Skip to content
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

No main.py file available #1

Open
JacksonWuxs opened this issue Mar 20, 2023 · 3 comments
Open

No main.py file available #1

JacksonWuxs opened this issue Mar 20, 2023 · 3 comments

Comments

@JacksonWuxs
Copy link

Dear Authors,

Congratulations to your publication on ICLR 2023!
I found SunGen is pretty interesting and I hope to use it in my research.

However, I cannot find the main.py from this repo, which should be a critical file to reproduce your experiments.
Would you please upload this file?

Best,
Xuansheng

@SumilerGAO
Copy link
Owner

Hi, Xuansheng

To run our code, you'll need to use the "run_reweight.py" file located in our repository. Please refer to the Run with generated data section in the Readme for instructions on how to use it.

For the "main.py" in the data generation part, please refer to the source code from ZeroGen.

Best,
Jiahui

@LindaLydia
Copy link

The file available at ZeroGen does not match your given running settings in README.md file, "--reload_model --input_file_type plain" are not provided arguments in ZeroGen/main.py. So could you please fix this miss-alignment?

@vmasrani
Copy link

vmasrani commented Sep 5, 2023

The file available at ZeroGen does not match your given running settings in README.md file, "--reload_model --input_file_type plain" are not provided arguments in ZeroGen/main.py. So could you please fix this miss-alignment?

Had the same problem, I got it running with this command:
python main.py --task_file ../SunGen/tasks/yelp/yelp-x1.json --output_dir yelp/output/yelp-x1-gen/ --model_name gpt2-xl --small_model_name distilbert --min_length 1 --max_length 5 --top_k 0 --top_p 0.9 --decay_constant 200 --batch_size 2048 --train_batch_size 32 --learning_rate 2e-5 --num_entries_per_input 500000

where I dropped --reload_model --input_file_type plain -base-uncased from the readme.md command. Can @SumilerGAO confirm this isn't going to cause problems downstream?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

4 participants