We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
There was an error while loading. Please reload this page.
1 parent 753915b commit f02e4a5Copy full SHA for f02e4a5
py/torch_tensorrt/_compile.py
@@ -597,7 +597,7 @@ def save(
597
output_format (str): Format to save the model. Options include exported_program | torchscript.
598
retrace (bool): When the module type is a fx.GraphModule, this option re-exports the graph using torch.export.export(strict=False) to save it.
599
This flag is experimental for now.
600
- pickle_protocol (int): The pickle protocol to use to save the model. Default is 2. Increase this to 4 or higher for large models (eg: SAM2)
+ pickle_protocol (int): The pickle protocol to use to save the model. Default is 2. Increase this to 4 or higher for large models
601
"""
602
if isinstance(module, CudaGraphsTorchTensorRTModule):
603
module = module.compiled_module
0 commit comments