-
It seems useful to collect instances of the bug minimizer being used, since I'm hopeful @achlipala and @Zimmi48 and I will be able to get a paper out of this.
Here are uses so far:
- https://githu…
-
Hi @tkipf ,
It seems _model.reconstructions_ includes all the positive and negative edges, no matter in training, validation or test set, whereas the labels used during training only have "1" entri…
-
Note: the issue was created automatically with bugzilla2github tool
Original bug ID: BZ#1671
From: @PierreCorbineau
Reported version: 8.5
CC: @pirbo
-
The following proof script checks fine when indices-matter is disabled, but gives an error when it is turned on.
``` coq
Inductive Empty : Type :=.
Inductive paths {A : Type} (a : A) : A -> Type :=
…
-
Sorry, I didn't consider the different categories of the issues earlier. I should have opened separate issues.Do you think the architecture for lip reading tasks has already become standardized? I m…
-
# Equivalence classes and quotienting
[https://lawrencecpaulson.github.io/2022/03/30/Quotienting.html](https://lawrencecpaulson.github.io/2022/03/30/Quotienting.html)
-
Hi,
I'm trying to use GAT in an inductive learning setting. I read from the previous discussions that this can be done by ClusterNodeGenerator with GAT. Can you please create a Demo notebook for th…
-
@sweirich in [a comment] on commit 4cd6596, about `unaxiomatize definition`:
> How about something a little more general, so that we can have "skip everything except for this" too.
> What I'm thin…
-
#### Version
`8.7.2`
#### Operating system
Linux (Ubuntu 16.04)
#### Description of the problem
```coq
Definition point2d : Type := nat * nat.
(** Error: Destructing let on …
bollu updated
6 years ago
-
我感觉您的test_set更应该像valid_set的功能，因为参与了idf和pmi的计算过程，所以论文中的performance应该是valid_set的。
但是，如果要另外分出一个真正的test_set去和其他论文比较多话(比如我将这模型应用在其他数据集)，公平起见的话，test_set不应该参与训练期间的idf和pmi的计算。所以这里有个难点，因为做评估的时候预测值是直接从node获得的，…