"negative dimensions" error

[V-JJ]

Good evening!

I'm running into this error from time to time, depending on the input matrix => ValueError: negative dimensions are not allowed
I don't understand where this error comes from.

Any help would be appreciated,
Many thanks!

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-20-775a9b82f10c> in <module>()
     18                 out_centrom_fig = "{}/{}/centromere_figures/{}.png".format(resolution_value, k_scf, os.path.basename(submatrix_path))
     19 
---> 20                 CentromesOnSubamtrix(submatrix_path, resolution_value, centromere_record_file, out_centrom_fig)
     21 

<ipython-input-18-5b5b26e1311b> in CentromesOnSubamtrix(i_submatrix_file, i_resolution, o_file, o_fig)
     63     # Call centromeres
     64     try:
---> 65         centromeres = centromeres_calls.centromeres_calls(counts, lengths, resolution=i_resolution)
     66         print(centromeres)
     67         # Figure

/media/vadim/DATA/Local4T/Programs/miniconda3/envs/centurion_py27/lib/python2.7/site-packages/centurion/centromeres_calls.pyc in centromeres_calls(counts, lengths, resolution, init, n_candidate, sigma, verbose, filter_candidates, normalize)
     21     if init is None:
     22         counts_40kb, lengths_40kb = utils.downsample_resolution(
---> 23             counts, lengths, coefficient=coef)
     24 
     25         if normalize:

/media/vadim/DATA/Local4T/Programs/miniconda3/envs/centurion_py27/lib/python2.7/site-packages/centurion/utils.pyc in downsample_resolution(counts, lengths, coefficient)
     26     target_lengths = np.ceil(lengths.astype(float) / coefficient).astype(int)
     27     target_counts = np.zeros((target_lengths.sum(),
---> 28                               target_lengths.sum()))
     29     begin_i, end_i = 0, 0
     30     target_begin_i, target_end_i = 0, 0

ValueError: negative dimensions are not allowed