train_dreambooth_inpaint.py throwing "returned non-zero exit status 1" error

[lakshman111]

Describe the bug

@thedarkzeno @patil-suraj

I've been running train_dreambooth_inpaint.py from https://github.com/huggingface/diffusers/tree/main/examples/research_projects/dreambooth_inpaint for the last few days with the same environment configurations. Today I ran it again and got:

Traceback (most recent call last):
File "train_dreambooth_inpaint.py", line 825, in
main()
File "train_dreambooth_inpaint.py", line 421, in main
accelerator = Accelerator(
TypeError: init() got an unexpected keyword argument 'accelerator_project_config'
Traceback (most recent call last):
File "/usr/local/envs/laksh/bin/accelerate", line 8, in
sys.exit(main())
File "/usr/local/envs/laksh/lib/python3.8/site-packages/accelerate/commands/accelerate_cli.py", line 45, in main
args.func(args)
File "/usr/local/envs/laksh/lib/python3.8/site-packages/accelerate/commands/launch.py", line 1097, in launch_command
simple_launcher(args)
File "/usr/local/envs/laksh/lib/python3.8/site-packages/accelerate/commands/launch.py", line 552, in simple_launcher
raise subprocess.CalledProcessError(returncode=process.returncode, cmd=cmd)
subprocess.CalledProcessError: Command '['/usr/local/envs/laksh/bin/python', 'train_dreambooth_inpaint.py', '--pretrained_model_name_or_path=runwayml/stable-diffusion-inpainting', '--instance_data_dir=/workspace/diffusers/examples/research_projects/dreambooth_inpaint/modern_512_images', '--output_dir=/workspace/diffusers/examples/research_projects/dreambooth_inpaint/modern_512_400steps_model', '--instance_prompt=a photo of mmodern furniture', '--resolution=512', '--train_batch_size=1', '--gradient_accumulation_steps=1', '--learning_rate=5e-6', '--lr_scheduler=constant', '--lr_warmup_steps=0', '--max_train_steps=400']' returned non-zero exit status 1.

Reproduction

1. SSH into GPU
2. Create a conda environment: conda create -n YOUR_ENV_NAME python=3.8
3. Activate the session: source activate YOUR_ENV_NAME
4. CD into https://github.com/huggingface/diffusers/tree/main/examples/research_projects/dreambooth_inpaint
5. pip install -r requirements.txt
6. pip install git+https://github.com/huggingface/diffusers
7. Run the first code snippet with custom inputs (my code is below)

`export MODEL_NAME="runwayml/stable-diffusion-inpainting"
export INSTANCE_DIR="/workspace/diffusers/examples/research_projects/dreambooth_inpaint/modern_512_images"
export OUTPUT_DIR="/workspace/diffusers/examples/research_projects/dreambooth_inpaint/modern_512_400steps_model"

accelerate launch train_dreambooth_inpaint.py
--pretrained_model_name_or_path=$MODEL_NAME
--instance_data_dir=$INSTANCE_DIR
--output_dir=$OUTPUT_DIR
--instance_prompt="a photo of mmodern furniture"
--resolution=512
--train_batch_size=1
--gradient_accumulation_steps=1
--learning_rate=5e-6
--lr_scheduler="constant"
--lr_warmup_steps=0
--max_train_steps=400`

Logs

No response

System Info

• diffusers version: 0.14.0.dev0
• Platform: Linux-5.15.0-47-generic-x86_64-with-glibc2.17
• Python version: 3.8.16
• PyTorch version (GPU?): 1.13.1+cu117 (True)
• Huggingface_hub version: 0.12.1
• Transformers version: 4.26.1
• Accelerate version: 0.16.0
• xFormers version: not installed
• Using GPU in script?: yes
• Using distributed or parallel set-up in script?: no

[lakshman111]

Solved! I changed "accelerator_project_config" to "project_config" and it worked. Looks like the variable Accelerate is looking for is "project_config": https://github.com/huggingface/accelerate/blob/main/src/accelerate/accelerator.py

Didn't get around to submitting a PR yet. I've never submitted one, so need to still read through the guidelines on how to do that.

[sayakpaul]

Thanks and glad that it got resolved!

[kadirnar]

Solved! I changed "accelerator_project_config" to "project_config" and it worked. Looks like the variable Accelerate is looking for is "project_config": https://github.com/huggingface/accelerate/blob/main/src/accelerate/accelerator.py

Didn't get around to submitting a PR yet. I've never submitted one, so need to still read through the guidelines on how to do that.

Thank you. This error should be fixed immediately. @thedarkzeno @patil-suraj

[sayakpaul]

@lakshman111 could we close this issue?

[kadirnar]

@lakshman111bu konuyu kapatabilir miyiz?

No. The error still persists. It needs to be fixed. If you want, I can do it.

[patrickvonplaten]

@lakshman111bu konuyu kapatabilir miyiz?

No. The error still persists. It needs to be fixed. If you want, I can do it.

That'd be very nice!

[github-actions]

This issue has been automatically marked as stale because it has not had recent activity. If you think this still needs to be addressed please comment on this thread.

Please note that issues that do not follow the contributing guidelines are likely to be ignored.