| Commit message (Collapse) | Author | Age | |
|---|---|---|---|
| * | Make the pruning of the graph in ahead-of-time compilation |  A. Unique TensorFlower | 2017-07-06 |
| | | | | | | | | | | | | also prune out dependencies of fed tensors. Remove special handling of control dependencies from the introduced placeholders. For a compilation going to XLA, those control dependencies would not really be doing anything anyway (and the current code was only handling Placeholder, not PlaceholderV2). PiperOrigin-RevId: 161157126 | ||
| * | Initial open-source release of XLA: Accelerated Linear Algebra. |  Peter Hawkins | 2017-01-09 |
| XLA is a compiler-based linear algebra execution engine that targets CPUs, GPUs and custom accelerators. XLA is still experimental; we are releasing it early to get the community involved. Change: 143990941 | |||
 |
index : tensorflow | |
| machine learning framework |
| aboutsummaryrefslogtreecommitdiffhomepage |