search

lhe17 / nebula

GNU General Public License v2.0
28 stars 6 forks source link

task 1 failed - "object 'cv2' not found" #21

Open h1hui opened 1 year ago

h1hui commented 1 year ago

Hello,

First of all, thank you for this great tool. However I ran into an error like this:

`> df = model.matrix(~type+well+batch+age, data=predictors)

head(df) (Intercept) typeCT well batch2 batch1 age AACCTTTTCTGAGTCA-1 1 0 1 0 1 78 AACGTCAAGCCTGAAG-1 1 0 1 0 1 78 AAGAACAGTATAATGG-1 1 0 1 0 1 78 AAGCGTTCAACCCTAA-1 1 0 1 0 1 78 AATGAAGAGATAGGGA-1 1 0 1 0 1 78 AATGAAGAGCGAGGAG-1 1 0 1 0 1 78 re = nebula(count,sub_sample$sample_id,pred=df,ncore=2) Remove 3463 genes having low expression. Analyzing 12978 genes with 30 subjects and 6524 cells. Error in { : task 1 failed - "object 'cv2' not found"`

I did found out that with different predictors, the error may or may not be there. I am wondering why? How can I fix it for good?

Thanks!

lhe17 commented 1 year ago

Dear Ashley,

Thank you for your questions, and sorry for the error. The error you reported here is new and had not been seen before, so I suspect that this bug might be introduced in Version 1.4.0 in which the parallel computing is used. To check whether the bug is related to parallel computing, could you please run your analysis again with ncore=1 or with the previous Version 1.3.0 and see whether the error disappears. Also, it would be very helpful for me to fix this bug if it is possible for you to share with me your design matrix 'df' and 'sub_sample$sample_id'. The count matrix is not needed.

Best regards, Liang

On Tue, May 2, 2023 at 11:49 PM Ashley Hui @.***> wrote:

Hello,

First of all, thank you for this great tool. However I ran into an error like this:

`> df = model.matrix(~type+well+batch+age, data=predictors)

head(df) (Intercept) typeCT well batch2 batch1 age AACCTTTTCTGAGTCA-1 1 0 1 0 1 78 AACGTCAAGCCTGAAG-1 1 0 1 0 1 78 AAGAACAGTATAATGG-1 1 0 1 0 1 78 AAGCGTTCAACCCTAA-1 1 0 1 0 1 78 AATGAAGAGATAGGGA-1 1 0 1 0 1 78 AATGAAGAGCGAGGAG-1 1 0 1 0 1 78 re = nebula(count,sub_sample$sample_id,pred=df,ncore=2) Remove 3463 genes having low expression. Analyzing 12978 genes with 30 subjects and 6524 cells. Error in { : task 1 failed - "object 'cv2' not found"`

I did found out that with different predictors, the error may or may not be there. I am wondering why? How can I fix it for good?

Thanks!

— Reply to this email directly, view it on GitHub https://github.com/lhe17/nebula/issues/21, or unsubscribe https://github.com/notifications/unsubscribe-auth/AGDISUU3ANCOHCHF3BF5U3LXEF6NFANCNFSM6AAAAAAXTTOESI . You are receiving this because you are subscribed to this thread.Message ID: @.***>

h1hui commented 1 year ago

Hello Liang,

In fact setting ncore=1 fixed the issue. I am attaching both files here to help you fix the bug.

Thanks, Ashley df.txt sid.txt

lhe17 commented 1 year ago

Thank you very much for sharing your files, Ashley.

I replicated this error, which is indeed related to parallel computing. The error has now been fixed in Version 1.4.1.

Best regards,

Liang

On 5/3/2023 10:37 PM, Ashley Hui wrote:

Hello Liang,

In fact setting ncore=1 fixed the issue. I am attaching both files here to help you fix the bug.

Thanks, Ashley df.txt https://github.com/lhe17/nebula/files/11391044/df.txt sid.txt https://github.com/lhe17/nebula/files/11391046/sid.txt

— Reply to this email directly, view it on GitHub https://github.com/lhe17/nebula/issues/21#issuecomment-1533707535, or unsubscribe https://github.com/notifications/unsubscribe-auth/AGDISUTQTP7BJZY4WZ7TUHDXEK62BANCNFSM6AAAAAAXTTOESI. You are receiving this because you commented.Message ID: @.***>