- 24 3月, 2017 1 次提交
-
-
由 annarev 提交于
-
- 23 3月, 2017 2 次提交
-
-
由 Amit Patankar 提交于
Cherrypick fix windows gpu build (#8603)
-
由 guschmue 提交于
-
- 22 3月, 2017 9 次提交
-
-
由 Martin Wicke 提交于
Cherrypicking Asim's changes.
-
由 Asim Shankar 提交于
This is one step towards #6926. The idea is that if the native libraries are packaged in a .jar file available in the classpath of the JVM, then they can be extracted and used. Change: 150697993
-
由 Amit Patankar 提交于
-
由 annarev 提交于
* Release notes for 1.1 * Indented RNNCell note * Removed duplicate note * Updated RNN cell note * Updated RNN cell note * Updated spectral comment, mac gpu support comment and fixed username * Reword tf.spectral message.
-
由 Sean Papay 提交于
* implemented partial shape inference for tensordot tensordot should now correctly infer the partial shape of the product, whenever possible * updated tensordot tests to test partial shape inference * removed TODO note
-
由 Yun Peng 提交于
In e8aefd21, a dependency: //tensorflow/tools/pip_package:build_pip_package //tensorflow/tools/pip_package:simple_console_for_windows //tensorflow/python/debug:debug_pip //tensorflow/python/debug:debug_examples //tensorflow/python/debug:debug_mnist //tensorflow/examples/tutorials/mnist:input_data //tensorflow/contrib/learn/python/learn/datasets:datasets //tensorflow/contrib/framework:framework_py //tensorflow/contrib/framework:python/ops/_variable_ops.so was introduced. Cut this dependency by excluding :debug_examples on Windows
-
由 Shanqing Cai 提交于
Fix some comments in the run() method of the BaseSession class.
-
由 Shanqing Cai 提交于
Comment referring to the wrong variable
-
由 A. Besir Kurtulmus 提交于
-
- 21 3月, 2017 28 次提交
-
-
由 Neal Wu 提交于
Fixed the order of arguments for softmax_loss_function in two places, including a semantic code change. Change: 150685391
-
由 A. Unique TensorFlower 提交于
Change: 150680902
-
由 A. Unique TensorFlower 提交于
This is calling Square<int32>(65280), the result doesn't fit into an int32. LLVM takes advantage of this and changes the behavior of the code, making the test fail when optimization is enabled (e.g. in msan and tsan mode). Change: 150672767
-
由 A. Unique TensorFlower 提交于
Change: 150671385
-
由 A. Unique TensorFlower 提交于
Change: 150670512
-
由 Shanqing Cai 提交于
Older versions of Windows (e.g., Windows 7) have lower temporal resolution in timestamps than other platforms. On such Windows machines, multiple dumps of the same or different tensors may end up with the same timestamp, causing 1) failures in the causality check in the constructor of DebugDumpDir, 2) occasional overwriting of previous tensor dumps and core metadata files, 3) loss of ordering information for debug watches of a tensor that is dumped multiple times during a Session::Run() call, e.g., due to While loops. This CL addresses those two issues: 1) DebugDumpDir._validate_dump_with_graphs now examines tensor dumps of the same timestamp to take into account the possibility of timestamp collision during dependency validation. 2) During the dumping of tensors, file path collisions are checked and if they occur, an index suffix (e.g., "-1", "-2") is appended to the file path. 3) The core metadata Events files now include session_run_count in their names, to avoid collision due to same timestamps. Change: 150670298
-
由 Shanqing Cai 提交于
Added three attributes to the debug op "DebugNumericSummary" used in tfdbg-based TensorBoard health pills: 1) lower_bound (type: float) 2) upper_bound (type: float) 3) mute_if_healthy (type: bool) lower_bound and upper_bound make it possible to customize thresholds beyond which tensor elements are counted as -inf or inf. mute_if_healthy makes it possible to mute a DebugNumericSummary op unless there are nan, -inf or inf elements in the watched tensor, which is useful for reducing the amount of health pill data. Changes are made in the C++ DebugNodeInserter class, so that these attributes can be directly set from Python methods such as tf_debug.watch_graph() using the following syntax in the debug_ops argument: debug_ops=["DebugNumericSummary(attribute_name=attribute_value)"] e.g., debug_ops=["DebugNumericSummary(lower_bound=-100.0; mute_if_healthy=true)"] Currently, string, float, int, and bool attribute value types are supported. Change: 150665493
-
由 A. Unique TensorFlower 提交于
Change: 150664674
-
由 Alexandre Passos 提交于
Change: 150664185
-
由 A. Unique TensorFlower 提交于
-
由 Yao Zhang 提交于
Change: 150663008
-
由 A. Unique TensorFlower 提交于
Change: 150659388
-
由 gunan 提交于
Branch 150655275
-
由 Shanqing Cai 提交于
Resolved conflicts in : tensorflow/compiler/xla/service/call_graph.cc tensorflow/compiler/xla/service/call_graph.h tensorflow/compiler/xla/service/call_graph_test.cc tensorflow/core/BUILD tensorflow/python/ops/image_ops_test.py
-
由 Yuefeng Zhou 提交于
Change: 150655275
-
由 Justin Lebar 提交于
Change: 150653208
-
由 A. Unique TensorFlower 提交于
Change: 150652150
-
由 Sukriti Ramesh 提交于
Change: 150651883
-
由 A. Unique TensorFlower 提交于
Change: 150650257
-
由 Shanqing Cai 提交于
Change: 150648734
-
由 Benoit Steiner 提交于
Change: 150648084
-
由 Jonathan Hseu 提交于
Change: 150646570
-
由 Pete Warden 提交于
Change: 150645199
-
由 Alexandre Passos 提交于
Change: 150642610
-
由 A. Unique TensorFlower 提交于
Include option to run TensorForest eval locally even if training is distributed, which is a common setup. Change: 150640056
-
由 Charles Nicholson 提交于
Change: 150633544
-
由 A. Unique TensorFlower 提交于
Fixes: #8047 Change: 150631022
-
由 A. Unique TensorFlower 提交于
Change: 150630961
-