lllliy

Hi @lucidrains, I found that the input of the En-Transformer is different from EGNN in your released repository that EGNN using `torch_geometric` lib to deal with the node and edge....

Hi, thanks for your coding. but i don't find the `downsample` operation.

question about sent_message()

23

Hi, i get the token through your step, and when i run the code, `print(client.get_chat_history())` can return, but when i run the part about `for chunk in client.send_message(bot, message):`, there...

train了50个epoch，突然报这个错误。请问了解不？ 

segmentation fault (core dumped)

3

大家训练过程中有遇到这个错误吗？训练中途，报错，segmentation fault (core dumped)...

运行过程中会出错

3

运行到n个epoch后，提示TypeError: 'Tensor' object is not iterable.不知道大家遇到过没？ `utils_fit.py` 中的 optimizer.step()产生的，但是训练了那么多epoch都没问题，为什么会突然出现问题呢？

Hi, thanks for your work! I wondering if you release the test data consistent with your paper？ how to test the model on the whole test data？ I found that...

Hi, could I use the demo to test editing online? I find that the URL is wrong. and the same issue #15 has not been solved.

Hi, I find that the cross-attention map is calculated between text and latent image representation. Why is the visualization between text and image pixels? (I find that the code in...

How to support VESDE

1

In the paper "[Score-Based Generative Modeling through Stochastic Differential Equations](https://openreview.net/pdf?id=PxTIG12RRHS)", they support VESDE, use reverse-diffusion (predictor) and Langevin (corrector) to sample data. How do we adapt this solver to VESDE...