[tf]中只有if else没有单独的if语句，所以要这么包装起来

# Define some large matrices
a = ...
b = ...
c = ...

pred = tf.placeholder(tf.bool)

def if_true():
  return tf.matmul(a, b)

def if_false():
  return tf.matmul(b, c)

# Will be `tf.cond()` in the next release.
from tensorflow.python.ops import control_flow_ops

result = tf.cond(pred, if_true, if_false)

sess = tf.Session()
sess.run(result, feed_dict={pred: True})   # ==> executes only (a x b)
sess.run(result, feed_dict={pred: False})  # ==> executes only (b x c)

def is_false():
    ori_loss = tf.cond(
                    tf.equal(mode,1),
                    lambda:tf.nn.sparse_softmax_cross_entropy_with_logits(labels=tf.argmax(action_one,1),logits=action_one),
                    lambda:tf.nn.sparse_softmax_cross_entropy_with_logits(labels=tf.argmax(action_two,1),logits=action_two),
                )
    gene_loss = reward_score * ori_loss
    train_gene_op = optimize(gene_loss, self.global_step)
    return gene_loss,train_gene_op

return tf.cond(tf.equal(mode,0),lambda:tf.zeros(shape=[]),tf.zeros(shape=[]),is_false)