Skip to content

Instantly share code, notes, and snippets.

@sriramch
Created May 30, 2019 22:53
Show Gist options
  • Select an option

  • Save sriramch/2174171d3bf8395b71222212fcb9594b to your computer and use it in GitHub Desktop.

Select an option

Save sriramch/2174171d3bf8395b71222212fcb9594b to your computer and use it in GitHub Desktop.
4478 cli regression
the dataset is ~ 80gb
external memory is turned on
i do not think the cache files are created on the backing store. it is trying to make the entire dataset resident
sample config:
booster = gbtree
objective = binary:logistic
verbosity = 3
#single_precision_histogram = true
#tree_method = gpu_hist
#n_gpus = -1
#gpu_batch_nrows = -1
tree_method = hist
#tree_method = approx
nthread = 32
# Tree Booster Parameters
# step size shrinkage
eta = 1.0
# minimum loss reduction required to make a further partition
gamma = 1.0
# minimum sum of instance weight(hessian) needed in a child
min_child_weight = 1
# maximum depth of a tree
max_depth = 8
# Task Parameters
# the number of round to do boosting
num_round = 3
# 0 means do not save any model except the final round model
save_period = 0
# The path of training data
data = "train_file.txt#trainbincache"
#data = "train_file.txt"
eval_train = 1
stack while it was running fwiw:
0x00469c55: void __gnu_cxx::new_allocator<xgboost::Entry>::construct<xgboost::Entry, unsigned int&, float&>(xgboost::Entry*, unsigned int&, float&) + 0x49 (7ffd4f68e290, 7ffd4f68e28c, 7f7e05d27630, f5cf4d0, 2841b52f4, 7ffd4f68e290)
0x00468aca: _ZNSt16allocator_traitsISaIN7xgboost5EntryEEE9constructIS1_IRjRfEEEvRS2_PT_DpOT0_ + 0x4a (7f892db40010, 7ffd4f68e290, 7ffd4f68e28c, f5cf4d0, 7ffd4f68e2a0, 7f8a841b4310)
0x00467587: void std::vector<xgboost::Entry, std::allocator<xgboost::Entry> >::emplace_back<unsigned int&, float&>(unsigned int&, float&) + 0x5f (f5ce8a0, f5cf150, 7ffd4f68e2b0, d00000000, 7f8943b40000, 0) + 210
0x0046bd6a: xgboost::data::SimpleCSRSource::CopyFrom(dmlc::Parser<unsigned int, float>*) + 0x436 (7ffd4f68e4e0, 2000000, 7ffd4f68e610, f5ce8a0, f5cf150, 467040) + 50
0x00463d62: xgboost::DMatrix::Create(dmlc::Parser<unsigned int, float>*, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, unsigned long) + 0x98 (2000000, 7ffd4f68fcb0, 7f004f68e500, 7ffd4f68ff00, 20302e302d0e8bd0, 100000000) + c60
0x00463591: xgboost::DMatrix::Load(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, bool, bool, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, unsigned long) + 0x987 (0, 7ffd4f68fef0, 0, 0, 0, 0) + c40
0x0040ef15: xgboost::CLITrain(xgboost::CLIParam const&) + 0x14e (7ffd4f6905d8, 2004208d6, 7ffd4f690030, 20f5c5680, f5cc4a0, f5cc860) + 630
0x004111cd: xgboost::CLIRunTask(int, char**) + 0x273 (7ffd4f6905d8, 200000000)
0x00411331: main + 0x20 (11c00, 7ffd4f6905d8, 24f6905f0, 411311, 0, d948194818624ec0) + ffff8002b11f2b10
0x7f8b2ba3f830: _fini + 0x7f8afef96d5c (e26e2d8d48550e6e, 8949f68949530e6e, 8ec8348e5294cd5, b87357e803fdc148, db312074ed8548ff, 841f0f) + 41ff894155b926c1
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment