-
Notifications
You must be signed in to change notification settings - Fork 0
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
I can't use vision command with same session #6
Comments
I tested v2.1.2 and is same. I can't use record session with vision command in Windows. Please, @duanemay, if I can help you, you can ask me. Thank you a lot. |
I guess I inadvertently added the session file as an option to the vision command, but did not plan to support that. The reason being that the encoded image data would be included in those files and they could get rather large. I could add the feature, knowing they will be larger than the original image, probably should not create the files by default or remove this unused option. PRs are welcome. |
Thank you very much for your reply.
From my point of view, it would be very useful to be able to add the images to the session file. You could warn in the readme that this can increase its size a lot.
Maybe you could add an option to add the full file or just the file name.
I also think it would be very useful to add an option to show at the end of each API call a line with the cost ($) of that call. I don't know, but it is possible that having more size for the images would increase the cost.
Something like the following would be very useful:
echo "The name of the man in this image is John" | chatgpt-cli -f john.jpg -c config.cfg >> out.txt
echo "The name of the woman in this image is Mary" | chatgpt-cli -f mary.jpg -c config.cfg >> out.txt
echo "Describe this image and say me who is in it" | chatgpt-cli -f image1.jpg -c config.cfg >> out.txt
Thanks a lot.
De: Duane May ***@***.***>
Enviado el: domingo, 19 de enero de 2025 1:44
Para: duanemay/chatgpt-cli ***@***.***>
CC: afr-ux ***@***.***>; Author ***@***.***>
Asunto: Re: [duanemay/chatgpt-cli] I can't use vision command with same session (Issue #6)
I guess I inadvertently added the session file as an option to the vision command, but did not plan to support that. The reason being that the encoded image data would be included in those files and they could get rather large. I could add the feature, knowing they will be larger than the original image, probably should not create the files by default or remove this unused option.
PRs are welcome.
—
Reply to this email directly, view it on GitHub <#6 (comment)> , or unsubscribe <https://github.com/notifications/unsubscribe-auth/AQ3N74Q4PVE6FXQDRHZABT32LLYLPAVCNFSM6AAAAABUYUS4QSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDMMBQGM3DANZUGI> .
You are receiving this because you authored the thread.Message ID: ***@***.***>
|
The use case you shared makes it clear what you are trying to use this for, and makes sense that the feature is needed for multiple images. I'll look into adding this. |
I am using:
echo "This is an image to describe" | chatgpt-cli vision -f 1.jpg -c afr.cfg
In afr.cfg file, I have:
API_KEY=sk-my-secret-key
TEMPERATURE=0.5
SESSION_FILE=afr.json
I can't use session file with vision command. In afr.json nothing is added when I use vision command.
I have chatgpt-clie v2.1.1 in Windows 10-64 bits.
This tool is wonderfull for describe images for blind people but I would like record session to teach some things in images and chatGPT remember it in other images.
Thanks.
The text was updated successfully, but these errors were encountered: