-
Use this code to reproduce published results for RNNs for text classification. Put the results and references in the README.
Nothing exhaustive, just reproduce something to show that it's working.
-
@gcp recently increased the training steps to 256k https://github.com/gcp/leela-zero/issues/78#issuecomment-354068002 but training an additional 128k steps takes 4 hours, which typically means no test…
-
先占个坑。[CVPR2016接收文章](http://www.cv-foundation.org/openaccess/CVPR2016.py)643篇，标题中含“video”的小于59篇（个别文章同一个标题里多次出现“video”一词），下面按[CVPR2016接收文章](http://www.cv-foundation.org/openaccess/CVPR2016.py)中的顺序逐一整理。
…
y-wan updated
7 years ago
-
The new PaddlePaddle core calls REGISTER_OP macro to fill in information (input, outputs, attributes) of an operator into a global variable of type [OpProto](https://github.com/PaddlePaddle/Paddle/blo…
-
Currently I train a network(20 blocks and 256 filters). It initialized from b8adb7da(78) n2n to 20b 256f. And it is now being trained to b3a80524(99). According to unofficial test by others, it is muc…
-
Based on weight items:
- Elo Rating
- Surface Elo Rating
- H2H
- Surface H2H
- H2H in last period
- Surface H2H in last period
- Vs Ranking
- Vs Hand
- Vs Backhand
- Winning Pct in Level
- …
-
There's a (linux) cheat that actually utilizes the in-game outlines that is normally used in demos and is able to simply change the memory for player entities in a particular struct to get a working w…
-
We'd thought that a DL framework should implement concepts like *model* and *cost*. But we realized that these are not flexible enough to describe deep learning problems. Instead, we need the concept …
-
It would be nice if scikit-learn showed which lines are covered by unit tests in our PRs. For example, Coveralls will show your test coverage line by line.
![image](https://cloud.githubusercontent.co…
-
Currently, the development of machine learning tools are in several different packages without little coordination. Consequently, some efforts are repetitive, while some important aspects remain lacki…