Output file management for batches

[MrFrankDavis]

I would like to see an update with the same file management system as deforum. I like to make batch outputs, and it would be nice if all the outputs weren't named vid.

[terbo]

Only way I've found to do this is through the API. A pretty simple python script can do this. I'm playing with one right now if you want the code.

[MrFrankDavis]

Yes please, that would be awesome!

…

________________________________ From: CB ***@***.***> Sent: Wednesday, October 18, 2023 10:12 AM To: kabachuha/sd-webui-text2video ***@***.***> Cc: MrFrankDavis ***@***.***>; Author ***@***.***> Subject: Re: [kabachuha/sd-webui-text2video] Output file management for batches (Discussion #231) Only way I've found to do this is through the API. A pretty simple python script can do this. I'm playing with one right now if you want the code. — Reply to this email directly, view it on GitHub<#231 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/A57USUZP3CAVOTL7BR7TDVLX77WV5AVCNFSM6AAAAAA5E6SWASVHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM3TGMJWGUYTI>. You are receiving this because you authored the thread.Message ID: ***@***.***>

[MrFrankDavis]

Try adding vivd color to the end of the prompt. I've seen it make a big difference!

[beepbeepplop]

That's not something I've had to do before. I will nuke my install and see if starting from scratch gets me back to working properly...

[beepbeepplop]

Tried with another install. Videos are definitely washed out compared to earlier videos. But seems noone is really interested in modelscope anymore?

[MrFrankDavis]

Could it have something to do with TensorRT? I think it started for me near the time I added the extension.

[beepbeepplop]

I worked it out. After all my moaning it was was the Sampling method. Set it to DDIM and it works like I expected.

[MrFrankDavis]

Thanks!

…

________________________________ From: CB ***@***.***> Sent: Wednesday, October 18, 2023 9:23 PM To: kabachuha/sd-webui-text2video ***@***.***> Cc: MrFrankDavis ***@***.***>; Author ***@***.***> Subject: Re: [kabachuha/sd-webui-text2video] Output file management for batches (Discussion #231) Here<https://github.com/terbo/sdscripts/blob/main/dev/t2v.py> ya go. Usage: Enable API access in automatic1111 (add --api to COMMANDLINE_ARGS in webui-user.bat/sh) In settings/text2video, select "All" under 'Keep model in VRAM between runs'. Run python t2v.py -p "your prompt" --api http://127.0.0.1:7860 Or python t2v.py -f prompts.txt ... Most of the options are available on the command line, however I haven't been able to get img2img/vid2vid to work, as the post requests keep failing. Have fun! — Reply to this email directly, view it on GitHub<#231 (reply in thread)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/A57USU52MNWDHNXGZMWJUG3YACFJLAVCNFSM6AAAAAA5E6SWASVHI2DSMVQWIX3LMV43SRDJONRXK43TNFXW4Q3PNVWWK3TUHM3TGMRRGU4DI>. You are receiving this because you authored the thread.Message ID: ***@***.***>