根据提示，在右侧编辑器补充代码，完成OvR类中的fit(self, train_datas, train_labels)与predict(self, test_datas)函数。PS:请使用代码中所提供的tiny_logistic_regression作为二分类分类器。

好的，了解。以下是代码：

from tiny_logistic_regression import BinaryLogisticRegression

class OvR:
    def __init__(self):
        self.classifiers = {}
    def fit(self, train_datas, train_labels):
        """
        Parameters:
            train_datas: list of list, shape = (n_samples, n_features)
                Training data.
            train_labels: list of int, shape = (n_samples,)
                Target classes.
        """
        classes = list(set(train_labels))
        for c in classes:
            binary_labels = [1 if label==c else 0 for label in train_labels]
            self.classifiers[c] = BinaryLogisticRegression()
            self.classifiers[c].fit(train_datas, binary_labels)

    def predict(self, test_datas):
        """
        Parameters:
            test_datas: list of list, shape = (n_samples, n_features)
                Testing data.
        Returns:
            list of int, shape = (n_samples,)
                Predicted classes.
        """
        predictions = []
        for data in test_datas:
            max_prob = 0
            max_class = None
            for c, clf in self.classifiers.items():
                prob = clf.predict_proba([data])[0][1]
                if prob > max_prob:
                    max_prob = prob
                    max_class = c
            predictions.append(max_class)
        return predictions

注意：本模型只能用于二分类问题的多类别分类。如有更复杂需求，请使用更适合的模型。