parameter tuning and bug fix

    def __init__(self, comp_hiddens, comp_activation,

            est_hiddens, est_activation, est_dropout_ratio=0.5,

            minibatch_size=1024, epoch_size=100,

            learning_rate=0.0001, lambda1=0.1, lambda2=0.005,

            learning_rate=0.0001, lambda1=0.1, lambda2=0.0001,

            normalize=True, random_seed=123):

        """

        Parameters

            # Calculate a cholesky decomposition of covariance in advance

            n_features = z.shape[1]

            min_vals = tf.diag(tf.ones(n_features, dtype=tf.float32)) * 1e-3

            min_vals = tf.diag(tf.ones(n_features, dtype=tf.float32)) * 1e-6

            self.L = tf.cholesky(sigma + min_vals[None,:,:])

        self.training = False

            log_det_sigma = 2.0 * tf.reduce_sum(tf.log(tf.matrix_diag_part(self.L)), axis=1)

            # To calculate energies, use "log-sum-exp" (different from orginal paper)

            d = z.get_shape().as_list()[1]

            logits = tf.log(self.phi[:,None]) - 0.5 * (tf.reduce_sum(tf.square(v), axis=1)

                + tf.log(2.0 * np.pi) + log_det_sigma[:,None])

                + d * tf.log(2.0 * np.pi) + log_det_sigma[:,None])

            energies = - tf.reduce_logsumexp(logits, axis=0)

        return energies