diff --git a/fluid/ocr/ctc_train.py b/fluid/ocr/ctc_train.py
index 66949d527ab4b439ea055fe6a1e726e0281aaeda..2e7cf5d53058093b8aa64d51f4421562621198e8 100644
--- a/fluid/ocr/ctc_train.py
+++ b/fluid/ocr/ctc_train.py
@@ -24,7 +24,7 @@ from utility import add_arguments, print_arguments
 parser = argparse.ArgumentParser(description=__doc__)
 add_arg = functools.partial(add_arguments, argparser=parser)
 # yapf: disable
-add_arg('batch_size',     int,   2,     "Minibatch size.")
+add_arg('batch_size',     int,   16,     "Minibatch size.")
 add_arg('pass_num',       int,   16,     "# of training epochs.")
 add_arg('learning_rate',  float, 1.0e-3, "Learning rate.")
 add_arg('l2',             float, 0.0005, "L2 regularizer.")
@@ -121,7 +121,7 @@ def train(args, data_reader=dummy_reader):
     avg_cost = fluid.layers.mean(x=cost)
     optimizer = fluid.optimizer.Momentum(
         learning_rate=args.learning_rate, momentum=args.momentum)
-    opts = optimizer.minimize(cost)
+    optimizer.minimize(avg_cost)
     # decoder and evaluator
     decoded_out = fluid.layers.ctc_greedy_decoder(
         input=fc_out, blank=num_classes)