A library for the collection of common low-level features used in learning-to-rank algorithms. These features are calculated using well-defined algorithms.
If you feel there are additional features that might be added to this library, create a new issue or submit a pull request.
High level features created using machine learning based methods can be found or added to my highletorfeats library.
The following categories of low-level features are collected using this library:
- TF-IDF
- Okapi BM25
- LMIR (TODO)
The recommended way to add this library to you project is by including the following to your CMakeLists.txt:
cmake_minimum_required(VERSION 3.13)
project(myProject)
include_directories("path/to/lowletorfeats/include")
add_subdirectory("path/to/lowletorfeats")
add_executable(myProject myProject_SOURCES)
# or `add_library(myProject myProject_SOURCES)`
target_link_libraries(myProject lowletorfeats)
The library has can be included in your source files with the following:
#include <lowletorfeats/FeatureCollector.hpp>
An example case can be found in tests/src/main.cpp.
We use SemVer for versioning. For the versions available, see the tags on this repository.
- Patrick Cox - paddy74
This project is licensed under the MIT License - see the LICENSE.md file for details.