An error of cnmf combine

[Lyu-Gaoyuan]

Hi,
Thank you for your wonderful work! I currently encountered the following error when running system('cnmf combine --output-dir tset_2 --name T1'):

Combining factorizations for k=7.
Combining factorizations for k=8.
Combining factorizations for k=9.
Combining factorizations for k=10.
Combining factorizations for k=11.
Traceback (most recent call last):
  File "/home/lgy/miniconda3/envs/cnmf/bin/cnmf", line 33, in <module>
    sys.exit(load_entry_point('cNMF', 'console_scripts', 'cnmf')())
  File "/home/lgy/miniconda3/envs/cnmf/lib/python3.7/site-packages/cnmf-master/cNMF/cnmf.py", line 1656, in main
    args.func(args)
  File "/home/lgy/miniconda3/envs/cnmf/lib/python3.7/site-packages/cnmf-master/cNMF/cnmf.py", line 1305, in combine
    cnmf_obj.combine_nmf(k)
  File "/home/lgy/miniconda3/envs/cnmf/lib/python3.7/site-packages/cnmf-master/cNMF/cnmf.py", line 793, in combine_nmf
    combined_spectra = combined_spectra.reshape(-1, combined_spectra.shape[-1])
AttributeError: 'NoneType' object has no attribute 'reshape'

my complete code are as follow:
use_condaenv(condaenv = "cnmf", required = T)
py_config()

python:         /home/lgy/miniconda3/envs/cnmf/bin/python
libpython:      /home/lgy/miniconda3/envs/cnmf/lib/libpython3.7m.so
pythonhome:     /home/lgy/miniconda3/envs/cnmf:/home/lgy/miniconda3/envs/cnmf
version:        3.7.12 | packaged by conda-forge | (default, Oct 26 2021, 06:08:21)  [GCC 9.4.0]
numpy:          /home/lgy/miniconda3/envs/cnmf/lib/python3.7/site-packages/numpy
numpy_version:  1.21.6
NOTE: Python version was forced by use_python() function

py_module_available('cnmf')
TRUE
system("cnmf prepare --output-dir tset_2 --name T1 count_data/T1.count.txt -k 3 4 5 6 7 8 9 10 --n-iter 200")

Reading in counts from count_data/T1.count.txt - 177 cells and 21337 genes
Removing 3022 genes with zero counts; final shape (177, 18315)
Selecting 2000 highly variable genes
Warning: 8 cells have zero counts of overdispersed genes - ignoring                 these cells for factorization.
/home/lgy/miniconda3/envs/cnmf/lib/python3.7/site-packages/cnmf-master/cNMF/cnmf.py:1186: FutureWarning: X.dtype being converted to np.float32 from int64. In the next version of anndata (0.9) conversion will not be automatic. Pass dtype explicitly to avoid this warning. Pass `AnnData(X, dtype=X.dtype, ...)` to get the future behavour.
  var=pd.DataFrame(index=input_counts.columns),

system("cnmf factorize --output-dir tset_2 --name T1 --worker-index 0")

.........
.........
[Worker 0]. Starting task 1596.
[Worker 0]. Starting task 1597.
[Worker 0]. Starting task 1598.
[Worker 0]. Starting task 1599.

system("cnmf combine --output-dir tset_2 --name T1")

Combining factorizations for k=7.
Combining factorizations for k=8.
Combining factorizations for k=9.
Combining factorizations for k=10.
Combining factorizations for k=11.
Traceback (most recent call last):
  File "/home/lgy/miniconda3/envs/cnmf/bin/cnmf", line 33, in <module>
    sys.exit(load_entry_point('cNMF', 'console_scripts', 'cnmf')())
  File "/home/lgy/miniconda3/envs/cnmf/lib/python3.7/site-packages/cnmf-master/cNMF/cnmf.py", line 1656, in main
    args.func(args)
  File "/home/lgy/miniconda3/envs/cnmf/lib/python3.7/site-packages/cnmf-master/cNMF/cnmf.py", line 1305, in combine
    cnmf_obj.combine_nmf(k)
  File "/home/lgy/miniconda3/envs/cnmf/lib/python3.7/site-packages/cnmf-master/cNMF/cnmf.py", line 793, in combine_nmf
    combined_spectra = combined_spectra.reshape(-1, combined_spectra.shape[-1])
AttributeError: 'NoneType' object has no attribute 'reshape'

Could you tell me how to fix this bug.
Looking forward to your reply.
Think you

[xyzheng123]

Hi @Lyu-Gaoyuan, have you been able to resolve the issue?