anyone can provide a dockerfile for finetuning the dreambooth model.

[Xiaoxiao-mobius]

Is your feature request related to a problem? Please describe. I am using this dockerfile to finetune stable-diffusion-v1-5 model. However, the training results is not good as I trained on the google colab

Describe the solution you'd like A clear and concise description of what you want to happen. I noticed there are weird warning message shows up using bitsandbytes:

BUG REPORT Welcome to bitsandbytes. For bug reports, please submit your error trace to: https://github.com/TimDettmers/bitsandbytes/issues For effortless bug reporting copy-paste your error into this form: https://docs.google.com/forms/d/e/1FAIpQLScPB8emS3Thkp66nvqwmjTEgxp8Y9ufuWTzFyr9kJ5AoI47dQ/viewform?usp=sf_link

/opt/conda/lib/python3.7/site-packages/bitsandbytes/cuda_setup/main.py:134: UserWarning: WARNING: The following directories listed in your path were found to be non-existent: {PosixPath('/usr/local/nvidia/lib'), PosixPath('/usr/local/nvidia/lib64')} warn(msg) /opt/conda/lib/python3.7/site-packages/bitsandbytes/cuda_setup/main.py:134: UserWarning: /usr/local/nvidia/lib:/usr/local/nvidia/lib64 did not contain libcudart.so as expected! Searching further paths... warn(msg) CUDA_SETUP: WARNING! libcudart.so not found in any environmental path. Searching /usr/local/cuda/lib64... CUDA SETUP: CUDA runtime path found: /usr/local/cuda/lib64/libcudart.so CUDA SETUP: Highest compute capability among GPUs detected: 7.5 CUDA SETUP: Detected CUDA version 113

I am not sure if this cause the difference for training results

Additional context If anyone built docker container for finetuning the stable-diffusion-v1-5 model, and produced the exact good results, please share the dockerfile.

[SU1199]

not a docker file but you can use paperspace's base docker image in runpod or vast.... I made a notebook to do the same : https://github.com/SU1199/fastBooth

[hamzaraouzi]

I have the same problem, I have been fine tuning , and generating images smoothly on Google colab and VM, and I was getting very good results, but within Docker container, the results were terrible even with Dockerfile provided with the implementation here: Dockerfile, I have built over 4 Docker Images From scratch and I was always getting the same bad results, whereas on the VM the results were perfect even with the Tuning Hyperparameters.

I will really appreciate, if you to provide me some hints about the solution.