Comments (4)
I uploaded the base model with modified vocab.json
and merges.txt
files. The issue should be fixed now.
from polish-roberta.
The vocabulary for the base model is indeed wrong. This is probably a bug in the conversion from Fairseq to Transformers. The last id in the vocab.json for the base model should be 49999, so the entry with id 50001 is unnecessary. Can you check whether removing the last entry from the vocab helps?
As for the difference in size of the model vocab and vocab.json file, Fairseq reserves the last entry in the vocabulary for a special <mask>
symbol. This symbol should not be used in tasks other than language modeling so it was not included in the vocab.json and it is not included in the original Fairseq vocabulary as well.
from polish-roberta.
The error is appearing during training and not for every parameters (which is weird). I have added the 50000th entry and it helped.
I will check your solution.
from polish-roberta.
I can confirm, that the new model is working properly.
from polish-roberta.
Related Issues (8)
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 polish-roberta.