[Breaking] Switch from rabit to the collective communicator (#8257)

* Switch from rabit to the collective communicator

* fix size_t specialization

* really fix size_t

* try again

* add include

* more include

* fix lint errors

* remove rabit includes

* fix pylint error

* return dict from communicator context

* fix communicator shutdown

* fix dask test

* reset communicator mocklist

* fix distributed tests

* do not save device communicator

* fix jvm gpu tests

* add python test for federated communicator

* Update gputreeshap submodule

Co-authored-by: Hyunsu Philip Cho <chohyu01@cs.washington.edu>
This commit is contained in:
Rong Ou
2022-10-05 15:39:01 -07:00
committed by GitHub
parent e47b3a3da3
commit 668b8a0ea4
79 changed files with 805 additions and 2212 deletions

View File

@@ -37,7 +37,6 @@ class SketchContainer {
private:
Monitor timer_;
std::shared_ptr<dh::AllReducer> reducer_;
HostDeviceVector<FeatureType> feature_types_;
bst_row_t num_rows_;
bst_feature_t num_columns_;
@@ -93,15 +92,12 @@ class SketchContainer {
* \param num_columns Total number of columns in dataset.
* \param num_rows Total number of rows in known dataset (typically the rows in current worker).
* \param device GPU ID.
* \param reducer Optional initialised reducer. Useful for speeding up testing.
*/
SketchContainer(HostDeviceVector<FeatureType> const &feature_types,
int32_t max_bin, bst_feature_t num_columns,
bst_row_t num_rows, int32_t device,
std::shared_ptr<dh::AllReducer> reducer = nullptr)
bst_row_t num_rows, int32_t device)
: num_rows_{num_rows},
num_columns_{num_columns}, num_bins_{max_bin}, device_{device},
reducer_(std::move(reducer)) {
num_columns_{num_columns}, num_bins_{max_bin}, device_{device} {
CHECK_GE(device, 0);
// Initialize Sketches for this dmatrix
this->columns_ptr_.SetDevice(device_);