Keras decision threshold for accuracy and recall

Question

I am doing binary classification using Keras(with Tensorflowbackend) and I have an accuracy of 76% and 70%. Now I want to try to play with a decision threshold. As far as I know, it Kerasuses a solution threshold of 0.5. Is there a way to Kerasuse a user threshold for precision decision making and challenge?

Thank you for your time!

Answer 1

create custom metrics as follows:

@Marcin: , threshold_value

def precision_threshold(threshold=0.5):
    def precision(y_true, y_pred):
        """Precision metric.
        Computes the precision over the whole batch using threshold_value.
        """
        threshold_value = threshold
        # Adaptation of the "round()" used before to get the predictions. Clipping to make sure that the predicted raw values are between 0 and 1.
        y_pred = K.cast(K.greater(K.clip(y_pred, 0, 1), threshold_value), K.floatx())
        # Compute the number of true positives. Rounding in prevention to make sure we have an integer.
        true_positives = K.round(K.sum(K.clip(y_true * y_pred, 0, 1)))
        # count the predicted positives
        predicted_positives = K.sum(y_pred)
        # Get the precision ratio
        precision_ratio = true_positives / (predicted_positives + K.epsilon())
        return precision_ratio
    return precision

def recall_threshold(threshold = 0.5):
    def recall(y_true, y_pred):
        """Recall metric.
        Computes the recall over the whole batch using threshold_value.
        """
        threshold_value = threshold
        # Adaptation of the "round()" used before to get the predictions. Clipping to make sure that the predicted raw values are between 0 and 1.
        y_pred = K.cast(K.greater(K.clip(y_pred, 0, 1), threshold_value), K.floatx())
        # Compute the number of true positives. Rounding in prevention to make sure we have an integer.
        true_positives = K.round(K.sum(K.clip(y_true * y_pred, 0, 1)))
        # Compute the number of positive targets.
        possible_positives = K.sum(K.clip(y_true, 0, 1))
        recall_ratio = true_positives / (possible_positives + K.epsilon())
        return recall_ratio
    return recall

model.compile(..., metrics = [precision_threshold(0.1), precision_threshold(0.2),precision_threshold(0.8), recall_threshold(0.2,...)])

, :)