ireval
This Python package provides an implementation of the most common information retrieval (IR) metrics.
Our goal is to return the same scores as trec_eval.
We achieve this by extensively comparing our implementations across many different datasets with their results.
ireval can be installed via
pip install ireval
Implemented metrics
The following metrics are currently implemented:
| Name | Function | Description |
|---|---|---|
| Precision@k | precision_at_k |
Precision is the fraction of retrieved documents that are relevant to the query. Precision@k considers only the documents with the highest k scores. |
| Precision@k% | precision_at_k_percent |
Precision is the fraction of retrieved documents that are relevant to the query. Precision@k% considers only the documents with the highest k% scores. |
| Recall@k | recall_at_k |
Recall is the fraction of the relevant documents that are successfully retrieved. Recall@k considers only the documents with the highest k scores. |
| Recall@k% | recall_at_k_percent |
Recall is the fraction of the relevant documents that are successfully retrieved. Recall@k% considers only the documents with the highest k% scores. |
| Average precision | average_precision |
Average precision is the area under the precision-recall curve. |
| R-precision | r_precision |
R-Precision is the precision after R documents have been retrieved, where R is the number of relevant documents for the topic. |
Usage
import ireval
relevancies = [1, 0, 1, 1, 0]
scores = [0.1, 0.4, 0.35, 0.8, .25]
p5 = ireval.precision_at_k(relevancies, scores, 5)
p5pct = ireval.precision_at_k_percent(relevancies, scores, 5)
r5 = ireval.recall_at_k(relevancies, scores, 5)
r5pct = ireval.recall_at_k_percent(relevancies, scores, 5)
ap = ireval.average_precision(relevancies, scores)
rprec = ireval.r_precision(relevancies, scores)
1 maintainer
Metadata
1 maintainer
Release files for ireval 0.1.1
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | |
|---|---|---|---|
| ireval-0.1.1.tar.gz | 4.0 kB | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| ireval-0.1.1-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 8.9 kB
Release files / ireval-0.1.1.tar.gz
| Download URL | ireval-0.1.1.tar.gz |
|---|---|
| Size | 4.0 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
bebf5faec983468672ca23a9ac20611550f374723d025c2459bf9265eb374b92
|
|
BLAKE2b-256 checksum How to use checksums |
b8d793e4fcf369685f9d354ac2c00051e8c81f37fb9eb529e087b2e5c131eafc
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
poetry/1.1.13 CPython/3.8.2 Windows/10
|
Release files / ireval-0.1.1-py3-none-any.whl
| Download URL | ireval-0.1.1-py3-none-any.whl |
|---|---|
| Size | 4.9 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
ad0d3680a1bd77b1718f12626d5af4d5ccae453dacf3edfcec20af72dfd7b6b5
|
|
BLAKE2b-256 checksum How to use checksums |
7d4ed6a6061cae1299c7d93583d15dd4ca7409351fb8e04ce6ac2a0e676292ed
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
poetry/1.1.13 CPython/3.8.2 Windows/10
|
