Added attention layer function

Updated test cases to test basic useage

Author: atroiano

R/layers_attention.R

@@ -0,0 +1,23 @@
+
+#' Applies Dropout to the input.
+#' 
+#' Dropout consists in randomly setting a fraction `rate` of input units to 0 at
+#' each update during training time, which helps prevent overfitting.
+#' 
+#' @inheritParams layer_dense
+#' @param inputs a list of inputs first should be the query tensor, the second the value tensor
+#' @param use_scale If True, will create a scalar variable to scale the attention scores.
+#' @param causal Boolean. Set to True for decoder self-attention. Adds a mask such that position i cannot attend to positions j > i. 
+#' This prevents the flow of information from the future towards the past.
+#'
+#' @family core layers
+#' @family attention layers   
+#'         
+#' @export
+layer_attention <- function(inputs, use_scale=FALSE, causal = FALSE) {
+  
+  create_layer(tf$keras$layers$Attention,object = inputs,args = list(use_scale = use_scale, causal = causal) )
+}
+
+
+

tests/testthat/test-layers.R

@@ -498,6 +498,12 @@ test_call_succeeds("layer_separable_conv_1d", required_version = "2.1.3", {
   }
 })
 
+test_call_succeeds('layer_attention',{
+input_1 = layer_input(shape=c(4,5))
+input_2 = layer_input(shape=c(4,5))
+layer_attention(c(input_1,input_2))
+})
+
 test_call_succeeds("layer_dense_features", required_version = "2.1.3", {
   if (is_tensorflow_implementation() && tensorflow::tf_version() >= "1.14") {