-
Notifications
You must be signed in to change notification settings - Fork 143
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
[QST]How Do I Solve the Problem that Missing Values Cannot Be Converted to Int Values? #1770
Comments
@gukejun1 can you provide more info about your env? how and where did you install merlin libraries? Are you using a docker image? if yes, which docker image? thanks. |
the code is same from this, I use docker images(nvcr.io/nvidia/merlin/merlin-tensorflow 22.12) |
@rnyak this is my full code
I install merlin libraries from the web of https://catalog.ngc.nvidia.com/orgs/nvidia/teams/merlin/containers/merlin-tensorflow |
@rnyak The training data is from the first 40 million rows of day_0 in the criteo data set, and the verification data is from the first 4 million rows of day_1.The following figure shows some parquet data visualization. |
@gukejun1 if you have null values, normally, when you apply the following lines in the NVT workflow the missing/null values should be filled..
can you share a subset of your parquet file like only couple hundreds rows, so that we can reproduce the issue? thanks. |
@rnyak day_1_100.parquet.txt |
@gukejun1 I used your small dataset with this notebook and all worked fine for me. I cannot reproduce your error.. are you able to reproduce your error only with this small parquet file? |
@rnyak Very strange. |
@gukejun1 please note that your screenshot shows that you are trying to read in a |
@rnyak It's the same. The only difference is that the file name extension is in the parquet format. Because GitHub cannot upload files with the parquet file name extension, the file name extension is changed to txt. |
@rnyak So, this code didn't work. |
@rnyak Because my graphics card supports up to cuda 11.3, so I reinstalled cupy-cuda to 113. Is it related to this? Does cupy-cuda 113 support populating missing values? |
@gukejun1 what's your graphic card? your sample set does not have any nulls in the
|
@rnyak The error is still reported. |
@gukejun1 cudf supports can you test if you are able to run the notebooks
thanks. |
@gukejun1 the error looks like because of pandas, and looks like you are running on CPU not on GPU... Please confirm that the
|
@rnyak For the movie_lens case, 01 / 02 is successful. 3、i use docker pull nvcr.io/nvidia/merlin/merlin-tensorflow:22.12 to get the docker images. |
|
When I run the case in , an error is reported.
Why is the error still reported that nan cannot be converted to an int value? The official website handles the problem of missing values.
How to solve this problem?
The text was updated successfully, but these errors were encountered: