# Copyright 2020 The HuggingFace Datasets Authors and the current dataset script contributor.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# 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.
"""TODO: MCC is a correlation coefficient between the observed and predicted binary classifications, and takes into account true and false positives and negatives."""

import evaluate
import datasets
from sklearn.metrics import matthews_corrcoef



# TODO: Add BibTeX citation
_CITATION = """\
@InProceedings{huggingface:module,
title = {MCC Metric},
authors={huggingface, Inc.},
year={2020}
}
"""

# TODO: Add description of the module here
_DESCRIPTION = """\
MCC (Matthews Correlation Coefficient) is a correlation coefficient between the observed and predicted binary classifications, and takes into account true and false positives and negatives. It can be computed with the equation:
MCC = (TP * TN - FP * FN) / sqrt((TP+FP) * (TP+FN) * (TN+FP) * (TN+FN))
Where TP is the true positives, TN is the true negatives, FP is the false positives, and FN is the false negatives.
"""


# TODO: Add description of the arguments of the module here
_KWARGS_DESCRIPTION = """
Calculates how good are predictions given some references, using certain scores
Args:
    - **predictions** (`list` of `int`): The predicted labels.
    - **references** (`list` of `int`): The ground truth labels.
Returns:
    - **mcc** (`float`): The MCC score. Minimum possible value is -1. Maximum possible value is 1. A higher MCC means that the predicted and observed binary classifications agree better, while a negative MCC means that they agree worse than chance.
Examples:
    Example 1-A simple example with some errors
        >>> mcc_metric = evaluate.load('mcc')
        >>> results = mcc_metric.compute(references=[0, 0, 1, 1, 1], predictions=[0, 1, 0, 1, 1])
        >>> print(results)
        {'mcc': 0.16666666666666666}
    Example 2-The same example as Example 1, but with some different labels
        >>> mcc_metric = evaluate.load('mcc')
        >>> results = mcc_metric.compute(references=[0, 1, 2, 2, 2], predictions=[0, 2, 2, 1, 2])
        >>> print(results)
        {'mcc': 0.2041241452319315}
"""

# TODO: Define external resources urls if needed
BAD_WORDS_URL = "https://scikit-learn.org/stable/modules/generated/sklearn.metrics.matthews_corrcoef.html"


@evaluate.utils.file_utils.add_start_docstrings(_DESCRIPTION, _KWARGS_DESCRIPTION)
class MCC(evaluate.Metric):
    """Compute MCC Scores"""

    def _info(self):
        return evaluate.MetricInfo(
            module_type="metric",
            description=_DESCRIPTION,
            citation=_CITATION,
            inputs_description=_KWARGS_DESCRIPTION,
            features=datasets.Features({
                'predictions': datasets.Value('int64'),
                'references': datasets.Value('int64'),
            }),
            # Homepage of the module for documentation
            homepage="https://huggingface.co/evaluate-metric?message=Request%20sent",
            # Additional links to the codebase or references
            codebase_urls=[],
            reference_urls=[]
        )

    def _compute(self, predictions, references):
        """Returns the mcc scores"""
        # Computes the MCC score using matthews_corrcoef from sklearn
        
        return {"mcc": matthews_corrcoef(references, predictions)}