2022-09-15 14:53:41 +00:00
|
|
|
---
|
|
|
|
title: Linux
|
|
|
|
---
|
2022-09-11 15:52:43 +00:00
|
|
|
|
2022-09-15 14:53:41 +00:00
|
|
|
1. You will need to install the following prerequisites if they are not already
|
|
|
|
available. Use your operating system's preferred installer.
|
2022-09-11 15:52:43 +00:00
|
|
|
|
2022-09-15 14:53:41 +00:00
|
|
|
- Python (version 3.8.5 recommended; higher may work)
|
|
|
|
- git
|
2022-09-11 15:52:43 +00:00
|
|
|
|
|
|
|
2. Install the Python Anaconda environment manager.
|
|
|
|
|
2022-09-15 14:53:41 +00:00
|
|
|
```bash
|
|
|
|
~$ wget https://repo.anaconda.com/archive/Anaconda3-2022.05-Linux-x86_64.sh
|
|
|
|
~$ chmod +x Anaconda3-2022.05-Linux-x86_64.sh
|
|
|
|
~$ ./Anaconda3-2022.05-Linux-x86_64.sh
|
|
|
|
```
|
2022-09-11 15:52:43 +00:00
|
|
|
|
2022-09-15 14:53:41 +00:00
|
|
|
After installing anaconda, you should log out of your system and log back in. If
|
|
|
|
the installation worked, your command prompt will be prefixed by the name of the
|
|
|
|
current anaconda environment - `(base)`.
|
2022-09-11 15:52:43 +00:00
|
|
|
|
|
|
|
3. Copy the stable-diffusion source code from GitHub:
|
|
|
|
|
2022-09-15 14:53:41 +00:00
|
|
|
```bash
|
|
|
|
(base) ~$ git clone https://github.com/lstein/stable-diffusion.git
|
|
|
|
```
|
2022-09-11 15:52:43 +00:00
|
|
|
|
2022-09-15 14:53:41 +00:00
|
|
|
This will create stable-diffusion folder where you will follow the rest of the
|
|
|
|
steps.
|
2022-09-11 15:52:43 +00:00
|
|
|
|
2022-09-15 14:53:41 +00:00
|
|
|
4. Enter the newly-created stable-diffusion folder. From this step forward make
|
|
|
|
sure that you are working in the stable-diffusion directory!
|
2022-09-11 15:52:43 +00:00
|
|
|
|
2022-09-15 14:53:41 +00:00
|
|
|
```bash
|
|
|
|
(base) ~$ cd stable-diffusion
|
|
|
|
(base) ~/stable-diffusion$
|
|
|
|
```
|
2022-09-11 15:52:43 +00:00
|
|
|
|
2022-09-15 14:53:41 +00:00
|
|
|
5. Use anaconda to copy necessary python packages, create a new python
|
|
|
|
environment named `ldm` and activate the environment.
|
2022-09-11 15:52:43 +00:00
|
|
|
|
2022-09-15 14:53:41 +00:00
|
|
|
```bash
|
|
|
|
(base) ~/stable-diffusion$ conda env create -f environment.yaml
|
|
|
|
(base) ~/stable-diffusion$ conda activate ldm
|
|
|
|
(ldm) ~/stable-diffusion$
|
|
|
|
```
|
2022-09-11 15:52:43 +00:00
|
|
|
|
2022-09-15 14:53:41 +00:00
|
|
|
After these steps, your command prompt will be prefixed by `(ldm)` as shown
|
|
|
|
above.
|
2022-09-11 15:52:43 +00:00
|
|
|
|
|
|
|
6. Load a couple of small machine-learning models required by stable diffusion:
|
|
|
|
|
2022-09-15 14:53:41 +00:00
|
|
|
```bash
|
|
|
|
(ldm) ~/stable-diffusion$ python3 scripts/preload_models.py
|
|
|
|
```
|
2022-09-11 15:52:43 +00:00
|
|
|
|
2022-09-15 14:53:41 +00:00
|
|
|
Note that this step is necessary because I modified the original just-in-time
|
|
|
|
model loading scheme to allow the script to work on GPU machines that are not
|
|
|
|
internet connected. See [Preload Models](../features/OTHER.md#preload-models)
|
2022-09-11 15:52:43 +00:00
|
|
|
|
|
|
|
7. Now you need to install the weights for the stable diffusion model.
|
|
|
|
|
2022-09-15 14:53:41 +00:00
|
|
|
- For running with the released weights, you will first need to set up an acount
|
|
|
|
with [Hugging Face](https://huggingface.co).
|
|
|
|
- Use your credentials to log in, and then point your browser [here](https://huggingface.co/CompVis/stable-diffusion-v-1-4-original.)
|
|
|
|
- You may be asked to sign a license agreement at this point.
|
|
|
|
- Click on "Files and versions" near the top of the page, and then click on the
|
|
|
|
file named "sd-v1-4.ckpt". You'll be taken to a page that prompts you to click
|
|
|
|
the "download" link. Save the file somewhere safe on your local machine.
|
2022-09-11 15:52:43 +00:00
|
|
|
|
2022-09-15 14:53:41 +00:00
|
|
|
Now run the following commands from within the stable-diffusion directory.
|
|
|
|
This will create a symbolic link from the stable-diffusion model.ckpt file, to
|
|
|
|
the true location of the `sd-v1-4.ckpt` file.
|
2022-09-11 15:52:43 +00:00
|
|
|
|
2022-09-15 14:53:41 +00:00
|
|
|
```bash
|
|
|
|
(ldm) ~/stable-diffusion$ mkdir -p models/ldm/stable-diffusion-v1
|
|
|
|
(ldm) ~/stable-diffusion$ ln -sf /path/to/sd-v1-4.ckpt models/ldm/stable-diffusion-v1/model.ckpt
|
|
|
|
```
|
2022-09-11 15:52:43 +00:00
|
|
|
|
|
|
|
8. Start generating images!
|
|
|
|
|
2022-09-15 14:53:41 +00:00
|
|
|
```bash
|
|
|
|
# for the pre-release weights use the -l or --liaon400m switch
|
|
|
|
(ldm) ~/stable-diffusion$ python3 scripts/dream.py -l
|
2022-09-11 15:52:43 +00:00
|
|
|
|
2022-09-15 14:53:41 +00:00
|
|
|
# for the post-release weights do not use the switch
|
|
|
|
(ldm) ~/stable-diffusion$ python3 scripts/dream.py
|
2022-09-11 15:52:43 +00:00
|
|
|
|
2022-09-15 14:53:41 +00:00
|
|
|
# for additional configuration switches and arguments, use -h or --help
|
|
|
|
(ldm) ~/stable-diffusion$ python3 scripts/dream.py -h
|
|
|
|
```
|
2022-09-11 15:52:43 +00:00
|
|
|
|
2022-09-15 14:53:41 +00:00
|
|
|
9. Subsequently, to relaunch the script, be sure to run "conda activate ldm"
|
|
|
|
(step 5, second command), enter the `stable-diffusion` directory, and then
|
|
|
|
launch the dream script (step 8). If you forget to activate the ldm
|
|
|
|
environment, the script will fail with multiple `ModuleNotFound` errors.
|
2022-09-11 15:52:43 +00:00
|
|
|
|
2022-09-15 14:53:41 +00:00
|
|
|
### Updating to newer versions of the script
|
2022-09-11 15:52:43 +00:00
|
|
|
|
2022-09-15 14:53:41 +00:00
|
|
|
This distribution is changing rapidly. If you used the `git clone` method
|
|
|
|
(step 5) to download the stable-diffusion directory, then to update to the
|
|
|
|
latest and greatest version, launch the Anaconda window, enter
|
|
|
|
`stable-diffusion` and type:
|
2022-09-11 15:52:43 +00:00
|
|
|
|
2022-09-15 14:53:41 +00:00
|
|
|
```bash
|
|
|
|
(ldm) ~/stable-diffusion$ git pull
|
|
|
|
```
|
2022-09-11 15:52:43 +00:00
|
|
|
|
2022-09-15 14:53:41 +00:00
|
|
|
This will bring your local copy into sync with the remote one.
|