ru-dalle
ru-dalle copied to clipboard
ai-forever
Generate images from texts. In Russian
For example, captcha with russian words from yandex or captcha where you need to place some kind of figure (triangle, oval, square) into their place on the picture. Its much...
The model uses an uploaded image as a background image. Is it possible to use image as initial state?
There is error in the Kaggle notebook with example code for "Emojich" finetuning, that published here: https://www.kaggle.com/shonenkov/emojich-rudall-e When finetuning with a custom dataset, on the first try the training was...
Hi there, I just wanted to say this is fantastic work you've done here, I'm amazed at the results and really look forward to how your team progresses with this....
Currently, the inference code creates the entire attention matrix and then masks it. Sparse attention implementations like Triton are more efficient. Does the pre-training code support sparse attention? Will it...
Hi there :) First of all, I really want to thank you for sharing such a great work to the public. Because of your works, it was able to experiment...
In 'freeze' method, there is trivial error i think. When user intend to freeze attention layers, user will set freeze_attn=True. But, there is no layers contained 'attn' in names of...
Hi Team, Thanks for the excellent contribution to open source. I've been trying to adapt your code. I'm mostly focused on getting image embeddings from the given image and train...
Ran this in the ruDALLE-image-prompts-A100.ipynb JupyterLab notebook and am getting this error: !pip install rudalle==0.0.1rc2 > /dev/null The system cannot find the path specified. How to fix?
Many generated images have istock watermark. How to remove/prevent this from happening?