Comments (2)
Akababa
commented on July 25, 2024
1
I'm haven't looked at the the gpt2simple finetuning implementation, but about 36000/50000 of the BPE encodings in encoding.json involve some unicode escape like \u00e6 which we can discard if our goal is only to output certain english text without stuff like ĂŠĆŠ emojis, etc.
Are these baked into the model pre-finetuning? If not then we should be able to preprocess the training data to get a far more compact BPE.
from aidungeon.
dyc3
commented on July 25, 2024
1
I know that spaCy has named entity detection and it's pretty accurate. It could probably be automated
from aidungeon.
Related Issues (20)
- [FEAT] Add /forget, /update commands HOT 4
- [BUG] Stories disappear too quickly
- [BUG] When switching to another app on Android, game restarts HOT 4
- [BUG] Use balking retries to handle "The AI is a little overloaded"
- [BUG] Using back button on web browser logs me out HOT 1
- [BUG] Buffer overflow when text+remember_text gets too long? HOT 1
- [FEAT] Translation HOT 2
- [FEAT] Add sorting the search results HOT 3
- [BUG] AI never uses double quotes, even when direct dialog is enabled HOT 3
- [BUG] Starting custom prompts with capital letters clears input
- [Q/A]
- [BUG] White screen HOT 1
- [Q/A] Why does the subscriptions websocket returns the entire story text every time? Why not just the diff? HOT 1
- [BUG] Bottom bar on phone overlaps the button
- [DOC] re-training on custom data q HOT 1
- [FEAT] Solve the input and output confusion and other functional requirements of multiplayer mode.
- [FEAT] Add a Pin/Favourite button to Stories
- [BUG] Quests disappear after playing a few turns HOT 1
- [FEAT] nsfw toggle in Explore
- [BUG] Flickering pixel column on mobile app HOT 1
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
đ 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. đđđ
-
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 â€ïž Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from aidungeon.