Comments (2)
TODO @mkossakowski19 :
- Refactor Q&A code
- Tests on English Squad and add example of usage https://huggingface.co/datasets/squad_v2
- Tests on Polish translated Squad
from embeddings.
TODO (17.02):
- Add tests for the QA evaluator
- Adjust tests for pipeline
Fix errors and add more detailed metrics checks. Currently, in tests, metrics are compared as0 < mt < 100
, but it should be changed to 7 decimal precision, e.g.,assert_almost_equal(mt, 0.1234567)
. - Move batch encoding kwargs to LightningBasicConfig class
CreateQALightningBasicConfig
that is more suited for QA (2 batch_encoding parameters and 1 QA-specific parameter (doc_stride
) - Fix linter
There are issues with styling and pyflakes
Optional tasks (lower priority):
- [optional] Adjust LightningModule class logic
Currently, there is code duplication between LightningModules (setup
,freeze_transformer
,forward
). However, it needs restructuration of the inheritance hierarchy asHuggingFaceLightningModule
are specific for Classification tasks. - [optional] Notebook with QA pipeline example
Add an example how "easy" it is to run a QA pipeline
from embeddings.
Related Issues (20)
- Pipeline to cleanup and normalise content of txt files HOT 1
- Prepare keyword extraction dataset
- Create streamlit-based dashboard for leaderboard
- UDALM domain adaptation pipeline
- Validate inference of the persisted models (after training) HOT 1
- Create submission examples using wandb for Polish Leaderboard HOT 1
- Add Political Advertising dataset to HF HOT 1
- Export HF models to ONNX format
- Create a first version of leaderboard view based on submissions
- Decide on meta data similar to HF for each dataset and model
- Extend result set by new field
- Baseline models training HOT 1
- Add mlflow logging
- Add default dataset info and dataset version if train subset is missing in HuggingFaceDataModule
- Updates to HF spaces
- Create documentation and library presentation
- Add Inference Pipelines
- Prepare converter from squad to beir HOT 1
- Fix leaderboard webpage
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 embeddings.