-
Notifications
You must be signed in to change notification settings - Fork 825
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Long context summarize demo #844
Merged
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
research4pan
requested changes
May 27, 2024
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Some revisions can be made before merge.
contrib/long-context/hf_sft_full_finetune.sh
- [Style] line 1: shebang line
#!/bin/bash
can be added to specify the default shell. - [Bug] line 2: avoid exporting
CUDA_VISIBLE_DEVICES
to make this script runnable on single-gpu servers. - [Bug] line 5: please use a general huggingface model name instead of a local name. Currently the script cannot be run on other's servers.
contrib/long-context/hf_sft_lora_flashattn.sh
- [Style] line 1: shebang line
#!/bin/bash
can be added to specify the default shell. - [Style]: The naming can be improved: why there is
flashattn
in the name? - [Bug] line 2, 5: similar to
contrib/long-context/hf_sft_full_finetune.sh
- [Architecture]: I didn't see lora options in the script. I am wondering if this script is really doing lora training?
contrib/long-context/sft_summarizer.py
- [Style] line 1: shebang line
#!/usr/bin/env python
can be added. - [Style] line 1: string encoding
# coding=utf-8
is highly recommended to be specified. - [Bug] line 7, 24: put them in main would be better. Otherwise other
.py
modules loading this one would potentially execute those codes as well, causing unintended and hard-to-locate bugs. - [Bug] line 29: should not include your own personal wandb key in a public repo, otherwise others' experiments will be messed up with your. Even worse, there can be a security issue for your own account.
- [Style] line 35-40, 56, 72, 90-91, 109-110, 112: use
logging
instead of print, and restrict the logging level to debug if this information is not necessary - [Style] line 98-102: the indentation here can be significantly improved, please refer to google coding style (https://google.github.io/styleguide/pyguide.html).
research4pan
approved these changes
Jun 5, 2024
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM, thanks!
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.
No description provided.