[jvm-packages] update rabit, surface new changes to spark, add parity and failure tests (#4876)
* Expose sets of rabit configurations to spark layer
This commit is contained in:
@@ -18,7 +18,7 @@ CONFIG = {
|
||||
"USE_HDFS": "OFF",
|
||||
"USE_AZURE": "OFF",
|
||||
"USE_S3": "OFF",
|
||||
|
||||
"RABIT_MOCK": "OFF",
|
||||
"USE_CUDA": "OFF",
|
||||
"JVM_BINDINGS": "ON"
|
||||
}
|
||||
@@ -68,6 +68,7 @@ def normpath(path):
|
||||
|
||||
|
||||
if __name__ == "__main__":
|
||||
CONFIG["RABIT_MOCK"] = str(sys.argv[1])
|
||||
if sys.platform == "darwin":
|
||||
# Enable of your compiler supports OpenMP.
|
||||
CONFIG["USE_OPENMP"] = "OFF"
|
||||
|
||||
Reference in New Issue
Block a user