Commit message (Collapse) | Author | Age | |
---|---|---|---|
* | Website: Add a reference to the Go API in the "API DOCS" section. | 2016-11-14 | |
| | | | | Change: 139120760 | ||
* | Add deprecation notices on learn/ops/* with suggested replacements. | 2016-11-14 | |
| | | | | Change: 139119637 | ||
* | Restore gradients test for string outputs. | 2016-11-14 | |
| | | | | | Remove test for op with no shape function. Change: 139116241 | ||
* | Fix bug when looking for a checkpoint in parent directory. | 2016-11-14 | |
| | | | | Change: 139116176 | ||
* | Add documentation and links for SavedModel constants. | 2016-11-14 | |
| | | | | Change: 139115405 | ||
* | XLA open-sourcing: remove dependency on util/process/subprocess.h. | 2016-11-14 | |
| | | | | | | | | | | | Add to TensorFlow a subset of the functionality of SubProcess to support the open-sourceable parts of XLA and support TensorFlow's own testing. On google3 let it use the internal SubProcess from google3. In the open-source world, use the TF implementation of the feature subset. Modify XLA and some TF tests to use the TF SubProcess. Change: 139114815 | ||
* | Reduce log level for Android Session variable fetching. | 2016-11-14 | |
| | | | | Change: 139112713 | ||
* | Move ValidateSparseTensor to common_shape_fns.h. | 2016-11-14 | |
| | | | | Change: 139112567 | ||
* | Correct format of input_fn wrapper example. | 2016-11-14 | |
| | | | | Change: 139111125 | ||
* | Updates docstring examples for Transformer class | 2016-11-14 | |
| | | | | Change: 139110997 | ||
* | Update the styling of the "Upload data" and "Publish" modal dialoges as per | 2016-11-14 | |
| | | | | | our design review. Change: 139109311 | ||
* | Adding support for dense tensors as input to sparse_column_* methods. | 2016-11-14 | |
| | | | | Change: 139108615 | ||
* | Embedding Projector: Send readable tensor names to client. | 2016-11-14 | |
| | | | | | Serve 'tensor_name' instead of 'tensor_name:0'. Change: 139105151 | ||
* | Update generated Python Op docs. | 2016-11-14 | |
| | | | | Change: 139103900 | ||
* | Updates docstring examples for input_from_feature_columns | 2016-11-14 | |
| | | | | Change: 139102879 | ||
* | Expand the documentation of tf.learn feature columns. | 2016-11-14 | |
| | | | | Change: 139102565 | ||
* | C API: Stick to a convention around use of size_t and int. | 2016-11-14 | |
| | | | | | | | | | | | | | The types size_t and int were being used inconsistently across different functions. This change documents the convention and fixes a few functions up to ensure that all functions adhere to it. Not quite sure if this convention (in particular using 'int' for array indices, as opposed to say 'size_t') is the best choice, but at least the choice has been made consistently. Change: 139100887 | ||
* | Updates docstring examples for weighted_sparse_column | 2016-11-14 | |
| | | | | Change: 139100782 | ||
* | Tune canvas label font rendering: use a stroke width of 3px and a fill width of | 2016-11-14 | |
| | | | | | 6px. Change: 139100540 | ||
* | Fix external compile and release new version of TB. | 2016-11-14 | |
| | | | | Change: 139100164 | ||
* | Update generated Python Op docs. | 2016-11-14 | |
| | | | | Change: 139097110 | ||
* | Automated rollback of change 138934248 | 2016-11-14 | |
| | | | | Change: 139096643 | ||
* | Plumb ModelFnOps through Estimator. | 2016-11-14 | |
| | | | | | This CL changes the signature of _get_train_ops, _get_eval_ops and _get_predict_ops for Estimator and BaseEstimator to return ModelFnOps instead of tuple. Change: 139095075 | ||
* | Switch ScatterNd to use C++ shape fn. | 2016-11-14 | |
| | | | | Change: 139094556 | ||
* | Add bookmarks functionality to the demo dataset provider, and add a bookmark | 2016-11-14 | |
| | | | | | for word2vec. Change: 139090881 | ||
* | Update generated Python Op docs. | 2016-11-14 | |
| | | | | Change: 139090080 | ||
* | Make the python default for shape fn be call_cpp_shape_fn. | 2016-11-14 | |
| | | | | | | | | Remove explicit RegisterShape calls that used call_cpp_shape_fn without input_tensors_needed arguments. Remove requirement that C++ shape functions be specified. Change: 139089667 | ||
* | Anonymize docstring for create_feature_spec_for_parsing | 2016-11-14 | |
| | | | | Change: 139087061 | ||
* | Add tests for export with input_feature_key arg set. | 2016-11-14 | |
| | | | | Change: 139084390 | ||
* | Update generated Python Op docs. | 2016-11-14 | |
| | | | | Change: 139083966 | ||
* | Simplified the code of some relu gradients | 2016-11-14 | |
| | | | | Change: 139083540 | ||
* | Remove tf.image.per_image_whitening() (which has been replaced with | 2016-11-14 | |
| | | | | | tf.image.per_image_standardization). Change: 139083060 | ||
* | Points in the data set all implemented an interface from ScatterPlot. Break the | 2016-11-14 | |
| | | | | | | dependency, there should be no relationship between DataSet points and ScatterPlot points. Change: 139082849 | ||
* | Fix markdown formatting. | 2016-11-14 | |
| | | | | Change: 139075656 | ||
* | Fix bug when looking for a checkpoint in parent directory. | 2016-11-14 | |
| | | | | Change: 139070284 | ||
* | Anonymize docstring snippets, and give more detailed examples to use | 2016-11-13 | |
| | | | | | MetricSpec. Change: 139022486 | ||
* | Add a small description for the BUILD file under sycl/sycl. | 2016-11-12 | |
| | | | | Change: 138980879 | ||
* | Fix typo in embedding tutorial | 2016-11-12 | |
| | | | | Change: 138976669 | ||
* | 1 char typo fix in rendezvous.h. | 2016-11-12 | |
| | | | | Change: 138974804 | ||
* | Update generated Python Op docs. | 2016-11-12 | |
| | | | | Change: 138972225 | ||
* | Seal *_ops interfaces. | 2016-11-12 | |
| | | | | Change: 138971763 | ||
* | Add dialog for upload data. | 2016-11-12 | |
| | | | | Change: 138971608 | ||
* | Update generated Python Op docs. | 2016-11-12 | |
| | | | | Change: 138971086 | ||
* | Update doc to use numpy name (tf.sub -> tf.subtract). | 2016-11-12 | |
| | | | | | Also remove mul and neg from the doc. Change: 138970263 | ||
* | Update generated Python Op docs. | 2016-11-11 | |
| | | | | Change: 138943977 | ||
* | Improve debug output for sparse_ops. | 2016-11-11 | |
| | | | | Change: 138942814 | ||
* | Fix class docstring example to be dataset independent. | 2016-11-11 | |
| | | | | Change: 138942464 | ||
* | Update generated Python Op docs. | 2016-11-11 | |
| | | | | Change: 138942367 | ||
* | Changing http_request to accept a vector for its output buffer. | 2016-11-11 | |
| | | | | Change: 138941640 | ||
* | Fix tf.learn classifier doc-strings to describe that the TF classifiers ↵ | 2016-11-11 | |
| | | | | | | expect class labels as indices (unlike scikit-learn which can accept arbitrary values as labels). Change: 138941065 |