Comments (6)
from ncrfpp.
Hi @azraelkuan , the reported value of the n-best is the same as the 1-best, i.e. the acc/p/r/f. They are all the performance of one best. The difference is the decoded text, you will see the decoded text includes 10 label sequences and their probabilities for each sentence.
from ncrfpp.
so, how can i use the decoded text with 10 label to get better acc/p/r/f?
what i do is that only choose the first label of 10 to do the evaluation, that is the same with 1 best..
from ncrfpp.
The n-best cannot give better performance in separate. It only provides more abundant predict results rather than the single one. For example, sometimes the gold label sequence is the second best rather than the one best.
The nbest decoding is normally used in reranking models to provide more available candidates. And the rerank system will learn how to choose the best sequence from the n-best candidate. The oracle results I reported is based on an assumption that the rerank system works perfectly to choose the best sequence from the n-best outputs. It is a standard metric to show the upbound of the rerank system. You may search the papers of reranking system for more details, or you can check our rerank paper: http://www.acl-bg.org/proceedings/2017/RANLP%202017/pdf/RANLP101.pdf.
In addition, the n-best decoding can also be used in other areas such as the search engine, which extracts all the recognized entities from the n-best outputs to give a better recommendation.
from ncrfpp.
thanks!
from ncrfpp.
hi jiesutd! What is the way you calculate the f1 and acc of n-best you reported in your paper?
from ncrfpp.
Related Issues (20)
- Conll 使用内存大 HOT 2
- nbest score HOT 1
- Provision for Custom Features? HOT 1
- Bug in forward method when calling _viterbi_decode, mask is not provided
- about requirement HOT 1
- Can I just use the CRF layer? HOT 1
- 大神你好,请教一些关于报错的问题 HOT 1
- About "tcmalloc: large alloc" message and training aborted HOT 3
- Difference between main_parse.py and main.py HOT 1
- Different results on CPU versus GPU HOT 1
- Test Data Format for Chunking
- Experimenting with Transformer models? HOT 1
- Decode config parameter when not using CRF HOT 1
- Reading fasttext word embeddings HOT 2
- About MAX_SENTENCE_LENGTH parameter HOT 2
- Decode error with nbest=0 when not using CRF
- Resume model training from a checkpoint
- 训练一段时间后,f1为-1 HOT 2
- Unknown labels in decoding
- Support trained model serialization to ONNX format 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 ncrfpp.