Skip to content

Commit 2de0e4a

Browse files
jan-elasticjan-elastic
committed
remove ununsed EmbeddingRequestChunker.EmbeddingType
1 parent 26c8395 commit 2de0e4a

File tree

17 files changed

+26
-187
lines changed

17 files changed

+26
-187
lines changed

‎x-pack/plugin/inference/src/main/java/org/elasticsearch/xpack/inference/chunking/EmbeddingRequestChunker.java‎

Lines changed: 5 additions & 33 deletions
Original file line numberDiff line numberDiff line change
@@ -10,7 +10,6 @@
1010
import org.elasticsearch.ElasticsearchStatusException;
1111
import org.elasticsearch.action.ActionListener;
1212
import org.elasticsearch.common.util.concurrent.AtomicArray;
13-
import org.elasticsearch.index.mapper.vectors.DenseVectorFieldMapper;
1413
import org.elasticsearch.inference.ChunkedInference;
1514
import org.elasticsearch.inference.ChunkingSettings;
1615
import org.elasticsearch.inference.InferenceServiceResults;
@@ -40,20 +39,6 @@
4039
*/
4140
public class EmbeddingRequestChunker {
4241

43-
public enum EmbeddingType {
44-
FLOAT,
45-
BYTE,
46-
SPARSE;
47-
48-
public static EmbeddingType fromDenseVectorElementType(DenseVectorFieldMapper.ElementType elementType) {
49-
return switch (elementType) {
50-
case BYTE -> EmbeddingType.BYTE;
51-
case FLOAT -> EmbeddingType.FLOAT;
52-
case BIT -> throw new IllegalArgumentException("Bit vectors are not supported");
53-
};
54-
}
55-
};
56-
5742
// Visible for testing
5843
record Request(int inputIndex, int chunkIndex, ChunkOffset chunk, List<String> inputs) {
5944
public String chunkText() {
@@ -76,34 +61,21 @@ public record BatchRequestAndListener(BatchRequest batch, ActionListener<Inferen
7661
private final List<List<Request>> requests;
7762
private final List<BatchRequest> batchRequests;
7863
private final AtomicInteger resultCount = new AtomicInteger();
79-
private final EmbeddingType embeddingType;
8064

8165
private final List<AtomicReferenceArray<EmbeddingResults.Embedding<?>>> results;
8266
private final AtomicArray<Exception> errors;
8367
private ActionListener<List<ChunkedInference>> finalListener;
8468

85-
public EmbeddingRequestChunker(List<String> inputs, int maxNumberOfInputsPerBatch, EmbeddingType embeddingType) {
86-
this(inputs, maxNumberOfInputsPerBatch, embeddingType, null);
69+
public EmbeddingRequestChunker(List<String> inputs, int maxNumberOfInputsPerBatch) {
70+
this(inputs, maxNumberOfInputsPerBatch, null);
8771
}
8872

89-
public EmbeddingRequestChunker(
90-
List<String> inputs,
91-
int maxNumberOfInputsPerBatch,
92-
int wordsPerChunk,
93-
int chunkOverlap,
94-
EmbeddingType embeddingType
95-
) {
96-
this(inputs, maxNumberOfInputsPerBatch, embeddingType, new WordBoundaryChunkingSettings(wordsPerChunk, chunkOverlap));
73+
public EmbeddingRequestChunker(List<String> inputs, int maxNumberOfInputsPerBatch, int wordsPerChunk, int chunkOverlap) {
74+
this(inputs, maxNumberOfInputsPerBatch, new WordBoundaryChunkingSettings(wordsPerChunk, chunkOverlap));
9775
}
9876

99-
public EmbeddingRequestChunker(
100-
List<String> inputs,
101-
int maxNumberOfInputsPerBatch,
102-
EmbeddingType embeddingType,
103-
ChunkingSettings chunkingSettings
104-
) {
77+
public EmbeddingRequestChunker(List<String> inputs, int maxNumberOfInputsPerBatch, ChunkingSettings chunkingSettings) {
10578
this.inputs = inputs;
106-
this.embeddingType = embeddingType;
10779
this.results = new ArrayList<>(inputs.size());
10880
this.errors = new AtomicArray<>(inputs.size());
10981

‎x-pack/plugin/inference/src/main/java/org/elasticsearch/xpack/inference/services/alibabacloudsearch/AlibabaCloudSearchService.java‎

Lines changed: 0 additions & 9 deletions
Original file line numberDiff line numberDiff line change
@@ -308,7 +308,6 @@ protected void doChunkedInfer(
308308
List<EmbeddingRequestChunker.BatchRequestAndListener> batchedRequests = new EmbeddingRequestChunker(
309309
inputs.getInputs(),
310310
EMBEDDING_MAX_BATCH_SIZE,
311-
getEmbeddingTypeFromTaskType(alibabaCloudSearchModel.getTaskType()),
312311
alibabaCloudSearchModel.getConfigurations().getChunkingSettings()
313312
).batchRequestsWithListeners(listener);
314313

@@ -318,14 +317,6 @@ protected void doChunkedInfer(
318317
}
319318
}
320319

321-
private EmbeddingRequestChunker.EmbeddingType getEmbeddingTypeFromTaskType(TaskType taskType) {
322-
return switch (taskType) {
323-
case TEXT_EMBEDDING -> EmbeddingRequestChunker.EmbeddingType.FLOAT;
324-
case SPARSE_EMBEDDING -> EmbeddingRequestChunker.EmbeddingType.SPARSE;
325-
default -> throw new IllegalArgumentException("Unsupported task type for chunking: " + taskType);
326-
};
327-
}
328-
329320
/**
330321
* For text embedding models get the embedding size and
331322
* update the service settings.

‎x-pack/plugin/inference/src/main/java/org/elasticsearch/xpack/inference/services/amazonbedrock/AmazonBedrockService.java‎

Lines changed: 0 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -132,7 +132,6 @@ protected void doChunkedInfer(
132132
List<EmbeddingRequestChunker.BatchRequestAndListener> batchedRequests = new EmbeddingRequestChunker(
133133
inputs.getInputs(),
134134
maxBatchSize,
135-
EmbeddingRequestChunker.EmbeddingType.FLOAT,
136135
baseAmazonBedrockModel.getConfigurations().getChunkingSettings()
137136
).batchRequestsWithListeners(listener);
138137

‎x-pack/plugin/inference/src/main/java/org/elasticsearch/xpack/inference/services/azureaistudio/AzureAiStudioService.java‎

Lines changed: 0 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -124,7 +124,6 @@ protected void doChunkedInfer(
124124
List<EmbeddingRequestChunker.BatchRequestAndListener> batchedRequests = new EmbeddingRequestChunker(
125125
inputs.getInputs(),
126126
EMBEDDING_MAX_BATCH_SIZE,
127-
EmbeddingRequestChunker.EmbeddingType.FLOAT,
128127
baseAzureAiStudioModel.getConfigurations().getChunkingSettings()
129128
).batchRequestsWithListeners(listener);
130129

‎x-pack/plugin/inference/src/main/java/org/elasticsearch/xpack/inference/services/azureopenai/AzureOpenAiService.java‎

Lines changed: 0 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -284,7 +284,6 @@ protected void doChunkedInfer(
284284
List<EmbeddingRequestChunker.BatchRequestAndListener> batchedRequests = new EmbeddingRequestChunker(
285285
inputs.getInputs(),
286286
EMBEDDING_MAX_BATCH_SIZE,
287-
EmbeddingRequestChunker.EmbeddingType.FLOAT,
288287
azureOpenAiModel.getConfigurations().getChunkingSettings()
289288
).batchRequestsWithListeners(listener);
290289

‎x-pack/plugin/inference/src/main/java/org/elasticsearch/xpack/inference/services/cohere/CohereService.java‎

Lines changed: 0 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -286,7 +286,6 @@ protected void doChunkedInfer(
286286
List<EmbeddingRequestChunker.BatchRequestAndListener> batchedRequests = new EmbeddingRequestChunker(
287287
inputs.getInputs(),
288288
EMBEDDING_MAX_BATCH_SIZE,
289-
EmbeddingRequestChunker.EmbeddingType.fromDenseVectorElementType(model.getServiceSettings().elementType()),
290289
cohereModel.getConfigurations().getChunkingSettings()
291290
).batchRequestsWithListeners(listener);
292291

‎x-pack/plugin/inference/src/main/java/org/elasticsearch/xpack/inference/services/elasticsearch/ElasticsearchInternalService.java‎

Lines changed: 0 additions & 18 deletions
Original file line numberDiff line numberDiff line change
@@ -728,7 +728,6 @@ public void chunkedInfer(
728728
List<EmbeddingRequestChunker.BatchRequestAndListener> batchedRequests = new EmbeddingRequestChunker(
729729
input,
730730
EMBEDDING_MAX_BATCH_SIZE,
731-
embeddingTypeFromTaskTypeAndSettings(model.getTaskType(), esModel.internalServiceSettings),
732731
esModel.getConfigurations().getChunkingSettings()
733732
).batchRequestsWithListeners(listener);
734733

@@ -944,23 +943,6 @@ boolean isDefaultId(String inferenceId) {
944943
return DEFAULT_ELSER_ID.equals(inferenceId) || DEFAULT_E5_ID.equals(inferenceId) || DEFAULT_RERANK_ID.equals(inferenceId);
945944
}
946945

947-
static EmbeddingRequestChunker.EmbeddingType embeddingTypeFromTaskTypeAndSettings(
948-
TaskType taskType,
949-
ElasticsearchInternalServiceSettings serviceSettings
950-
) {
951-
return switch (taskType) {
952-
case SPARSE_EMBEDDING -> EmbeddingRequestChunker.EmbeddingType.SPARSE;
953-
case TEXT_EMBEDDING -> serviceSettings.elementType() == null
954-
? EmbeddingRequestChunker.EmbeddingType.FLOAT
955-
: EmbeddingRequestChunker.EmbeddingType.fromDenseVectorElementType(serviceSettings.elementType());
956-
default -> throw new ElasticsearchStatusException(
957-
"Chunking is not supported for task type [{}]",
958-
RestStatus.BAD_REQUEST,
959-
taskType
960-
);
961-
};
962-
}
963-
964946
private void validateAgainstDeployment(
965947
String modelId,
966948
String deploymentId,

‎x-pack/plugin/inference/src/main/java/org/elasticsearch/xpack/inference/services/googleaistudio/GoogleAiStudioService.java‎

Lines changed: 0 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -331,7 +331,6 @@ protected void doChunkedInfer(
331331
List<EmbeddingRequestChunker.BatchRequestAndListener> batchedRequests = new EmbeddingRequestChunker(
332332
inputs.getInputs(),
333333
EMBEDDING_MAX_BATCH_SIZE,
334-
EmbeddingRequestChunker.EmbeddingType.FLOAT,
335334
googleAiStudioModel.getConfigurations().getChunkingSettings()
336335
).batchRequestsWithListeners(listener);
337336

‎x-pack/plugin/inference/src/main/java/org/elasticsearch/xpack/inference/services/googlevertexai/GoogleVertexAiService.java‎

Lines changed: 0 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -231,7 +231,6 @@ protected void doChunkedInfer(
231231
List<EmbeddingRequestChunker.BatchRequestAndListener> batchedRequests = new EmbeddingRequestChunker(
232232
inputs.getInputs(),
233233
EMBEDDING_MAX_BATCH_SIZE,
234-
EmbeddingRequestChunker.EmbeddingType.FLOAT,
235234
googleVertexAiModel.getConfigurations().getChunkingSettings()
236235
).batchRequestsWithListeners(listener);
237236

‎x-pack/plugin/inference/src/main/java/org/elasticsearch/xpack/inference/services/huggingface/HuggingFaceService.java‎

Lines changed: 0 additions & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -130,7 +130,6 @@ protected void doChunkedInfer(
130130
List<EmbeddingRequestChunker.BatchRequestAndListener> batchedRequests = new EmbeddingRequestChunker(
131131
inputs.getInputs(),
132132
EMBEDDING_MAX_BATCH_SIZE,
133-
EmbeddingRequestChunker.EmbeddingType.FLOAT,
134133
huggingFaceModel.getConfigurations().getChunkingSettings()
135134
).batchRequestsWithListeners(listener);
136135

0 commit comments

Comments
 (0)