automatic VS stable-diffusion-webui-wd14-tagger

> it won't ever be a viable option

For production workloads, I generally agree. It's an unsupported hack with a questionable future, I wouldn't do anything money-making with it.

However, for tinkering and consumer workloads, it already works pretty well. Enough of cuDNN and cuBLAS work to run PyTorch and in turn, Stable Diffusion with https://github.com/lshqqytiger/ZLUDA - there's even a fairly user-friendly setup process already in https://github.com/vladmandic/automatic .

I was able to get a personal non-ML related project working on my AMD card in just a few minutes, which saved me a lot of development time before I then deployed the production workload on NV hardware (this is probably why AMD pulled the plug on the project - it's almost more of a boost to NV than anything else, AMD really need people to be writing code on ROCm to deploy on AMD datacenter hardware).

While I currently use SD.Next[1], I have tested ComfyUI locally with my AMD card. The UI can be daunting, but you learn quite a great deal about how a Stable Diffusion pipeline works. In addition some innovations and advances find their way into ComfyUI first.

[1] https://github.com/vladmandic/automatic

I used SD.NEXT ( https://github.com/vladmandic/automatic ) and https://civitai.com/models/82098/add-more-details-detail-enhancer-tweaker-lora and epicphotogasm_lastUnicorn

I found this thread on GitHub talking about how this was fixed in the latest version with an optional setting. I tried enabling it, as they mentioned, but it just resulted in an immediate CUDA out of memory error when starting generation. So it seems I'm actually needing the shared memory, which I assume is my issue.

As always, do check out our more detailed changelog, give us a quick install from our Repo, and stop by our Discord Server for any questions or help you may need.

SD.Next

Today I finally wanted to give SDXL a chance, so I set everythin up according to Vladmandic's Wiki https://github.com/vladmandic/automatic/wiki/SD-XL

I'm trying to get SDXL working on Vlad's SDNext, but I keep getting the error in the title when trying to run basic operations. I'm not sure what's going on, I followed his guide for it to a T.

We used an inference container based on SDNext, along with a custom worker written in Typescript that implemented the job processing pipeline. The worker used HTTP to communicate with both the SDNext container and with our batch framework.

Create an extension like 'send pictures' that uses the WD14 tagger which is way more detailed and has options for nsfw etc. Its used in Automatic1111 and Koyha ss so there's extensions you can probably implement from. https://github.com/toriato/stable-diffusion-webui-wd14-tagger

If anyone is interested I made some changes to toriato's wd14-tagger, now it works also on vladmandic webui, repo here. You can do a new installation, or use your old automatic1111 one changing 3 files, instructions on my repo. The lora files also work (there were some problems in the vlad issue page). I'm not a programmer and it's not perfect though, in fact for now if you don't like the default tagger model you have to change it manually (instructions in the repo), and since it is basically a fork of toriato's version, if there were errors there, there will be here too.

Automatic captioning tools that can be used as an initial point for captions: this tool or this one.

wd14-tagger, to describe anime images and get a prompt idea

Use WD 1.4 tagger (https://github.com/toriato/stable-diffusion-webui-wd14-tagger) to extract prompt words from each frame (threshold 0.65), then use the dataset tag editor (https://github.com/toshiaki1729/stable-diffusion-webui-dataset-tag-editor) for batch editing, mainly:

I recommend using EveryDream2. You'll need an 11GB VRAM GPU. There's no need to crop or resize images, just caption them, which can be done automatically with CLIP Interrogator or WD14 taggers. Make sure to add the trigger word for your subject. It's not a Dreambooth script; it's actual training, so it shouldn't be as destructive to the model as Dreambooth. Typically, using an LR of 1e-6 with a cosine scheduler over two epochs and a batch size of 4 works fine. This script supports validation, so you can actually watch in real-time whether the training is going well or if you're overfitting. I got very good results using it.

In my current process, I use CLIP Interrogator to produce a high level caption and wd14 tagger for more granular booru tags. Typically in that order, because you can append the results from the latter to the former. Both tools perform with greater accuracy than the standard interrogators in img2img and give you more flexibility and features as well. You still have to do some manual adjustments, but I generally prefer this process over starting from scratch.