search

rmarkello / snfpy

Similarity network fusion in Python
https://snfpy.readthedocs.io
GNU Lesser General Public License v3.0
75 stars 19 forks source link

value error #18

Open QUEST2179 opened 4 years ago

QUEST2179 commented 4 years ago

Dear developer,

I tried a real dataset GBM from original paper but got the following error. Could you please help? Thanks!

mirna = pd.read_csv('GLIO_Mirna_Expression.txt', index_col=0, sep='\t') exp = pd.read_csv('GLIO_Gene_Expression.txt', index_col=0, sep='\t') methy = pd.read_csv('GLIO_Methy_Expression.txt', index_col=0, sep='\t') affinity_networks = snf.make_affinity([mirna.T, exp.T, methy.T], metric='euclidean', K=20, mu=0.5) fused_network = snf.snf(affinity_networks, K=20)

Traceback (most recent call last): File "testSNF.py", line 27, in fused_network = snf.snf(affinity_networks, K=20) File "C:\Users\Miniconda3\lib\site-packages\snf\compute.py", line 392, in snf aff = _check_SNF_inputs(aff) File "C:\Users\Miniconda3\lib\site-packages\snf\compute.py", line 446, in _check_SNF_inputs ac = check_array(a, force_all_finite=True, copy=True) File "C:\Users\Miniconda3\lib\site-packages\sklearn\utils\validation.py", line 73, in inner_f return f(**kwargs) File "C:\Users\Miniconda3\lib\site-packages\sklearn\utils\validation.py", line 646, in check_array allow_nan=force_all_finite == 'allow-nan') File "C:\Users\Miniconda3\lib\site-packages\sklearn\utils\validation.py", line 100, in _assert_all_finite msg_dtype if msg_dtype is not None else X.dtype) ValueError: Input contains NaN, infinity or a value too large for dtype('float64').

lillepeder commented 3 years ago

The fault is in the dataset. One patient (ID=TCGA-12-0780-01A...) has only missing values for every modality. I dropped this patient from the dataframes and it runs just fine.