metrics.py

import tensorflow as tf


slim = tf.contrib.slim

def concordance_cc2(prediction, ground_truth):
    """Defines concordance metric for model evaluation. 

    Args:
       prediction: prediction of the model.
       ground_truth: ground truth values.
    Returns:
       The concordance value.
    """
    
    names_to_values, names_to_updates = slim.metrics.aggregate_metric_map({
        'eval/mean_pred':slim.metrics.streaming_mean(prediction),
        'eval/mean_lab':slim.metrics.streaming_mean(ground_truth),
        'eval/cov_pred':slim.metrics.streaming_covariance(prediction, prediction),
        'eval/cov_lab':slim.metrics.streaming_covariance(ground_truth, ground_truth),
        'eval/cov_lab_pred':slim.metrics.streaming_covariance(prediction, ground_truth)
    })

    metrics = dict()
    for name, value in names_to_values.items():
      metrics[name] = value

    mean_pred = metrics['eval/mean_pred']
    var_pred = metrics['eval/cov_pred']
    mean_lab = metrics['eval/mean_lab']
    var_lab = metrics['eval/cov_lab']
    var_lab_pred = metrics['eval/cov_lab_pred']

    denominator = (var_pred + var_lab + (mean_pred - mean_lab) ** 2)

    concordance_cc2 = (2 * var_lab_pred) / denominator

    return concordance_cc2, names_to_values, names_to_updates