Lin (2004)#
Publication#
Repositories#
n/a
Available Models#
ROUGE
Description: A wrapper around the original Perl implementation of ROUGE
Name:
lin2004-rouge
Usage:
from repro.models.lin2004 import ROUGE model = ROUGE() inputs = [ {"candidate": "The candidate summary", "references": ["The first reference", "The second"]} ] macro, micro = model.predict_batch(inputs)
The
macro
results are the ROUGE scores averaged over theinputs
. Themicro
results are the ROUGE results for each item ininputs
.
Implementation Notes#
Docker Information#
Image name:
lin2004
Docker Hub: https://hub.docker.com/repository/docker/danieldeutsch/lin2004
Build command:
repro setup lin2004 [--silent]
Requires network: No
Testing#
repro setup lin2004
pytest models/lin2004/tests
Status#
[x] Regression unit tests pass
See here[ ] Correctness unit tests pass
[ ] Model runs on full test dataset
[ ] Predictions approximately replicate results reported in the paper
n/a[ ] Predictions exactly replicate results reported in the paper
n/a