}

void TF_OperationAllInputs(TF_Operation* oper, TF_Output* inputs,
                           int max_inputs) {
  for (auto* edge : oper->node.in_edges()) {
    if (edge->dst_input() >= 0 && edge->dst_input() < max_inputs) {
      inputs[edge->dst_input()] = {ToOperation(edge->src()),
                                   edge->src_output()};
    }
  }
}

int TF_OperationOutputNumConsumers(TF_Output oper_out) {

import com.google.common.math.IntMath;
import com.google.errorprone.annotations.CanIgnoreReturnValue;
import java.io.ByteArrayInputStream;
import java.io.ByteArrayOutputStream;
import java.io.DataInput;
import java.io.DataInputStream;
import java.io.DataOutput;
import java.io.DataOutputStream;
import java.io.EOFException;
import java.io.FilterInputStream;
import java.io.IOException;
import java.io.InputStream;

`tf.split(value, num_or_size_splits, axis)`. * `tf.sparse_split` now takes
arguments in reversed order and with different keywords. In particular we now
match NumPy order as `tf.sparse_split(sp_input, num_split, axis)`. NOTE: we have
temporarily made `tf.sparse_split` require keyword arguments. * `tf.concat` now
takes arguments in reversed order and with different keywords. In particular we