-
Hi, this looks like really nice work, and I'm impressed with the speedups you're getting over Zygote AD.
Have you thought about reversible random number generation? Random number generation is reve…
-
My question is actually because I have a normal(theta,sigma) distribution where theta is dnorm(mu,tau) and then I assign mu dunif(-10,10).
I want to have dunif (-10,10), how can this work? shoul…
-
Hi Team,
I have been trying to train MViT on the custom dataset, but it happens to overfit very quickly on a dataset of size 10k images. It reached 5%< error in the first iteration for the trainse…
nik13 updated
2 years ago
-
I wanted to note a discrepancy between sequencing data with and without duplicates. I know this is mentioned in the manuscript, Issue #31, and issue #24, but from what I can tell it seems that preseq…
-
Let's start with Sobol sequences, only because we currently have code for these. If we're in `d` dimensions, the Sobol sequence is a sequence of length-d vectors.
Actually, let's make this even mo…
-
@hpieper14 @ChrisRackauckas @frankschae @KirillZubov
Design discussions here so that they don't get slurped up by slack.
-
Hi Xenko team. I tried several ways and it seems to be impossible.
The following code compiles, but fails at runtime -
shader TextureColor : ComputeColor, Texturing
{
overri…
-
### 🐛 Describe the bug
The `uniform_` function is single-threaded and slow, causing `nn.Linear` to be slow during initialization. Since the first step to loading a model is to create an empty model, …
-
## ❓ Questions & Help
I've found only **one** `torch.nn.Embedding` layer in `torch_geometric.nn.models.node2vec `, which is not consistent with the offcial [previous implementation](https://github.co…
-
hello there, i'm getting the following error:
```---------------------------------------------------------------------------
ValueError Traceback (most recent call l…