[easy] allow multithreading in load_dataset #301
Merged
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
|
doh I didn't know about this, thank you!! |
|
looks like we might have to revert this :( while it works well on Linux machines i tried it on, when I tried to download the data on Macbook it hangs somewhere in the library (unless I set num_proc=1) |
|
another option is to make |
|
hopefully no need to revert. 542ac51 -- the issue was supposedly that multiprocessing module implementation on that platform cannot find entry point unless you define 'main'. I'll test it and create new PR |
carmocca
added a commit
to Lightning-AI/litgpt
that referenced
this pull request
Jun 21, 2023
ken-viper
added a commit
to ken-viper/litgpt
that referenced
this pull request
Sep 5, 2024
gkielian
pushed a commit
to gkielian/ReaLLMASIC_nanogpt
that referenced
this pull request
Sep 5, 2024
[easy] allow multithreading in load_dataset
gkielian
added a commit
to gkielian/ReaLLMASIC_nanogpt
that referenced
this pull request
Dec 5, 2024
Expand chess features
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Outdated suggestions cannot be applied.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.
Both loading from the network and generating train split can benefit from multithreading (especially on larger instances/faster network).
On 30 CPU + single A100 lambda instance it brings down the time to get the data from ~35min (28 threads for tokenizing, 1 thread for load_dataset) to ~8min (28 threads for both).
CPU util looks like this:
28 thread load_dataset (finished):
1 thread load_dataset (still running):
