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

[CLI] segfaults when training #6845

Closed
903019003 opened this issue Feb 28, 2025 · 5 comments
Closed

[CLI] segfaults when training #6845

903019003 opened this issue Feb 28, 2025 · 5 comments
Labels

Comments

@903019003
Copy link

903019003 commented Feb 28, 2025

Image

Description

segfaults when training,the log shows that the data has been loaded,but segfaults
[LightGBM] [Warning] eval_at is set with ndcg_eval_at=1,3,5,10, will be overridden by map_eval_at=1,3,5,10. Current value: eval_at=1,3,5,10
[LightGBM] [Info] Finished loading parameters
[LightGBM] [Info] Using column number 0 as label
[LightGBM] [Info] Calculating query boundaries...
[LightGBM] [Info] Construct bin mappers from text data time 52.62 seconds
[LightGBM] [Info] Making second pass...
[LightGBM] [Info] Total groups: 433515, total data: 11835033
[LightGBM] [Info] Calculating query boundaries...
[LightGBM] [Info] Total groups: 51561, total data: 1410272
[LightGBM] [Info] Finished loading data in 1718.841160 seconds

train.sh: line 4:2374791 $egmentation fault
(core dumped) lightgbm config=train land.conf output model=./models/rank stage2 fuse
Igb ${datadate}.model>./logs/${datadate} land.log

Reproducible example

Environment info

LightGBM version or commit hash:

Command(s) you used to install LightGBM

Additional Comments

@jameslamb
Copy link
Collaborator

Thanks for using LightGBM.

It's not possible for us to help you based only on an error message and a small snippet of logs. If you'd like some help, please provide some of the details that were asked for in the template that came up when you clicked "new issue".

  • version of LightGBM
  • commands you used to install it
  • operating system

And are you able to provide a minimal, reproducible example? (Stack Overflow docs explaining what that is)

Or at least, can you please provide the full set of parameters you used and the exact command you used to invoke the CLI?

@903019003
Copy link
Author

Thanks for using LightGBM.

It's not possible for us to help you based only on an error message and a small snippet of logs. If you'd like some help, please provide some of the details that were asked for in the template that came up when you clicked "new issue".

  • version of LightGBM
  • commands you used to install it
  • operating system

And are you able to provide a minimal, reproducible example? (Stack Overflow docs explaining what that is)

Or at least, can you please provide the full set of parameters you used and the exact command you used to invoke the CLI?

version of LightGBM:
4.3.0.99
operating system:
CentOS Linux release 8.5.2111
Linux version 4.18.0-348.el8.x86_64

para:
task = train
boosting_type = gbdt
objective = lambdarank
metric = ndcg,map
ndcg_eval_at = 1,3,5,10
map_eval_at = 1,3,5,10
metric_freq = 1
is_training_metric = true
label_column = 0
max_bin = 255
data = train_data/stage2.train
valid_data = /train_data/stage2.valid
early_stopping = 100
num_trees = 1000
feature_fraction = 1.0
bagging_fraction = 0.8
learning_rate = 0.1
max_depth = 5
num_leaves = 32
two_round=true
num_threads=100
verbosity = 1

@jameslamb
Copy link
Collaborator

version of LightGBM:
4.3.0.99

That is not a release... it means you're building from some commit between v4.3.0 and v4.4.0. What commit are you building from?

And is it possible to upgrade to the latest release (v4.6.0)?

@jameslamb
Copy link
Collaborator

Please also provide this:

commands you used to install it

@903019003
Copy link
Author

version of LightGBM:
4.3.0.99

That is not a release... it means you're building from some commit between v4.3.0 and v4.4.0. What commit are you building from?

And is it possible to upgrade to the latest release (v4.6.0)?

Thanks for your help :) !! I found that the problem is caused by the memory being full

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

No branches or pull requests

2 participants