class ExternalCheckpointManager extends AnyRef
This class contains the methods that are required for managing the state of the training process. The training state is stored in a distributed file system, that consists of UBJ (Universal Binary JSON) model files. The class provides methods for saving, loading and cleaning up checkpoints.
- Alphabetic
- By Inheritance
- ExternalCheckpointManager
- AnyRef
- Any
- Hide All
- Show All
- Public
- All
Instance Constructors
- new ExternalCheckpointManager(checkpointPath: String, fs: FileSystem)
Value Members
-
final
def
!=(arg0: Any): Boolean
- Definition Classes
- AnyRef → Any
-
final
def
##(): Int
- Definition Classes
- AnyRef → Any
-
final
def
==(arg0: Any): Boolean
- Definition Classes
- AnyRef → Any
-
final
def
asInstanceOf[T0]: T0
- Definition Classes
- Any
-
def
cleanPath(): Unit
This method cleans all the directories and files that are present in the checkpoint path.
This method cleans all the directories and files that are present in the checkpoint path.
- Exceptions thrown
IOExceptionexception that is thrown when there is an error deleting the checkpoint path.
-
def
cleanUpHigherVersions(currentRound: Int): Unit
This method cleans up all the checkpoint versions that are higher than the current round.
This method cleans up all the checkpoint versions that are higher than the current round. This is useful when multiple training instances are running and we want to make sure that only the checkpoints from the current training instance are retained.
- currentRound
The current round of training.
- Exceptions thrown
IOExceptionAny exception that occurs when deleting the checkpoint files.
-
def
clone(): AnyRef
- Attributes
- protected[lang]
- Definition Classes
- AnyRef
- Annotations
- @throws( ... ) @native()
-
final
def
eq(arg0: AnyRef): Boolean
- Definition Classes
- AnyRef
-
def
equals(arg0: Any): Boolean
- Definition Classes
- AnyRef → Any
-
def
finalize(): Unit
- Attributes
- protected[lang]
- Definition Classes
- AnyRef
- Annotations
- @throws( classOf[java.lang.Throwable] )
-
def
getCheckpointRounds(firstRound: Int, checkpointInterval: Int, numOfRounds: Int): List[Integer]
Get a list of iterations that need checkpointing.
Get a list of iterations that need checkpointing.
- firstRound
The first round of training.
- checkpointInterval
The interval at which checkpoints are to be saved.
- numOfRounds
The number of rounds to be trained.
- returns
A list of integer rounds that need checkpointing.
- Exceptions thrown
IOExceptionAny exception that occurs when getting the list of rounds.
-
final
def
getClass(): Class[_]
- Definition Classes
- AnyRef → Any
- Annotations
- @native()
-
def
hashCode(): Int
- Definition Classes
- AnyRef → Any
- Annotations
- @native()
-
final
def
isInstanceOf[T0]: Boolean
- Definition Classes
- Any
-
def
loadCheckpointAsBooster(): Booster
Read the checkpoint from the checkpoint path.
Read the checkpoint from the checkpoint path. Once the checkpoint path is read, we get the latest version of the checkpoint from all the checkpoint versions and lead it into the booster for the purpose of making predictions.
- returns
The booster object that is used for making predictions.
- Exceptions thrown
IOExceptionAny expection that occurs when reading the checkpoint path.XGBoostErrorAny exception that occurs when loading the model into the booster.
-
final
def
ne(arg0: AnyRef): Boolean
- Definition Classes
- AnyRef
-
final
def
notify(): Unit
- Definition Classes
- AnyRef
- Annotations
- @native()
-
final
def
notifyAll(): Unit
- Definition Classes
- AnyRef
- Annotations
- @native()
-
final
def
synchronized[T0](arg0: ⇒ T0): T0
- Definition Classes
- AnyRef
-
def
toString(): String
- Definition Classes
- AnyRef → Any
-
def
updateCheckpoint(boosterToCheckpoint: Booster): Unit
This method updates the booster checkpoint to the the latest or current version and deleted all the previous versions of the checkpoint.
This method updates the booster checkpoint to the the latest or current version and deleted all the previous versions of the checkpoint.
- boosterToCheckpoint
The booster object that is to be checkpointed and saved as a model file.
- Exceptions thrown
IOExceptionAny exception that occurs when writing the model file to the checkpoint path.XGBoostErrorAny exception that occurs when saving the model from the booster.
-
final
def
wait(): Unit
- Definition Classes
- AnyRef
- Annotations
- @throws( ... )
-
final
def
wait(arg0: Long, arg1: Int): Unit
- Definition Classes
- AnyRef
- Annotations
- @throws( ... )
-
final
def
wait(arg0: Long): Unit
- Definition Classes
- AnyRef
- Annotations
- @throws( ... ) @native()