def pearson_correlation_metric_fn(
y_true: tf.Tensor,
y_pred: tf.Tensor,
) -> tf.Tensor:
"""
Pearson correlation metric function.
https://github.com/WenYanger/Keras_Metrics
Args:
y_true (tf.Tensor): y_true
y_pred (tf.Tensor): y_pred
Returns:
tf.contrib.metrics: pearson correlation
"""
x = y_true
y = y_pred
mx = K.mean(x, axis=0)
my = K.mean(y, axis=0)
xm, ym = x - mx, y - my
r_num = K.sum(xm * ym)
x_square_sum = K.sum(xm * xm)
y_square_sum = K.sum(ym * ym)
r_den = K.sqrt(x_square_sum * y_square_sum) + 1e-12
r = r_num / r_den
return K.mean(r)
def vae_loss(self, x, z_decoded):
x = K.flatten(x)
z_decoded = K.flatten(z_decoded)
# Reconstruction loss (as we used sigmoid activation we can use binarycrossentropy)
recon_loss = keras.metrics.binary_crossentropy(x, z_decoded)
# KL divergence
kl_loss = -5e-4 * K.mean(1 + z_sigma - K.square(z_mu) - K.exp(z_sigma), axis=-1)
return K.mean(recon_loss + kl_loss)
def loss(y_true, y_pred):
PPO_LOSS_CLIPPING = 0.2
PPO_ENTROPY_LOSS = 5 * 1e-3 # Does not converge without entropy penalty
log_pdf_new = get_log_probability_density(y_pred, y_true)
log_pdf_old = get_log_probability_density(old_prediction, y_true)
ratio = K.exp(log_pdf_new - log_pdf_old)
surrogate1 = ratio * advantage
clip_ratio = K.clip(ratio,
min_value=(1 - PPO_LOSS_CLIPPING),
max_value=(1 + PPO_LOSS_CLIPPING))
surrogate2 = clip_ratio * advantage
loss_actor = -K.mean(K.minimum(surrogate1, surrogate2))
sigma = y_pred[:, 2:]
variance = K.square(sigma)
loss_entropy = PPO_ENTROPY_LOSS * K.mean(
-(K.log(2 * np.pi * variance) + 1) / 2)
return loss_actor + loss_entropy
def contrastive_loss(y_true, y_pred):
return K.mean(y_true * K.square(K.maximum(y_pred - m_pos, 0)) +
(1 - y_true) * K.square(K.maximum(m_neg - y_pred, 0)))
def mean_pred(y_true, y_pred):
return -K.mean(y_true * K.log(y_pred + 1.e-7) +
(1 - y_true) * K.log(1 - y_pred + 1.e-7)) * 10
def triplet_loss(_, y_diff):
return K.mean(K.maximum(y_diff + m, 0))
def root_mean_squared_error(y_true, y_pred):
return K.sqrt(K.mean(K.square(y_pred - y_true)))
def __cosine(x, y):
x = K.l2_normalize(x, axis=-1)
y = K.l2_normalize(y, axis=-1)
c = K.mean(x * y, axis=-1)
return -c