diff --git a/fluid/ocr_recognition/.run.sh b/fluid/ocr_recognition/.run.sh
new file mode 100644
index 0000000000000000000000000000000000000000..8bb5b9cca9e995d68fb18532c35ab8bdc494ceda
--- /dev/null
+++ b/fluid/ocr_recognition/.run.sh
@@ -0,0 +1 @@
+python ctc_train.py --batch_size=128 --total_step=10000 --use_gpu=True
diff --git a/fluid/ocr_recognition/.run_ce.sh b/fluid/ocr_recognition/.run_ce.sh
new file mode 100644
index 0000000000000000000000000000000000000000..17d36384b02eec0345eb529a31f65fd43326f5df
--- /dev/null
+++ b/fluid/ocr_recognition/.run_ce.sh
@@ -0,0 +1 @@
+python ctc_train.py --batch_size=128 --total_step=10000 --use_gpu=True | python _ce.py
diff --git a/fluid/ocr_recognition/ctc_train.py b/fluid/ocr_recognition/ctc_train.py
index dde07e51887ab6d7724f9b8893ae49479ee7b9a7..dc68cc2e2e9f6e98d2331ff926109d5df56d1df6 100644
--- a/fluid/ocr_recognition/ctc_train.py
+++ b/fluid/ocr_recognition/ctc_train.py
@@ -98,6 +98,10 @@ def train(args, data_reader=ctc_reader):
         print "\nTime: %s; Iter[%d]; Test seq error: %s.\n" % (
             time.time(), iter_num, str(test_seq_error[0]))
 
+        #Note: The following logs are special for CE monitoring.
+        #Other situations do not need to care about these logs.
+        print("kpis	test_seq_error	%f" % test_seq_error[0])
+
     def save_model(args, exe, iter_num):
         filename = "model_%05d" % iter_num
         fluid.io.save_params(