Comments (4)
It depends how your training data looks like.
MULTI_NEGATIVES_RANKING_LOSS assumes that the batch looks like this:
x_1, y_1
x_2, y_2
....
x_n, y_n
(x_i, y_i) is treated as a positive example, for example given query x_i and the correct answer y_i, all other possible pairs (i.e. x_i, y_j for i != j) is treated as negative examples, i.e., y_j would be the wrong answer for query x_i.
from sentence-transformers.
Thank you! Are there any other loss function for doc ranking problem?
from sentence-transformers.
Doc ranking is currently not the core of this framework. You could use triplet loss, with providing a positive example and a negative example (positive example should rank higher than negative example).
Doc ranking is not my expertise, so I sadly don't know if there are any other good / well established doc ranking losses that are useful.
from sentence-transformers.
Thank you!
from sentence-transformers.
Related Issues (20)
- Segfault when using sentence-transformers (2.7.0) with Sklearn (1.3.0) HOT 12
- SimCSE dropout parameter HOT 2
- [`v3`] How do you envision this working with local datasets? HOT 3
- sentence-transformers 2.2.2 pulling in nvidia packages HOT 18
- MS Marco Cross Encoder Evaluation?
- MS Marco Cross Encoder Training
- Cross Encoder Recommendation for RAG
- Loading quantized models with SentenceTransformers HOT 3
- Tokenization issue when training: OverflowError: can't convert negative int to unsigned HOT 3
- Weird Behaviour for Finetuning Embeddings HOT 4
- Model.fit not saving the config.json file correctly for Crossencoders HOT 1
- HTTP 416 when trying to download model HOT 2
- Can SentenceTransformer load a large model using two GPUs?
- Using `LayerNorm` before PCA while performing embedding dimensionality reduction. HOT 2
- Refactoring `BinaryClassificationEvaluator.py`
- cannot import name '_sentencepiece' from partially initialized module 'sentencepiece' (most likely due to a circular import) HOT 1
- Constantly zero (or near zero) elements of the vectors of some of the older models HOT 8
- No sentence-transformers model found with name sentence-transformers/all-MiniLM-L6-v2. HOT 1
- [integration] [feature request] Add LLM2Vec to sentence-transformers
- Error training GPU and doing inference in cpu HOT 2
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 sentence-transformers.