-
Very interesting piece here on comparing Python against C++ and fortran implementations of classic physics simulations: https://www.nature.com/articles/s41550-021-01342-y
We should certainly mentio…
-
We are planning to implement Praos+Genesis in a slightly different way than the paper suggests (#219)
We need to verify that if there are independent implementations of genesis (in particular, the Ru…
edsko updated
10 months ago
-
**What would you like to be added**:
I would like to be able to enforce authentication & (limited) authorization at the Gateway or HTTPRoute level, so that I can safely route traffic to apps.
**…
-
Hi, thanks for your insightful work! But it seems that some settings and implementations of baselines are not so proper and fair?
1. Window size. In `run_longbench.py: 221`, PyramidKV uses window s…
-
As said in your paper, the nonlocal block is initialized as zero weights, while with its shortcut connection, it will not influence the original network. But in your implementation, nonlocal block is …
-
### Planed TODO
- [ ] KL + Power - Single speaker
### Properties not specified in the paper
- Sampling number for the loss (We may have some limitation for GPU)
- Number of mixture for IAF layer…
-
Taking inspiration from [this paper](https://spcl.inf.ethz.ch/Publications/.pdf/fblas.pdf), implement some level 1 and 2 BLAS kernels and parameterize them similarly. Specifically:
- [ ] a `GEMV` imp…
-
**Describe the feature**
**Motivation**
A clear and concise description of the motivation of the feature.
Ex1. It is inconvenient when \[....\].
Ex2. There is a recent paper \[....\], which …
-
Hi, thanks for open-sourcing the great work, which is very helpful for sparse deep learning workloads. I notice there is a `𝑛:𝑚:𝑔` sparsity layout in your paper, but I could not find the `GroupedNMSpa…
-
Hi. Thanks for your great work!
I would like to ask regarding the models uploaded in google drive for att2in2 and top-down. What modifications from the authors paper have you used in both to get thes…