microsoft / k-adapter Goto Github PK
View Code? Open in Web Editor NEWLicense: MIT License
License: MIT License
Hi, thanks for your great work!
I wonder why the hyperparameters inside the run_finetune_figer_adapter.sh script differs from the best hyperparameters you mentioned in the paper's supplementary section?
Thanks for sharing the code. Can you also release the data for downstream tasks?
Just curious what the purpose of this would be, as I don't think I've seen it before and would like to know what the motivation is. This causes many data sample to be discarded when creating the examples. Thanks.
Just curious, because I tried running the code but realized evaluation wasn't taking place. Uncommenting the evaluation block also leads to an error and I'm wondering if that's to be expected.
I have followed the code here and generate all 3 tsv files under DisExtract/data/books/ALL18_2019jan02_[valid, train, test].tsv. However the format is not aligned with the required json file to run pretraining for Factual Adapter. The format of the tsv is also different than the required json format as well.
The content format of generated tsv file after executing python producer.py
is as follows:
[Sentence 1]\t[Sentence 2]\t[Marker]
...
The required json file format should be as follows:
{ "sent" : "Sentence 1", "tokens": "sentence 2", "pairs" : [ ... ] }
...
Is there a conversion script that convert generated tsv format to json?
I download the opentity dataset from its website and copy ./cloud in it to /data. However ,when I try to "bash run_finetune_openentity_adapter.sh",I got an Error:"json.decoder.JSONDecodeError: Extra data“
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 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. 📊📈🎉
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 ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.