Skip to content
This repository was archived by the owner on Sep 18, 2024. It is now read-only.
This repository was archived by the owner on Sep 18, 2024. It is now read-only.

How to create a 'trans_from_jit_to_python'? #4654

Open
@maxin-cn

Description

@maxin-cn

Describe the issue:
I found an important function in NNI called trans_from_jit_to_python. I was pruning some networks and found some operations that were not in the list, such as torch.randn. Thus, how do I add support for operations like torch. randn to this list? Thanks~

Environment:

  • NNI version: 2.6
  • Training service (local|remote|pai|aml|etc): local
  • Client OS: centos
  • Server OS (for remote mode only):
  • Python version: 3.7
  • PyTorch/TensorFlow version: 1.7
  • Is conda/virtualenv/venv used?: conda
  • Is running in Docker?: no

Metadata

Metadata

Assignees

Type

No type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions