-
Updated
Oct 3, 2020 - Jupyter Notebook
transformer
Here are 801 public repositories matching this topic...
-
Updated
Oct 10, 2020 - Python
chooses 15% of token
PositionalEmbedding
-
Updated
Jul 4, 2020 - Python
-
Updated
Sep 25, 2020 - Python
-
Updated
Aug 17, 2020 - Python
-
Updated
May 3, 2017 - Java
-
Updated
Sep 23, 2020 - JavaScript
-
Updated
Jun 1, 2020 - Jupyter Notebook
-
Updated
Feb 7, 2019 - Python
-
Updated
Sep 20, 2020 - Python
-
Updated
Oct 9, 2020 - Python
-
Updated
May 16, 2020 - Python
-
Updated
Sep 24, 2020 - Java
-
Updated
Jul 26, 2019 - Python
-
Updated
May 12, 2020 - Java
-
Updated
Jun 21, 2020 - Python
-
Updated
Sep 3, 2020 - Python
Hi, I am so interesting in your project, and wonder if you need contributor and how could I make my own contribution?
-
Updated
Jun 1, 2020
-
Updated
Nov 15, 2019 - Python
-
Updated
May 30, 2017 - Java
-
Updated
Jun 27, 2020 - Jupyter Notebook
-
Updated
Jul 16, 2020 - TypeScript
-
Updated
May 7, 2020 - Python
-
Updated
Oct 4, 2020 - Python
-
Updated
Jul 3, 2020 - Python
-
Updated
Oct 9, 2020
-
Updated
Oct 8, 2020 - JavaScript
Improve this page
Add a description, image, and links to the transformer topic page so that developers can more easily learn about it.
Add this topic to your repo
To associate your repository with the transformer topic, visit your repo's landing page and select "manage topics."
From paper, it mentioned
It means that 15% of token will be choose for sure.
From https://github.com/codertimo/BERT-pytorch/blob/master/bert_pytorch/dataset/dataset.py#L68,
for every single token, it has 15% of chance that go though the followup procedure.