Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Stuck in SVD-EM #28

Open
aakrosh opened this issue Jul 24, 2023 · 1 comment
Open

Stuck in SVD-EM #28

aakrosh opened this issue Jul 24, 2023 · 1 comment
Assignees
Labels
bug Something isn't working sumo run issue concerns "run" mode

Comments

@aakrosh
Copy link
Contributor

aakrosh commented Jul 24, 2023

In some cases with missing assays with samples, it seems that SUMO get stuck in the svdEM routine. Here is the output from one such run of sumo run

#Loading file: cnprofile.tsv.gz 
Number of dropped rows/features: 0
Number of dropped columns/samples: 0
Data shape: (6756, 344)
#Loading file: gene_expression.tsv.gz 
Number of dropped rows/features: 0
Number of dropped columns/samples: 0
Data shape: (2712, 300)
#Loading file: methylation.tsv.gz 
Number of dropped rows/features: 18
Number of dropped columns/samples: 0
Data shape: (5361, 337)
#Setting all 'missing' parameters to 0.1
#Total number of unique samples: 345
#Layer: 0
Feature matrix: (345 samples x 6756 features)
Adjacency matrix (345, 345) created [similarity method: euclidean]
Adjacency matrix plot saved to expr_meth_cn/adjacency_0.png
#Layer: 1
Feature matrix: (345 samples x 2712 features)
Adjacency matrix (345, 345) created [similarity method: euclidean]
Adjacency matrix plot saved to expr_meth_cn/adjacency_1.png
#Layer: 2
Feature matrix: (345 samples x 5361 features)
Adjacency matrix (345, 345) created [similarity method: euclidean]
Adjacency matrix plot saved to expr_meth_cn/adjacency_2.png
#Output file expr_meth_cn/prepare.npz created
Directory 'expr_meth_cn/results' already exist and will be overwritten.
Number of clusters ('k'): [2]
Sparsity values ('eta'): [0.1]
#Loading file: expr_meth_cn/prepare.npz
#Number of found graph layers: 3
#Loading sample labels file: labels.tsv
#Found 2 unique sample labels for 90/345 samples
#Number of samples randomly removed in each run: 17 out of 345
#SVD-EM for imputation of missing values in average adjacency matrix
- Initialized 8 missing sample-sample similarities (~0%)
 - Iteration(0):	Relative change in trace norm: 1.4764676362149612e+18
 - Iteration(1):	Relative change in trace norm: 0.0

SUMO gets stuck here and does not seem to proceed.

@sienkie
Copy link
Collaborator

sienkie commented Jul 31, 2023

Hi @aakrosh, would you be able to share the input .npz file you are using for sumo run? It looks like a stop condition issue, I will take a look.

@sienkie sienkie self-assigned this Jul 31, 2023
@sienkie sienkie added bug Something isn't working sumo run issue concerns "run" mode labels Jul 31, 2023
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working sumo run issue concerns "run" mode
Projects
None yet
Development

No branches or pull requests

2 participants