Skip to content

patch out bug in collective primitive in TensorFlow 2.2.0 #11175

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
boegel merged 1 commit into from
Sep 25, 2020
Merged

patch out bug in collective primitive in TensorFlow 2.2.0 #11175

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
22 changes: 22 additions & 0 deletions easybuild/easyconfigs/t/TensorFlow/TensorFlow-2.1.0_fix-collective-all-reduce-strategy.patch
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,22 @@
Fix for AttributeError: 'CollectiveAllReduceExtended' object has no attribute '_cfer_fn_cache'
See https://github.com/tensorflow/tensorflow/issues/39417

Author: Alexander Grund
--- tensorflow-2.1.0-orig/tensorflow/python/distribute/collective_all_reduce_strategy.py 2020-05-11 18:14:03.682085572 +0200
+++ tensorflow-2.1.0/tensorflow/python/distribute/collective_all_reduce_strategy.py 2020-05-11 18:15:41.623409641 +0200
@@ -19,6 +19,7 @@
from __future__ import print_function

import copy
+import weakref

from tensorflow.core.protobuf import config_pb2
from tensorflow.core.protobuf import rewriter_config_pb2
@@ -172,6 +173,7 @@
cross_device_ops_lib.CollectiveCommunication)
self._communication = communication
self._initialize_strategy(cluster_resolver)
+ self._cfer_fn_cache = weakref.WeakKeyDictionary()
assert isinstance(self._get_cross_device_ops(),
cross_device_ops_lib.CollectiveAllReduce)

3 changes: 3 additions & 0 deletions easybuild/easyconfigs/t/TensorFlow/TensorFlow-2.2.0-foss-2019b-Python-3.7.4.eb
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -129,6 +129,7 @@ exts_list = [
'patches': [
'TensorFlow-1.14.0_swig-env.patch',
'TensorFlow-2.1.0_fix-cuda-build.patch',
'TensorFlow-2.1.0_fix-collective-all-reduce-strategy.patch',
],
'source_tmpl': 'v%(version)s.tar.gz',
'source_urls': ['https://github.com/tensorflow/tensorflow/archive/'],
Expand All @@ -137,6 +138,8 @@ exts_list = [
'69cd836f87b8c53506c4f706f655d423270f5a563b76dc1cfa60fbc3184185a3', # v2.2.0.tar.gz
'b83cce6b91c7d19b8b320158ffc50fb4b2de454f5ac191c58d704234a1bf9005', # TensorFlow-1.14.0_swig-env.patch
'78c20aeaa7784b8ceb46238a81e8c2461137d28e0b576deeba8357d23fbe1f5a', # TensorFlow-2.1.0_fix-cuda-build.patch
# TensorFlow-2.1.0_fix-collective-all-reduce-strategy.patch
'093f4dd3ec372a82d50dffe32eea6821025cd1c406911a746c4367a40bc38486',
],
}),
]
Expand Down
3 changes: 3 additions & 0 deletions easybuild/easyconfigs/t/TensorFlow/TensorFlow-2.2.0-fosscuda-2019b-Python-3.7.4.eb
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -131,6 +131,7 @@ exts_list = [
'patches': [
'TensorFlow-1.14.0_swig-env.patch',
'TensorFlow-2.1.0_fix-cuda-build.patch',
'TensorFlow-2.1.0_fix-collective-all-reduce-strategy.patch',
],
'source_tmpl': 'v%(version)s.tar.gz',
'source_urls': ['https://github.com/tensorflow/tensorflow/archive/'],
Expand All @@ -139,6 +140,8 @@ exts_list = [
'69cd836f87b8c53506c4f706f655d423270f5a563b76dc1cfa60fbc3184185a3', # v2.2.0.tar.gz
'b83cce6b91c7d19b8b320158ffc50fb4b2de454f5ac191c58d704234a1bf9005', # TensorFlow-1.14.0_swig-env.patch
'78c20aeaa7784b8ceb46238a81e8c2461137d28e0b576deeba8357d23fbe1f5a', # TensorFlow-2.1.0_fix-cuda-build.patch
# TensorFlow-2.1.0_fix-collective-all-reduce-strategy.patch
'093f4dd3ec372a82d50dffe32eea6821025cd1c406911a746c4367a40bc38486',
],
}),
]
Expand Down