saymrwulf/onnxruntime
1
0
Fork 0
mirror of https://github.com/saymrwulf/onnxruntime.git synced 2026-07-30 20:18:08 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

cuda_mem_limit has been renamed to gpu_mem_limit (#8284)

Browse source 
`cuda_mem_limit` was renamed to `gpu_mem_limit` in nighlty build `nightly-1.7.0.dev202104231`.

https://github.com/microsoft/onnxruntime/issues/7457
This commit is contained in:
Matthew Nelson - Silverark 2021-08-05 22:15:03 +01:00 committed by GitHub
parent 8b62250fcd
commit 59f6271da1
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
1 changed files with 3 additions and 3 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

6
docs/reference/execution-providers/CUDA-ExecutionProvider.md
View file

@ -45,7 +45,7 @@ The device ID.
Default value: 0
### cuda_mem_limit
### gpu_mem_limit
The size limit of the device memory arena in bytes. This size limit is only for the execution provider's arena. The total device memory usage may be higher.
s: max value of C++ size_t type (effectively unlimited)
@ -88,7 +88,7 @@ providers = [
('CUDAExecutionProvider', {
'device_id': 0,
'arena_extend_strategy': 'kNextPowerOfTwo',
'cuda_mem_limit': 2 * 1024 * 1024 * 1024,
'gpu_mem_limit': 2 * 1024 * 1024 * 1024,
'cudnn_conv_algo_search': 'EXHAUSTIVE',
'do_copy_in_default_stream': True,
}),
@ -106,7 +106,7 @@ OrtSessionOptions* session_options = /* ... */;
OrtCUDAProviderOptions options;
options.device_id = 0;
options.arena_extend_strategy = 0;
options.cuda_mem_limit = 2 * 1024 * 1024 * 1024;
options.gpu_mem_limit = 2 * 1024 * 1024 * 1024;
options.cudnn_conv_algo_search = OrtCudnnConvAlgoSearch::EXHAUSTIVE;
options.do_copy_in_default_stream = 1;