diffusers
diffusers copied to clipboard
huggingface
🤗 Diffusers: State-of-the-art diffusion models for image, video, and audio generation in PyTorch.
I have encountered a memory leak problem, we need to store and switch the model of stable diffusion in memory, but when we convert the model in memory to gpu,...
### Model/Pipeline/Scheduler description Hi, Looking to see if we have any inputs on getting the LDM with inpaint, as that is advanced today with cleaning of image given a mask...
During Dreambooth or Textual Inversion training, I want to see periodic sample images to see how the training is going. There should be an option to save every X steps...
As seen in this PR: https://github.com/huggingface/diffusers/pull/2193 our tests can all pass even if dependencies are added to PRs that are no hard dependencies. The reason for this is that we...
convert this notebook to doc page https://colab.research.google.com/github/huggingface/notebooks/blob/main/diffusers/diffusers_intro.ipynb#scrollTo=jDk3_zhMsfPs
This PR exposes the `ff_multiplier` option of the Attention so that the UNet is able to set it. cc @sayakpaul
`attention.py` has at the moment two concurrent attention implementations which essentially do the exact same thing: - https://github.com/huggingface/diffusers/blob/62608a9102e423ad0fe79f12a8ceb1710d2027b2/src/diffusers/models/attention.py#L256 and - https://github.com/huggingface/diffusers/blob/62608a9102e423ad0fe79f12a8ceb1710d2027b2/src/diffusers/models/cross_attention.py#L30 Both https://github.com/huggingface/diffusers/blob/62608a9102e423ad0fe79f12a8ceb1710d2027b2/src/diffusers/models/cross_attention.py#L30 and https://github.com/huggingface/diffusers/blob/62608a9102e423ad0fe79f12a8ceb1710d2027b2/src/diffusers/models/attention.py#L256 are already used for "simple"...
So I was trying to feed the stablediffusionimg2imgpipeline with a video frame turned into torch tensor instead of a PIL image as is typical, and I was... struggling with garbled...
### Model/Pipeline/Scheduler description T2I-Adapter is a simple and small (~70M parameters, ~300M storage space) network that can provide extra guidance to pre-trained text-to-image models while freezing the original large text-to-image...
Added --save_every option to the dreambooth training script. In contast to --checkpointing_steps, it saves the model where the final model will be stored and does so independently of checkpointing settings....
Metadata
Owner
Metadata
🤗 Diffusers: State-of-the-art diffusion models for image, video, and audio generation in PyTorch.