IntenseWebs/openvino
3
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Note ONNX FE in mem_optimization_guide.md (#18777)

Browse Source
This commit is contained in:
Vitaliy Urusovskij 2023-07-27 10:42:35 +04:00 committed by GitHub
parent 416fdddd34
commit f8f6d4a5b5
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
docs/optimization_guide/memory_optimization_guide.md
View File

@ -11,7 +11,7 @@ The most RAM-consuming OpenVINO stage is model compilation. It may cause several
* Not enough memory to compile a model. To decrease memory requirement, the following options may be applied:
* Weights mapping - memory mapping (using ``mmap``) has been introduced as the default way to work
with weights. Currently, this feature is supported by the IR frontend.
with weights. Currently, this feature is supported by the IR and ONNX frontends.
Mapping may be switched by specifying the ``ov::enable_mmap(BOOL)`` property for the ``ov::Core``.
Because of its "memory-on-demand" nature, there is no need to store all weights
in RAM. Storing just the data that is needed at the moment lowers the amount of memory