tensorflow

TensorFlow bindings for OCaml
IN THIS PACKAGE
type float32_tensor = ( float, Bigarray.float32_elt ) Tensorflow_core.Tensor.t
type t = {
train_images : float32_tensor;
train_labels : float32_tensor;
test_images : float32_tensor;
test_labels : float32_tensor;
}
val read_files : ?train_image_file:string -> ?train_label_file:string -> ?test_image_file:string -> ?test_label_file:string -> unit -> t
val train_batch : t -> batch_size:int -> batch_idx:int -> float32_tensor * float32_tensor
val image_dim : int
val label_count : int
val batch_accuracy : ?samples:int -> t -> [ `train | `test ] -> batch_size:int -> predict:( float32_tensor -> float32_tensor ) -> float

batch_accuracy ?samples t ~batch_size ~predict computes the accuracy of the predict function on test images using batches of size at most batch_size. The average is computed on samples images.