Open zackees opened 1 year ago
Dockerfile
FROM pytorch/pytorch:1.12.1-cuda11.3-cudnn8-devel
RUN apt update && apt install --yes git ffmpeg libsm6 libxext6
# Create conda environment
RUN git clone https://github.com/Stability-AI/stablediffusion.git
WORKDIR /workspace/stablediffusion/
# Overwrite the environment.yaml file
COPY environment.yaml .
RUN conda env create -f environment.yaml
# Make RUN commands use the new environment:
SHELL ["conda", "run", "-n", "ldm", "/bin/bash", "-c"]
# Install xformers for memory efficient flash attention
RUN conda install xformers -c xformers/label/dev
RUN conda init bash
RUN echo "conda activate ldm" >> $HOME/.bashrc
Modified environment file .
name: ldm
channels:
- pytorch
- defaults
dependencies:
# Changed to Python 3.9 to make use of precompiled binaries for xformers
- python=3.9
- pip
- cudatoolkit=11.3
- pytorch=1.12.1
- torchvision=0.13.1
- numpy=1.23.1
- pip:
- albumentations==1.3.0
- opencv-python==4.6.0.66
- imageio==2.9.0
- imageio-ffmpeg==0.4.2
- pytorch-lightning==1.4.2
- omegaconf==2.1.1
- test-tube>=0.7.5
- streamlit==1.12.1
- einops==0.3.0
- transformers==4.19.2
- webdataset==0.2.5
- kornia==0.6
- open_clip_torch==2.0.2
- invisible-watermark>=0.1.5
- streamlit-drawable-canvas==0.8.0
- torchmetrics==0.6.0
- -e .
Sorry @prasantpoudel & @xiankgx for standing on your toes ! I guess I have just made another one. Pull request here: #326 Does yours have the same issues as mine?
maybe someone need, just use sd:
or, someone need better performance sd-webui:
This would allow people to spin up an instance very fast.