Merge pull request #2463 from peastman/pearson (361207f8) · Commits · 钟慕尧 / deepchem

deepchem/metrics/score_function.py

+20 −2

Original line number	Diff line number	Diff line
		"""Evaluation metrics."""

		import numpy as np
		import scipy.stats
		from sklearn.metrics import matthews_corrcoef # noqa
		from sklearn.metrics import recall_score # noqa
		from sklearn.metrics import cohen_kappa_score
		@@ -15,12 +16,29 @@ from sklearn.metrics import f1_score
		from sklearn.metrics import roc_auc_score # noqa
		from sklearn.metrics import accuracy_score # noqa
		from sklearn.metrics import balanced_accuracy_score # noqa
		from scipy.stats import pearsonr

		# kappa_score is an alias for `sklearn.metrics.cohen_kappa_score`
		kappa_score = cohen_kappa_score


		def pearsonr(y: np.ndarray, y_pred: np.ndarray) -> float:
		"""Computes Pearson correlation coefficient.

		Parameters
		----------
		y: np.ndarray
		ground truth array
		y_pred: np.ndarray
		predicted array

		Returns
		-------
		float
		The Pearson correlation coefficient.
		"""
		return scipy.stats.pearsonr(y, y_pred)[0]


		def pearson_r2_score(y: np.ndarray, y_pred: np.ndarray) -> float:
		"""Computes Pearson R^2 (square of Pearson correlation).

		@@ -36,7 +54,7 @@ def pearson_r2_score(y: np.ndarray, y_pred: np.ndarray) -> float:
		float
		The Pearson-R^2 score.
		"""
		return pearsonr(y, y_pred)[0]**2
		return scipy.stats.pearsonr(y, y_pred)[0]**2


		def jaccard_index(y: np.ndarray, y_pred: np.ndarray) -> float:

+14 −0

Original line number	Diff line number	Diff line
		@@ -34,6 +34,20 @@ def test_one_sample():
		_ = metric.compute_singletask_metric(y_true, y_pred, w)


		def test_pearsonr():
		"""Test the Pearson correlation coefficient is correct."""
		metric = dc.metrics.Metric(dc.metrics.pearsonr)
		r = metric.compute_metric(
		np.array([1.0, 2.0, 3.0]), np.array([2.0, 3.0, 4.0]))
		np.testing.assert_almost_equal(1.0, r)
		r = metric.compute_metric(
		np.array([1.0, 2.0, 3.0]), np.array([-2.0, -3.0, -4.0]))
		np.testing.assert_almost_equal(-1.0, r)
		r = metric.compute_metric(
		np.array([1.0, 2.0, 3.0, 4.0]), np.array([1.0, 2.0, 2.0, 1.0]))
		np.testing.assert_almost_equal(0.0, r)


		def test_r2_score():
		"""Test that R^2 metric passes basic sanity tests"""
		np.random.seed(123)