thibaudart

Training datasets

27

First of all, **Thanks** a lot for your work on this amazing tool! Could you share the datasets used for training? With that we can make the training on sd2.1!

Is it possible to resume a training from a checkpoint?

Add ZoeDepth as annotator

6

Add ZoeDepth as annotator. The training of the model (for SD2.X) is almost done and we'll need this annotator to use it.

Option for Chat memory

1

Maybe we can use LangChain and their method to have conversational memory. It could help when we have long discussion and can lower the costs. https://www.pinecone.io/learn/langchain-conversational-memory/

`load_in_8bit_fp32_cpu_offload=True

4

Any idea how to solve this: Some modules are dispatched on the CPU or the disk. Make sure you have enough GPU RAM to fit the quantized model. If you...

integration with other workflow

1

hi your repo works well but the issue is, it's not compatible with other workflow (to have other controlnet, other ipadapter and use custom sampler). the "all in one" node...

Hi Do you plan to release the training script? I didn't find it.

ModuleNotFoundError: No module named 'diffusers.models.activations'

3

I'm trying to install it but even after restarting webui, I got: *** Error loading script: rich_text_on_tab.py Traceback (most recent call last): File "E:\automatic_xl\modules\scripts.py", line 382, in load_scripts script_module =...