Assistance Needed with Code Execution

[chujun225]

Hi!

I'm making a concerted effort to replicate your impressive workflow, but I've encountered some difficulties that I can't resolve on my own.

I ran the following code:

python
Anchor = 'CD4+ T cells'
ntrain_per_class = 12
K = 50
k = 10
labels = ['1', '2', 'rand1', 'rand2']
classes = [0, 1, 2, 2]
path = '/'.join(os.path.abspath(os.getcwd()).split('/')[:-1]) + '/CRCs'
nrun = 50
nset_thr = 0.9
background = True
ms.run_scima(Anchor, ntrain_per_class, K, k, nset_thr, labels, classes, path, nrun, background)

However, I keep encountering the following error:

---

ValueError Traceback (most recent call last)
Cell In[102], line 16
14 nset_thr = 0.9
15 background = True
---> 16 ms.run_scima(Anchor, ntrain_per_class, K, k, nset_thr, labels, classes, path, nrun, background)

File ~/S3-CIMA/S3-CIMA-main/S3CIMA/modules_scima.py:246, in run_scima(Anchor, ntrain_per_class, K, k, nset_thr, labels, classes, path, nrun, background)
244 for i in range(0, int(len(labels)/2)):
245 lab = str(labels[i])
--> 246 train_idx[lab] = list(np.random.choice(group[lab], size=ntrain_per_class, replace=False))
247 test_idx[lab] = [j for j in group[lab] if j not in train_idx[lab]]
248 rlab = str(labels[i + int(len(labels)/2)])

File mtrand.pyx:934, in numpy.random.mtrand.RandomState.choice()

ValueError: 'a' cannot be empty unless no samples are taken

The code in S3CIMA_example.ipynb runs successfully up until the line ms.run_scima(Anchor, ntrain_per_class, K, k, nset_thr, labels, classes, path, nrun, background).

I've tried various approaches to fix this issue, but unfortunately, I haven't been able to resolve it. I would greatly appreciate any assistance or guidance you could provide.

Thank you very much!

Feel free to adjust any specific details as needed!