NVIDIA / NVIDIA/apex

How to disable the defalf print when we use the apex?

Open
#960 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Python
Stars
9k
Forks
1.5k
Avg merge
2d 4h
Merged PRs (30d)
3

Description

In my code, I have to make the following definition in every epoch:

self.model, self.optimizer = amp.initialize(self.model, self.optimizer, opt_level="O1")

Every time I use the apex, the following part will be printed:

Selected optimization level O1: Insert automatic casts around Pytorch functions and Tensor methods.
Defaults for this optimization level are:
enabled : True
opt_level : O1
cast_model_type : None
patch_torch_functions : True
keep_batchnorm_fp32 : None
master_weights : None
loss_scale : dynamic
Processing user overrides (additional kwargs that are not None)...
After processing overrides, optimization options are:
enabled : True
opt_level : O1
cast_model_type : None
patch_torch_functions : True
keep_batchnorm_fp32 : None
master_weights : None
loss_scale : dynamic

My confusion is that: is there any possible way to disable this print?

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start at the amp.initialize entry point shown in the issue and trace where the optimization-level details are printed. Determine whether the initialization output can be disabled without changing the reported settings, then verify that using opt_level="O1" no longer emits the configuration text when suppression is requested.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, pytorch
Domain
machine-learning
Issue type
Feature
Difficulty
2/5
Estimated time
1-3 hours
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.