Skip to content
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

segmentation fault when using dart with continue training 2 #1867

Closed
gwuillou opened this issue Nov 22, 2018 · 6 comments
Closed

segmentation fault when using dart with continue training 2 #1867

gwuillou opened this issue Nov 22, 2018 · 6 comments

Comments

@gwuillou
Copy link

gwuillou commented Nov 22, 2018

Hello,
My problem is related to #984. I have the last version and I am using mac os.
I have also the problem of segmentation fault when I am trying to use dart with continue training:
I am doing:

        gbm = lgb.train(params,
                        lgb_train,
                        num_boost_round=10,
                        valid_sets=lgb_train,
                        feature_name=header)
        y_val = gbm.predict(testset.x)
        y_train = gbm.predict(trainingset.x)
        ...
        nb_tree_by_it = 100
        nb_it = 100
        for jj in range(1, nb_it):
            lgb_train.init_score = None
            gbm = lgb.train(params,
                            lgb_train,
                            num_boost_round=nb_tree_by_it,
                            valid_sets=lgb_train,
                            feature_name=header,
                            verbose_eval=100,
                            init_model=gbm
                            )
             ...

What I am doing wrong? I also tried examples/python-guide/advanced_example.py, I get also a segmentation fault.

@gwuillou gwuillou changed the title segmentation false when using dart with continue training 2 segmentation fault when using dart with continue training 2 Nov 22, 2018
@guolinke
Copy link
Collaborator

Can you try the latest version ?
It will be better if you provide a reproduce example with randomly generated data.

@guolinke
Copy link
Collaborator

@gwuillou
we have tests for the "advanced_example", so it should not fail.

@gwuillou
Copy link
Author

Hello,

I have the version lightgbm 2.2.2.
When I try the "advanced_example", the Segmentation fault: 11 appears randomly. sometimes I have also this error:
python(14274,0x7fff8985f380) malloc: *** error for object 0x3fd3b3c07d68: pointer being freed was not allocated
*** set a breakpoint in malloc_error_break to debug

...

I am using mac osx high sierra,

Thank you,

@guolinke
Copy link
Collaborator

I guess the problem is related to binaries release.
@gwuillou can you try to install from source ?
pip install --no-binary :all:

may related to: #1743

@StrikerRUS
Copy link
Collaborator

ping @gwuillou

@StrikerRUS
Copy link
Collaborator

Closing due to inactivity.
Feel free to reopen if the error persists after installing from sources.

@lock lock bot locked as resolved and limited conversation to collaborators Mar 11, 2020
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Projects
None yet
Development

No branches or pull requests

3 participants