Carsten Ditzel
Carsten Ditzel
What if the lib itself would be structured into several subprojects by means of the add_subdirectory() command in the top-level CMakeLists.txt. Can you maybe show, how to proceed then? Where...
Upon heavy reshaping and dimension manipulations, it is necessary from time to time to call .contiguous() on the resulting tensors to straighten out the memory layout. Does einops account for...
I believe that there is at least one 1x1 conv missing. In the paper on p. 3 they mention the crucial importance of those but I could only find a...
do I see it correctly that the code fragments provided by OpenAI and the way you binded it in the vae.py file means that there is no actual codebook in...
Been trying to get a grasp of the DALLE code recently. However, there are a couple of things, I cant quite wrap my head around and since the paper is...
in the train_vae script the kl_loss is set to zero via the weight parameter and also in my elaborate runs of experiments, I found that including the KL term does...
https://github.com/lucidrains/DALLE-pytorch/blob/995bfe1789243cbc838943cdc748daab406aae3e/dalle_pytorch/dalle_pytorch.py#L195 I am fairly certain that this should instead read `logits = rearrange(logits, 'b n h w -> (b h w) n') ` since we are summing over the latent...
According to your paper you use a large batch size of ~32k samples which means that the raw untrained network initially has a chance of ~1/32k of predicting the correct...
Hi John and first of all thank you for what you have given the Emacs community. Appreciate it. - does this package precede [ipython](https://github.com/jkitchin/ob-ipython)? - Whats the relation to https://github.com/gregsexton/ob-ipython?...
Can someone explain to me the difference between both approaches? They both are generating image content given a text input do they not? And what is then the difference between...