def _download_and_prepare(self, dl_manager):
import nltk
nltk.download("wordnet")
if NLTK_VERSION >= version.Version("3.6.5"):
nltk.download("punkt")
if NLTK_VERSION >= version.Version("3.6.6"):
nltk.download("omw-1.4")
def _compute(self, predictions, references, alpha=0.9, beta=3, gamma=0.5):
if NLTK_VERSION >= version.Version("3.6.5"):
scores = [
meteor_score.single_meteor_score(word_tokenize(ref),
word_tokenize(pred),
alpha=alpha,
beta=beta,
gamma=gamma)
for ref, pred in zip(references, predictions)
]
else:
scores = [
meteor_score.single_meteor_score(ref,
pred,
alpha=alpha,
beta=beta,
gamma=gamma)
for ref, pred in zip(references, predictions)
]
return {"meteor": np.mean(scores)}
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
""" METEOR metric. """
import numpy as np
from nltk.translate import meteor_score
import datasets
from datasets.config import importlib_metadata, version
NLTK_VERSION = version.parse(importlib_metadata.version("nltk"))
if NLTK_VERSION >= version.Version("3.6.4"):
from nltk import word_tokenize
_CITATION = """\
@inproceedings{banarjee2005,
title = {{METEOR}: An Automatic Metric for {MT} Evaluation with Improved Correlation with Human Judgments},
author = {Banerjee, Satanjeev and Lavie, Alon},
booktitle = {Proceedings of the {ACL} Workshop on Intrinsic and Extrinsic Evaluation Measures for Machine Translation and/or Summarization},
month = jun,
year = {2005},
address = {Ann Arbor, Michigan},
publisher = {Association for Computational Linguistics},
url = {https://www.aclweb.org/anthology/W05-0909},
pages = {65--72},
}
"""