def testCachedPredictionOnEmptyEnsemble(self):
"""Tests that prediction on a dummy ensemble does not fail."""
with self.test_session() as session:
# Create a dummy ensemble.
tree_ensemble = boosted_trees_ops.TreeEnsemble('ensemble',
serialized_proto='')
tree_ensemble_handle = tree_ensemble.resource_handle
resources.initialize_resources(resources.shared_resources()).run()
# No previous cached values.
cached_tree_ids = [0, 0]
cached_node_ids = [0, 0]
# We have two features: 0 and 1. Values don't matter here on a dummy
# ensemble.
feature_0_values = [67, 5]
feature_1_values = [9, 17]
# Grow tree ensemble.
predict_op = boosted_trees_ops.training_predict(
tree_ensemble_handle,
cached_tree_ids=cached_tree_ids,
cached_node_ids=cached_node_ids,
bucketized_features=[feature_0_values, feature_1_values],
logits_dimension=1)
logits_updates, new_tree_ids, new_node_ids = session.run(
predict_op)
# Nothing changed.
self.assertAllClose(cached_tree_ids, new_tree_ids)
self.assertAllClose(cached_node_ids, new_node_ids)
self.assertAllClose([[0], [0]], logits_updates)
def testContribsForOnlyABiasNode(self):
"""Tests case when, after training, only left with a bias node.
For example, this could happen if the final ensemble contains one tree that
got pruned up to the root.
"""
with self.test_session() as session:
tree_ensemble_config = boosted_trees_pb2.TreeEnsemble()
text_format.Merge(
"""
trees {
nodes {
leaf {
scalar: 1.72
}
}
}
tree_weights: 0.1
tree_metadata: {
num_layers_grown: 0
}
""", tree_ensemble_config)
tree_ensemble = boosted_trees_ops.TreeEnsemble(
'ensemble',
serialized_proto=tree_ensemble_config.SerializeToString())
tree_ensemble_handle = tree_ensemble.resource_handle
resources.initialize_resources(resources.shared_resources()).run()
# All features are unused.
feature_0_values = [36, 32]
feature_1_values = [13, -29]
feature_2_values = [11, 27]
# Expected logits are computed by traversing the logit path and
# subtracting child logits from parent logits.
bias = 1.72 * 0.1 # Root node of tree_0.
expected_feature_ids = ((), ())
expected_logits_paths = ((bias, ), (bias, ))
bucketized_features = [
feature_0_values, feature_1_values, feature_2_values
]
debug_op = boosted_trees_ops.example_debug_outputs(
tree_ensemble_handle,
bucketized_features=bucketized_features,
logits_dimension=1)
serialized_examples_debug_outputs = session.run(debug_op)
feature_ids = []
logits_paths = []
for example in serialized_examples_debug_outputs:
example_debug_outputs = boosted_trees_pb2.DebugOutput()
example_debug_outputs.ParseFromString(example)
feature_ids.append(example_debug_outputs.feature_ids)
logits_paths.append(example_debug_outputs.logits_path)
self.assertAllClose(feature_ids, expected_feature_ids)
self.assertAllClose(logits_paths, expected_logits_paths)
def testCreate(self):
with self.test_session():
ensemble = boosted_trees_ops.TreeEnsemble('ensemble')
resources.initialize_resources(resources.shared_resources()).run()
stamp_token = ensemble.get_stamp_token()
self.assertEqual(0, stamp_token.eval())
(_, num_trees, num_finalized_trees,
num_attempted_layers) = ensemble.get_states()
self.assertEqual(0, num_trees.eval())
self.assertEqual(0, num_finalized_trees.eval())
self.assertEqual(0, num_attempted_layers.eval())
def testCreate(self):
with self.cached_session():
ensemble = boosted_trees_ops.TreeEnsemble('ensemble')
resources.initialize_resources(resources.shared_resources()).run()
stamp_token = ensemble.get_stamp_token()
self.assertEqual(0, self.evaluate(stamp_token))
(_, num_trees, num_finalized_trees, num_attempted_layers,
nodes_range) = ensemble.get_states()
self.assertEqual(0, self.evaluate(num_trees))
self.assertEqual(0, self.evaluate(num_finalized_trees))
self.assertEqual(0, self.evaluate(num_attempted_layers))
self.assertAllEqual([0, 1], self.evaluate(nodes_range))
def testPredictionOnEmptyEnsemble(self):
"""Tests that prediction on a empty ensemble does not fail."""
with self.cached_session() as session:
# Create an empty ensemble.
tree_ensemble = boosted_trees_ops.TreeEnsemble('ensemble',
serialized_proto='')
tree_ensemble_handle = tree_ensemble.resource_handle
resources.initialize_resources(resources.shared_resources()).run()
feature_0_values = [36, 32]
feature_1_values = [11, 27]
expected_logits = [[0.0], [0.0]]
# Prediction should work fine.
predict_op = boosted_trees_ops.predict(
tree_ensemble_handle,
bucketized_features=[feature_0_values, feature_1_values],
logits_dimension=1)
logits = session.run(predict_op)
self.assertAllClose(expected_logits, logits)
def testPredictionMultipleTreeMultiClass(self):
"""Tests the predictions work when we have multiple trees."""
with self.cached_session() as session:
tree_ensemble_config = boosted_trees_pb2.TreeEnsemble()
text_format.Merge(
"""
trees {
nodes {
bucketized_split {
feature_id: 1
threshold: 28
left_id: 1
right_id: 2
}
metadata {
gain: 7.62
}
}
nodes {
leaf {
vector: {
value: 0.51
}
vector: {
value: 1.14
}
}
}
nodes {
leaf {
vector: {
value: 1.29
}
vector: {
value: 8.79
}
}
}
}
trees {
nodes {
bucketized_split {
feature_id: 1
threshold: 26
left_id: 1
right_id: 2
}
}
nodes {
bucketized_split {
feature_id: 0
threshold: 50
left_id: 3
right_id: 4
}
}
nodes {
leaf {
vector: {
value: -4.33
}
vector: {
value: 7.0
}
}
}
nodes {
leaf {
vector: {
value: 0.2
}
vector: {
value: 5.0
}
}
}
nodes {
leaf {
vector: {
value: -4.1
}
vector: {
value: 6.0
}
}
}
}
trees {
nodes {
bucketized_split {
feature_id: 0
threshold: 34
left_id: 1
right_id: 2
}
}
nodes {
leaf {
vector: {
value: 2.0
}
vector: {
value: -7.0
}
}
}
nodes {
leaf {
vector: {
value: 6.3
}
vector: {
value: 5.0
}
}
}
}
tree_weights: 0.1
tree_weights: 0.2
tree_weights: 1.0
""", tree_ensemble_config)
# Create existing ensemble with one root split
tree_ensemble = boosted_trees_ops.TreeEnsemble(
'ensemble',
serialized_proto=tree_ensemble_config.SerializeToString())
tree_ensemble_handle = tree_ensemble.resource_handle
resources.initialize_resources(resources.shared_resources()).run()
feature_0_values = [36, 32]
feature_1_values = [11, 27]
# Example 1: tree 0: (0.51, 1.14), tree 1: (0.2, 5.0), tree 2: (6.3, 5.0)
#
# logits = (0.1*0.51+0.2*0.2+1*6.3,
# 0.1*1.14+0.2*5.0+1*5)
# Example 2: tree 0: (0.51, 1.14), tree 1: (-4.33, 7.0), tree 2: (2.0, -7)
#
# logits = (0.1*0.51+0.2*-4.33+1*2.0,
# 0.1*1.14+0.2*7.0+1*-7)
logits_dimension = 2
expected_logits = [[6.391, 6.114], [1.185, -5.486]]
# Prediction should work fine.
predict_op = boosted_trees_ops.predict(
tree_ensemble_handle,
bucketized_features=[feature_0_values, feature_1_values],
logits_dimension=logits_dimension)
logits = session.run(predict_op)
self.assertAllClose(expected_logits, logits)
def testMetadataWhenCantSplitDueToEmptySplits(self):
"""Test that the metadata is updated even though we can't split."""
with self.test_session() as session:
tree_ensemble_config = boosted_trees_pb2.TreeEnsemble()
text_format.Merge("""
trees {
nodes {
bucketized_split {
feature_id: 4
left_id: 1
right_id: 2
}
metadata {
gain: 7.62
}
}
nodes {
leaf {
scalar: 0.714
}
}
nodes {
leaf {
scalar: -0.4375
}
}
}
tree_weights: 1.0
tree_metadata {
num_layers_grown: 1
is_finalized: false
}
growing_metadata {
num_trees_attempted: 1
num_layers_attempted: 1
}
""", tree_ensemble_config)
# Create existing ensemble with one root split
tree_ensemble = boosted_trees_ops.TreeEnsemble(
'ensemble', serialized_proto=tree_ensemble_config.SerializeToString())
tree_ensemble_handle = tree_ensemble.resource_handle
resources.initialize_resources(resources.shared_resources()).run()
# Prepare feature inputs.
# feature 1 only has a candidate for node 1, feature 2 has candidates
# for both nodes and feature 3 only has a candidate for node 2.
# Grow tree ensemble.
grow_op = boosted_trees_ops.update_ensemble(
tree_ensemble_handle,
learning_rate=0.1,
pruning_mode=boosted_trees_ops.PruningMode.NO_PRUNING,
max_depth=2,
# No splits are available.
feature_ids=[],
node_ids=[],
gains=[],
thresholds=[],
left_node_contribs=[],
right_node_contribs=[])
session.run(grow_op)
# Expect no new splits created, but attempted (global) stats updated. Meta
# data for this tree should not be updated (we didn't succeed building a
# layer.
new_stamp, serialized = session.run(tree_ensemble.serialize())
tree_ensemble = boosted_trees_pb2.TreeEnsemble()
tree_ensemble.ParseFromString(serialized)
expected_result = """
trees {
nodes {
bucketized_split {
feature_id: 4
left_id: 1
right_id: 2
}
metadata {
gain: 7.62
}
}
nodes {
leaf {
scalar: 0.714
}
}
nodes {
leaf {
scalar: -0.4375
}
}
}
tree_weights: 1.0
tree_metadata {
num_layers_grown: 1
is_finalized: false
}
growing_metadata {
num_trees_attempted: 1
num_layers_attempted: 2
}
"""
self.assertEqual(new_stamp, 1)
self.assertProtoEquals(expected_result, tree_ensemble)
def testCachedPredictionTheWholeTreeWasPruned(self):
"""Tests that prediction based on previous node in the tree works."""
with self.test_session() as session:
tree_ensemble_config = boosted_trees_pb2.TreeEnsemble()
text_format.Merge(
"""
trees {
nodes {
leaf {
scalar: 0.00
}
}
}
tree_weights: 1.0
tree_metadata {
num_layers_grown: 1
is_finalized: true
post_pruned_nodes_meta {
new_node_id: 0
logit_change: 0.0
}
post_pruned_nodes_meta {
new_node_id: 0
logit_change: -6.0
}
post_pruned_nodes_meta {
new_node_id: 0
logit_change: 5.0
}
}
growing_metadata {
num_trees_attempted: 1
num_layers_attempted: 1
}
""", tree_ensemble_config)
# Create existing ensemble.
tree_ensemble = boosted_trees_ops.TreeEnsemble(
'ensemble',
serialized_proto=tree_ensemble_config.SerializeToString())
tree_ensemble_handle = tree_ensemble.resource_handle
resources.initialize_resources(resources.shared_resources()).run()
cached_tree_ids = [
0,
0,
]
# The predictions were cached in 1 and 2, both were pruned to the root.
cached_node_ids = [1, 2]
# We have two features: 0 and 1.These are not going to be used anywhere.
feature_0_values = [12, 17]
feature_1_values = [12, 12]
# Grow tree ensemble.
predict_op = boosted_trees_ops.training_predict(
tree_ensemble_handle,
cached_tree_ids=cached_tree_ids,
cached_node_ids=cached_node_ids,
bucketized_features=[feature_0_values, feature_1_values],
logits_dimension=1)
logits_updates, new_tree_ids, new_node_ids = session.run(
predict_op)
# We are in the last tree.
self.assertAllClose([0, 0], new_tree_ids)
self.assertAllClose([0, 0], new_node_ids)
self.assertAllClose([[-6.0], [5.0]], logits_updates)
def _bt_model_fn(
features,
labels,
mode,
head,
feature_columns,
tree_hparams,
n_batches_per_layer,
config,
closed_form_grad_and_hess_fn=None,
example_id_column_name=None,
# TODO(youngheek): replace this later using other options.
train_in_memory=False,
name='TreeEnsembleModel'):
"""Gradient Boosted Decision Tree model_fn.
Args:
features: dict of `Tensor`.
labels: `Tensor` of shape [batch_size, 1] or [batch_size] labels of
dtype `int32` or `int64` in the range `[0, n_classes)`.
mode: Defines whether this is training, evaluation or prediction.
See `ModeKeys`.
head: A `head_lib._Head` instance.
feature_columns: Iterable of `feature_column._FeatureColumn` model inputs.
tree_hparams: TODO. collections.namedtuple for hyper parameters.
n_batches_per_layer: A `Tensor` of `int64`. Each layer is built after at
least n_batches_per_layer accumulations.
config: `RunConfig` object to configure the runtime settings.
closed_form_grad_and_hess_fn: a function that accepts logits and labels
and returns gradients and hessians. By default, they are created by
tf.gradients() from the loss.
example_id_column_name: Name of the feature for a unique ID per example.
Currently experimental -- not exposed to public API.
train_in_memory: `bool`, when true, it assumes the dataset is in memory,
i.e., input_fn should return the entire dataset as a single batch, and
also n_batches_per_layer should be set as 1.
name: Name to use for the model.
Returns:
An `EstimatorSpec` instance.
Raises:
ValueError: mode or params are invalid, or features has the wrong type.
"""
is_single_machine = (config.num_worker_replicas == 1)
if train_in_memory:
assert n_batches_per_layer == 1, (
'When train_in_memory is enabled, input_fn should return the entire '
'dataset as a single batch, and n_batches_per_layer should be set as '
'1.')
worker_device = control_flow_ops.no_op().device
# maximum number of splits possible in the whole tree =2^(D-1)-1
# TODO(youngheek): perhaps storage could be optimized by storing stats with
# the dimension max_splits_per_layer, instead of max_splits (for the entire
# tree).
max_splits = (1 << tree_hparams.max_depth) - 1
with ops.name_scope(name) as name:
# Prepare.
global_step = training_util.get_or_create_global_step()
input_feature_list, num_buckets = _get_transformed_features(
features, feature_columns)
if train_in_memory and mode == model_fn.ModeKeys.TRAIN:
input_feature_list = [
_keep_as_local_variable(feature)
for feature in input_feature_list
]
num_features = len(input_feature_list)
cache = None
if mode == model_fn.ModeKeys.TRAIN:
if train_in_memory and is_single_machine: # maybe just train_in_memory?
batch_size = array_ops.shape(input_feature_list[0])[0]
cache = _CacheTrainingStatesUsingVariables(
batch_size, head.logits_dimension)
elif example_id_column_name:
example_ids = features[example_id_column_name]
cache = _CacheTrainingStatesUsingHashTable(
example_ids, head.logits_dimension)
# Create Ensemble resources.
if is_single_machine:
tree_ensemble = boosted_trees_ops.TreeEnsemble(name=name)
local_tree_ensemble = tree_ensemble
ensemble_reload = control_flow_ops.no_op()
else:
tree_ensemble = boosted_trees_ops.TreeEnsemble(name=name)
with ops.device(worker_device):
local_tree_ensemble = boosted_trees_ops.TreeEnsemble(
name=name + '_local', is_local=True)
# TODO(soroush): Do partial updates if this becomes a bottleneck.
ensemble_reload = local_tree_ensemble.deserialize(
*tree_ensemble.serialize())
# Create logits.
if mode != model_fn.ModeKeys.TRAIN:
logits = boosted_trees_ops.predict(
tree_ensemble_handle=local_tree_ensemble.resource_handle,
bucketized_features=input_feature_list,
logits_dimension=head.logits_dimension,
max_depth=tree_hparams.max_depth)
else:
if cache:
cached_tree_ids, cached_node_ids, cached_logits = cache.lookup(
)
else:
# Always start from the beginning when no cache is set up.
batch_size = array_ops.shape(input_feature_list[0])[0]
cached_tree_ids, cached_node_ids, cached_logits = (
array_ops.zeros([batch_size], dtype=dtypes.int32),
array_ops.zeros([batch_size], dtype=dtypes.int32),
array_ops.zeros([batch_size, head.logits_dimension],
dtype=dtypes.float32))
with ops.control_dependencies([ensemble_reload]):
(stamp_token, num_trees, num_finalized_trees,
num_attempted_layers) = local_tree_ensemble.get_states()
summary.scalar('ensemble/num_trees', num_trees)
summary.scalar('ensemble/num_finalized_trees',
num_finalized_trees)
summary.scalar('ensemble/num_attempted_layers',
num_attempted_layers)
partial_logits, tree_ids, node_ids = boosted_trees_ops.training_predict(
tree_ensemble_handle=local_tree_ensemble.resource_handle,
cached_tree_ids=cached_tree_ids,
cached_node_ids=cached_node_ids,
bucketized_features=input_feature_list,
logits_dimension=head.logits_dimension,
max_depth=tree_hparams.max_depth)
logits = cached_logits + partial_logits
# Create training graph.
def _train_op_fn(loss):
"""Run one training iteration."""
train_op = []
if cache:
train_op.append(cache.insert(tree_ids, node_ids, logits))
if closed_form_grad_and_hess_fn:
gradients, hessians = closed_form_grad_and_hess_fn(
logits, labels)
else:
gradients = gradients_impl.gradients(loss,
logits,
name='Gradients')[0]
hessians = gradients_impl.gradients(gradients,
logits,
name='Hessians')[0]
stats_summary_list = [
array_ops.squeeze(boosted_trees_ops.make_stats_summary(
node_ids=node_ids,
gradients=gradients,
hessians=hessians,
bucketized_features_list=[input_feature_list[f]],
max_splits=max_splits,
num_buckets=num_buckets),
axis=0) for f in range(num_features)
]
def grow_tree_from_stats_summaries(stats_summary_list):
"""Updates ensemble based on the best gains from stats summaries."""
(node_ids_per_feature, gains_list, thresholds_list,
left_node_contribs_list, right_node_contribs_list) = (
boosted_trees_ops.calculate_best_gains_per_feature(
node_id_range=array_ops.stack([
math_ops.reduce_min(node_ids),
math_ops.reduce_max(node_ids)
]),
stats_summary_list=stats_summary_list,
l1=tree_hparams.l1,
l2=tree_hparams.l2,
tree_complexity=tree_hparams.tree_complexity,
max_splits=max_splits))
grow_op = boosted_trees_ops.update_ensemble(
# Confirm if local_tree_ensemble or tree_ensemble should be used.
tree_ensemble.resource_handle,
feature_ids=math_ops.range(0,
num_features,
dtype=dtypes.int32),
node_ids=node_ids_per_feature,
gains=gains_list,
thresholds=thresholds_list,
left_node_contribs=left_node_contribs_list,
right_node_contribs=right_node_contribs_list,
learning_rate=tree_hparams.learning_rate,
max_depth=tree_hparams.max_depth,
pruning_mode=boosted_trees_ops.PruningMode.NO_PRUNING)
return grow_op
if train_in_memory and is_single_machine:
train_op.append(state_ops.assign_add(global_step, 1))
train_op.append(
grow_tree_from_stats_summaries(stats_summary_list))
else:
summary_accumulator = data_flow_ops.ConditionalAccumulator(
dtype=dtypes.float32,
# The stats consist of gradients and hessians (the last dimension).
shape=[num_features, max_splits, num_buckets, 2],
shared_name='stats_summary_accumulator')
apply_grad = summary_accumulator.apply_grad(
array_ops.stack(stats_summary_list, axis=0), stamp_token)
def grow_tree_from_accumulated_summaries_fn():
"""Updates the tree with the best layer from accumulated summaries."""
# Take out the accumulated summaries from the accumulator and grow.
stats_summary_list = array_ops.unstack(
summary_accumulator.take_grad(1), axis=0)
grow_op = grow_tree_from_stats_summaries(
stats_summary_list)
return grow_op
with ops.control_dependencies([apply_grad]):
train_op.append(state_ops.assign_add(global_step, 1))
if config.is_chief:
train_op.append(
control_flow_ops.cond(
math_ops.greater_equal(
summary_accumulator.num_accumulated(),
n_batches_per_layer),
grow_tree_from_accumulated_summaries_fn,
control_flow_ops.no_op,
name='wait_until_n_batches_accumulated'))
return control_flow_ops.group(train_op, name='train_op')
estimator_spec = head.create_estimator_spec(features=features,
mode=mode,
labels=labels,
train_op_fn=_train_op_fn,
logits=logits)
if mode == model_fn.ModeKeys.TRAIN:
# Add an early stop hook.
estimator_spec = estimator_spec._replace(
training_hooks=estimator_spec.training_hooks +
(StopAtNumTreesHook(num_trees, tree_hparams.n_trees), ))
return estimator_spec
def testPostPruningOfAllNodes(self):
"""Test growing an ensemble with post-pruning, with all nodes are pruned."""
with self.test_session() as session:
# Create empty ensemble.
# Create empty ensemble.
tree_ensemble_config = boosted_trees_pb2.TreeEnsemble()
tree_ensemble = boosted_trees_ops.TreeEnsemble(
'ensemble', serialized_proto=tree_ensemble_config.SerializeToString())
tree_ensemble_handle = tree_ensemble.resource_handle
resources.initialize_resources(resources.shared_resources()).run()
# Prepare inputs. All have negative gains.
feature_ids = [0, 1]
feature1_nodes = np.array([0], dtype=np.int32)
feature1_gains = np.array([-1.3], dtype=np.float32)
feature1_thresholds = np.array([7], dtype=np.int32)
feature1_left_node_contribs = np.array([[0.013]], dtype=np.float32)
feature1_right_node_contribs = np.array([[0.0143]], dtype=np.float32)
feature2_nodes = np.array([0], dtype=np.int32)
feature2_gains = np.array([-0.62], dtype=np.float32)
feature2_thresholds = np.array([33], dtype=np.int32)
feature2_left_node_contribs = np.array([[0.01]], dtype=np.float32)
feature2_right_node_contribs = np.array([[0.0143]], dtype=np.float32)
# Grow tree ensemble.
grow_op = boosted_trees_ops.update_ensemble(
tree_ensemble_handle,
learning_rate=1.0,
pruning_mode=boosted_trees_ops.PruningMode.POST_PRUNING,
max_depth=2,
feature_ids=feature_ids,
node_ids=[feature1_nodes, feature2_nodes],
gains=[feature1_gains, feature2_gains],
thresholds=[feature1_thresholds, feature2_thresholds],
left_node_contribs=[
feature1_left_node_contribs, feature2_left_node_contribs
],
right_node_contribs=[
feature1_right_node_contribs, feature2_right_node_contribs
])
session.run(grow_op)
# Expect the split from feature 2 to be chosen despite the negative gain.
# The grown tree should not be finalized as max tree depth is 2 so no
# pruning occurs.
new_stamp, serialized = session.run(tree_ensemble.serialize())
res_ensemble = boosted_trees_pb2.TreeEnsemble()
res_ensemble.ParseFromString(serialized)
expected_result = """
trees {
nodes {
bucketized_split {
feature_id: 1
threshold: 33
left_id: 1
right_id: 2
}
metadata {
gain: -0.62
}
}
nodes {
leaf {
scalar: 0.01
}
}
nodes {
leaf {
scalar: 0.0143
}
}
}
tree_weights: 1.0
tree_metadata {
num_layers_grown: 1
}
growing_metadata {
num_trees_attempted: 1
num_layers_attempted: 1
}
"""
self.assertEqual(new_stamp, 1)
self.assertProtoEquals(expected_result, res_ensemble)
# Prepare inputs.
# All have negative gain.
feature_ids = [3]
feature1_nodes = np.array([1, 2], dtype=np.int32)
feature1_gains = np.array([-0.2, -0.5], dtype=np.float32)
feature1_thresholds = np.array([77, 79], dtype=np.int32)
feature1_left_node_contribs = np.array([[0.023], [0.3]], dtype=np.float32)
feature1_right_node_contribs = np.array(
[[0.012343], [24]], dtype=np.float32)
grow_op = boosted_trees_ops.update_ensemble(
tree_ensemble_handle,
learning_rate=1.0,
pruning_mode=boosted_trees_ops.PruningMode.POST_PRUNING,
max_depth=2,
feature_ids=feature_ids,
node_ids=[feature1_nodes],
gains=[feature1_gains],
thresholds=[feature1_thresholds],
left_node_contribs=[feature1_left_node_contribs],
right_node_contribs=[feature1_right_node_contribs])
session.run(grow_op)
# Expect the split from feature 1 to be chosen despite the negative gain.
# The grown tree should be finalized. Since all nodes have negative gain,
# the whole tree is pruned.
new_stamp, serialized = session.run(tree_ensemble.serialize())
res_ensemble = boosted_trees_pb2.TreeEnsemble()
res_ensemble.ParseFromString(serialized)
# Expect the ensemble to be empty as post-pruning will prune
# the entire finalized tree.
self.assertEqual(new_stamp, 2)
self.assertProtoEquals("""
trees {
nodes {
leaf {
}
}
}
trees {
nodes {
leaf {
}
}
}
tree_weights: 1.0
tree_weights: 1.0
tree_metadata{
num_layers_grown: 2
is_finalized: true
post_pruned_nodes_meta {
new_node_id: 0
logit_change: 0.0
}
post_pruned_nodes_meta {
new_node_id: 0
logit_change: -0.01
}
post_pruned_nodes_meta {
new_node_id: 0
logit_change: -0.0143
}
post_pruned_nodes_meta {
new_node_id: 0
logit_change: -0.033
}
post_pruned_nodes_meta {
new_node_id: 0
logit_change: -0.022343
}
post_pruned_nodes_meta {
new_node_id: 0
logit_change: -0.3143
}
post_pruned_nodes_meta {
new_node_id: 0
logit_change: -24.0143
}
}
tree_metadata {
}
growing_metadata {
num_trees_attempted: 1
num_layers_attempted: 2
}
""", res_ensemble)
def testCachedPredictionFromThePreviousTreeWithPostPrunedNodes(self):
"""Tests that prediction based on previous node in the tree works."""
with self.test_session() as session:
tree_ensemble_config = boosted_trees_pb2.TreeEnsemble()
text_format.Merge(
"""
trees {
nodes {
bucketized_split {
feature_id:0
threshold: 33
left_id: 1
right_id: 2
}
metadata {
gain: -0.2
}
}
nodes {
leaf {
scalar: 0.01
}
}
nodes {
bucketized_split {
feature_id: 1
threshold: 5
left_id: 3
right_id: 4
}
metadata {
gain: 0.5
original_leaf {
scalar: 0.0143
}
}
}
nodes {
leaf {
scalar: 0.0553
}
}
nodes {
leaf {
scalar: 0.0783
}
}
}
trees {
nodes {
leaf {
scalar: 0.55
}
}
}
tree_weights: 1.0
tree_weights: 1.0
tree_metadata {
num_layers_grown: 3
is_finalized: true
post_pruned_nodes_meta {
new_node_id: 0
logit_change: 0.0
}
post_pruned_nodes_meta {
new_node_id: 1
logit_change: 0.0
}
post_pruned_nodes_meta {
new_node_id: 2
logit_change: 0.0
}
post_pruned_nodes_meta {
new_node_id: 1
logit_change: -0.07
}
post_pruned_nodes_meta {
new_node_id: 1
logit_change: -0.083
}
post_pruned_nodes_meta {
new_node_id: 3
logit_change: 0.0
}
post_pruned_nodes_meta {
new_node_id: 4
logit_change: 0.0
}
post_pruned_nodes_meta {
new_node_id: 1
logit_change: -0.22
}
post_pruned_nodes_meta {
new_node_id: 1
logit_change: -0.57
}
}
tree_metadata {
num_layers_grown: 1
is_finalized: false
}
growing_metadata {
num_trees_attempted: 2
num_layers_attempted: 4
}
""", tree_ensemble_config)
# Create existing ensemble.
tree_ensemble = boosted_trees_ops.TreeEnsemble(
'ensemble',
serialized_proto=tree_ensemble_config.SerializeToString())
tree_ensemble_handle = tree_ensemble.resource_handle
resources.initialize_resources(resources.shared_resources()).run()
cached_tree_ids = [0, 0, 0, 0, 0, 0]
# Leaves 3,4, 7 and 8 got deleted during post-pruning, leaves 5 and 6
# changed the ids to 3 and 4 respectively.
cached_node_ids = [3, 4, 5, 6, 7, 8]
# We have two features: 0 and 1.
feature_0_values = [12, 17, 35, 36, 23, 11]
feature_1_values = [12, 12, 17, 18, 123, 24]
# Grow tree ensemble.
predict_op = boosted_trees_ops.training_predict(
tree_ensemble_handle,
cached_tree_ids=cached_tree_ids,
cached_node_ids=cached_node_ids,
bucketized_features=[feature_0_values, feature_1_values],
logits_dimension=1)
logits_updates, new_tree_ids, new_node_ids = session.run(
predict_op)
# We are in the last tree.
self.assertAllClose([1, 1, 1, 1, 1, 1], new_tree_ids)
# Examples from leaves 3,4,7,8 should be in leaf 1, examples from leaf 5
# and 6 in leaf 3 and 4 in tree 0. For tree 1, all of the examples are in
# the root node.
self.assertAllClose([0, 0, 0, 0, 0, 0], new_node_ids)
cached_values = [[0.08], [0.093], [0.0553], [0.0783],
[0.15 + 0.08], [0.5 + 0.08]]
root = 0.55
self.assertAllClose(
[[root + 0.01], [root + 0.01], [root + 0.0553],
[root + 0.0783], [root + 0.01], [root + 0.01]],
logits_updates + cached_values)
def testPostPruningChangesNothing(self):
"""Test growing an ensemble with post-pruning with all gains >0."""
with self.test_session() as session:
# Create empty ensemble.
tree_ensemble_config = boosted_trees_pb2.TreeEnsemble()
tree_ensemble = boosted_trees_ops.TreeEnsemble(
'ensemble', serialized_proto=tree_ensemble_config.SerializeToString())
tree_ensemble_handle = tree_ensemble.resource_handle
resources.initialize_resources(resources.shared_resources()).run()
# Prepare inputs.
# Second feature has larger (but still negative gain).
feature_ids = [3, 4]
feature1_nodes = np.array([0], dtype=np.int32)
feature1_gains = np.array([7.62], dtype=np.float32)
feature1_thresholds = np.array([52], dtype=np.int32)
feature1_left_node_contribs = np.array([[-4.375]], dtype=np.float32)
feature1_right_node_contribs = np.array([[7.143]], dtype=np.float32)
feature2_nodes = np.array([0], dtype=np.int32)
feature2_gains = np.array([0.63], dtype=np.float32)
feature2_thresholds = np.array([23], dtype=np.int32)
feature2_left_node_contribs = np.array([[-0.6]], dtype=np.float32)
feature2_right_node_contribs = np.array([[0.24]], dtype=np.float32)
# Grow tree ensemble.
grow_op = boosted_trees_ops.update_ensemble(
tree_ensemble_handle,
learning_rate=1.0,
pruning_mode=boosted_trees_ops.PruningMode.POST_PRUNING,
max_depth=1,
feature_ids=feature_ids,
node_ids=[feature1_nodes, feature2_nodes],
gains=[feature1_gains, feature2_gains],
thresholds=[feature1_thresholds, feature2_thresholds],
left_node_contribs=[
feature1_left_node_contribs, feature2_left_node_contribs
],
right_node_contribs=[
feature1_right_node_contribs, feature2_right_node_contribs
])
session.run(grow_op)
# Expect the split from the first feature to be chosen.
# Pruning got triggered but changed nothing.
new_stamp, serialized = session.run(tree_ensemble.serialize())
res_ensemble = boosted_trees_pb2.TreeEnsemble()
res_ensemble.ParseFromString(serialized)
expected_result = """
trees {
nodes {
bucketized_split {
feature_id: 3
threshold: 52
left_id: 1
right_id: 2
}
metadata {
gain: 7.62
}
}
nodes {
leaf {
scalar: -4.375
}
}
nodes {
leaf {
scalar: 7.143
}
}
}
trees {
nodes {
leaf {
}
}
}
tree_weights: 1.0
tree_weights: 1.0
tree_metadata {
num_layers_grown: 1
is_finalized: true
}
tree_metadata {
}
growing_metadata {
num_trees_attempted: 1
num_layers_attempted: 1
}
"""
self.assertEqual(new_stamp, 1)
self.assertProtoEquals(expected_result, res_ensemble)
def testCachedPredictionIsCurrent(self):
"""Tests that prediction based on previous node in the tree works."""
with self.test_session() as session:
tree_ensemble_config = boosted_trees_pb2.TreeEnsemble()
text_format.Merge(
"""
trees {
nodes {
bucketized_split {
feature_id: 1
threshold: 15
left_id: 1
right_id: 2
}
metadata {
gain: 7.62
original_leaf {
scalar: -2
}
}
}
nodes {
leaf {
scalar: 1.14
}
}
nodes {
leaf {
scalar: 8.79
}
}
}
tree_weights: 0.1
tree_metadata {
is_finalized: true
num_layers_grown: 2
}
growing_metadata {
num_trees_attempted: 1
num_layers_attempted: 2
}
""", tree_ensemble_config)
# Create existing ensemble with one root split
tree_ensemble = boosted_trees_ops.TreeEnsemble(
'ensemble',
serialized_proto=tree_ensemble_config.SerializeToString())
tree_ensemble_handle = tree_ensemble.resource_handle
resources.initialize_resources(resources.shared_resources()).run()
# Two examples, one was cached in node 1 first, another in node 0.
cached_tree_ids = [0, 0]
cached_node_ids = [1, 2]
# We have two features: 0 and 1. Values don't matter because trees didn't
# change.
feature_0_values = [67, 5]
feature_1_values = [9, 17]
# Grow tree ensemble.
predict_op = boosted_trees_ops.training_predict(
tree_ensemble_handle,
cached_tree_ids=cached_tree_ids,
cached_node_ids=cached_node_ids,
bucketized_features=[feature_0_values, feature_1_values],
logits_dimension=1)
logits_updates, new_tree_ids, new_node_ids = session.run(
predict_op)
# Nothing changed.
self.assertAllClose(cached_tree_ids, new_tree_ids)
self.assertAllClose(cached_node_ids, new_node_ids)
self.assertAllClose([[0], [0]], logits_updates)
def testContribsMultipleTree(self):
"""Tests that the contribs work when we have multiple trees."""
with self.cached_session() as session:
tree_ensemble_config = boosted_trees_pb2.TreeEnsemble()
text_format.Merge(
"""
trees {
nodes {
bucketized_split {
feature_id: 2
threshold: 28
left_id: 1
right_id: 2
}
metadata {
gain: 7.62
original_leaf: {scalar: 2.1}
}
}
nodes {
leaf {
scalar: 1.14
}
}
nodes {
leaf {
scalar: 8.79
}
}
}
trees {
nodes {
bucketized_split {
feature_id: 2
threshold: 26
left_id: 1
right_id: 2
}
}
nodes {
bucketized_split {
feature_id: 0
threshold: 50
left_id: 3
right_id: 4
}
metadata {
original_leaf: {scalar: 5.5}
}
}
nodes {
leaf {
scalar: 7.0
}
}
nodes {
leaf {
scalar: 5.0
}
}
nodes {
leaf {
scalar: 6.0
}
}
}
trees {
nodes {
bucketized_split {
feature_id: 0
threshold: 34
left_id: 1
right_id: 2
}
}
nodes {
leaf {
scalar: -7.0
}
}
nodes {
leaf {
scalar: 5.0
}
}
}
tree_weights: 0.1
tree_weights: 0.2
tree_weights: 1.0
tree_metadata: {
num_layers_grown: 1}
tree_metadata: {
num_layers_grown: 2}
tree_metadata: {
num_layers_grown: 1}
""", tree_ensemble_config)
tree_ensemble = boosted_trees_ops.TreeEnsemble(
'ensemble',
serialized_proto=tree_ensemble_config.SerializeToString())
tree_ensemble_handle = tree_ensemble.resource_handle
resources.initialize_resources(resources.shared_resources()).run()
feature_0_values = [36, 32]
feature_1_values = [13, -29
] # Unused. Feature is not in above ensemble.
feature_2_values = [11, 27]
# Expected logits are computed by traversing the logit path and
# subtracting child logits from parent logits.
bias = 2.1 * 0.1 # Root node of tree_0.
expected_feature_ids = ((2, 2, 0, 0), (2, 2, 0))
# example_0 : (bias, 0.1 * 1.14, 0.2 * 5.5 + .114, 0.2 * 5. + .114,
# 1.0 * 5.0 + 0.2 * 5. + .114)
# example_1 : (bias, 0.1 * 1.14, 0.2 * 7 + .114,
# 1.0 * -7. + 0.2 * 7 + .114)
expected_logits_paths = ((bias, 0.114, 1.214, 1.114, 6.114),
(bias, 0.114, 1.514, -5.486))
bucketized_features = [
feature_0_values, feature_1_values, feature_2_values
]
debug_op = boosted_trees_ops.example_debug_outputs(
tree_ensemble_handle,
bucketized_features=bucketized_features,
logits_dimension=1)
serialized_examples_debug_outputs = session.run(debug_op)
feature_ids = []
logits_paths = []
for example in serialized_examples_debug_outputs:
example_debug_outputs = boosted_trees_pb2.DebugOutput()
example_debug_outputs.ParseFromString(example)
feature_ids.append(example_debug_outputs.feature_ids)
logits_paths.append(example_debug_outputs.logits_path)
self.assertAllClose(feature_ids, expected_feature_ids)
self.assertAllClose(logits_paths, expected_logits_paths)
def testPrePruning(self):
"""Test growing an existing ensemble with pre-pruning."""
with self.test_session() as session:
tree_ensemble_config = boosted_trees_pb2.TreeEnsemble()
text_format.Merge("""
trees {
nodes {
bucketized_split {
feature_id: 4
left_id: 1
right_id: 2
}
metadata {
gain: 7.62
}
}
nodes {
leaf {
scalar: 7.14
}
}
nodes {
leaf {
scalar: -4.375
}
}
}
tree_weights: 0.1
tree_metadata {
num_layers_grown: 1
is_finalized: false
}
growing_metadata {
num_trees_attempted: 1
num_layers_attempted: 1
}
""", tree_ensemble_config)
# Create existing ensemble with one root split
tree_ensemble = boosted_trees_ops.TreeEnsemble(
'ensemble', serialized_proto=tree_ensemble_config.SerializeToString())
tree_ensemble_handle = tree_ensemble.resource_handle
resources.initialize_resources(resources.shared_resources()).run()
# Prepare feature inputs.
# For node 1, the best split is on feature 2 (gain -0.63), but the gain
# is negative so node 1 will not be split.
# For node 2, the best split is on feature 3, gain is positive.
feature_ids = [0, 1, 0]
feature1_nodes = np.array([1], dtype=np.int32)
feature1_gains = np.array([-1.4], dtype=np.float32)
feature1_thresholds = np.array([21], dtype=np.int32)
feature1_left_node_contribs = np.array([[-6.0]], dtype=np.float32)
feature1_right_node_contribs = np.array([[1.65]], dtype=np.float32)
feature2_nodes = np.array([1, 2], dtype=np.int32)
feature2_gains = np.array([-0.63, 2.7], dtype=np.float32)
feature2_thresholds = np.array([23, 7], dtype=np.int32)
feature2_left_node_contribs = np.array([[-0.6], [-1.5]], dtype=np.float32)
feature2_right_node_contribs = np.array([[0.24], [2.3]], dtype=np.float32)
feature3_nodes = np.array([2], dtype=np.int32)
feature3_gains = np.array([2.8], dtype=np.float32)
feature3_thresholds = np.array([3], dtype=np.int32)
feature3_left_node_contribs = np.array([[-0.75]], dtype=np.float32)
feature3_right_node_contribs = np.array([[1.93]], dtype=np.float32)
# Grow tree ensemble.
grow_op = boosted_trees_ops.update_ensemble(
tree_ensemble_handle,
learning_rate=0.1,
pruning_mode=boosted_trees_ops.PruningMode.PRE_PRUNING,
max_depth=3,
feature_ids=feature_ids,
node_ids=[feature1_nodes, feature2_nodes, feature3_nodes],
gains=[feature1_gains, feature2_gains, feature3_gains],
thresholds=[
feature1_thresholds, feature2_thresholds, feature3_thresholds
],
left_node_contribs=[
feature1_left_node_contribs, feature2_left_node_contribs,
feature3_left_node_contribs
],
right_node_contribs=[
feature1_right_node_contribs, feature2_right_node_contribs,
feature3_right_node_contribs
])
session.run(grow_op)
# Expect the split for node 1 to be chosen from feature 1 and
# the split for node 2 to be chosen from feature 2.
# The grown tree should not be finalized as max tree depth is 3 and
# it's only grown 2 layers.
new_stamp, serialized = session.run(tree_ensemble.serialize())
tree_ensemble = boosted_trees_pb2.TreeEnsemble()
tree_ensemble.ParseFromString(serialized)
expected_result = """
trees {
nodes {
bucketized_split {
feature_id: 4
left_id: 1
right_id: 2
}
metadata {
gain: 7.62
}
}
nodes {
leaf {
scalar: 7.14
}
}
nodes {
bucketized_split {
feature_id: 0
threshold: 3
left_id: 3
right_id: 4
}
metadata {
gain: 2.8
original_leaf {
scalar: -4.375
}
}
}
nodes {
leaf {
scalar: -4.45
}
}
nodes {
leaf {
scalar: -4.182
}
}
}
tree_weights: 0.1
tree_metadata {
is_finalized: false
num_layers_grown: 2
}
growing_metadata {
num_trees_attempted: 1
num_layers_attempted: 2
}
"""
self.assertEqual(new_stamp, 1)
self.assertProtoEquals(expected_result, tree_ensemble)
def testGrowExistingEnsembleTreeFinalized(self):
"""Test growing an existing ensemble with the last tree finalized."""
with self.test_session() as session:
tree_ensemble_config = boosted_trees_pb2.TreeEnsemble()
text_format.Merge("""
trees {
nodes {
bucketized_split {
feature_id: 4
left_id: 1
right_id: 2
}
metadata {
gain: 7.62
}
}
nodes {
leaf {
scalar: 7.14
}
}
nodes {
leaf {
scalar: -4.375
}
}
}
trees {
nodes {
leaf {
scalar: 0.0
}
}
}
tree_weights: 0.15
tree_weights: 1.0
tree_metadata {
num_layers_grown: 1
is_finalized: true
}
tree_metadata {
}
growing_metadata {
num_trees_attempted: 1
num_layers_attempted: 1
}
""", tree_ensemble_config)
# Create existing ensemble with one root split
tree_ensemble = boosted_trees_ops.TreeEnsemble(
'ensemble', serialized_proto=tree_ensemble_config.SerializeToString())
tree_ensemble_handle = tree_ensemble.resource_handle
resources.initialize_resources(resources.shared_resources()).run()
# Prepare feature inputs.
feature_ids = [75]
feature1_nodes = np.array([0], dtype=np.int32)
feature1_gains = np.array([-1.4], dtype=np.float32)
feature1_thresholds = np.array([21], dtype=np.int32)
feature1_left_node_contribs = np.array([[-6.0]], dtype=np.float32)
feature1_right_node_contribs = np.array([[1.65]], dtype=np.float32)
# Grow tree ensemble.
grow_op = boosted_trees_ops.update_ensemble(
tree_ensemble_handle,
pruning_mode=boosted_trees_ops.PruningMode.NO_PRUNING,
learning_rate=0.1,
max_depth=2,
feature_ids=feature_ids,
node_ids=[feature1_nodes],
gains=[feature1_gains],
thresholds=[feature1_thresholds],
left_node_contribs=[feature1_left_node_contribs],
right_node_contribs=[feature1_right_node_contribs])
session.run(grow_op)
# Expect a new tree added, with a split on feature 75
new_stamp, serialized = session.run(tree_ensemble.serialize())
tree_ensemble = boosted_trees_pb2.TreeEnsemble()
tree_ensemble.ParseFromString(serialized)
expected_result = """
trees {
nodes {
bucketized_split {
feature_id: 4
left_id: 1
right_id: 2
}
metadata {
gain: 7.62
}
}
nodes {
leaf {
scalar: 7.14
}
}
nodes {
leaf {
scalar: -4.375
}
}
}
trees {
nodes {
bucketized_split {
feature_id: 75
threshold: 21
left_id: 1
right_id: 2
}
metadata {
gain: -1.4
}
}
nodes {
leaf {
scalar: -0.6
}
}
nodes {
leaf {
scalar: 0.165
}
}
}
tree_weights: 0.15
tree_weights: 1.0
tree_metadata {
num_layers_grown: 1
is_finalized: true
}
tree_metadata {
num_layers_grown: 1
is_finalized: false
}
growing_metadata {
num_trees_attempted: 2
num_layers_attempted: 2
}
"""
self.assertEqual(new_stamp, 1)
self.assertProtoEquals(expected_result, tree_ensemble)
def testGrowWithEmptyEnsemble(self):
"""Test growing an empty ensemble."""
with self.test_session() as session:
# Create empty ensemble.
tree_ensemble = boosted_trees_ops.TreeEnsemble('ensemble')
tree_ensemble_handle = tree_ensemble.resource_handle
resources.initialize_resources(resources.shared_resources()).run()
feature_ids = [0, 2, 6]
# Prepare feature inputs.
# Note that features 1 & 3 have the same gain but different splits.
feature1_nodes = np.array([0], dtype=np.int32)
feature1_gains = np.array([7.62], dtype=np.float32)
feature1_thresholds = np.array([52], dtype=np.int32)
feature1_left_node_contribs = np.array([[-4.375]], dtype=np.float32)
feature1_right_node_contribs = np.array([[7.143]], dtype=np.float32)
feature2_nodes = np.array([0], dtype=np.int32)
feature2_gains = np.array([0.63], dtype=np.float32)
feature2_thresholds = np.array([23], dtype=np.int32)
feature2_left_node_contribs = np.array([[-0.6]], dtype=np.float32)
feature2_right_node_contribs = np.array([[0.24]], dtype=np.float32)
# Feature split with the highest gain.
feature3_nodes = np.array([0], dtype=np.int32)
feature3_gains = np.array([7.65], dtype=np.float32)
feature3_thresholds = np.array([7], dtype=np.int32)
feature3_left_node_contribs = np.array([[-4.89]], dtype=np.float32)
feature3_right_node_contribs = np.array([[5.3]], dtype=np.float32)
# Grow tree ensemble.
grow_op = boosted_trees_ops.update_ensemble(
tree_ensemble_handle,
learning_rate=0.1,
pruning_mode=boosted_trees_ops.PruningMode.NO_PRUNING,
# Tree will be finalized now, since we will reach depth 1.
max_depth=1,
feature_ids=feature_ids,
node_ids=[feature1_nodes, feature2_nodes, feature3_nodes],
gains=[feature1_gains, feature2_gains, feature3_gains],
thresholds=[
feature1_thresholds, feature2_thresholds, feature3_thresholds
],
left_node_contribs=[
feature1_left_node_contribs, feature2_left_node_contribs,
feature3_left_node_contribs
],
right_node_contribs=[
feature1_right_node_contribs, feature2_right_node_contribs,
feature3_right_node_contribs
])
session.run(grow_op)
new_stamp, serialized = session.run(tree_ensemble.serialize())
tree_ensemble = boosted_trees_pb2.TreeEnsemble()
tree_ensemble.ParseFromString(serialized)
# Note that since the tree is finalized, we added a new dummy tree.
expected_result = """
trees {
nodes {
bucketized_split {
feature_id: 6
threshold: 7
left_id: 1
right_id: 2
}
metadata {
gain: 7.65
}
}
nodes {
leaf {
scalar: -0.489
}
}
nodes {
leaf {
scalar: 0.53
}
}
}
trees {
nodes {
leaf {
scalar: 0.0
}
}
}
tree_weights: 1.0
tree_weights: 1.0
tree_metadata {
num_layers_grown: 1
is_finalized: true
}
tree_metadata {
}
growing_metadata {
num_trees_attempted: 1
num_layers_attempted: 1
}
"""
self.assertEqual(new_stamp, 1)
self.assertProtoEquals(expected_result, tree_ensemble)
def testGrowExistingEnsembleTreeNotFinalized(self):
"""Test growing an existing ensemble with the last tree not finalized."""
with self.test_session() as session:
tree_ensemble_config = boosted_trees_pb2.TreeEnsemble()
text_format.Merge("""
trees {
nodes {
bucketized_split {
feature_id: 4
left_id: 1
right_id: 2
}
metadata {
gain: 7.62
}
}
nodes {
leaf {
scalar: 0.714
}
}
nodes {
leaf {
scalar: -0.4375
}
}
}
tree_weights: 1.0
tree_metadata {
num_layers_grown: 1
is_finalized: false
}
growing_metadata {
num_trees_attempted: 1
num_layers_attempted: 1
}
""", tree_ensemble_config)
# Create existing ensemble with one root split
tree_ensemble = boosted_trees_ops.TreeEnsemble(
'ensemble', serialized_proto=tree_ensemble_config.SerializeToString())
tree_ensemble_handle = tree_ensemble.resource_handle
resources.initialize_resources(resources.shared_resources()).run()
# Prepare feature inputs.
# feature 1 only has a candidate for node 1, feature 2 has candidates
# for both nodes and feature 3 only has a candidate for node 2.
feature_ids = [0, 1, 0]
feature1_nodes = np.array([1], dtype=np.int32)
feature1_gains = np.array([1.4], dtype=np.float32)
feature1_thresholds = np.array([21], dtype=np.int32)
feature1_left_node_contribs = np.array([[-6.0]], dtype=np.float32)
feature1_right_node_contribs = np.array([[1.65]], dtype=np.float32)
feature2_nodes = np.array([1, 2], dtype=np.int32)
feature2_gains = np.array([0.63, 2.7], dtype=np.float32)
feature2_thresholds = np.array([23, 7], dtype=np.int32)
feature2_left_node_contribs = np.array([[-0.6], [-1.5]], dtype=np.float32)
feature2_right_node_contribs = np.array([[0.24], [2.3]], dtype=np.float32)
feature3_nodes = np.array([2], dtype=np.int32)
feature3_gains = np.array([1.7], dtype=np.float32)
feature3_thresholds = np.array([3], dtype=np.int32)
feature3_left_node_contribs = np.array([[-0.75]], dtype=np.float32)
feature3_right_node_contribs = np.array([[1.93]], dtype=np.float32)
# Grow tree ensemble.
grow_op = boosted_trees_ops.update_ensemble(
tree_ensemble_handle,
learning_rate=0.1,
pruning_mode=boosted_trees_ops.PruningMode.NO_PRUNING,
# tree is going to be finalized now, since we reach depth 2.
max_depth=2,
feature_ids=feature_ids,
node_ids=[feature1_nodes, feature2_nodes, feature3_nodes],
gains=[feature1_gains, feature2_gains, feature3_gains],
thresholds=[
feature1_thresholds, feature2_thresholds, feature3_thresholds
],
left_node_contribs=[
feature1_left_node_contribs, feature2_left_node_contribs,
feature3_left_node_contribs
],
right_node_contribs=[
feature1_right_node_contribs, feature2_right_node_contribs,
feature3_right_node_contribs
])
session.run(grow_op)
# Expect the split for node 1 to be chosen from feature 1 and
# the split for node 2 to be chosen from feature 2.
# The grown tree should be finalized as max tree depth is 2 and we have
# grown 2 layers.
new_stamp, serialized = session.run(tree_ensemble.serialize())
tree_ensemble = boosted_trees_pb2.TreeEnsemble()
tree_ensemble.ParseFromString(serialized)
expected_result = """
trees {
nodes {
bucketized_split {
feature_id: 4
left_id: 1
right_id: 2
}
metadata {
gain: 7.62
}
}
nodes {
bucketized_split {
threshold: 21
left_id: 3
right_id: 4
}
metadata {
gain: 1.4
original_leaf {
scalar: 0.714
}
}
}
nodes {
bucketized_split {
feature_id: 1
threshold: 7
left_id: 5
right_id: 6
}
metadata {
gain: 2.7
original_leaf {
scalar: -0.4375
}
}
}
nodes {
leaf {
scalar: 0.114
}
}
nodes {
leaf {
scalar: 0.879
}
}
nodes {
leaf {
scalar: -0.5875
}
}
nodes {
leaf {
scalar: -0.2075
}
}
}
trees {
nodes {
leaf {
scalar: 0.0
}
}
}
tree_weights: 1.0
tree_weights: 1.0
tree_metadata {
is_finalized: true
num_layers_grown: 2
}
tree_metadata {
}
growing_metadata {
num_trees_attempted: 1
num_layers_attempted: 2
}
"""
self.assertEqual(new_stamp, 1)
self.assertProtoEquals(expected_result, tree_ensemble)
def testSerializeDeserialize(self):
with self.cached_session():
# Initialize.
ensemble = boosted_trees_ops.TreeEnsemble('ensemble',
stamp_token=5)
resources.initialize_resources(resources.shared_resources()).run()
(stamp_token, num_trees, num_finalized_trees, num_attempted_layers,
nodes_range) = ensemble.get_states()
self.assertEqual(5, self.evaluate(stamp_token))
self.assertEqual(0, self.evaluate(num_trees))
self.assertEqual(0, self.evaluate(num_finalized_trees))
self.assertEqual(0, self.evaluate(num_attempted_layers))
self.assertAllEqual([0, 1], self.evaluate(nodes_range))
# Deserialize.
ensemble_proto = boosted_trees_pb2.TreeEnsemble()
text_format.Merge(
"""
trees {
nodes {
bucketized_split {
feature_id: 75
threshold: 21
left_id: 1
right_id: 2
}
metadata {
gain: -1.4
}
}
nodes {
leaf {
scalar: -0.6
}
}
nodes {
leaf {
scalar: 0.165
}
}
}
tree_weights: 0.5
tree_metadata {
num_layers_grown: 4 # it's fake intentionally.
is_finalized: false
}
growing_metadata {
num_trees_attempted: 1
num_layers_attempted: 5
last_layer_node_start: 3
last_layer_node_end: 7
}
""", ensemble_proto)
with ops.control_dependencies([
ensemble.deserialize(
stamp_token=3,
serialized_proto=ensemble_proto.SerializeToString())
]):
(stamp_token, num_trees, num_finalized_trees,
num_attempted_layers, nodes_range) = ensemble.get_states()
self.assertEqual(3, self.evaluate(stamp_token))
self.assertEqual(1, self.evaluate(num_trees))
# This reads from metadata, not really counting the layers.
self.assertEqual(5, self.evaluate(num_attempted_layers))
self.assertEqual(0, self.evaluate(num_finalized_trees))
self.assertAllEqual([3, 7], self.evaluate(nodes_range))
# Serialize.
new_ensemble_proto = boosted_trees_pb2.TreeEnsemble()
new_stamp_token, new_serialized = ensemble.serialize()
self.assertEqual(3, self.evaluate(new_stamp_token))
new_ensemble_proto.ParseFromString(new_serialized.eval())
self.assertProtoEquals(ensemble_proto, new_ensemble_proto)
def testMetadataWhenCantSplitDuePrePruning(self):
"""Test metadata is updated correctly when no split due to prepruning."""
with self.test_session() as session:
tree_ensemble_config = boosted_trees_pb2.TreeEnsemble()
text_format.Merge("""
trees {
nodes {
bucketized_split {
feature_id: 4
left_id: 1
right_id: 2
}
metadata {
gain: 7.62
}
}
nodes {
leaf {
scalar: 7.14
}
}
nodes {
leaf {
scalar: -4.375
}
}
}
tree_weights: 1.0
tree_metadata {
num_layers_grown: 1
is_finalized: false
}
growing_metadata {
num_trees_attempted: 1
num_layers_attempted: 1
}
""", tree_ensemble_config)
# Create existing ensemble with one root split
tree_ensemble = boosted_trees_ops.TreeEnsemble(
'ensemble', serialized_proto=tree_ensemble_config.SerializeToString())
tree_ensemble_handle = tree_ensemble.resource_handle
resources.initialize_resources(resources.shared_resources()).run()
# Prepare feature inputs.
feature_ids = [0, 1, 0]
# All the gains are negative.
feature1_nodes = np.array([1], dtype=np.int32)
feature1_gains = np.array([-1.4], dtype=np.float32)
feature1_thresholds = np.array([21], dtype=np.int32)
feature1_left_node_contribs = np.array([[-6.0]], dtype=np.float32)
feature1_right_node_contribs = np.array([[1.65]], dtype=np.float32)
feature2_nodes = np.array([1, 2], dtype=np.int32)
feature2_gains = np.array([-0.63, -2.7], dtype=np.float32)
feature2_thresholds = np.array([23, 7], dtype=np.int32)
feature2_left_node_contribs = np.array([[-0.6], [-1.5]], dtype=np.float32)
feature2_right_node_contribs = np.array([[0.24], [2.3]], dtype=np.float32)
feature3_nodes = np.array([2], dtype=np.int32)
feature3_gains = np.array([-2.8], dtype=np.float32)
feature3_thresholds = np.array([3], dtype=np.int32)
feature3_left_node_contribs = np.array([[-0.75]], dtype=np.float32)
feature3_right_node_contribs = np.array([[1.93]], dtype=np.float32)
# Grow tree ensemble.
grow_op = boosted_trees_ops.update_ensemble(
tree_ensemble_handle,
learning_rate=0.1,
pruning_mode=boosted_trees_ops.PruningMode.PRE_PRUNING,
max_depth=3,
feature_ids=feature_ids,
node_ids=[feature1_nodes, feature2_nodes, feature3_nodes],
gains=[feature1_gains, feature2_gains, feature3_gains],
thresholds=[
feature1_thresholds, feature2_thresholds, feature3_thresholds
],
left_node_contribs=[
feature1_left_node_contribs, feature2_left_node_contribs,
feature3_left_node_contribs
],
right_node_contribs=[
feature1_right_node_contribs, feature2_right_node_contribs,
feature3_right_node_contribs
])
session.run(grow_op)
# Expect that no new split was created because all the gains were negative
# Global metadata should be updated, tree metadata should not be updated.
new_stamp, serialized = session.run(tree_ensemble.serialize())
tree_ensemble = boosted_trees_pb2.TreeEnsemble()
tree_ensemble.ParseFromString(serialized)
expected_result = """
trees {
nodes {
bucketized_split {
feature_id: 4
left_id: 1
right_id: 2
}
metadata {
gain: 7.62
}
}
nodes {
leaf {
scalar: 7.14
}
}
nodes {
leaf {
scalar: -4.375
}
}
}
tree_weights: 1.0
tree_metadata {
num_layers_grown: 1
is_finalized: false
}
growing_metadata {
num_trees_attempted: 1
num_layers_attempted: 2
}
"""
self.assertEqual(new_stamp, 1)
self.assertProtoEquals(expected_result, tree_ensemble)
def testCreateWithProto(self):
with self.cached_session():
ensemble_proto = boosted_trees_pb2.TreeEnsemble()
text_format.Merge(
"""
trees {
nodes {
bucketized_split {
feature_id: 4
left_id: 1
right_id: 2
}
metadata {
gain: 7.62
}
}
nodes {
bucketized_split {
threshold: 21
left_id: 3
right_id: 4
}
metadata {
gain: 1.4
original_leaf {
scalar: 7.14
}
}
}
nodes {
bucketized_split {
feature_id: 1
threshold: 7
left_id: 5
right_id: 6
}
metadata {
gain: 2.7
original_leaf {
scalar: -4.375
}
}
}
nodes {
leaf {
scalar: 6.54
}
}
nodes {
leaf {
scalar: 7.305
}
}
nodes {
leaf {
scalar: -4.525
}
}
nodes {
leaf {
scalar: -4.145
}
}
}
trees {
nodes {
bucketized_split {
feature_id: 75
threshold: 21
left_id: 1
right_id: 2
}
metadata {
gain: -1.4
}
}
nodes {
leaf {
scalar: -0.6
}
}
nodes {
leaf {
scalar: 0.165
}
}
}
tree_weights: 0.15
tree_weights: 1.0
tree_metadata {
num_layers_grown: 2
is_finalized: true
}
tree_metadata {
num_layers_grown: 1
is_finalized: false
}
growing_metadata {
num_trees_attempted: 2
num_layers_attempted: 6
last_layer_node_start: 16
last_layer_node_end: 19
}
""", ensemble_proto)
ensemble = boosted_trees_ops.TreeEnsemble(
'ensemble',
stamp_token=7,
serialized_proto=ensemble_proto.SerializeToString())
resources.initialize_resources(resources.shared_resources()).run()
(stamp_token, num_trees, num_finalized_trees, num_attempted_layers,
nodes_range) = ensemble.get_states()
self.assertEqual(7, self.evaluate(stamp_token))
self.assertEqual(2, self.evaluate(num_trees))
self.assertEqual(1, self.evaluate(num_finalized_trees))
self.assertEqual(6, self.evaluate(num_attempted_layers))
self.assertAllEqual([16, 19], self.evaluate(nodes_range))
def testPostPruningOfSomeNodes(self):
"""Test growing an ensemble with post-pruning."""
with self.test_session() as session:
# Create empty ensemble.
tree_ensemble_config = boosted_trees_pb2.TreeEnsemble()
tree_ensemble = boosted_trees_ops.TreeEnsemble(
'ensemble', serialized_proto=tree_ensemble_config.SerializeToString())
tree_ensemble_handle = tree_ensemble.resource_handle
resources.initialize_resources(resources.shared_resources()).run()
# Prepare inputs.
# Second feature has larger (but still negative gain).
feature_ids = [0, 1]
feature1_nodes = np.array([0], dtype=np.int32)
feature1_gains = np.array([-1.3], dtype=np.float32)
feature1_thresholds = np.array([7], dtype=np.int32)
feature1_left_node_contribs = np.array([[0.013]], dtype=np.float32)
feature1_right_node_contribs = np.array([[0.0143]], dtype=np.float32)
feature2_nodes = np.array([0], dtype=np.int32)
feature2_gains = np.array([-0.2], dtype=np.float32)
feature2_thresholds = np.array([33], dtype=np.int32)
feature2_left_node_contribs = np.array([[0.01]], dtype=np.float32)
feature2_right_node_contribs = np.array([[0.0143]], dtype=np.float32)
# Grow tree ensemble.
grow_op = boosted_trees_ops.update_ensemble(
tree_ensemble_handle,
learning_rate=1.0,
pruning_mode=boosted_trees_ops.PruningMode.POST_PRUNING,
max_depth=3,
feature_ids=feature_ids,
node_ids=[feature1_nodes, feature2_nodes],
gains=[feature1_gains, feature2_gains],
thresholds=[feature1_thresholds, feature2_thresholds],
left_node_contribs=[
feature1_left_node_contribs, feature2_left_node_contribs
],
right_node_contribs=[
feature1_right_node_contribs, feature2_right_node_contribs
])
session.run(grow_op)
# Expect the split from second features to be chosen despite the negative
# gain.
# No pruning happened just yet.
new_stamp, serialized = session.run(tree_ensemble.serialize())
res_ensemble = boosted_trees_pb2.TreeEnsemble()
res_ensemble.ParseFromString(serialized)
expected_result = """
trees {
nodes {
bucketized_split {
feature_id: 1
threshold: 33
left_id: 1
right_id: 2
}
metadata {
gain: -0.2
}
}
nodes {
leaf {
scalar: 0.01
}
}
nodes {
leaf {
scalar: 0.0143
}
}
}
tree_weights: 1.0
tree_metadata {
num_layers_grown: 1
is_finalized: false
}
growing_metadata {
num_trees_attempted: 1
num_layers_attempted: 1
}
"""
self.assertEqual(new_stamp, 1)
self.assertProtoEquals(expected_result, res_ensemble)
# Prepare the second layer.
# Note that node 1 gain is negative and node 2 gain is positive.
feature_ids = [3]
feature1_nodes = np.array([1, 2], dtype=np.int32)
feature1_gains = np.array([-0.2, 0.5], dtype=np.float32)
feature1_thresholds = np.array([7, 5], dtype=np.int32)
feature1_left_node_contribs = np.array(
[[0.07], [0.041]], dtype=np.float32)
feature1_right_node_contribs = np.array(
[[0.083], [0.064]], dtype=np.float32)
# Grow tree ensemble.
grow_op = boosted_trees_ops.update_ensemble(
tree_ensemble_handle,
learning_rate=1.0,
pruning_mode=boosted_trees_ops.PruningMode.POST_PRUNING,
max_depth=3,
feature_ids=feature_ids,
node_ids=[feature1_nodes],
gains=[feature1_gains],
thresholds=[feature1_thresholds],
left_node_contribs=[feature1_left_node_contribs],
right_node_contribs=[feature1_right_node_contribs])
session.run(grow_op)
# After adding this layer, the tree will not be finalized
new_stamp, serialized = session.run(tree_ensemble.serialize())
res_ensemble = boosted_trees_pb2.TreeEnsemble()
res_ensemble.ParseFromString(serialized)
expected_result = """
trees {
nodes {
bucketized_split {
feature_id:1
threshold: 33
left_id: 1
right_id: 2
}
metadata {
gain: -0.2
}
}
nodes {
bucketized_split {
feature_id: 3
threshold: 7
left_id: 3
right_id: 4
}
metadata {
gain: -0.2
original_leaf {
scalar: 0.01
}
}
}
nodes {
bucketized_split {
feature_id: 3
threshold: 5
left_id: 5
right_id: 6
}
metadata {
gain: 0.5
original_leaf {
scalar: 0.0143
}
}
}
nodes {
leaf {
scalar: 0.08
}
}
nodes {
leaf {
scalar: 0.093
}
}
nodes {
leaf {
scalar: 0.0553
}
}
nodes {
leaf {
scalar: 0.0783
}
}
}
tree_weights: 1.0
tree_metadata {
num_layers_grown: 2
is_finalized: false
}
growing_metadata {
num_trees_attempted: 1
num_layers_attempted: 2
}
"""
self.assertEqual(new_stamp, 2)
self.assertProtoEquals(expected_result, res_ensemble)
# Now split the leaf 3, again with negative gain. After this layer, the
# tree will be finalized, and post-pruning happens. The leafs 3,4,7,8 will
# be pruned out.
# Prepare the third layer.
feature_ids = [92]
feature1_nodes = np.array([3], dtype=np.int32)
feature1_gains = np.array([-0.45], dtype=np.float32)
feature1_thresholds = np.array([11], dtype=np.int32)
feature1_left_node_contribs = np.array([[0.15]], dtype=np.float32)
feature1_right_node_contribs = np.array([[0.5]], dtype=np.float32)
# Grow tree ensemble.
grow_op = boosted_trees_ops.update_ensemble(
tree_ensemble_handle,
learning_rate=1.0,
pruning_mode=boosted_trees_ops.PruningMode.POST_PRUNING,
max_depth=3,
feature_ids=feature_ids,
node_ids=[feature1_nodes],
gains=[feature1_gains],
thresholds=[feature1_thresholds],
left_node_contribs=[feature1_left_node_contribs],
right_node_contribs=[feature1_right_node_contribs])
session.run(grow_op)
# After adding this layer, the tree will be finalized
new_stamp, serialized = session.run(tree_ensemble.serialize())
res_ensemble = boosted_trees_pb2.TreeEnsemble()
res_ensemble.ParseFromString(serialized)
# Node that nodes 3, 4, 7 and 8 got deleted, so metadata stores has ids
# mapped to their parent node 1, with the respective change in logits.
expected_result = """
trees {
nodes {
bucketized_split {
feature_id:1
threshold: 33
left_id: 1
right_id: 2
}
metadata {
gain: -0.2
}
}
nodes {
leaf {
scalar: 0.01
}
}
nodes {
bucketized_split {
feature_id: 3
threshold: 5
left_id: 3
right_id: 4
}
metadata {
gain: 0.5
original_leaf {
scalar: 0.0143
}
}
}
nodes {
leaf {
scalar: 0.0553
}
}
nodes {
leaf {
scalar: 0.0783
}
}
}
trees {
nodes {
leaf {
}
}
}
tree_weights: 1.0
tree_weights: 1.0
tree_metadata {
num_layers_grown: 3
is_finalized: true
post_pruned_nodes_meta {
new_node_id: 0
logit_change: 0.0
}
post_pruned_nodes_meta {
new_node_id: 1
logit_change: 0.0
}
post_pruned_nodes_meta {
new_node_id: 2
logit_change: 0.0
}
post_pruned_nodes_meta {
new_node_id: 1
logit_change: -0.07
}
post_pruned_nodes_meta {
new_node_id: 1
logit_change: -0.083
}
post_pruned_nodes_meta {
new_node_id: 3
logit_change: 0.0
}
post_pruned_nodes_meta {
new_node_id: 4
logit_change: 0.0
}
post_pruned_nodes_meta {
new_node_id: 1
logit_change: -0.22
}
post_pruned_nodes_meta {
new_node_id: 1
logit_change: -0.57
}
}
tree_metadata {
}
growing_metadata {
num_trees_attempted: 1
num_layers_attempted: 3
}
"""
self.assertEqual(new_stamp, 3)
self.assertProtoEquals(expected_result, res_ensemble)
def testCachedPredictionFromTheSameTree(self):
"""Tests that prediction based on previous node in the tree works."""
with self.test_session() as session:
tree_ensemble_config = boosted_trees_pb2.TreeEnsemble()
text_format.Merge(
"""
trees {
nodes {
bucketized_split {
feature_id: 1
threshold: 15
left_id: 1
right_id: 2
}
metadata {
gain: 7.62
original_leaf {
scalar: -2
}
}
}
nodes {
bucketized_split {
feature_id: 1
threshold: 7
left_id: 3
right_id: 4
}
metadata {
gain: 1.4
original_leaf {
scalar: 7.14
}
}
}
nodes {
bucketized_split {
feature_id: 0
threshold: 7
left_id: 5
right_id: 6
}
metadata {
gain: 2.7
original_leaf {
scalar: -4.375
}
}
}
nodes {
leaf {
scalar: 1.14
}
}
nodes {
leaf {
scalar: 8.79
}
}
nodes {
leaf {
scalar: -5.875
}
}
nodes {
leaf {
scalar: -2.075
}
}
}
tree_weights: 0.1
tree_metadata {
is_finalized: true
num_layers_grown: 2
}
growing_metadata {
num_trees_attempted: 1
num_layers_attempted: 2
}
""", tree_ensemble_config)
# Create existing ensemble with one root split
tree_ensemble = boosted_trees_ops.TreeEnsemble(
'ensemble',
serialized_proto=tree_ensemble_config.SerializeToString())
tree_ensemble_handle = tree_ensemble.resource_handle
resources.initialize_resources(resources.shared_resources()).run()
# Two examples, one was cached in node 1 first, another in node 0.
cached_tree_ids = [0, 0]
cached_node_ids = [1, 0]
# We have two features: 0 and 1.
feature_0_values = [67, 5]
feature_1_values = [9, 17]
# Grow tree ensemble.
predict_op = boosted_trees_ops.training_predict(
tree_ensemble_handle,
cached_tree_ids=cached_tree_ids,
cached_node_ids=cached_node_ids,
bucketized_features=[feature_0_values, feature_1_values],
logits_dimension=1)
logits_updates, new_tree_ids, new_node_ids = session.run(
predict_op)
# We are still in the same tree.
self.assertAllClose([0, 0], new_tree_ids)
# When using the full tree, the first example will end up in node 4,
# the second in node 5.
self.assertAllClose([4, 5], new_node_ids)
# Full predictions for each instance would be 8.79 and -5.875,
# so an update from the previous cached values lr*(7.14 and -2) would be
# 1.65 and -3.875, and then multiply them by 0.1 (lr)
self.assertAllClose([[0.1 * 1.65], [0.1 * -3.875]], logits_updates)
def testCategoricalSplits(self):
"""Tests the training prediction work for categorical splits."""
with self.cached_session() as session:
tree_ensemble_config = boosted_trees_pb2.TreeEnsemble()
text_format.Merge(
"""
trees {
nodes {
categorical_split {
feature_id: 1
value: 2
left_id: 1
right_id: 2
}
}
nodes {
categorical_split {
feature_id: 0
value: 13
left_id: 3
right_id: 4
}
}
nodes {
leaf {
scalar: 7.0
}
}
nodes {
leaf {
scalar: 5.0
}
}
nodes {
leaf {
scalar: 6.0
}
}
}
tree_weights: 1.0
tree_metadata {
is_finalized: true
}
""", tree_ensemble_config)
# Create existing ensemble with one root split
tree_ensemble = boosted_trees_ops.TreeEnsemble(
'ensemble', serialized_proto=tree_ensemble_config.SerializeToString())
tree_ensemble_handle = tree_ensemble.resource_handle
resources.initialize_resources(resources.shared_resources()).run()
feature_0_values = [13, 1, 3]
feature_1_values = [2, 2, 1]
# No previous cached values.
cached_tree_ids = [0, 0, 0]
cached_node_ids = [0, 0, 0]
# Grow tree ensemble.
predict_op = boosted_trees_ops.training_predict(
tree_ensemble_handle,
cached_tree_ids=cached_tree_ids,
cached_node_ids=cached_node_ids,
bucketized_features=[feature_0_values, feature_1_values],
logits_dimension=1)
logits_updates, new_tree_ids, new_node_ids = session.run(predict_op)
self.assertAllClose([0, 0, 0], new_tree_ids)
self.assertAllClose([3, 4, 2], new_node_ids)
self.assertAllClose([[5.], [6.], [7.]], logits_updates)
def testCachedPredictionFromPreviousTree(self):
"""Tests the predictions work when we have cache from previous trees."""
with self.test_session() as session:
tree_ensemble_config = boosted_trees_pb2.TreeEnsemble()
text_format.Merge(
"""
trees {
nodes {
bucketized_split {
feature_id: 1
threshold: 28
left_id: 1
right_id: 2
}
metadata {
gain: 7.62
}
}
nodes {
leaf {
scalar: 1.14
}
}
nodes {
leaf {
scalar: 8.79
}
}
}
trees {
nodes {
bucketized_split {
feature_id: 1
threshold: 26
left_id: 1
right_id: 2
}
}
nodes {
bucketized_split {
feature_id: 0
threshold: 50
left_id: 3
right_id: 4
}
}
nodes {
leaf {
scalar: 7
}
}
nodes {
leaf {
scalar: 5
}
}
nodes {
leaf {
scalar: 6
}
}
}
trees {
nodes {
bucketized_split {
feature_id: 0
threshold: 34
left_id: 1
right_id: 2
}
}
nodes {
leaf {
scalar: -7.0
}
}
nodes {
leaf {
scalar: 5.0
}
}
}
tree_metadata {
is_finalized: true
}
tree_metadata {
is_finalized: true
}
tree_metadata {
is_finalized: false
}
tree_weights: 0.1
tree_weights: 0.1
tree_weights: 0.1
""", tree_ensemble_config)
# Create existing ensemble with one root split
tree_ensemble = boosted_trees_ops.TreeEnsemble(
'ensemble',
serialized_proto=tree_ensemble_config.SerializeToString())
tree_ensemble_handle = tree_ensemble.resource_handle
resources.initialize_resources(resources.shared_resources()).run()
# Two examples, one was cached in node 1 first, another in node 2.
cached_tree_ids = [0, 0]
cached_node_ids = [1, 0]
# We have two features: 0 and 1.
feature_0_values = [36, 32]
feature_1_values = [11, 27]
# Grow tree ensemble.
predict_op = boosted_trees_ops.training_predict(
tree_ensemble_handle,
cached_tree_ids=cached_tree_ids,
cached_node_ids=cached_node_ids,
bucketized_features=[feature_0_values, feature_1_values],
logits_dimension=1)
logits_updates, new_tree_ids, new_node_ids = session.run(
predict_op)
# Example 1 will get to node 3 in tree 1 and node 2 of tree 2
# Example 2 will get to node 2 in tree 1 and node 1 of tree 2
# We are in the last tree.
self.assertAllClose([2, 2], new_tree_ids)
# When using the full tree, the first example will end up in node 4,
# the second in node 5.
self.assertAllClose([2, 1], new_node_ids)
# Example 1: tree 0: 8.79, tree 1: 5.0, tree 2: 5.0 = >
# change = 0.1*(5.0+5.0)
# Example 2: tree 0: 1.14, tree 1: 7.0, tree 2: -7 = >
# change= 0.1(1.14+7.0-7.0)
self.assertAllClose([[1], [0.114]], logits_updates)
def testCategoricalSplits(self):
"""Tests the predictions work for categorical splits."""
with self.cached_session() as session:
tree_ensemble_config = boosted_trees_pb2.TreeEnsemble()
text_format.Merge(
"""
trees {
nodes {
categorical_split {
feature_id: 1
value: 2
left_id: 1
right_id: 2
}
}
nodes {
categorical_split {
feature_id: 0
value: 13
left_id: 3
right_id: 4
}
}
nodes {
leaf {
scalar: 7.0
}
}
nodes {
leaf {
scalar: 5.0
}
}
nodes {
leaf {
scalar: 6.0
}
}
}
tree_weights: 1.0
""", tree_ensemble_config)
# Create existing ensemble with one root split
tree_ensemble = boosted_trees_ops.TreeEnsemble(
'ensemble', serialized_proto=tree_ensemble_config.SerializeToString())
tree_ensemble_handle = tree_ensemble.resource_handle
resources.initialize_resources(resources.shared_resources()).run()
feature_0_values = [13, 1, 3]
feature_1_values = [2, 2, 1]
expected_logits = [[5.], [6.], [7.]]
# Prediction should work fine.
predict_op = boosted_trees_ops.predict(
tree_ensemble_handle,
bucketized_features=[feature_0_values, feature_1_values],
logits_dimension=1)
logits = session.run(predict_op)
self.assertAllClose(expected_logits, logits)
def testNoCachedPredictionButTreeExists(self):
"""Tests that predictions are updated once trees are added."""
with self.test_session() as session:
tree_ensemble_config = boosted_trees_pb2.TreeEnsemble()
text_format.Merge(
"""
trees {
nodes {
bucketized_split {
feature_id: 0
threshold: 15
left_id: 1
right_id: 2
}
metadata {
gain: 7.62
}
}
nodes {
leaf {
scalar: 1.14
}
}
nodes {
leaf {
scalar: 8.79
}
}
}
tree_weights: 0.1
tree_metadata {
is_finalized: true
num_layers_grown: 1
}
growing_metadata {
num_trees_attempted: 1
num_layers_attempted: 2
}
""", tree_ensemble_config)
# Create existing ensemble with one root split
tree_ensemble = boosted_trees_ops.TreeEnsemble(
'ensemble',
serialized_proto=tree_ensemble_config.SerializeToString())
tree_ensemble_handle = tree_ensemble.resource_handle
resources.initialize_resources(resources.shared_resources()).run()
# Two examples, none were cached before.
cached_tree_ids = [0, 0]
cached_node_ids = [0, 0]
feature_0_values = [67, 5]
# Grow tree ensemble.
predict_op = boosted_trees_ops.training_predict(
tree_ensemble_handle,
cached_tree_ids=cached_tree_ids,
cached_node_ids=cached_node_ids,
bucketized_features=[feature_0_values],
logits_dimension=1)
logits_updates, new_tree_ids, new_node_ids = session.run(
predict_op)
# We are in the first tree.
self.assertAllClose([0, 0], new_tree_ids)
self.assertAllClose([2, 1], new_node_ids)
self.assertAllClose([[0.1 * 8.79], [0.1 * 1.14]], logits_updates)
def _bt_model_fn(
features,
labels,
mode,
head,
feature_columns,
tree_hparams,
n_batches_per_layer,
config,
closed_form_grad_and_hess_fn=None,
example_id_column_name=None,
# TODO(youngheek): replace this later using other options.
train_in_memory=False,
name='boosted_trees'):
"""Gradient Boosted Trees model_fn.
Args:
features: dict of `Tensor`.
labels: `Tensor` of shape [batch_size, 1] or [batch_size] labels of
dtype `int32` or `int64` in the range `[0, n_classes)`.
mode: Defines whether this is training, evaluation or prediction.
See `ModeKeys`.
head: A `head_lib._Head` instance.
feature_columns: Iterable of `feature_column._FeatureColumn` model inputs.
tree_hparams: TODO. collections.namedtuple for hyper parameters.
n_batches_per_layer: A `Tensor` of `int64`. Each layer is built after at
least n_batches_per_layer accumulations.
config: `RunConfig` object to configure the runtime settings.
closed_form_grad_and_hess_fn: a function that accepts logits and labels
and returns gradients and hessians. By default, they are created by
tf.gradients() from the loss.
example_id_column_name: Name of the feature for a unique ID per example.
Currently experimental -- not exposed to public API.
train_in_memory: `bool`, when true, it assumes the dataset is in memory,
i.e., input_fn should return the entire dataset as a single batch, and
also n_batches_per_layer should be set as 1.
name: Name to use for the model.
Returns:
An `EstimatorSpec` instance.
Raises:
ValueError: mode or params are invalid, or features has the wrong type.
"""
is_single_machine = (config.num_worker_replicas <= 1)
sorted_feature_columns = sorted(feature_columns, key=lambda tc: tc.name)
if train_in_memory:
assert n_batches_per_layer == 1, (
'When train_in_memory is enabled, input_fn should return the entire '
'dataset as a single batch, and n_batches_per_layer should be set as '
'1.')
if (not config.is_chief or config.num_worker_replicas > 1 or
config.num_ps_replicas > 0):
raise ValueError('train_in_memory is supported only for '
'non-distributed training.')
worker_device = control_flow_ops.no_op().device
# maximum number of splits possible in the whole tree =2^(D-1)-1
# TODO(youngheek): perhaps storage could be optimized by storing stats with
# the dimension max_splits_per_layer, instead of max_splits (for the entire
# tree).
max_splits = (1 << tree_hparams.max_depth) - 1
train_op = []
with ops.name_scope(name) as name:
# Prepare.
global_step = training_util.get_or_create_global_step()
bucket_size_list, feature_ids_list = _group_features_by_num_buckets(
sorted_feature_columns)
# Extract input features and set up cache for training.
training_state_cache = None
if mode == model_fn.ModeKeys.TRAIN and train_in_memory:
# cache transformed features as well for in-memory training.
batch_size = array_ops.shape(labels)[0]
input_feature_list, input_cache_op = (
_cache_transformed_features(features, sorted_feature_columns,
batch_size))
train_op.append(input_cache_op)
training_state_cache = _CacheTrainingStatesUsingVariables(
batch_size, head.logits_dimension)
else:
input_feature_list = _get_transformed_features(features,
sorted_feature_columns)
if mode == model_fn.ModeKeys.TRAIN and example_id_column_name:
example_ids = features[example_id_column_name]
training_state_cache = _CacheTrainingStatesUsingHashTable(
example_ids, head.logits_dimension)
# Create Ensemble resources.
tree_ensemble = boosted_trees_ops.TreeEnsemble(name=name)
# Create logits.
if mode != model_fn.ModeKeys.TRAIN:
logits = boosted_trees_ops.predict(
# For non-TRAIN mode, ensemble doesn't change after initialization,
# so no local copy is needed; using tree_ensemble directly.
tree_ensemble_handle=tree_ensemble.resource_handle,
bucketized_features=input_feature_list,
logits_dimension=head.logits_dimension)
else:
if is_single_machine:
local_tree_ensemble = tree_ensemble
ensemble_reload = control_flow_ops.no_op()
else:
# Have a local copy of ensemble for the distributed setting.
with ops.device(worker_device):
local_tree_ensemble = boosted_trees_ops.TreeEnsemble(
name=name + '_local', is_local=True)
# TODO(soroush): Do partial updates if this becomes a bottleneck.
ensemble_reload = local_tree_ensemble.deserialize(
*tree_ensemble.serialize())
if training_state_cache:
cached_tree_ids, cached_node_ids, cached_logits = (
training_state_cache.lookup())
else:
# Always start from the beginning when no cache is set up.
batch_size = array_ops.shape(labels)[0]
cached_tree_ids, cached_node_ids, cached_logits = (
array_ops.zeros([batch_size], dtype=dtypes.int32),
array_ops.zeros([batch_size], dtype=dtypes.int32),
array_ops.zeros(
[batch_size, head.logits_dimension], dtype=dtypes.float32))
with ops.control_dependencies([ensemble_reload]):
(stamp_token, num_trees, num_finalized_trees, num_attempted_layers,
last_layer_nodes_range) = local_tree_ensemble.get_states()
summary.scalar('ensemble/num_trees', num_trees)
summary.scalar('ensemble/num_finalized_trees', num_finalized_trees)
summary.scalar('ensemble/num_attempted_layers', num_attempted_layers)
partial_logits, tree_ids, node_ids = boosted_trees_ops.training_predict(
tree_ensemble_handle=local_tree_ensemble.resource_handle,
cached_tree_ids=cached_tree_ids,
cached_node_ids=cached_node_ids,
bucketized_features=input_feature_list,
logits_dimension=head.logits_dimension)
logits = cached_logits + partial_logits
# Create training graph.
def _train_op_fn(loss):
"""Run one training iteration."""
if training_state_cache:
train_op.append(training_state_cache.insert(tree_ids, node_ids, logits))
if closed_form_grad_and_hess_fn:
gradients, hessians = closed_form_grad_and_hess_fn(logits, labels)
else:
gradients = gradients_impl.gradients(loss, logits, name='Gradients')[0]
hessians = gradients_impl.gradients(
gradients, logits, name='Hessians')[0]
stats_summaries_list = []
for i, feature_ids in enumerate(feature_ids_list):
num_buckets = bucket_size_list[i]
summaries = [
array_ops.squeeze(
boosted_trees_ops.make_stats_summary(
node_ids=node_ids,
gradients=gradients,
hessians=hessians,
bucketized_features_list=[input_feature_list[f]],
max_splits=max_splits,
num_buckets=num_buckets),
axis=0) for f in feature_ids
]
stats_summaries_list.append(summaries)
accumulators = []
def grow_tree_from_stats_summaries(stats_summaries_list,
feature_ids_list):
"""Updates ensemble based on the best gains from stats summaries."""
node_ids_per_feature = []
gains_list = []
thresholds_list = []
left_node_contribs_list = []
right_node_contribs_list = []
all_feature_ids = []
assert len(stats_summaries_list) == len(feature_ids_list)
for i, feature_ids in enumerate(feature_ids_list):
(numeric_node_ids_per_feature, numeric_gains_list,
numeric_thresholds_list, numeric_left_node_contribs_list,
numeric_right_node_contribs_list) = (
boosted_trees_ops.calculate_best_gains_per_feature(
node_id_range=last_layer_nodes_range,
stats_summary_list=stats_summaries_list[i],
l1=tree_hparams.l1,
l2=tree_hparams.l2,
tree_complexity=tree_hparams.tree_complexity,
min_node_weight=tree_hparams.min_node_weight,
max_splits=max_splits))
all_feature_ids += feature_ids
node_ids_per_feature += numeric_node_ids_per_feature
gains_list += numeric_gains_list
thresholds_list += numeric_thresholds_list
left_node_contribs_list += numeric_left_node_contribs_list
right_node_contribs_list += numeric_right_node_contribs_list
grow_op = boosted_trees_ops.update_ensemble(
# Confirm if local_tree_ensemble or tree_ensemble should be used.
tree_ensemble.resource_handle,
feature_ids=all_feature_ids,
node_ids=node_ids_per_feature,
gains=gains_list,
thresholds=thresholds_list,
left_node_contribs=left_node_contribs_list,
right_node_contribs=right_node_contribs_list,
learning_rate=tree_hparams.learning_rate,
max_depth=tree_hparams.max_depth,
pruning_mode=boosted_trees_ops.PruningMode.NO_PRUNING)
return grow_op
if train_in_memory and is_single_machine:
train_op.append(distribute_lib.increment_var(global_step))
train_op.append(
grow_tree_from_stats_summaries(stats_summaries_list,
feature_ids_list))
else:
dependencies = []
for i, feature_ids in enumerate(feature_ids_list):
stats_summaries = stats_summaries_list[i]
accumulator = data_flow_ops.ConditionalAccumulator(
dtype=dtypes.float32,
# The stats consist of grads and hessians (the last dimension).
shape=[len(feature_ids), max_splits, bucket_size_list[i], 2],
shared_name='numeric_stats_summary_accumulator_' + str(i))
accumulators.append(accumulator)
apply_grad = accumulator.apply_grad(
array_ops.stack(stats_summaries, axis=0), stamp_token)
dependencies.append(apply_grad)
def grow_tree_from_accumulated_summaries_fn():
"""Updates the tree with the best layer from accumulated summaries."""
# Take out the accumulated summaries from the accumulator and grow.
stats_summaries_list = []
stats_summaries_list = [
array_ops.unstack(accumulator.take_grad(1), axis=0)
for accumulator in accumulators
]
grow_op = grow_tree_from_stats_summaries(stats_summaries_list,
feature_ids_list)
return grow_op
with ops.control_dependencies(dependencies):
train_op.append(distribute_lib.increment_var(global_step))
if config.is_chief:
min_accumulated = math_ops.reduce_min(
array_ops.stack(
[acc.num_accumulated() for acc in accumulators]))
train_op.append(
control_flow_ops.cond(
math_ops.greater_equal(min_accumulated,
n_batches_per_layer),
grow_tree_from_accumulated_summaries_fn,
control_flow_ops.no_op,
name='wait_until_n_batches_accumulated'))
return control_flow_ops.group(train_op, name='train_op')
estimator_spec = head.create_estimator_spec(
features=features,
mode=mode,
labels=labels,
train_op_fn=_train_op_fn,
logits=logits)
if mode == model_fn.ModeKeys.TRAIN:
# Add an early stop hook.
estimator_spec = estimator_spec._replace(
training_hooks=estimator_spec.training_hooks +
(_StopAtAttemptsHook(num_finalized_trees, num_attempted_layers,
tree_hparams.n_trees, tree_hparams.max_depth),))
return estimator_spec
def testPredictionMultipleTree(self):
"""Tests the predictions work when we have multiple trees."""
with self.test_session() as session:
tree_ensemble_config = boosted_trees_pb2.TreeEnsemble()
text_format.Merge(
"""
trees {
nodes {
bucketized_split {
feature_id: 1
threshold: 28
left_id: 1
right_id: 2
}
metadata {
gain: 7.62
}
}
nodes {
leaf {
scalar: 1.14
}
}
nodes {
leaf {
scalar: 8.79
}
}
}
trees {
nodes {
bucketized_split {
feature_id: 1
threshold: 26
left_id: 1
right_id: 2
}
}
nodes {
bucketized_split {
feature_id: 0
threshold: 50
left_id: 3
right_id: 4
}
}
nodes {
leaf {
scalar: 7.0
}
}
nodes {
leaf {
scalar: 5.0
}
}
nodes {
leaf {
scalar: 6.0
}
}
}
trees {
nodes {
bucketized_split {
feature_id: 0
threshold: 34
left_id: 1
right_id: 2
}
}
nodes {
leaf {
scalar: -7.0
}
}
nodes {
leaf {
scalar: 5.0
}
}
}
tree_weights: 0.1
tree_weights: 0.2
tree_weights: 1.0
""", tree_ensemble_config)
# Create existing ensemble with one root split
tree_ensemble = boosted_trees_ops.TreeEnsemble(
'ensemble',
serialized_proto=tree_ensemble_config.SerializeToString())
tree_ensemble_handle = tree_ensemble.resource_handle
resources.initialize_resources(resources.shared_resources()).run()
feature_0_values = [36, 32]
feature_1_values = [11, 27]
# Example 1: tree 0: 1.14, tree 1: 5.0, tree 2: 5.0 = >
# logit = 0.1*5.0+0.2*5.0+1*5
# Example 2: tree 0: 1.14, tree 1: 7.0, tree 2: -7 = >
# logit= 0.1*1.14+0.2*7.0-1*7.0
expected_logits = [[6.114], [-5.486]]
# Do with parallelization, e.g. EVAL
predict_op = boosted_trees_ops.predict(
tree_ensemble_handle,
bucketized_features=[feature_0_values, feature_1_values],
logits_dimension=1)
logits = session.run(predict_op)
self.assertAllClose(expected_logits, logits)
# Do without parallelization, e.g. INFER - the result is the same
predict_op = boosted_trees_ops.predict(
tree_ensemble_handle,
bucketized_features=[feature_0_values, feature_1_values],
logits_dimension=1)
logits = session.run(predict_op)
self.assertAllClose(expected_logits, logits)
def testContribsMultipleTreeWhenFirstTreeIsABiasNode(self):
"""Tests case when, after training, first tree contains only a bias node."""
with self.test_session() as session:
tree_ensemble_config = boosted_trees_pb2.TreeEnsemble()
text_format.Merge(
"""
trees {
nodes {
leaf {
scalar: 1.72
}
}
}
trees {
nodes {
bucketized_split {
feature_id: 2
threshold: 26
left_id: 1
right_id: 2
}
}
nodes {
bucketized_split {
feature_id: 0
threshold: 50
left_id: 3
right_id: 4
}
metadata {
original_leaf: {scalar: 5.5}
}
}
nodes {
leaf {
scalar: 7.0
}
}
nodes {
leaf {
scalar: 5.0
}
}
nodes {
leaf {
scalar: 6.0
}
}
}
tree_weights: 1.
tree_weights: 0.1
tree_metadata: {
num_layers_grown: 0
}
tree_metadata: {
num_layers_grown: 1
}
""", tree_ensemble_config)
tree_ensemble = boosted_trees_ops.TreeEnsemble(
'ensemble',
serialized_proto=tree_ensemble_config.SerializeToString())
tree_ensemble_handle = tree_ensemble.resource_handle
resources.initialize_resources(resources.shared_resources()).run()
feature_0_values = [36, 32]
feature_1_values = [13, -29] # Unused feature.
feature_2_values = [11, 27]
# Expected logits are computed by traversing the logit path and
# subtracting child logits from parent logits.
expected_feature_ids = ((2, 0), (2, ))
# bias = 1.72 * 1. # Root node of tree_0.
# example_0 : (bias, 0.1 * 5.5 + bias, 0.1 * 5. + bias)
# example_1 : (bias, 0.1 * 7. + bias )
expected_logits_paths = ((1.72, 2.27, 2.22), (1.72, 2.42))
bucketized_features = [
feature_0_values, feature_1_values, feature_2_values
]
debug_op = boosted_trees_ops.example_debug_outputs(
tree_ensemble_handle,
bucketized_features=bucketized_features,
logits_dimension=1)
serialized_examples_debug_outputs = session.run(debug_op)
feature_ids = []
logits_paths = []
for example in serialized_examples_debug_outputs:
example_debug_outputs = boosted_trees_pb2.DebugOutput()
example_debug_outputs.ParseFromString(example)
feature_ids.append(example_debug_outputs.feature_ids)
logits_paths.append(example_debug_outputs.logits_path)
self.assertAllClose(feature_ids, expected_feature_ids)
self.assertAllClose(logits_paths, expected_logits_paths)