merlin: Specified dimension not compatible with data

Hi, I am implementing a base-model for DNN based TTS system using merlin frontend. Currently I’m using 50 wav files from arctic database and I have 280 binary features. However, I’m unable to proceed with output feature composition with the following error-

2016-09-17 13:32:25,526 INFO      acoustic_comp: processing file    1 of   50 : /home/neo/work/speech_iiitd/dnn_test2/data/nn_mgc_lf0_vuv_bap_199/arctic_a0001.cmp
Traceback (most recent call last):
  File "../dnn/dnn_tts/run_dnn.py", line 970, in <module>
    main_function(cfg)
  File "../dnn/dnn_tts/run_dnn.py", line 635, in main_function
    acoustic_worker.prepare_nn_data(in_file_list_dict, nn_cmp_file_list, cfg.in_dimension_dict, cfg.out_dimension_dict)
  File "/home/neo/work/speech_iiitd/dnn/dnn_tts/frontend/acoustic_base.py", line 122, in prepare_nn_data
    self.prepare_data(in_file_list_dict, out_file_list, in_dimension_dict, out_dimension_dict)
  File "/home/neo/work/speech_iiitd/dnn/dnn_tts/frontend/acoustic_composition.py", line 126, in prepare_data
    features, frame_number = io_funcs.load_binary_file_frame(in_file_name, in_feature_dim)
  File "/home/neo/work/speech_iiitd/dnn/dnn_tts/io_funcs/binary_io.py", line 67, in load_binary_file_frame
    assert features.size % float(dimension) == 0.0,'specified dimension %s not compatible with data'%(dimension)
AssertionError: specified dimension 5 not compatible with data

I’m new to dnn based speech synthesis. Kindly give some suggestions. Also, why do we normalize the features with [0,1]. Thanks.

My conf file-

[Labels]
question_file_name:path/to/data/resources/questions_dnn_unilex-rpx_quinphone.hed 

[Outputs]
mgc    : 60
dmgc   : 180
bap    : 5  
dbap   : 15
lf0    : 1  
dlf0   : 3  

[Data]
train_file_number: 30
valid_file_number: 10
test_file_number : 10
buffer_size: 200000

[Processes]
NORMLAB  : False
MAKECMP  : True
NORMCMP  : True
TRAINDNN : False
DNNGEN   : False
GENWAV   : False
CALMCD   : False