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
Fails to Convert Categorical Columns on Big Dataset and Identity Column #3132
Comments
Marek Novotny commented: [~accountid:5f915eee9c31840076f0f317] Thanks for reporting the problem, can you also share schema of your dataset, you can anonymize the column names if needed. |
Juan Campos commented: I’ve attached in a csv file the schema (variables(1).csv), it’s big, with 2995 columns. [^variables(1).csv] The dataset has 18M rows. I’ve detected the problem is in “obfs_bkuuid“, this one is a id , (each row has a different value). if I remove this column from sql before asH2OFrame , it works. |
Marek Novotny commented: I think it will be related to the changes i did for [https://h2oai.atlassian.net/jira/software/c/projects/SW/issues/SW-2449|https://h2oai.atlassian.net/jira/software/c/projects/SW/issues/SW-2449?filter=reportedbyme]. This functionality is covered by tests, but it’s not all internal edge cases are covered. |
JIRA Issue Migration Info Jira Issue: SW-2470 Linked PRs from JIRA Attachments From Jira Attachment Name: exception.txt Attachment Name: variables(1).csv |
JIRA Issue Migration Info Cont'd Jira Issue Created Date: 2020-10-22T05:42:21.243-0700 |
When we try to upload a big table (18M rows) using
train_age<- hc$asH2OFrame(wide,"wide") fails.. Throws the following exception:
Error: ai.h2o.sparkling.backend.exceptions.RestApiCommunicationException: H2O node 10.100.90.29:54321 responded with
Status code: 500 : Server Error
probably caused by a NullPointeException in ai.h2o.sparkling.extensions.internals.UpdateCategoricalIndicesTask.map(UpdateCategoricalIndicesTask.java:67)"
I have attached a file with more information
The text was updated successfully, but these errors were encountered: