Module org.elasticsearch.xcore

Package org.elasticsearch.xpack.core.ml

Class MachineLearningField

java.lang.Object

org.elasticsearch.xpack.core.ml.MachineLearningField

public final class MachineLearningField
extends Object

Field Summary

Fields

Modifier and Type	Field	Description
static final Setting<Boolean>	AUTODETECT_PROCESS
static final Setting<Integer>	MAX_LAZY_ML_NODES
static final Setting<ByteSizeValue>	MAX_MODEL_MEMORY_LIMIT
static final MlConfigVersion	MIN_SUPPORTED_SNAPSHOT_VERSION
static final LicensedFeature.Momentary	ML_API_FEATURE
static final String	ML_FEATURE_FAMILY
static final Setting<Boolean>	MODEL_GRAPH_VALIDATION_ENABLED	When set to false, the pytorch_inference process skips TorchScript model graph validation (the operation allowlist/forbidden list check).
static final Setting<List<String>>	MODEL_PLATFORM_ARCHITECTURES	When set, overrides automatic ML node platform detection for built-in model variant selection (e.g.
static final TimeValue	STATE_PERSIST_RESTORE_TIMEOUT
static final Setting<Boolean>	USE_AUTO_MACHINE_MEMORY_PERCENT	This boolean value indicates if `max_machine_memory_percent` should be ignored and an automatic calculation is used instead.

Method Summary

Modifier and Type	Method	Description
static String	valuesToId(String... values)

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Details

AUTODETECT_PROCESS

public static final Setting<Boolean> AUTODETECT_PROCESS

MAX_MODEL_MEMORY_LIMIT

public static final Setting<ByteSizeValue> MAX_MODEL_MEMORY_LIMIT

MAX_LAZY_ML_NODES

public static final Setting<Integer> MAX_LAZY_ML_NODES

USE_AUTO_MACHINE_MEMORY_PERCENT

public static final Setting<Boolean> USE_AUTO_MACHINE_MEMORY_PERCENT

This boolean value indicates if `max_machine_memory_percent` should be ignored and an automatic calculation is used instead. This calculation takes into account total node size and the size of the JVM on that node. If the calculation fails, we fall back to `max_machine_memory_percent`.

MODEL_PLATFORM_ARCHITECTURES

public static final Setting<List<String>> MODEL_PLATFORM_ARCHITECTURES

When set, overrides automatic ML node platform detection for built-in model variant selection (e.g. ELSER, E5). The configured list of architectures is used in place of querying ML node OS info. When empty (default), architectures are auto-detected from running ML nodes. Example values: ["linux-x86_64"], ["linux-aarch64"].

MODEL_GRAPH_VALIDATION_ENABLED

public static final Setting<Boolean> MODEL_GRAPH_VALIDATION_ENABLED

When set to false, the pytorch_inference process skips TorchScript model graph validation (the operation allowlist/forbidden list check). This is an emergency escape hatch — disabling validation removes the security check that blocks models with dangerous operations.

STATE_PERSIST_RESTORE_TIMEOUT

public static final TimeValue STATE_PERSIST_RESTORE_TIMEOUT

ML_FEATURE_FAMILY

public static final String ML_FEATURE_FAMILY

See Also:

• Constant Field Values

ML_API_FEATURE

public static final LicensedFeature.Momentary ML_API_FEATURE

MIN_SUPPORTED_SNAPSHOT_VERSION

public static final MlConfigVersion MIN_SUPPORTED_SNAPSHOT_VERSION

Method Details

valuesToId

public static String valuesToId(String... values)