microsoft / kid Goto Github PK
View Code? Open in Web Editor NEWKnowledge Infused Decoding
License: MIT License
Knowledge Infused Decoding
License: MIT License
In compute_reward()
function in kid_policy.py
kl penalty is calculated as below.
kl_penalty = self.kl_scale * ( (kid_last_probs * (kid_last_probs / norm_last_probs).log()).sum() )
But the equation in the paper is
I am a little bit confused because, as the comment of the function says norm_last_probs
is pi, it is considered kid_last_probs
and norm_last_probs
should be changed. If I am wrong, please let me know.
Nice work. Would the training code be open sourced?
Hi,
This paper is very interesting, can you share the evaluation scripts?
I find some trouble in the evaluation stage.
Thanks
Hi, I try to use Spacy V3, but I got a warning that "The component 'entity_ruler' does not have any patterns defined."
So, should I specify a specific pattern for entity_ruler?
Thanks
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.