remove redundant code

fluid/sequence_tagging_for_ner/train.py

@@ -78,8 +78,6 @@ def main(train_data_file, test_data_file, vocab_file, target_file, emb_file,
     embedding_param = fluid.global_scope().find_var(embedding_name).get_tensor()
     embedding_param.set(word_vector_values, place)
 
-    print fluid.default_main_program()
-
     batch_id = 0
     for pass_id in xrange(num_passes):
         chunk_evaluator.reset(exe)
@@ -89,21 +87,20 @@ def main(train_data_file, test_data_file, vocab_file, target_file, emb_file,
                 feed=feeder.feed(data),
                 fetch_list=[avg_cost] + chunk_evaluator.metrics)
             if batch_id % 5 == 0:
-                print(
-                    "Pass " + str(pass_id) + ", Batch " + str(batch_id) +
-                    ", Cost " + str(cost[0]) + ", Precision " +
-                    str(batch_precision[0]) + ", Recall " + str(batch_recall[0])
+                print("Pass " + str(pass_id) + ", Batch " + str(
+                    batch_id) + ", Cost " + str(cost[0]) + ", Precision " + str(
+                        batch_precision[0]) + ", Recall " + str(batch_recall[0])
                       + ", F1_score" + str(batch_f1_score[0]))
             batch_id = batch_id + 1
 
         pass_precision, pass_recall, pass_f1_score = chunk_evaluator.eval(exe)
-        print("[TrainSet] pass_id:" + str(pass_id) + " pass_precision:" +
-              str(pass_precision) + " pass_recall:" + str(pass_recall) +
+        print("[TrainSet] pass_id:" + str(pass_id) + " pass_precision:" + str(
+            pass_precision) + " pass_recall:" + str(pass_recall) +
               " pass_f1_score:" + str(pass_f1_score))
         pass_precision, pass_recall, pass_f1_score = test(
             exe, chunk_evaluator, inference_program, test_reader, place)
-        print("[TestSet] pass_id:" + str(pass_id) + " pass_precision:" +
-              str(pass_precision) + " pass_recall:" + str(pass_recall) +
+        print("[TestSet] pass_id:" + str(pass_id) + " pass_precision:" + str(
+            pass_precision) + " pass_recall:" + str(pass_recall) +
               " pass_f1_score:" + str(pass_f1_score))
 
         save_dirname = os.path.join(model_save_dir, "params_pass_%d" % pass_id)