Comments (8)
The short answer is "yes"!
We are working to consolidate the split and should be able to release something soon -- hopefully in a week or two.
FYI @yizhongw
from natural-instructions.
Hi @danyaljj, thanks for this work! Can I clarify - since this issue was a few weeks ago, and I saw the paper released recently, were you referring to the paper being released or is there something else coming?
In particular, I am looking for
- The evaluation split of tasks
- Code to reproduce the results of the paper (e.g. for finetuning and evaluating on this dataset)
If they are not yet released, it would be helpful if you could give a rough idea for when that would be released (I am trying to decide whether to include this excellent benchmark in my own work but that would depend on the timeline).
Thank you!
Thanks for your interest! We are planning to release the splits and evaluation scripts next week (hopefully by Wed). Does this work for your schedule?
from natural-instructions.
Hi @yizhongw, any update on this? Sorry to rush you, but for our planning it'll be helpful to know whether we will get to include this or not.
from natural-instructions.
Hi @yizhongw, any update on this? Sorry to rush you, but for our planning it'll be helpful to know whether we will get to include this or not.
@JunShern Sorry for being late on this! We have merged this PR just now, and now the official evaluation splits can be found here. You can follow the instructions there for setting up your experiments.
We also released our code for Tk-Instruct model and checkpoints. Feel free to let me know if you find any issue!
from natural-instructions.
Thanks for the heads-up!
from natural-instructions.
Hi @danyaljj, thanks for this work! Can I clarify - since this issue was a few weeks ago, and I saw the paper released recently, were you referring to the paper being released or is there something else coming?
In particular, I am looking for
- The evaluation split of tasks
- Code to reproduce the results of the paper (e.g. for finetuning and evaluating on this dataset)
If they are not yet released, it would be helpful if you could give a rough idea for when that would be released (I am trying to decide whether to include this excellent benchmark in my own work but that would depend on the timeline).
Thank you!
from natural-instructions.
Amazing, yes next week should work for us! Thanks for the quick response and all the best. :)
from natural-instructions.
Amazing, thank you! I'll be looking into it :)
from natural-instructions.
Related Issues (20)
- Task1578 and Task288 HOT 6
- Additional classification tasks HOT 1
- Indirect responses
- The problem with task 839 HOT 2
- Dialog Tasks HOT 2
- Data-to-text tasks
- Tasks 475 and 746 seem to be duplicates?
- Readme file has problems HOT 1
- Duplicate tasks: 1508 and 1509 HOT 3
- Ungrammatical Instructions for Some Tasks HOT 1
- More crowdworker feedback HOT 10
- Task languages HOT 1
- Data leakage with tasks 1295 and 1640 HOT 4
- Sequence tagging HOT 5
- Task 183 - is it correct? HOT 1
- Spacing/tokenization issues HOT 4
- Task 199 (MNLI): is the data correct?
- Multi-class classification HOT 1
- about original dataset templates
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from natural-instructions.