[AutoTuner] Add auto tuner to obtain optima configuration #54460
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
你的PR提交成功，感谢你对开源项目的贡献! |
Caozhou1995
force-pushed
the
auto_configurator
branch
from
90ea799
to
6d47943
Compare
XieYunshen
approved these changes
Jun 14, 2023
zhiqiu
approved these changes
Jun 14, 2023
|
|
||
| process = subprocess.Popen(cmd) | ||
| process.wait() | ||
| self.assertEqual(process.returncode, 0) |
Comment on lines
+297
to
+305
| import copy | ||
| import json | ||
| import signal | ||
| import sys | ||
| import time | ||
|
|
||
| from ..auto_tuner.tuner import AutoTuner | ||
| from ..auto_tuner.utils import gen_new_args | ||
| from . import controllers |
There was a problem hiding this comment.
Better import at the top of the file
| cur_cfg = auto_tuner.search_once() | ||
|
|
||
| # get max time per task run | ||
| max_time_per_task = tuner_cfg.get("max_time_per_task", 1800) |
There was a problem hiding this comment.
max_time_per_task -> max_time_in_seconds_per_task?
|
|
||
| def __init__(self, tuner_cfg): | ||
| self.cur_task_id = 1 | ||
| self.task_limit = tuner_cfg.get("task_limit", 100) |
There was a problem hiding this comment.
DEFAULT_MAX_TASK_LIMIT = 100 ?
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
PR types
New features
PR changes
Others
Description
Pcard-72023
The optimal configuration for large model with distributed training/inference often requires designing multiple sets of experiments based on experiences (network, parameter size, gpu memory or flops, etc.), and comparing the results to determine the optimal configuration. This process heavily relies on human experience, and the determined optimal configuration may not be the global optimal configuration. When any condition changes, the above process needs to be repeated repeatedly, resulting in poor usability of large models.
To address the above issues, we have implemented AutoTuner based on Profiling, with the main modules as follows:
At present, we have built-in grid search support for 8 dimensions, including dp degree, mp degree, pp degree, mbs, sharding degree, sharding stage, recompute, and recompute granularity. The example JSON is as follows:
The usage is as follows:
python -m paddle.distributed.launch --devices "0,1,2,3,4,5,6,7" --auto_tuner_json=test.json your_train.py your_args
NOTE: Since the auto_tuner is non-invasive, users need to expose args in their script to enable the configuration generated by auto_tuner be executed.