diff --git a/paddlehub/finetune/evaluate.py b/paddlehub/finetune/evaluate.py
index 448a127e173ae4b70e0a865133716f6b05d32656..758a0b665bb4341b06afdddc0f94bdf1a1c267af 100644
--- a/paddlehub/finetune/evaluate.py
+++ b/paddlehub/finetune/evaluate.py
@@ -20,6 +20,8 @@ import time
 
 import paddle.fluid as fluid
 import paddlehub as hub
+import numpy as np
+
 from paddlehub.common.logger import logger
 
 
diff --git a/paddlehub/finetune/finetune.py b/paddlehub/finetune/finetune.py
index 114f6d468c974d252ea24724f2c959ad5176185c..367cbf903ce06096e14810bfe6042bd3b45d8b7d 100644
--- a/paddlehub/finetune/finetune.py
+++ b/paddlehub/finetune/finetune.py
@@ -124,13 +124,13 @@ def _finetune_seq_label_task(task,
                         exe=exe)
 
                 if do_eval and global_step % config.eval_interval == 0:
-                    evaluate_seq_label_task(
+                    evaluate_seq_labeling_task(
                         task,
                         data_reader,
                         feed_list,
                         phase="test",
                         config=config)
-                    evaluate_seq_label_task(
+                    evaluate_seq_labeling_task(
                         task,
                         data_reader,
                         feed_list,
@@ -146,9 +146,9 @@ def _finetune_seq_label_task(task,
             exe=exe)
 
         if do_eval:
-            evaluate_seq_label_task(
+            evaluate_seq_labeling_task(
                 task, data_reader, feed_list, phase="dev", config=config)
-            evaluate_seq_label_task(
+            evaluate_seq_labeling_task(
                 task, data_reader, feed_list, phase="test", config=config)
         logger.info("PaddleHub finetune finished.")