Package ai.vespa.embedding.config
Class GgufEmbedderConfig
java.lang.Object
com.yahoo.config.Node
com.yahoo.config.InnerNode
com.yahoo.config.ConfigInstance
ai.vespa.embedding.config.GgufEmbedderConfig
public final class GgufEmbedderConfig
extends com.yahoo.config.ConfigInstance
This class represents the root node of gguf-embedder
-
Nested Class Summary
Nested ClassesModifier and TypeClassDescriptionstatic final classstatic final classThis class represents gguf-embedder.poolingType Set pooling type for embeddings (default: UNSPECIFIED, UNSPECIFIED = model default).static interface -
Field Summary
Fields -
Constructor Summary
Constructors -
Method Summary
Methods inherited from class com.yahoo.config.ConfigInstance
getDefName, getDefNamespace, serialize, serializeMethods inherited from class com.yahoo.config.InnerNode
equals, getAllDescendantLeafNodes, getChildren, getChildrenWithVectorsFlattened, hashCode, postInitialize, toStringMethods inherited from class com.yahoo.config.Node
clone
-
Field Details
-
CONFIG_DEF_MD5
- See Also:
-
CONFIG_DEF_NAME
- See Also:
-
CONFIG_DEF_NAMESPACE
- See Also:
-
CONFIG_DEF_SCHEMA
-
-
Constructor Details
-
GgufEmbedderConfig
-
-
Method Details
-
getDefMd5
-
getDefName
-
getDefNamespace
-
embeddingModel
- Returns:
- gguf-embedder.embeddingModel
-
embeddingModelReference
public com.yahoo.config.ModelReference embeddingModelReference()- Returns:
- gguf-embedder.embeddingModel ModelReference
-
gpuLayers
public int gpuLayers()- Returns:
- gguf-embedder.gpuLayers
-
poolingType
- Returns:
- gguf-embedder.poolingType
-
physicalMaxBatchSize
public int physicalMaxBatchSize()- Returns:
- gguf-embedder.physicalMaxBatchSize
-
logicalMaxBatchSize
public int logicalMaxBatchSize()- Returns:
- gguf-embedder.logicalMaxBatchSize
-
continuousBatching
public boolean continuousBatching()- Returns:
- gguf-embedder.continuousBatching
-
contextSize
public int contextSize()- Returns:
- gguf-embedder.contextSize
-
maxPromptTokens
public int maxPromptTokens()- Returns:
- gguf-embedder.maxPromptTokens
-
seed
public int seed()- Returns:
- gguf-embedder.seed
-
parallel
public int parallel()- Returns:
- gguf-embedder.parallel
-
threads
public double threads()- Returns:
- gguf-embedder.threads
-
batchThreads
public double batchThreads()- Returns:
- gguf-embedder.batchThreads
-
prependQuery
- Returns:
- gguf-embedder.prependQuery
-
prependDocument
- Returns:
- gguf-embedder.prependDocument
-
normalize
public boolean normalize()- Returns:
- gguf-embedder.normalize
-