Use this crossentropy loss function when there are two or more label
classes. We expect labels to be provided in a one_hot representation. If
you want to provide labels as integers, please use
SparseCategoricalCrossentropy loss. There should be num_classes floating
point values per feature, i.e., the shape of both y_pred and y_true are
[batch_size, num_classes].
loss_categorical_crossentropy(
y_true,
y_pred,
from_logits = FALSE,
label_smoothing = 0,
axis = -1L,
...,
reduction = "sum_over_batch_size",
name = "categorical_crossentropy",
dtype = NULL
)Categorical crossentropy loss value.
Tensor of one-hot true targets.
Tensor of predicted targets.
Whether y_pred is expected to be a logits tensor. By
default, we assume that y_pred encodes a probability distribution.
Float in [0, 1]. When > 0, label values are smoothed,
meaning the confidence on label values are relaxed. For example, if
0.1, use 0.1 / num_classes for non-target labels and
0.9 + 0.1 / num_classes for target labels.
The axis along which to compute crossentropy (the features
axis). Defaults to -1.
For forward/backward compatability.
Type of reduction to apply to the loss. In almost all cases
this should be "sum_over_batch_size". Supported options are
"sum", "sum_over_batch_size", "mean",
"mean_with_sample_weight" or NULL. "sum" sums the loss,
"sum_over_batch_size" and "mean" sum the loss and divide by the
sample size, and "mean_with_sample_weight" sums the loss and
divides by the sum of the sample weights. "none" and NULL
perform no aggregation. Defaults to "sum_over_batch_size".
Optional name for the loss instance.
The dtype of the loss's computations. Defaults to NULL, which
means using config_floatx(). config_floatx() is a
"float32" unless set to different value
(via config_set_floatx()). If a keras$DTypePolicy is
provided, then the compute_dtype will be utilized.
y_true <- rbind(c(0, 1, 0), c(0, 0, 1))
y_pred <- rbind(c(0.05, 0.95, 0), c(0.1, 0.8, 0.1))
loss <- loss_categorical_crossentropy(y_true, y_pred)
loss
## tf.Tensor([0.05129329 2.30258509], shape=(2), dtype=float64)Standalone usage:
y_true <- rbind(c(0, 1, 0), c(0, 0, 1))
y_pred <- rbind(c(0.05, 0.95, 0), c(0.1, 0.8, 0.1))
# Using 'auto'/'sum_over_batch_size' reduction type.
cce <- loss_categorical_crossentropy()
cce(y_true, y_pred)
## tf.Tensor(1.1769392, shape=(), dtype=float32)# Calling with 'sample_weight'.
cce(y_true, y_pred, sample_weight = op_array(c(0.3, 0.7)))
## tf.Tensor(0.8135988, shape=(), dtype=float32)# Using 'sum' reduction type.
cce <- loss_categorical_crossentropy(reduction = "sum")
cce(y_true, y_pred)
## tf.Tensor(2.3538785, shape=(), dtype=float32)# Using 'none' reduction type.
cce <- loss_categorical_crossentropy(reduction = NULL)
cce(y_true, y_pred)
## tf.Tensor([0.05129331 2.3025851 ], shape=(2), dtype=float32)Usage with the compile() API:
model %>% compile(optimizer = 'sgd',
loss=loss_categorical_crossentropy())
Other losses:
Loss()
loss_binary_crossentropy()
loss_binary_focal_crossentropy()
loss_categorical_focal_crossentropy()
loss_categorical_hinge()
loss_circle()
loss_cosine_similarity()
loss_ctc()
loss_dice()
loss_hinge()
loss_huber()
loss_kl_divergence()
loss_log_cosh()
loss_mean_absolute_error()
loss_mean_absolute_percentage_error()
loss_mean_squared_error()
loss_mean_squared_logarithmic_error()
loss_poisson()
loss_sparse_categorical_crossentropy()
loss_squared_hinge()
loss_tversky()
metric_binary_crossentropy()
metric_binary_focal_crossentropy()
metric_categorical_crossentropy()
metric_categorical_focal_crossentropy()
metric_categorical_hinge()
metric_hinge()
metric_huber()
metric_kl_divergence()
metric_log_cosh()
metric_mean_absolute_error()
metric_mean_absolute_percentage_error()
metric_mean_squared_error()
metric_mean_squared_logarithmic_error()
metric_poisson()
metric_sparse_categorical_crossentropy()
metric_squared_hinge()