_parms._autoencoder
? "Auto-Encoder"
: isClassifier() ? "Classification" : "Regression"));
final long model_size = model.model_info().size();
Log.info(
"Number of model parameters (weights/biases): " + String.format("%,d", model_size));
model.write_lock(_job);
_job.update(0, "Setting up training data...");
final DeepLearningParameters mp = model.model_info().get_params();
// temporary frames of the same "name" as the orig _train/_valid (asking the parameter's
// Key, not the actual frame)
// Note: don't put into DKV or they would overwrite the _train/_valid frames!
Frame tra_fr = new Frame(mp._train, _train.names(), _train.vecs());
Frame val_fr = _valid != null ? new Frame(mp._valid, _valid.names(), _valid.vecs()) : null;
train = tra_fr;
if (model._output.isClassifier() && mp._balance_classes) {
_job.update(0, "Balancing class distribution of training data...");
float[] trainSamplingFactors =
new float
[train
.lastVec()
.domain()
.length]; // leave initialized to 0 -> will be filled up below
if (mp._class_sampling_factors != null) {
if (mp._class_sampling_factors.length != train.lastVec().domain().length)
throw new IllegalArgumentException(
"class_sampling_factors must have "
+ train.lastVec().domain().length
+ " elements");
trainSamplingFactors =
mp._class_sampling_factors.clone(); // clone: don't modify the original
}
train =
sampleFrameStratified(
train,
train.lastVec(),
train.vec(model._output.weightsName()),
trainSamplingFactors,
(long) (mp._max_after_balance_size * train.numRows()),
mp._seed,
true,
false);
Vec l = train.lastVec();
Vec w = train.vec(model._output.weightsName());
MRUtils.ClassDist cd = new MRUtils.ClassDist(l);
model._output._modelClassDist =
_weights != null ? cd.doAll(l, w).rel_dist() : cd.doAll(l).rel_dist();
}
model.training_rows = train.numRows();
if (_weights != null && _weights.min() == 0 && _weights.max() == 1 && _weights.isInt()) {
model.training_rows = Math.round(train.numRows() * _weights.mean());
Log.warn(
"Not counting "
+ (train.numRows() - model.training_rows)
+ " rows with weight=0 towards an epoch.");
}
Log.info("One epoch corresponds to " + model.training_rows + " training data rows.");
trainScoreFrame =
sampleFrame(
train,
mp._score_training_samples,
mp._seed); // training scoring dataset is always sampled uniformly from the training
// dataset
if (trainScoreFrame != train) Scope.track(trainScoreFrame);
if (!_parms._quiet_mode)
Log.info("Number of chunks of the training data: " + train.anyVec().nChunks());
if (val_fr != null) {
model.validation_rows = val_fr.numRows();
// validation scoring dataset can be sampled in multiple ways from the given validation
// dataset
if (model._output.isClassifier()
&& mp._balance_classes
&& mp._score_validation_sampling
== DeepLearningParameters.ClassSamplingMethod.Stratified) {
_job.update(0, "Sampling validation data (stratified)...");
validScoreFrame =
sampleFrameStratified(
val_fr,
val_fr.lastVec(),
val_fr.vec(model._output.weightsName()),
null,
mp._score_validation_samples > 0
? mp._score_validation_samples
: val_fr.numRows(),
mp._seed + 1,
false /* no oversampling */,
false);
} else {
_job.update(0, "Sampling validation data...");
validScoreFrame = sampleFrame(val_fr, mp._score_validation_samples, mp._seed + 1);
if (validScoreFrame != val_fr) Scope.track(validScoreFrame);
}
if (!_parms._quiet_mode)
Log.info(
"Number of chunks of the validation data: " + validScoreFrame.anyVec().nChunks());
}
// Set train_samples_per_iteration size (cannot be done earlier since this depends on
// whether stratified sampling is done)
model.actual_train_samples_per_iteration =
computeTrainSamplesPerIteration(mp, model.training_rows, model);
// Determine whether shuffling is enforced
if (mp._replicate_training_data
&& (model.actual_train_samples_per_iteration
== model.training_rows * (mp._single_node_mode ? 1 : H2O.CLOUD.size()))
&& !mp._shuffle_training_data
&& H2O.CLOUD.size() > 1
&& !mp._reproducible) {
if (!mp._quiet_mode)
Log.info(
"Enabling training data shuffling, because all nodes train on the full dataset (replicated training data).");
mp._shuffle_training_data = true;
}
if (!mp._shuffle_training_data
&& model.actual_train_samples_per_iteration == model.training_rows
&& train.anyVec().nChunks() == 1) {
if (!mp._quiet_mode)
Log.info(
"Enabling training data shuffling to avoid training rows in the same order over and over (no Hogwild since there's only 1 chunk).");
mp._shuffle_training_data = true;
}
// if (!mp._quiet_mode) Log.info("Initial model:\n" + model.model_info());
long now = System.currentTimeMillis();
model._timeLastIterationEnter = now;
if (_parms._autoencoder) {
_job.update(0, "Scoring null model of autoencoder...");
if (!mp._quiet_mode) Log.info("Scoring the null model of the autoencoder.");
model.doScoring(
trainScoreFrame,
validScoreFrame,
_job._key,
0,
false); // get the null model reconstruction error
}
// put the initial version of the model into DKV
model.update(_job);
model.total_setup_time_ms += now - _job.start_time();
Log.info("Total setup time: " + PrettyPrint.msecs(model.total_setup_time_ms, true));
Log.info("Starting to train the Deep Learning model.");
_job.update(0, "Training...");
// main loop
for (; ; ) {
model.iterations++;
model.set_model_info(
mp._epochs == 0
? model.model_info()
: H2O.CLOUD.size() > 1 && mp._replicate_training_data
? (mp._single_node_mode
? new DeepLearningTask2(
_job._key,
train,
model.model_info(),
rowFraction(train, mp, model),
model.iterations)
.doAll(Key.make(H2O.SELF))
.model_info()
: // replicated data + single node mode
new DeepLearningTask2(
_job._key,
train,
model.model_info(),
rowFraction(train, mp, model),
model.iterations)
.doAllNodes()
.model_info())
: // replicated data + multi-node mode
new DeepLearningTask(
_job._key,
model.model_info(),
rowFraction(train, mp, model),
model.iterations)
.doAll(train)
.model_info()); // distributed data (always in multi-node mode)
if (stop_requested() && !timeout()) break; // cancellation
if (!model.doScoring(
trainScoreFrame, validScoreFrame, _job._key, model.iterations, false))
break; // finished training (or early stopping or convergence)
if (timeout()) break; // stop after scoring
}
// replace the model with the best model so far (if it's better)
if (!stop_requested()
&& _parms._overwrite_with_best_model
&& model.actual_best_model_key != null
&& _parms._nfolds == 0) {
DeepLearningModel best_model = DKV.getGet(model.actual_best_model_key);
if (best_model != null
&& best_model.loss() < model.loss()
&& Arrays.equals(best_model.model_info().units, model.model_info().units)) {
if (!_parms._quiet_mode)
Log.info("Setting the model to be the best model so far (based on scoring history).");
DeepLearningModelInfo mi = best_model.model_info().deep_clone();
// Don't cheat - count full amount of training samples, since that's the amount of
// training it took to train (without finding anything better)
mi.set_processed_global(model.model_info().get_processed_global());
mi.set_processed_local(model.model_info().get_processed_local());
model.set_model_info(mi);
model.update(_job);
model.doScoring(trainScoreFrame, validScoreFrame, _job._key, model.iterations, true);
assert (best_model.loss() == model.loss());
}
}
// store coefficient names for future use
// possibly change
model.model_info().data_info().coefNames();
if (!_parms._quiet_mode) {
Log.info(
"==============================================================================================================================================================================");
if (stop_requested()) {
Log.info("Deep Learning model training was interrupted.");
} else {
Log.info("Finished training the Deep Learning model.");
Log.info(model);
}
Log.info(
"==============================================================================================================================================================================");
}
} finally {
if (model != null) {
model.deleteElasticAverageModels();
model.unlock(_job);
if (model.actual_best_model_key != null) {
assert (model.actual_best_model_key != model._key);
DKV.remove(model.actual_best_model_key);
}
}
}
return model;
}
@Override
protected void compute2() {
_model = null; // Resulting model!
try {
Scope.enter(); // Cleanup temp keys
init(true); // Do any expensive tests & conversions now
// Do lock even before checking the errors, since this block is finalized by unlock
// (not the best solution, but the code is more readable)
_parms.read_lock_frames(SharedTree.this); // Fetch & read-lock input frames
if (error_count() > 0)
throw H2OModelBuilderIllegalArgumentException.makeFromBuilder(SharedTree.this);
// New Model? Or continuing from a checkpoint?
if (_parms._checkpoint && DKV.get(_parms._model_id) != null) {
_model = DKV.get(_dest).get();
_model.write_lock(_key); // do not delete previous model; we are extending it
} else { // New Model
// Compute the zero-tree error - guessing only the class distribution.
// MSE is stddev squared when guessing for regression.
// For classification, guess the largest class.
_model =
makeModel(
_dest,
_parms,
initial_MSE(_response, _response),
_valid == null
? Double.NaN
: initial_MSE(_response, _vresponse)); // Make a fresh model
_model.delete_and_lock(_key); // and clear & write-lock it (smashing any prior)
_model._output._init_f = _initialPrediction;
}
// Compute the response domain; makes for nicer printouts
String[] domain = _response.domain();
assert (_nclass > 1 && domain != null) || (_nclass == 1 && domain == null);
if (_nclass == 1) domain = new String[] {"r"}; // For regression, give a name to class 0
// Compute class distribution, used to for initial guesses and to
// upsample minority classes (if asked for).
if (_nclass > 1) { // Classification?
// Handle imbalanced classes by stratified over/under-sampling.
// initWorkFrame sets the modeled class distribution, and
// model.score() corrects the probabilities back using the
// distribution ratios
if (_model._output.isClassifier() && _parms._balance_classes) {
float[] trainSamplingFactors =
new float
[_train
.lastVec()
.domain()
.length]; // leave initialized to 0 -> will be filled up below
if (_parms._class_sampling_factors != null) {
if (_parms._class_sampling_factors.length != _train.lastVec().domain().length)
throw new IllegalArgumentException(
"class_sampling_factors must have "
+ _train.lastVec().domain().length
+ " elements");
trainSamplingFactors =
_parms._class_sampling_factors.clone(); // clone: don't modify the original
}
Frame stratified =
water.util.MRUtils.sampleFrameStratified(
_train,
_train.lastVec(),
_train.vec(_model._output.weightsName()),
trainSamplingFactors,
(long) (_parms._max_after_balance_size * _train.numRows()),
_parms._seed,
true,
false);
if (stratified != _train) {
_train = stratified;
_response = stratified.vec(_parms._response_column);
_weights = stratified.vec(_parms._weights_column);
// Recompute distribution since the input frame was modified
MRUtils.ClassDist cdmt2 =
_weights != null
? new MRUtils.ClassDist(_nclass).doAll(_response, _weights)
: new MRUtils.ClassDist(_nclass).doAll(_response);
_model._output._distribution = cdmt2.dist();
_model._output._modelClassDist = cdmt2.rel_dist();
}
}
Log.info("Prior class distribution: " + Arrays.toString(_model._output._priorClassDist));
Log.info("Model class distribution: " + Arrays.toString(_model._output._modelClassDist));
}
// Also add to the basic working Frame these sets:
// nclass Vecs of current forest results (sum across all trees)
// nclass Vecs of working/temp data
// nclass Vecs of NIDs, allowing 1 tree per class
// Current forest values: results of summing the prior M trees
for (int i = 0; i < _nclass; i++) _train.add("Tree_" + domain[i], _response.makeZero());
// Initial work columns. Set-before-use in the algos.
for (int i = 0; i < _nclass; i++) _train.add("Work_" + domain[i], _response.makeZero());
// One Tree per class, each tree needs a NIDs. For empty classes use a -1
// NID signifying an empty regression tree.
for (int i = 0; i < _nclass; i++)
_train.add(
"NIDs_" + domain[i],
_response.makeCon(
_model._output._distribution == null
? 0
: (_model._output._distribution[i] == 0 ? -1 : 0)));
// Tag out rows missing the response column
new ExcludeNAResponse().doAll(_train);
// Variable importance: squared-error-improvement-per-variable-per-split
_improvPerVar = new float[_ncols];
// Sub-class tree-model-builder specific build code
buildModel();
done(); // Job done!
} catch (Throwable t) {
Job thisJob = DKV.getGet(_key);
if (thisJob._state == JobState.CANCELLED) {
Log.info("Job cancelled by user.");
} else {
t.printStackTrace();
failed(t);
throw t;
}
} finally {
if (_model != null) _model.unlock(_key);
_parms.read_unlock_frames(SharedTree.this);
if (_model == null) Scope.exit();
else {
Scope.exit(
_model._key,
ModelMetrics.buildKey(_model, _parms.train()),
ModelMetrics.buildKey(_model, _parms.valid()));
}
}
tryComplete();
}