| Commit message (Collapse) | Author | Age | |
|---|---|---|---|
| * | Create SDCAOptimizerV2 op to fix the "adaptative" typo. |  Yuefeng Zhou | 2018-10-09 |
| | | | | | PiperOrigin-RevId: 216370193 | ||
| * | Add the poisson log loss to the SDCA optimizer. |  A. Unique TensorFlower | 2018-08-31 |
| | | | | | PiperOrigin-RevId: 211116606 | ||
| * | Convert a local variable and mutex to a struct so GUARDED_BY annotation ↵ | A. Unique TensorFlower | 2018-04-19 |
| | | | | | | | works correctly. PiperOrigin-RevId: 193584438 | ||
| * | Expose the adaptive sampling option for SDCA and shuffle the data when ↵ | A. Unique TensorFlower | 2018-04-05 |
| | | | | | | | adaptive sampling is off. PiperOrigin-RevId: 191836004 | ||
| * | Make adaptive SDCA the default. | A. Unique TensorFlower | 2018-03-08 |
| | | | | | PiperOrigin-RevId: 188380039 | ||
| * | Cleanup: Ran clang-format on all *.{cc,h} files in tensorflow/core/kernels. | A. Unique TensorFlower | 2018-01-26 |
| | | | | | PiperOrigin-RevId: 183423961 | ||
| * | Make single-parameter constructors explicit | A. Unique TensorFlower | 2017-05-31 |
| | | | | | PiperOrigin-RevId: 157628970 | ||
| * | Fix code that ignores tensorflow::Status. |  Peter Hawkins | 2017-02-13 |
| | | | | | | Add a new tensorflow::Status::IgnoreError() method to mark call sites where a Status has been intentionally ignored. Change: 147402405 | ||
| * | Fixed the signature of the Sdca*::Compute() methods |  Benoit Steiner | 2016-12-06 |
| | | | | | Change: 141203556 | ||
| * | bug fix. | A. Unique TensorFlower | 2016-11-21 |
| | | | | | Change: 139861812 | ||
| * | Refactor the common code in sdca_ops.cc out for multiclass SDCA. | A. Unique TensorFlower | 2016-11-21 |
| | | | | | Change: 139854348 | ||
| * | Change SDCA to use DenseMutableHashtable. | A. Unique TensorFlower | 2016-10-27 |
| | | | | | | | | Change SdcaFprint to return a two int64 values instead of one string to save memory given that DenseMutableHashtable supports non-scalar keys. Corresponding changes to the private sharded hashtable wrapper to support non-scalar keys. Change: 137464256 | ||
| * | Do not use Eigen Tensor multiplication for binary SDCA, since that has very | A. Unique TensorFlower | 2016-10-27 |
| | | | | | | poor performance. Also added a new bench mark test for this. Change: 137413950 | ||
| * | Change the interface of helper classes/methods in sdca_ops.cc to support ↵ | A. Unique TensorFlower | 2016-10-20 |
| | | | | | | | multiclass SDCA. Change: 136756962 | ||
| * | Move SDCA optimizer kernels to core. | A. Unique TensorFlower | 2016-09-27 |
| See: tensorflow/contrib/linear_optimizer/kernels/g3doc/readme.md Change: 134452636 | |||
 |
index : tensorflow | |
| machine learning framework |
| aboutsummaryrefslogtreecommitdiffhomepage |