aboutsummaryrefslogtreecommitdiffhomepage
Commit message (Collapse)AuthorAge
* fixes #3933 virtualenv using python 2.6 siteGravatar Olivia Nordquist2016-08-23
| | | | Change: 131069999
* Automated rollback of change 130904334Gravatar Sergio Guadarrama2016-08-23
| | | | Change: 131069022
* Initial version of the Go API. The API is subject to change.Gravatar Jonathan Hseu2016-08-23
| | | | | | | | Remaining work to do: - Generated ops. - Generated protocol buffers. - A few calls requiring protocol buffers aren't in this change. Change: 131066649
* Fix a bug introduced in cl/130951050. Second attempt to make sure that no ↵Gravatar Yuan Yu2016-08-23
| | | | | | control edges by an outer control dependency context are added to nodes inside cond/while. Change: 131062259
* Automated rollback of change 131059518Gravatar A. Unique TensorFlower2016-08-23
| | | | Change: 131061856
* Improve error reporting for invalid schedule values.Gravatar A. Unique TensorFlower2016-08-23
| | | | Change: 131059518
* Update generated Python Op docs.Gravatar A. Unique TensorFlower2016-08-23
| | | | Change: 131058624
* TensorShape and Queue API fixesGravatar Eugene Brevdo2016-08-23
| | | | | | * Fix handling of TensorShape(TensorShape(None)); before it would loop forever * Add shape merging for QueueBase.from_list Change: 131058295
* Allow limiting epochs for memory sourcesGravatar David Soergel2016-08-23
| | | | Change: 131054894
* Added config property to BaseEstimator to allow retrieval of RunConfig.Gravatar A. Unique TensorFlower2016-08-23
| | | | Change: 131054755
* Wire num_threads from in_memory_source to enqueue_dataGravatar David Soergel2016-08-23
| | | | Change: 131047076
* Fixed bug in _PartitionedVariable that prevents it from being created within ↵Gravatar A. Unique TensorFlower2016-08-23
| | | | | | cond()/while_loop() and then subsequently used outside it. The error is: InvalidArgumentError: All inputs to node add must be from the same frame. Change: 131036071
* Edit random ops test to minimize the use of use_gpu arg in session.Gravatar Gunhan Gulsoy2016-08-22
| | | | Change: 131012545
* Begin transition to use NodeDef in FunctionDef instead ofGravatar A. Unique TensorFlower2016-08-22
| | | | | FunctionDef.Node. Change: 131009401
* Support engineering notation in sort algorithmGravatar Justine Tunney2016-08-22
| | | | Change: 131007072
* Exceptions thrown in the session destructor get ignored and reported to ↵Gravatar A. Unique TensorFlower2016-08-22
| | | | | | | | | STDERR. For example calling session.Reset causes the session destructor to throw an error on sess.close() which gets reported in STDERR as: "Exception google3.third_party.tensorflow.python.framework.errors.AbortedError: AbortedError() in <bound method Session.__del__ of <google3.third_party.tensorflow.python.client.session.Session object at 0x7f7f020e1750>> ignored". Since the exception is always ignored, catching and passing on the exception in the destructor will make for clean STDERR logs. Change: 131006510
* Update generated Python Op docs.Gravatar A. Unique TensorFlower2016-08-22
| | | | Change: 131003245
* Fixing spacing issue in docs for tf.meshgrid as per b/30845368.Gravatar A. Unique TensorFlower2016-08-22
| | | | Change: 131001422
* Update generated Python Op docs.Gravatar A. Unique TensorFlower2016-08-22
| | | | Change: 130998646
* RNN API updates and reduce_min/reduce_max bugfixes.Gravatar Eugene Brevdo2016-08-22
| | | | | | * Expose dynamic_bidirectional_rnn and raw_rnn to the public api. * Add GPU ops for reduce_max/reduce_min, so rnn code will work even with hard device placement. Change: 130997558
* Bugfixes for softplusGravatar Eugene Brevdo2016-08-22
| | | | | | * Fix underflow issues in softplus * Fix overflow issues in softplus for float16s. Change: 130997105
* Update makefiles for recent change splitting NodeDef out ofGravatar A. Unique TensorFlower2016-08-22
| | | | | graph.proto into node_def.proto. Change: 130993653
* Update generated Python Op docs.Gravatar A. Unique TensorFlower2016-08-22
| | | | Change: 130993625
* Fix up documentation for sample_n which allows for `n` to be int64.Gravatar A. Unique TensorFlower2016-08-22
| | | | Change: 130992111
* Update generated Python Op docs.Gravatar A. Unique TensorFlower2016-08-22
| | | | Change: 130983143
* Split NodeDef out of graph.proto into node_def.proto. Needed so weGravatar A. Unique TensorFlower2016-08-22
| | | | | can use NodeDef in FunctionDef. Change: 130982373
* Remove missing `keep_keys` arg from pydoc.Gravatar A. Unique TensorFlower2016-08-22
| | | | Change: 130981665
* Implement ShapeManager, which will be used to perform graph-levelGravatar Vijay Vasudevan2016-08-22
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | shape inference in C++ using the C++ shape functions. The interface for this class follows the way Graph building works: you pass a Node at a time, and it creates an InferenceContext per Node. On every call to AddNode, it requests the shapes from the InferenceContexts of all the inputs, and it creates the InferenceContext and runs the shape function in C++ to generate the new node's InferenceContext data. One possible use of this class will be to perform a ReverseDFS (data-dep order) traversal of a Graph, calling AddNode each time, while another use of this function will be to add nodes as they are created in python, etc -- the only requirement is that the nodes be added in the data dependency order, so that a node's input shapes have been computed before the node. As a start for doing as effectively as python does, we try to materialize the input_tensor values by statically analyzing the graph in the same way python's constant_value works, but I've only implemented "Const" for now. I'll do the rest in a follow up CL to bring parity to the C++ implementation. Since we do not know which inputs are needed before the shape function runs, we run the inference function once to find out which tensors are needed, and then we call it again once we materialize the tensors. (I tried to get rid of the existing InferenceContext constructor, but it's difficult: moving the creation of shapes from strings relies on an InferenceContext in the first place! There's probably a way to do this, but this can be an internal refactor). This is only a small step towards where we want to be with C++ Shape inference. This does not yet implement lazy-running of the graphs to materialize as much of the graph as is computable, and does not yet support iterative shape inference (forward and backward inference until convergence). These features will be built incrementally as we discover any subtle nuances that make these features challenging. Change: 130981592
* Specifies individual tensor ranks in sparse_placeholder.Gravatar A. Unique TensorFlower2016-08-22
| | | | Change: 130980786
* Use asciinumerical sort for tooltips and moreGravatar Justine Tunney2016-08-22
| | | | | | In order to accomplish this, I refactored the new sorting algorithm out into a separate tf-sorting component. Change: 130979195
* Reduced some unnecessary abstraction layers in worker call sending path.Gravatar A. Unique TensorFlower2016-08-22
| | | | | | | | | Also cleaned up a bunch of boilerplate. Reduced allocations in Call tag management by inlining Tags into Call. Reduces CPU usage for an RPC intensive benchmark by ~2% Change: 130971469
* Autogenerated Change: Release TensorBoard at TAG: 26Gravatar Dan Mané2016-08-22
| | | | Change: 130970371
* Wrap all ops created by report_uninitialized_variables in a name_scope. ↵Gravatar A. Unique TensorFlower2016-08-22
| | | | | | Previously ops were created outside of a scope leading to the ugly occurrence of bare ops in TensorBoard. Change: 130970109
* Update generated Python Op docs.Gravatar A. Unique TensorFlower2016-08-22
| | | | Change: 130967884
* Add tests for _stochastic_dependencies_mapGravatar A. Unique TensorFlower2016-08-22
| | | | Change: 130967459
* Add option to log chrome trace in slim's train functionGravatar A. Unique TensorFlower2016-08-22
| | | | Change: 130964761
* Update generated Python Op docs.Gravatar A. Unique TensorFlower2016-08-22
| | | | Change: 130962360
* Make summary names for linear models uniqueGravatar A. Unique TensorFlower2016-08-22
| | | | Change: 130962334
* Adding assign_from_checkpoint_fn and assign_from_values_fn to tf.contrib.Gravatar Nathan Silberman2016-08-22
| | | | Change: 130962238
* Typo in docstring.Gravatar A. Unique TensorFlower2016-08-22
| | | | Change: 130961962
* Update comment about the optimizer.Gravatar A. Unique TensorFlower2016-08-22
| | | | Change: 130960947
* Increase bytes limit in ParseFrom().Gravatar A. Unique TensorFlower2016-08-22
| | | | Change: 130957917
* Improved the formatting of the codeGravatar Benoit Steiner2016-08-22
| | | | Change: 130957662
* Replace load_variable from checkpoint in gmm and kmeans as it will be ↵Gravatar A. Unique TensorFlower2016-08-22
| | | | | | deprecated soon. Change: 130957334
* Make sure that no control edges by an outer control dependency context are ↵Gravatar Yuan Yu2016-08-22
| | | | | | added to nodes inside cond/while. Change: 130951050
* Update generated Python Op docs.Gravatar A. Unique TensorFlower2016-08-22
| | | | Change: 130950555
* Fix L2Normalize when passing a list of dims. Fixes #3932.Gravatar Vijay Vasudevan2016-08-22
| | | | Change: 130947456
* Adding ResNets to TF-Slim network modules.Gravatar Nathan Silberman2016-08-22
| | | | Change: 130947454
* Rewording crossed column descriptions.Gravatar A. Unique TensorFlower2016-08-22
| | | | Change: 130943697
* Removing reference to TF-Example style guide.Gravatar Nathan Silberman2016-08-22
| | | | Change: 130943569
generated by cgit v1.2.3 (git 2.39.1) at 2024-07-02 21:55:59 +0000