InvokeAI
InvokeAI copied to clipboard
invoke-ai
[bug]: the NSFW checker isn't being disabled
Is there an existing issue for this?
- [X] I have searched the existing issues
OS
Windows
GPU
cuda
VRAM
8GB
What happened?
I tried to get Stable Diffusion to stop blurring some of my images, I checked the documentation for how to disable it and I tried '--nsfw_checker' and '--nsfw_checker' and '--no-nsfw_checker' but the command terminal could not recognize either of my commands by giving me this error:
invoke.py: error: unrecognized arguments: --no-nsfw_checker
invoke>
Screenshots
No response
Additional context
No response
Contact Details
No response
--safety-checker is the flag. It's off ("false") by default. Please provide additional information like the version you're running and the prompt, sampler, etc. for reproduction if it's still an issue.
You are actually running InvokeAI 2.1. The --no-nsfw_checker argument was added in 2.2. If you don't provide any arguments, the safety checker should be turned off. Could you paste in invoke.py's startup messages?
I believe I am using Invoke 2.2 because I have the unified canvas on the webui.
Do you want to generate images using the
1. command-line
2. browser-based UI
OR
3. open the developer console
Please enter 1, 2 or 3: 1
Starting the InvokeAI command-line.
* Initializing, be patient...
>> Initialization file C:\Users\ninja/.invokeai found. Loading...
>> InvokeAI runtime directory is "C:\Users\ninja\invokeai"
>> Patchmatch initialized
>> GFPGAN Initialized
>> CodeFormer Initialized
>> ESRGAN Initialized
>> Using device_type cuda
>> Initializing safety checker
>> Current VRAM usage: 1.22G
>> Scanning Model: stable-diffusion-1.5
>> Model Scanned. OK!!
>> Loading stable-diffusion-1.5 from C:\Users\ninja\invokeai\models\ldm\stable-diffusion-v1\v1-5-pruned-emaonly.ckpt
| LatentDiffusion: Running in eps-prediction mode
| DiffusionWrapper has 859.52 M params.
| Making attention of type 'vanilla' with 512 in_channels
| Working with z of shape (1, 4, 32, 32) = 4096 dimensions.
| Making attention of type 'vanilla' with 512 in_channels
| Using faster float16 precision
| Loading VAE weights from: C:\Users\ninja\invokeai\models\ldm\stable-diffusion-v1\vae-ft-mse-840000-ema-pruned.ckpt
>> Model loaded in 11.66s
>> Max VRAM used to load the model: 3.38G
>> Current VRAM usage:3.38G
>> Current embedding manager terms: *
>> Setting Sampler to k_heun
* Initialization done! Awaiting your command (-h for help, 'q' to quit)
invoke>
I'm just going to delete and reinstall.
Edit: I can't download it anymore.
Same issue on Windows, version 2.2.3
Safety checker enabled by default, it causes 35-40% of all requests with words man/woman/human/etc blocked. Docs says,
...stored in the InvokeAI initialization file (usually
`.invokeai` in your home directory). You can change the default at any
time by opening this file in a text editor and commenting or
uncommenting the line `--nsfw_checker`.
But I haven't found such a file in HOME or InvokeAI file.
Same issue on Windows, version 2.2.3
Safety checker enabled by default, it causes 35-40% of all requests with words man/woman/human/etc blocked. Docs says,
...stored in the InvokeAI initialization file (usually `.invokeai` in your home directory). You can change the default at any time by opening this file in a text editor and commenting or uncommenting the line `--nsfw_checker`.But I haven't found such a file in HOME or InvokeAI file.
For Windows the path is: C:\Users\[YOUR_USERNAME]\.invokeai Enable the setting that you can see hidden files to see files starting with a dot
Same issue on Windows, version 2.2.3 Safety checker enabled by default, it causes 35-40% of all requests with words man/woman/human/etc blocked. Docs says,
...stored in the InvokeAI initialization file (usually `.invokeai` in your home directory). You can change the default at any time by opening this file in a text editor and commenting or uncommenting the line `--nsfw_checker`.But I haven't found such a file in HOME or InvokeAI file.
For Windows the path is: C:\Users[YOUR_USERNAME].invokeai Enable the setting that you can see hidden files to see files starting with a dot
How do I do a complete reset of InvokeAI? I deleted the files and tried to reinstall it an everything went wrong and commandline and webui doesn't open, citing that there's no ldm module.
I also get this error
CommandNotFoundError: Your shell has not been properly configured to use 'conda activate'.
If using 'conda activate' from a batch script, change your
invocation to 'CALL conda.bat activate'.
it works for me on Linux, the file was invokeai.init in the folder home/user/invokeai, the same place where outputs are saved
