Comments (2)
My understanding from the code is that during training the system uses gold label spans and generates all negative spans, then does some downsampling for the negative spans (a hyperparameter). During inference/evaluation, however, it generates all possible spans and selects the best spans based on how the model prioritizing them i.e. what has it learned through the training. The authors might confirm this.
For the second question, that is exactly what this system does. You don't need pre-annotated entities or relation labels for inference. This is a 'relation extraction' task.
from spert.
Thanks for answering this issue @avipartho. You are right, during inference we classify all spans up to length 10 (this is a hyperparameter - you can adjust it in the config file). Any span assigned to the 'None' class is then filtered.
from spert.
Related Issues (20)
- How to easily use this model for inference HOT 3
- Can't make predictions following the example HOT 8
- Help! Help! HOT 1
- Help, HOT 6
- How to call only the relation classifier on a pair of entities? HOT 2
- What is the meaning of the dataset tensors? HOT 1
- Simple example issue HOT 1
- Parts of entities are recognised separately HOT 3
- How does span filtering work? HOT 3
- Runtime Error HOT 1
- RuntimeError: copy_if failed to synchronize: cudaErrorAssert: device-side assert triggered HOT 4
- Does SpERT work with GPT models? HOT 1
- How to prepare dataset for training the model? HOT 9
- Can't download datasets
- TypeError: 'NoneType' object is not callable HOT 1
- Can't make train following the example
- Trained model : Relation classification is bad
- HELP HOT 1
- Extract entities and relation from Spacy tokens?
- [WARNI] NaN or Inf found in input tensor. HOT 1
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 spert.