Barry G Becker

I am seeing the same problem. Execution failed for task ':minifyCss'. > java.lang.StackOverflowError

Is this a feature or a bug? If it is a feature, it should be documented. What is the default density if it is not specified?

@maxpumperla, thanks for the quickfix, but it didn't seem to work. I pulled in the fix to my local repo, and did `$ pip install dlgo` but still get a...

Thanks. Running `python code/setup.py develop` allowed it to get further. Now the error is the following. ``` running develop running egg_info creating dlgo.egg-info writing dlgo.egg-info\PKG-INFO writing dependency_links to dlgo.egg-info\dependency_links.txt writing...

I switched from gomill to sgfmill and get > Installed c:\users\barry\appdata\local\programs\python\python37\lib\site-packages\sgfmill-1.1.1-py3.7.egg > error: keras-preprocessing 1.0.2 is installed but keras-preprocessing>=1.0.5 is required by {'tensorflow'} > You mentioned that the keras version...

Thanks. I'm not quite ready to look at it at the moment but hope to get to it eventually.

We have noticed a small memory leak in the SparkSubmit task also. We are using spark 1.6.2 and the latest build of job-server (0.7-SNHAPSHOT). We used to run spark-submit (i.e....

Thanks for the tip. I followed the instructions [here|https://github.com/spark-jobserver/spark-jobserver/blob/master/doc/contexts.md] to figure out how to do it. I modified my local config to have ``` spark { contexts { sql-context {...

I've tried it both ways now. Same gradual memory leak in both cases.

Some more info. When running spart 1.6.2, we often seem messages like 16/10/10 10:31:50 ERROR Executor: Managed memory leak detected; size = 6034784 bytes, TID = 213 16/10/10 10:31:50 ERROR...