-
Thanks for the great work!
I have a question about :Is this kind of MoE suitable for autoregressive model?
-
Hi there!
I try hard to replicate the original paper for MMOE Arch while have no idea to get the package run.
Same as the original paper, my data is Synthetic Data like x is a random vector, y1,…
-
I have downloaded the newest version of FuxiCTR library and decided to integrate the work from https://github.com/SkylerLinn/Understanding-the-Ranking-Loss
Specifically their DCNv2PositiveWeight and …
-
-
result of the main(I only changed the size of the input)
I create the test data of (4,8), the output was the same,
task_output tensor([[-0.0373, -0.0265],
[-0.0373, -0.0265],
[-0…
-
It seems that after installing this package with the latest version of the MMOkit causes the build to crash (and subsequently delete itself). The only change I made was instlal this UMA package and u…
Vygar updated
6 months ago
-
-
Hi, @tankche1, thanks for your nice work!
btw, how can I import this one?
```
from moe import MoE as MMoE
from moe import cvMoE
```
https://github.com/UMass-Foundation-Model/Mod-Squad/blob…
-
MGDA-UB原文中求representation的梯度,对于PLE和MMOE这类模型来说,representation是经过gate加权求和后的representation?
-
我实验下来也这样,在我的网络里提升有限。
大佬,你怎么看这篇paper
https://arxiv.org/pdf/2209.11379.pdf