diff options
| author |  Allen Lavoie <allenl@google.com> | 2018-09-04 14:03:08 -0700 |
|---|---|---|
| committer |  TensorFlower Gardener <gardener@tensorflow.org> | 2018-09-04 14:26:01 -0700 |
| commit | d29eb6d1c9d1e4b2f601864f53878674f219fe6f (patch) | |
| tree | 4292df644dad3f34e813ddb742c31aa730e64fba /RELEASE.md | |
| parent | 06e8109af2e5ae5bc149e25fc64fbf66d6c8b817 (diff) | |
Remove reference cycles when constructing distribution objects
self -> _parameters -> self cycles were creating work for Python's garbage collector in training loops, where Distribution objects may be created repeatedly when executing eagerly. This CL just fixes that narrow memory issue; I'm not convinced dict(locals()) is super efficient, so we may want to follow up on that for performance.
Adds a few unit tests tests with run_test_in_graph_and_eager_modes(assert_no_eager_garbage=True). It'd be nice to expand this coverage over time.
Includes a small test_util simplification to support this (TFP tests don't like reset_default_graph for some reason). Testing for cycles in the TFP repo will need to wait on the Normal changes from the TF repo syncing.
PiperOrigin-RevId: 211520394
Diffstat (limited to 'RELEASE.md')
0 files changed, 0 insertions, 0 deletions