[bug]: from_file documentation is ambiguous

[Neosettler]

Is there an existing issue for this?

• [X] I have searched the existing issues

OS

Windows

GPU

cuda

VRAM

No response

What happened?

You may read a series of prompts from standard input by providing a filename of -:
(invokeai) ~/stable-diffusion$ echo "a beautiful day" | python3 scripts/invoke.py --from_file -

This example doesn't show how to provide several prompts. Does it need any other arguments?

A better example would start with: (invokeai) ~/stable-diffusion$ echo "a beautiful day" | "a rainy day" python3 scripts/invoke.py --from_file -

In addition, this feature requires to load the script/model every time it is being called. Not very practical. Consider migrating this to render group and/or implementing a proper render queue instead, which would tied up nicely with | operator/matrices

https://github.com/invoke-ai/InvokeAI/discussions/473 https://github.com/invoke-ai/InvokeAI/issues/566 https://github.com/invoke-ai/InvokeAI/issues/1008

Screenshots

No response

Additional context

No response

Contact Details

No response

[Neosettler]

Could anyone confirm how to use --from_file - with multiple prompts? Also, could this feature be used in interactive mode instead?

thank you,

[lstein]

Write a text file with one prompt per line. Name it my-prompts.txt. Then feed it into invoke.py like this:

invoke.py --from_file=my-prompts.txt

[Neosettler]

@lstein I'm talking about the other from_file function:

You may read a series of prompts from standard input by providing a filename of -: (invokeai) ~/stable-diffusion$ echo "a beautiful day" | python3 scripts/invoke.py --from_file -

Which indicate that read/write to file is not required but it's not clear what's the proper syntax.