aboutsummaryrefslogtreecommitdiffhomepage
path: root/tensorflow/contrib/boosted_trees
Commit message (Collapse)AuthorAge
* Export feature importance for oblivious tree nodes.Gravatar A. Unique TensorFlower2018-10-09
| | | | PiperOrigin-RevId: 216422334
* Automated rollback of commit 70f071f7afb2deffddbd9937d7a76b1e1c0b2b75Gravatar Revan Sopher2018-09-28
| | | | PiperOrigin-RevId: 215026418
* Updating the V2 variables API for boosted_trees.Gravatar A. Unique TensorFlower2018-09-28
| | | | PiperOrigin-RevId: 214952666
* Fix failing test.Gravatar Revan Sopher2018-09-27
| | | | PiperOrigin-RevId: 214875840
* Updating the V2 variables API.Gravatar Alexandre Passos2018-09-27
| | | | PiperOrigin-RevId: 214824023
* Move from deprecated self.test_session() to self.cached_session().Gravatar A. Unique TensorFlower2018-09-25
| | | | | | | | self.test_session() has been deprecated in 9962eb5e84b15e309410071b06c2ed2d6148ed44 as its name confuses readers of the test. Moving to cached_session() instead which is more explicit about: * the fact that the session may be reused. * the session is not closed even when doing a "with self.test_session()" statement. PiperOrigin-RevId: 214444907
* Fix bias feature being selected for splitting. The previous logic was broken ↵Gravatar A. Unique TensorFlower2018-09-21
| | | | | | for cases where all the examples in the last partition just had missing values. In those cases, the range that was selected for the leaf previous to the last included the bias value for the last leaf. PiperOrigin-RevId: 214046965
* Move from deprecated self.test_session() to self.cached_session().Gravatar A. Unique TensorFlower2018-09-21
| | | | | | | | self.test_session() has been deprecated in 9962eb5e84b15e309410071b06c2ed2d6148ed44 as its name confuses readers of the test. Moving to cached_session() instead which is more explicit about: * the fact that the session may be reused. * the session is not closed even when doing a "with self.test_session()" statement. PiperOrigin-RevId: 213944932
* Remove the CHECK added for debugging.Gravatar A. Unique TensorFlower2018-09-19
| | | | PiperOrigin-RevId: 213681549
* Increase test timeout for dnn_tree_combined_estimator_test to de-flake.Gravatar A. Unique TensorFlower2018-09-17
| | | | PiperOrigin-RevId: 213363558
* Improve the error messages in custom_export_strategy.Gravatar A. Unique TensorFlower2018-09-17
| | | | PiperOrigin-RevId: 213334465
* Move from deprecated self.test_session() to self.cached_session().Gravatar A. Unique TensorFlower2018-09-17
| | | | | | | | self.test_session() has been deprecated in 9962eb5e84b15e309410071b06c2ed2d6148ed44 as its name confuses readers of the test. Moving to cached_session() instead which is more explicit about: * the fact that the session may be reused. * the session is not closed even when doing a "with self.test_session()" statement. PiperOrigin-RevId: 213326581
* Add some debugging checks for categorical split handler.Gravatar A. Unique TensorFlower2018-09-13
| | | | | | Also use MIN_INT64 for the bias feature accumulation since categorical_feature_with_xyz use -1 for out of vocab features. PiperOrigin-RevId: 212855656
* Change HandleFromInput() to return a `const ResourceHandle&` and avoid ↵Gravatar Derek Murray2018-09-12
| | | | | | | | copying that type. This avoids unnecessary string copies and deallocations in the ReadVariableOp, and similar ops. PiperOrigin-RevId: 212652588
* Move from deprecated self.test_session() to self.cached_session().Gravatar A. Unique TensorFlower2018-09-10
| | | | | | | | self.test_session() has been deprecated in 9962eb5e84b15e309410071b06c2ed2d6148ed44 as its name confuses readers of the test. Moving to cached_session() instead which is more explicit about: * the fact that the session may be reused. * the session is not closed even when doing a "with self.test_session()" statement. PiperOrigin-RevId: 212336464
* Move from deprecated self.test_session() to self.cached_session().Gravatar A. Unique TensorFlower2018-09-10
| | | | | | | | self.test_session() has been deprecated in 9962eb5e84b15e309410071b06c2ed2d6148ed44 as its name confuses readers of the test. Moving to cached_session() instead which is more explicit about: * the fact that the session may be reused. * the session is not closed even when doing a "with self.test_session()" statement. PiperOrigin-RevId: 212336321
* Make num_quantiles configurable; update the epsilon value as well since ↵Gravatar A. Unique TensorFlower2018-09-06
| | | | | | epsilon controls the maximum number of quantiles generated. PiperOrigin-RevId: 211914388
* Fix categorical feature handler accumulator to use high precision 64 bit ↵Gravatar A. Unique TensorFlower2018-09-05
| | | | | | accumulator. PiperOrigin-RevId: 211642436
* Added support for categorical features.Gravatar A. Unique TensorFlower2018-08-28
| | | | | | Ops are now interconnected to support oblivious decision trees. PiperOrigin-RevId: 210642692
* Fixed a bug in the dense split handler ops.Gravatar A. Unique TensorFlower2018-08-27
| | | | PiperOrigin-RevId: 210412659
* Remove memset() for SummaryEntry in weighted quantiles. This is a bug ↵Gravatar A. Unique TensorFlower2018-08-23
| | | | | | whenever one of the template arguments is not a POD, in particular for types such as string. A memset on a struct that contains a string invalidates the internals of that class (in particular the pointer that stores the characters of the string) and causes segfaults. This allows using weighted quantiles beyond int/float like template parameters. The upstream issue with memcopy of alignment bytes used for padding has been solved separately in the meantime. PiperOrigin-RevId: 209906298
* Fix the interaction between the split handler ops and the grow tree ensemble ↵Gravatar A. Unique TensorFlower2018-08-22
| | | | | | ops when we have nodes with no examples. PiperOrigin-RevId: 209830550
* Move from deprecated self.test_session() to self.cached_session().Gravatar A. Unique TensorFlower2018-08-21
| | | | | | | | self.test_session() has been deprecated in 9962eb5e84b15e309410071b06c2ed2d6148ed44 as its name confuses readers of the test. Moving to cached_session() instead which is more explicit about: * the fact that the session may be reused. * the session is not closed even when doing a "with self.test_session()" statement. PiperOrigin-RevId: 209700634
* Move from deprecated self.test_session() to self.session() when a graph is set.Gravatar A. Unique TensorFlower2018-08-21
| | | | | | self.test_session() has been deprecated in cl/208545396 as its behavior confuses readers of the test. Moving to self.session() instead. PiperOrigin-RevId: 209696110
* fix C++ header guards.Gravatar A. Unique TensorFlower2018-08-21
| | | | PiperOrigin-RevId: 209679086
* Merged commit includes the following changes:Gravatar Yifei Feng2018-08-21
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 209663919 by yifeif<yifeif@google.com>: Internal change. -- 209663914 by amitpatankar<amitpatankar@google.com>: Fix the topk_op_test for numpy>1.15. -- 209660476 by jdduke<jdduke@google.com>: Fix model lifetime for TensorFlow Lite C# bindings Ensure the model's existence for the duration of the interpreter, as per API requirements. -- 209655960 by scottzhu<scottzhu@google.com>: Unify RNN Cell interface between TF and Keras. -- 209655731 by A. Unique TensorFlower<gardener@tensorflow.org>: Added tests for PredictionOps and PartitionExamplesOps -- 209655291 by nolivia<nolivia@google.com>: adding rate class so that we can save global_step/sec using tf.contrib.summary. The function takes the rate in relation to any tensors provided that the numerator and denominator are broadcastable and have dtypes that can be cast to float64 -- 209654655 by kramerb<kramerb@google.com>: [XLA] Switch from tensorflow::gtl::InlinedVector to absl::InlinedVector This one comes with extra goodies like a move constructor. -- 209653851 by A. Unique TensorFlower<gardener@tensorflow.org>: Internal build specification change -- PiperOrigin-RevId: 209663919
* 1) Update the proto files for oblivious trees.Gravatar A. Unique TensorFlower2018-08-21
| | | | | | 2) Grow a new layer of an oblivious tree. PiperOrigin-RevId: 209633300
* Adding weighted categorical feature column support.Gravatar A. Unique TensorFlower2018-08-16
| | | | PiperOrigin-RevId: 209058978
* Create a tf.constant for the weak_learner_type that's shared across all theGravatar A. Unique TensorFlower2018-08-14
| | | | | | handlers to avoid duplicate constant creation. PiperOrigin-RevId: 208755852
* First iteration of oblivious tree split handling for dense features.Gravatar A. Unique TensorFlower2018-08-14
| | | | PiperOrigin-RevId: 208705535
* Do not bucketize when buckets are empty. This is for situations when for a ↵Gravatar A. Unique TensorFlower2018-08-08
| | | | | | rare feature, no buckets were found. PiperOrigin-RevId: 207920196
* Estimator test for multiclass, core head for multiclass.Gravatar A. Unique TensorFlower2018-08-03
| | | | PiperOrigin-RevId: 207268708
* Add a default value of None for the newly added parameter.Gravatar A. Unique TensorFlower2018-08-02
| | | | PiperOrigin-RevId: 207135538
* Allow to set global step to a particular value, after the early stopping ↵Gravatar A. Unique TensorFlower2018-08-01
| | | | | | triggered by the number of trees fired. PiperOrigin-RevId: 207024504
* Store max_tree_depth as a Variable which can be changed after graph ↵Gravatar A. Unique TensorFlower2018-07-29
| | | | | | construction. PiperOrigin-RevId: 206497794
* Adding core estimator for ranking.Gravatar A. Unique TensorFlower2018-07-27
| | | | PiperOrigin-RevId: 206318440
* Adding core estimator for a fusion model.Gravatar A. Unique TensorFlower2018-07-26
| | | | PiperOrigin-RevId: 206183643
* fixing bug for when num of trees is not setGravatar A. Unique TensorFlower2018-07-26
| | | | PiperOrigin-RevId: 206171508
* Improve the documentation and error checking for invalid values of epsilon ↵Gravatar A. Unique TensorFlower2018-07-25
| | | | | | in weighted quantile stream. PiperOrigin-RevId: 206080755
* Adding an opportunity to skip the check for input tensor placements.Gravatar A. Unique TensorFlower2018-07-25
| | | | PiperOrigin-RevId: 206031015
* Adding core interface to a contrib versionGravatar A. Unique TensorFlower2018-07-23
| | | | PiperOrigin-RevId: 205728990
* Add leaf index modes as an argument, which specifies when output leaf indices.Gravatar A. Unique TensorFlower2018-07-20
| | | | PiperOrigin-RevId: 205443722
* Merge pull request #20807 from yongtang:20776-bostonGravatar TensorFlower Gardener2018-07-17
|\ | | | | | | PiperOrigin-RevId: 204928483
* | Use the safe sparse tensor API that returns errors rather than crashingGravatar A. Unique TensorFlower2018-07-16
| | | | | | | | | | | | in all TensorFlow core kernels. PiperOrigin-RevId: 204782675
| * Pylint fixGravatar Yong Tang2018-07-14
| | | | | | | | Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
| * Fix bug in custom_export_strategy.py for python 3Gravatar Yong Tang2018-07-14
| | | | | | | | | | | | by add compat.as_bytes Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
| * Fix bug inside boston.py in boosted_trees with python 3Gravatar Yong Tang2018-07-14
|/ | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This fix tries to address the issue raised in 20776 where run the sample with boston.py in boosted_trees throws out error with python 3: ``` $ python3 boston.py \ > --batch_size=404 --output_dir="/tmp/boston" --depth=4 --learning_rate=0.1 \ > --num_eval_steps=1 --num_trees=500 --l2=0.001 \ > --vmodule=training_ops=1 ...... ...... Traceback (most recent call last): File "boston.py", line 169, in <module> tf.app.run(main=main, argv=[sys.argv[0]] + unparsed) ...... ...... File "/usr/lib/python3.5/genericpath.py", line 145, in _check_arg_types raise TypeError("Can't mix strings and bytes in path components") from None TypeError: Can't mix strings and bytes in path components ``` The reason for the error was because in python 3, the export_dir was returned as bytes. This fix as the `compat.as_bytes` which is consistent with other places in tensorflow. This fix fixes 20776 Signed-off-by: Yong Tang <yong.tang.github@outlook.com>
* Avoid breakages of the TF weighted quantiles library if used with values ↵Gravatar A. Unique TensorFlower2018-07-11
| | | | | | being non-numeric types such as strings. PiperOrigin-RevId: 204087922
* Adding ranking support (over paired data for train and eval and unpaired ↵Gravatar A. Unique TensorFlower2018-07-09
| | | | | | data for inference). PiperOrigin-RevId: 203791296
* Adds reset op as an optional return from update_stats.Gravatar A. Unique TensorFlower2018-06-28
| | | | PiperOrigin-RevId: 202528760
generated by cgit v1.2.3 (git 2.39.1) at 2024-06-14 16:04:31 +0000