docs: A comment for field api_key_config in message `.google.cloud.…

…aiplatform.v1beta1.JiraSource` is changed docs: A comment for field `partial_failure_gcs_sink` in message `.google.cloud.aiplatform.v1beta1.ImportRagFilesConfig` is changed docs: A comment for field `partial_failure_bigquery_sink` in message `.google.cloud.aiplatform.v1beta1.ImportRagFilesConfig` is changed docs: A comment for field `rag_file_parsing_config` in message `.google.cloud.aiplatform.v1beta1.ImportRagFilesConfig` is changed docs: A comment for field `source_uri` in message `.google.cloud.aiplatform.v1beta1.RagContexts` is changed feat: Introduce HybridSearch and Ranking configuration for RAG feat: Introduce VertexAiSearch integration for RAG PiperOrigin-RevId: 699296635
googleapis · Nov 22, 2024 · 4fefaf1 · 4fefaf1
1 parent 1c69885
commit 4fefaf1
Show file tree

Hide file tree

Showing 5 changed files with 374 additions and 25 deletions.
diff --git a/google/cloud/aiplatform/v1beta1/aiplatform_v1beta1.yaml b/google/cloud/aiplatform/v1beta1/aiplatform_v1beta1.yaml
@@ -924,7 +924,7 @@ authentication:
     oauth:
       canonical_scopes: |-
         https://www.googleapis.com/auth/cloud-platform
-  - selector: google.cloud.aiplatform.v1beta1.VertexRagService.RetrieveContexts
+  - selector: 'google.cloud.aiplatform.v1beta1.VertexRagService.*'
     oauth:
       canonical_scopes: |-
         https://www.googleapis.com/auth/cloud-platform

diff --git a/google/cloud/aiplatform/v1beta1/io.proto b/google/cloud/aiplatform/v1beta1/io.proto
@@ -193,8 +193,8 @@ message JiraSource {
 
     // Required. The SecretManager secret version resource name (e.g.
     // projects/{project}/secrets/{secret}/versions/{version}) storing the
-    // Jira API key
-    // (https://support.atlassian.com/atlassian-account/docs/manage-api-tokens-for-your-atlassian-account/).
+    // Jira API key. See [Manage API tokens for your Atlassian
+    // account](https://support.atlassian.com/atlassian-account/docs/manage-api-tokens-for-your-atlassian-account/).
     ApiAuth.ApiKeyConfig api_key_config = 7
         [(google.api.field_behavior) = REQUIRED];
   }

diff --git a/google/cloud/aiplatform/v1beta1/tool.proto b/google/cloud/aiplatform/v1beta1/tool.proto
@@ -281,11 +281,16 @@ message VertexRagStore {
       [(google.api.field_behavior) = OPTIONAL];
 
   // Optional. Number of top k results to return from the selected corpora.
-  optional int32 similarity_top_k = 2 [(google.api.field_behavior) = OPTIONAL];
+  optional int32 similarity_top_k = 2
+      [deprecated = true, (google.api.field_behavior) = OPTIONAL];
 
   // Optional. Only return results with vector distance smaller than the
   // threshold.
   optional double vector_distance_threshold = 3
+      [deprecated = true, (google.api.field_behavior) = OPTIONAL];
+
+  // Optional. The retrieval config for the Rag query.
+  RagRetrievalConfig rag_retrieval_config = 6
       [(google.api.field_behavior) = OPTIONAL];
 }
 
@@ -361,3 +366,73 @@ message FunctionCallingConfig {
   repeated string allowed_function_names = 2
       [(google.api.field_behavior) = OPTIONAL];
 }
+
+// Specifies the context retrieval config.
+message RagRetrievalConfig {
+  // Config for Hybrid Search.
+  message HybridSearch {
+    // Optional. Alpha value controls the weight between dense and sparse vector
+    // search results. The range is [0, 1], while 0 means sparse vector search
+    // only and 1 means dense vector search only. The default value is 0.5 which
+    // balances sparse and dense vector search equally.
+    optional float alpha = 1 [(google.api.field_behavior) = OPTIONAL];
+  }
+
+  // Config for filters.
+  message Filter {
+    // Filter contexts retrieved from the vector DB based on either vector
+    // distance or vector similarity.
+    oneof vector_db_threshold {
+      // Optional. Only returns contexts with vector distance smaller than the
+      // threshold.
+      double vector_distance_threshold = 3
+          [(google.api.field_behavior) = OPTIONAL];
+
+      // Optional. Only returns contexts with vector similarity larger than the
+      // threshold.
+      double vector_similarity_threshold = 4
+          [(google.api.field_behavior) = OPTIONAL];
+    }
+
+    // Optional. String for metadata filtering.
+    string metadata_filter = 2 [(google.api.field_behavior) = OPTIONAL];
+  }
+
+  // Config for ranking and reranking.
+  message Ranking {
+    // Config for Rank Service.
+    message RankService {
+      // Optional. The model name of the rank service.
+      // Format: `semantic-ranker-512@latest`
+      optional string model_name = 1 [(google.api.field_behavior) = OPTIONAL];
+    }
+
+    // Config for LlmRanker.
+    message LlmRanker {
+      // Optional. The model name used for ranking.
+      // Format: `gemini-1.5-pro`
+      optional string model_name = 1 [(google.api.field_behavior) = OPTIONAL];
+    }
+
+    // Config options for ranking. Currently only Rank Service is supported.
+    oneof ranking_config {
+      // Optional. Config for Rank Service.
+      RankService rank_service = 1 [(google.api.field_behavior) = OPTIONAL];
+
+      // Optional. Config for LlmRanker.
+      LlmRanker llm_ranker = 3 [(google.api.field_behavior) = OPTIONAL];
+    }
+  }
+
+  // Optional. The number of contexts to retrieve.
+  int32 top_k = 1 [(google.api.field_behavior) = OPTIONAL];
+
+  // Optional. Config for Hybrid Search.
+  HybridSearch hybrid_search = 2 [(google.api.field_behavior) = OPTIONAL];
+
+  // Optional. Config for filters.
+  Filter filter = 3 [(google.api.field_behavior) = OPTIONAL];
+
+  // Optional. Config for ranking and reranking.
+  Ranking ranking = 4 [(google.api.field_behavior) = OPTIONAL];
+}
diff --git a/google/cloud/aiplatform/v1beta1/vertex_rag_data.proto b/google/cloud/aiplatform/v1beta1/vertex_rag_data.proto
@@ -180,6 +180,12 @@ message RagVectorDbConfig {
 
   // Authentication config for the chosen Vector DB.
   ApiAuth api_auth = 5;
+
+  // Optional. Immutable. The embedding model config of the Vector DB.
+  RagEmbeddingModelConfig rag_embedding_model_config = 7 [
+    (google.api.field_behavior) = OPTIONAL,
+    (google.api.field_behavior) = IMMUTABLE
+  ];
 }
 
 // RagFile status.
@@ -204,6 +210,15 @@ message FileStatus {
   string error_status = 2 [(google.api.field_behavior) = OUTPUT_ONLY];
 }
 
+// Config for the Vertex AI Search.
+message VertexAiSearchConfig {
+  // Vertex AI Search Serving Config resource full name. For example,
+  // `projects/{project}/locations/{location}/collections/{collection}/engines/{engine}/servingConfigs/{serving_config}`
+  // or
+  // `projects/{project}/locations/{location}/collections/{collection}/dataStores/{data_store}/servingConfigs/{serving_config}`.
+  string serving_config = 1;
+}
+
 // RagCorpus status.
 message CorpusStatus {
   // RagCorpus life state.
@@ -252,12 +267,14 @@ message RagCorpus {
 
   // Optional. Immutable. The embedding model config of the RagCorpus.
   RagEmbeddingModelConfig rag_embedding_model_config = 6 [
+    deprecated = true,
     (google.api.field_behavior) = OPTIONAL,
     (google.api.field_behavior) = IMMUTABLE
   ];
 
   // Optional. Immutable. The Vector DB config of the RagCorpus.
   RagVectorDbConfig rag_vector_db_config = 7 [
+    deprecated = true,
     (google.api.field_behavior) = OPTIONAL,
     (google.api.field_behavior) = IMMUTABLE
   ];
@@ -272,6 +289,22 @@ message RagCorpus {
 
   // Output only. RagCorpus state.
   CorpusStatus corpus_status = 8 [(google.api.field_behavior) = OUTPUT_ONLY];
+
+  // The backend config of the RagCorpus.
+  // It can be data store and/or retrieval engine.
+  oneof backend_config {
+    // Optional. Immutable. The config for the Vector DBs.
+    RagVectorDbConfig vector_db_config = 9 [
+      (google.api.field_behavior) = OPTIONAL,
+      (google.api.field_behavior) = IMMUTABLE
+    ];
+
+    // Optional. Immutable. The config for the Vertex AI Search.
+    VertexAiSearchConfig vertex_ai_search_config = 10 [
+      (google.api.field_behavior) = OPTIONAL,
+      (google.api.field_behavior) = IMMUTABLE
+    ];
+  }
 }
 
 // A RagFile contains user data for chunking, embedding and indexing.
@@ -353,23 +386,80 @@ message RagFile {
 
 // Specifies the size and overlap of chunks for RagFiles.
 message RagFileChunkingConfig {
+  // Specifies the fixed length chunking config.
+  message FixedLengthChunking {
+    // The size of the chunks.
+    int32 chunk_size = 1;
+
+    // The overlap between chunks.
+    int32 chunk_overlap = 2;
+  }
+
+  // Specifies the chunking config for RagFiles.
+  oneof chunking_config {
+    // Specifies the fixed length chunking config.
+    FixedLengthChunking fixed_length_chunking = 3;
+  }
+
   // The size of the chunks.
-  int32 chunk_size = 1;
+  int32 chunk_size = 1 [deprecated = true];
 
   // The overlap between chunks.
-  int32 chunk_overlap = 2;
+  int32 chunk_overlap = 2 [deprecated = true];
+}
+
+// Specifies the transformation config for RagFiles.
+message RagFileTransformationConfig {
+  // Specifies the chunking config for RagFiles.
+  RagFileChunkingConfig rag_file_chunking_config = 1;
 }
 
 // Specifies the parsing config for RagFiles.
 message RagFileParsingConfig {
+  // Specifies the advanced parsing for RagFiles.
+  message AdvancedParser {
+    // Whether to use advanced PDF parsing.
+    bool use_advanced_pdf_parsing = 1;
+  }
+
+  // Document AI Layout Parser config.
+  message LayoutParser {
+    // The full resource name of a Document AI processor or processor version.
+    // The processor must have type `LAYOUT_PARSER_PROCESSOR`. If specified, the
+    // `additional_config.parse_as_scanned_pdf` field must be false.
+    // Format:
+    // * `projects/{project_id}/locations/{location}/processors/{processor_id}`
+    // * `projects/{project_id}/locations/{location}/processors/{processor_id}/processorVersions/{processor_version_id}`
+    string processor_name = 1;
+
+    // The maximum number of requests the job is allowed to make to the Document
+    // AI processor per minute. Consult
+    // https://cloud.google.com/document-ai/quotas and the Quota page for your
+    // project to set an appropriate value here. If unspecified, a default value
+    // of 120 QPM would be used.
+    int32 max_parsing_requests_per_min = 2;
+  }
+
+  // The parser to use for RagFiles.
+  oneof parser {
+    // The Advanced Parser to use for RagFiles.
+    AdvancedParser advanced_parser = 3;
+
+    // The Layout Parser to use for RagFiles.
+    LayoutParser layout_parser = 4;
+  }
+
   // Whether to use advanced PDF parsing.
-  bool use_advanced_pdf_parsing = 2;
+  bool use_advanced_pdf_parsing = 2 [deprecated = true];
 }
 
 // Config for uploading RagFile.
 message UploadRagFileConfig {
   // Specifies the size and overlap of chunks after uploading RagFile.
-  RagFileChunkingConfig rag_file_chunking_config = 1;
+  RagFileChunkingConfig rag_file_chunking_config = 1 [deprecated = true];
+
+  // Specifies the transformation config for RagFiles.
+  RagFileTransformationConfig rag_file_transformation_config = 3;
 }
 
 // Config for importing RagFiles.
@@ -397,24 +487,32 @@ message ImportRagFilesConfig {
   }
 
   // Optional. If provided, all partial failures are written to the sink.
+  // Deprecated. Prefer to use the `import_result_sink`.
   oneof partial_failure_sink {
     // The Cloud Storage path to write partial failures to.
-    GcsDestination partial_failure_gcs_sink = 11;
+    // Deprecated. Prefer to use `import_result_gcs_sink`.
+    GcsDestination partial_failure_gcs_sink = 11 [deprecated = true];
 
     // The BigQuery destination to write partial failures to. It should be a
     // bigquery table resource name (e.g.
-    // "bq://projectId.bqDatasetId.bqTableId"). If the dataset id does not
-    // exist, it will be created. If the table does not exist, it will be
-    // created with the expected schema. If the table exists, the schema will be
-    // validated and data will be added to this existing table.
-    BigQueryDestination partial_failure_bigquery_sink = 12;
+    // "bq://projectId.bqDatasetId.bqTableId"). The dataset must exist. If the
+    // table does not exist, it will be created with the expected schema. If the
+    // table exists, the schema will be validated and data will be added to this
+    // existing table.
+    // Deprecated. Prefer to use `import_result_bq_sink`.
+    BigQueryDestination partial_failure_bigquery_sink = 12 [deprecated = true];
   }
 
   // Specifies the size and overlap of chunks after importing RagFiles.
-  RagFileChunkingConfig rag_file_chunking_config = 4;
+  RagFileChunkingConfig rag_file_chunking_config = 4 [deprecated = true];
+
+  // Specifies the transformation config for RagFiles.
+  RagFileTransformationConfig rag_file_transformation_config = 16;
 
-  // Specifies the parsing config for RagFiles.
-  RagFileParsingConfig rag_file_parsing_config = 8;
+  // Optional. Specifies the parsing config for RagFiles.
+  // RAG will use the default parser if this field is not set.
+  RagFileParsingConfig rag_file_parsing_config = 8
+      [(google.api.field_behavior) = OPTIONAL];
 
   // Optional. The max number of queries per minute that this job is allowed to
   // make to the embedding model specified on the corpus. This value is specific