Xiaoguang Sun 2ae56ca84f Use int32_t explicitly when serializing version (#2389) ...

Use int32_t explicitly when serializing version field of dmatrix in binary
format. On ILP64 architectures, although very little, size of int is 64 bits.

2017-06-07 10:03:42 -07:00

..

data.cc

Use int32_t explicitly when serializing version (#2389)

2017-06-07 10:03:42 -07:00

simple_csr_source.cc

Disable invalid check for completely sparse batch that results in failed assertion for issue #1827 (#2213)

2017-04-21 09:28:02 -07:00

simple_csr_source.h

[IO] Enable external memory

2016-01-16 10:24:01 -08:00

simple_dmatrix.cc

Changing omp_get_num_threads to omp_get_max_threads (#1831)

2016-12-04 11:26:45 -08:00

simple_dmatrix.h

Use bst_float consistently throughout (#1824)

2016-11-30 10:02:10 -08:00

sparse_batch_page.h

Fix compilation on OS X with GCC 7 (#2256)

2017-05-19 22:04:07 -07:00

sparse_page_dmatrix.cc

[UPDATE] Update rabit and threadlocal (#2114)

2017-03-16 18:48:37 -07:00

sparse_page_dmatrix.h

Update sparse_page_dmatrix.h (#2139)

2017-03-23 11:01:40 -07:00

sparse_page_raw_format.cc

[UPDATE] Update rabit and threadlocal (#2114)

2017-03-16 18:48:37 -07:00

sparse_page_source.cc

[UPDATE] Update rabit and threadlocal (#2114)

2017-03-16 18:48:37 -07:00

sparse_page_source.h

Add dump_format=json option (#1726)

2016-11-04 09:55:25 -07:00

sparse_page_writer.cc

[UPDATE] Update rabit and threadlocal (#2114)

2017-03-16 18:48:37 -07:00