FlexNeuART (flex-noo-art) is a Flexible classic and NeurAl Retrieval Toolkit
Reason this release was yanked:
this was a testing release, please, upgrade
Project description
FlexNeuART (flex-noo-art)
Flexible classic and NeurAl Retrieval Toolkit, or shortly FlexNeuART
(intended pronunciation flex-noo-art)
is a substantially reworked knn4qa
package. The overview can be found in our EMNLP OSS workshop paper:
Flexible retrieval with NMSLIB and FlexNeuART, 2020. Leonid Boytsov, Eric Nyberg.
In Aug-Dec 2020, we used this framework to generate best traditional and/or neural runs in the MSMARCO Document ranking task. In fact, our best traditional (non-neural) run slightly outperformed a couple of neural submissions. The code for the best-performing neural model will be published within 2-3 months. This model is described in our ECIR 2021 paper: Boytsov, Leonid, and Zico Kolter. "Exploring Classic and Neural Lexical Translation Models for Information Retrieval: Interpretability, Effectiveness, and Efficiency Benefits." ECIR 2021.
FlexNeuART
is under active development. More detailed description and documentaion is to appear. Currently we have:
- The installation instructions
- Usage notebooks covering most functionality (including Python API demo)
- Former life (as a knn4qa package), including acknowledgements and publications
- Collection-specific (older version the library):
The framework supports data in generic JSONL format. We provide conversion (and in some cases download) scripts for the following collections:
- MS MARCO data v1 and v2 (documents and passages)
- Wikipedia DPR (Natural Questions, SQuAD)
- Yahoo Answers collections
- Cranfield (a small toy collection)
For neural network training FlexNeuART incorporates a re-worked variant of CEDR (MacAvaney et al' 2019).
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for flexneuart-0.14-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 7480b1b2f29f3b8e4a8e2236afdb5c9d21308d9d2a2f8852e8760d822579b58d |
|
MD5 | 563c444e8cada5bf00ea13821d1a52e3 |
|
BLAKE2b-256 | 01b020c165b6d94430a816236fd8f76a997b41af3f30a451300aeb7d774b1b8a |