I made a submission today and around half an hour later I received an e-mail saying Hello ysngrmz, Your submission (syn22300172) is invalid, below are the invalid reasons: Your model did not generate a predictions.csv file. The log files of my submission are available here: https://www.synapse.org/#!Synapse:syn22300224 The docker container I submitted is docker.synapse.org/syn22283510/covid_46 I checked the log files but could not figure out the main reason. It is definitely not the time quota. There are some lines like STDERR: 2020-08-12T12:17:03.336124854Z 4756792863d8 2020-08-12 12:17:03,335 MainThread INFO toil.leader: Issued job 'https://raw.githubusercontent.com/Sage-Bionetworks/ChallengeWorkflowTemplates/v2.7/upload_to_synapse.cwl' python3 kind-https_raw.githubusercontent.com_Sage-Bionetworks_ChallengeWorkflowTemplates_v2.7_upload_to_synapse.cwl/instance-c1y8dur9 with job batch system ID: 25 and cores: 1, disk: 1.0 G, and memory: 100.0 M My code uses Python 2.7 and does not use any threading. Can there be a problem with memory limits, disk limits or Python version? When I tested the docker container locally it was able to generate a predictions.csv file without giving any errors. I would appreciate if you can help me to find the reason. Thanks.

Created by Yasin Görmez ysngrmz
I have a similar problem in my submission. The log files are here https://www.synapse.org/#!Synapse:syn22302701 The submission ID is 9706724. There are no errors on synthetic data but probably it gave error on real data. I will appreciate if you can help me on this one. p.s. I didn't get any e-mail from Timothy about why the submission is invalid yet.
Thank you so much. This information was useful to fix the problem and I was able to resubmit without any errors.
Hi @ysngrmz , So there was an error thrown in the inference stage. ```sys:1: DtypeWarning: Columns (5) have mixed types. Specify dtype option on import or set low_memory=False. Traceback (most recent call last): File "/app/infer.py", line 463, in log_likelihoods_gaussian = compute_log_likelihoods_gaussian(X_continuous_selected_2_i, means, variances) File "/app/infer.py", line 32, in compute_log_likelihoods_gaussian log_likelihood += math.log(norm.pdf(X_i[j], means[i][j], math.sqrt(variances[i][j]))) ValueError: math domain error ``` It's hard to tell from the log file whether this broke the final output because more logs were generated after this error. Hope this helps! @trberg

Model is invalidated after being accepted page is loading…