* Performs image binarization with corresponding threshold * (depends on the method)value, which creates a binary image from a grayscale. * @param i

* Construct a tensor by repeating it the number of times given by reps. * * This operation creates a new tensor by replicating `input` `reps` * tim

* Time one model inference with parallel compilation feature, based on the * current backend. * * The inference time contains the time spent by bot

* Collect all elements of this dataset into an array. * * Obviously this will succeed only for small datasets that fit in memory. * Useful fo

* Collect all remaining elements of a bounded stream into an array. * Obviously this will succeed only for small streams that fit in memory. * U

* Collect all elements of this dataset into an array with prefetching 100 * elements. This is useful for testing, because the prefetch changes the

* Collect all elements of this dataset into an array with prefetching 100 * elements. This is useful for testing, because the prefetch changes the

* A `tf.layers.embedding` instance for embedding token ids.

* Convert an integer id to a string token.

* The tokenizer used to tokenize strings.