[GitHub] [incubator-tvm] leandron edited a comment on issue #4953: [CI][Docker] xgboost 1.0.1 causes segfault on test_autotvm_xgboost_model.py

[GitBox <gi...@apache.org>]

leandron edited a comment on issue #4953: [CI][Docker] xgboost 1.0.1 causes segfault on test_autotvm_xgboost_model.py
URL: https://github.com/apache/incubator-tvm/issues/4953#issuecomment-608599550
 
 
   Hi, I only managed to investigate it further, today. XGBoost now is version `1.0.2`, and I can still reproduce this issue.
   
   To give some context, this is the function call that triggers the issue:
   https://github.com/apache/incubator-tvm/blob/54975a3fd24fa45b815be39075f4614e53009444/python/tvm/autotvm/tuner/xgboost_cost_model.py#L262-L272
   
   I tried just creating pickle files of some inputs (`self.xgb_params` and `dtrain`) and simplifying the function call, but this is not enough to reproduce the issue. The issue seems to be in the context  `custom_callback`, below:
   
   https://github.com/apache/incubator-tvm/blob/54975a3fd24fa45b815be39075f4614e53009444/python/tvm/autotvm/tuner/xgboost_cost_model.py#L419-L526
   
   Now, something that could help me a bit to narrow down where the problem is if I run XGBoost in debug mode. @hcho3 what is the simplest way I can do that?
   

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
users@infra.apache.org


With regards,
Apache Git Services