Function for weighted categorical cross entropy

ported from this implementation:

weighted_categorical_crossentropy(y_true, y_pred, weights)

Arguments

y_true	True labels (Tensor)
y_pred	Predictions (Tensor of the same shape as y_true)
weights	weights for each class

Value

function value

Details

https://gist.github.com/wassname/ce364fddfc8a025bfab4348cf5de852d

Author

Tustison NJ

Examples

library( ANTsRNet )
library( keras )

model <- createUnetModel2D( c( 64, 64, 1 ), numberOfOutputs = 2 )
#> Error in py_discover_config(required_module, use_environment): Python specified in RETICULATE_PYTHON (/Users/ntustison/anaconda3/envs/antsx/bin/python3) does not exist

model %>% compile( loss = weighted_categorical_crossentropy( weights = c( 1, 1 ) ),
 optimizer = optimizer_adam( lr = 0.0001 ),
   metrics = "accuracy" )
#> Error in compile(., loss = weighted_categorical_crossentropy(weights = c(1,     1)), optimizer = optimizer_adam(lr = 1e-04), metrics = "accuracy"): object 'model' not found