mirror of
https://github.com/saymrwulf/onnxruntime.git
synced 2026-05-22 22:01:08 +00:00
* Enable static memory planning for pipeline. 1. We fix a bug when resolving symbolic shape for scalars. 2. We pass the original inputs to all pipeline stages so that the symbolic shapes can be resolved. * Further Improvements 1. Address comments. 2. Further reduce activation size by ~50% when pipeline is on. This is done by removing all but one gradient tensor from the last RecordEvent in the backward pass. * Address a comment * Fix Windows build |
||
|---|---|---|
| .. | ||
| contrib_ops | ||
| dml_ops | ||
| featurizers_ops | ||
| function.cc | ||
| function.h | ||
| function_container.h | ||
| function_impl.h | ||
| function_inliner.h | ||
| graph.cc | ||
| graph_utils.cc | ||
| graph_utils.h | ||
| graph_viewer.cc | ||
| model.cc | ||
| model.h | ||
| op.cc | ||
| op.h | ||
| record.h | ||
| schema_registry.cc | ||