|
| 1 | +/* |
| 2 | + * Licensed to the Apache Software Foundation (ASF) under one |
| 3 | + * or more contributor license agreements. See the NOTICE file |
| 4 | + * distributed with this work for additional information |
| 5 | + * regarding copyright ownership. The ASF licenses this file |
| 6 | + * to you under the Apache License, Version 2.0 (the |
| 7 | + * "License"); you may not use this file except in compliance |
| 8 | + * with the License. You may obtain a copy of the License at |
| 9 | + * |
| 10 | + * http://www.apache.org/licenses/LICENSE-2.0 |
| 11 | + * |
| 12 | + * Unless required by applicable law or agreed to in writing, |
| 13 | + * software distributed under the License is distributed on an |
| 14 | + * "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY |
| 15 | + * KIND, either express or implied. See the License for the |
| 16 | + * specific language governing permissions and limitations |
| 17 | + * under the License. |
| 18 | + */ |
| 19 | + |
| 20 | +package io.milvus.v2; |
| 21 | + |
| 22 | +import com.google.gson.JsonArray; |
| 23 | +import com.google.gson.JsonObject; |
| 24 | +import io.milvus.common.utils.JsonUtils; |
| 25 | +import io.milvus.v1.CommonUtils; |
| 26 | +import io.milvus.v2.client.ConnectConfig; |
| 27 | +import io.milvus.v2.client.MilvusClientV2; |
| 28 | +import io.milvus.v2.common.ConsistencyLevel; |
| 29 | +import io.milvus.v2.common.DataType; |
| 30 | +import io.milvus.v2.common.IndexParam; |
| 31 | +import io.milvus.v2.service.collection.request.AddFieldReq; |
| 32 | +import io.milvus.v2.service.collection.request.CreateCollectionReq; |
| 33 | +import io.milvus.v2.service.collection.request.DropCollectionReq; |
| 34 | +import io.milvus.v2.service.collection.request.LoadCollectionReq; |
| 35 | +import io.milvus.v2.service.index.request.CreateIndexReq; |
| 36 | +import io.milvus.v2.service.vector.request.InsertReq; |
| 37 | +import io.milvus.v2.service.vector.request.QueryReq; |
| 38 | +import io.milvus.v2.service.vector.request.SearchReq; |
| 39 | +import io.milvus.v2.service.vector.request.data.BaseVector; |
| 40 | +import io.milvus.v2.service.vector.request.data.EmbeddingList; |
| 41 | +import io.milvus.v2.service.vector.request.data.FloatVec; |
| 42 | +import io.milvus.v2.service.vector.response.InsertResp; |
| 43 | +import io.milvus.v2.service.vector.response.QueryResp; |
| 44 | +import io.milvus.v2.service.vector.response.SearchResp; |
| 45 | + |
| 46 | +import java.util.*; |
| 47 | + |
| 48 | +public class StructExample { |
| 49 | + private static final MilvusClientV2 client; |
| 50 | + static { |
| 51 | + client = new MilvusClientV2(ConnectConfig.builder() |
| 52 | + .uri("http://localhost:19530") |
| 53 | + .build()); |
| 54 | + } |
| 55 | + |
| 56 | + private static final String COLLECTION_NAME = "java_sdk_example_struct_v2"; |
| 57 | + private static final String ID_FIELD = "id"; |
| 58 | + private static final String NAME_FIELD = "film_name"; |
| 59 | + private static final String STRUCT_FIELD = "clips"; |
| 60 | + private static final String FRAME_FIELD = "frame_number"; |
| 61 | + private static final String CLIP_VECTOR_FIELD = "clip_embedding"; |
| 62 | + private static final String DESC_FIELD = "clip_desc"; |
| 63 | + private static final String DESC_VECTOR_FIELD = "description_embedding"; |
| 64 | + private static final Integer VECTOR_DIM = 4; |
| 65 | + |
| 66 | + private static void createCollection() { |
| 67 | + CreateCollectionReq.CollectionSchema collectionSchema = CreateCollectionReq.CollectionSchema.builder() |
| 68 | + .build(); |
| 69 | + collectionSchema.addField(AddFieldReq.builder() |
| 70 | + .fieldName(ID_FIELD) |
| 71 | + .dataType(DataType.Int64) |
| 72 | + .isPrimaryKey(true) |
| 73 | + .build()); |
| 74 | + collectionSchema.addField(AddFieldReq.builder() |
| 75 | + .fieldName(NAME_FIELD) |
| 76 | + .dataType(DataType.VarChar) |
| 77 | + .maxLength(1024) |
| 78 | + .build()); |
| 79 | + // define struct field schema, note that each name of sub-field must be unique in the entire collection |
| 80 | + collectionSchema.addField(AddFieldReq.builder() |
| 81 | + .fieldName(STRUCT_FIELD) |
| 82 | + .description("clips of a film") |
| 83 | + .dataType(DataType.Array) |
| 84 | + .elementType(DataType.Struct) |
| 85 | + .maxCapacity(100) |
| 86 | + .addStructField(AddFieldReq.builder() |
| 87 | + .fieldName(FRAME_FIELD) |
| 88 | + .description("from which frame this clip begin") |
| 89 | + .dataType(DataType.Int32) |
| 90 | + .build()) |
| 91 | + .addStructField(AddFieldReq.builder() |
| 92 | + .fieldName(CLIP_VECTOR_FIELD) |
| 93 | + .description("embedding of a clip") |
| 94 | + .dataType(DataType.FloatVector) |
| 95 | + .dimension(VECTOR_DIM) |
| 96 | + .build()) |
| 97 | + .addStructField(AddFieldReq.builder() |
| 98 | + .fieldName(DESC_FIELD) |
| 99 | + .description("description of a clip") |
| 100 | + .dataType(DataType.VarChar) |
| 101 | + .maxLength(1024) |
| 102 | + .build()) |
| 103 | + .addStructField(AddFieldReq.builder() |
| 104 | + .fieldName(DESC_VECTOR_FIELD) |
| 105 | + .description("embedding of description") |
| 106 | + .dataType(DataType.FloatVector) |
| 107 | + .dimension(VECTOR_DIM) |
| 108 | + .build()) |
| 109 | + .build()); |
| 110 | + |
| 111 | + client.dropCollection(DropCollectionReq.builder() |
| 112 | + .collectionName(COLLECTION_NAME) |
| 113 | + .build()); |
| 114 | + |
| 115 | + CreateCollectionReq requestCreate = CreateCollectionReq.builder() |
| 116 | + .collectionName(COLLECTION_NAME) |
| 117 | + .collectionSchema(collectionSchema) |
| 118 | + .build(); |
| 119 | + client.createCollection(requestCreate); |
| 120 | + |
| 121 | + // struct vector uses special index/metric type |
| 122 | + List<IndexParam> indexParams = new ArrayList<>(); |
| 123 | + indexParams.add(IndexParam.builder() |
| 124 | + .fieldName(CLIP_VECTOR_FIELD) |
| 125 | + .indexName("index_1") |
| 126 | + .indexType(IndexParam.IndexType.EMB_LIST_HNSW) |
| 127 | + .metricType(IndexParam.MetricType.MAX_SIM) |
| 128 | + .build()); |
| 129 | + indexParams.add(IndexParam.builder() |
| 130 | + .fieldName(DESC_VECTOR_FIELD) |
| 131 | + .indexName("index_2") |
| 132 | + .indexType(IndexParam.IndexType.EMB_LIST_HNSW) |
| 133 | + .metricType(IndexParam.MetricType.MAX_SIM) |
| 134 | + .build()); |
| 135 | + client.createIndex(CreateIndexReq.builder() |
| 136 | + .collectionName(COLLECTION_NAME) |
| 137 | + .indexParams(indexParams) |
| 138 | + .build()); |
| 139 | + client.loadCollection(LoadCollectionReq.builder() |
| 140 | + .collectionName(COLLECTION_NAME) |
| 141 | + .build()); |
| 142 | + System.out.println("Collection created: " + COLLECTION_NAME); |
| 143 | + } |
| 144 | + |
| 145 | + private static void insertData(int rowCount) { |
| 146 | + final int batchSize = 300; |
| 147 | + int insertedCount = 0; |
| 148 | + Random ran = new Random(); |
| 149 | + while (insertedCount < rowCount) { |
| 150 | + int nextBatch = batchSize; |
| 151 | + int leftCount = rowCount - insertedCount; |
| 152 | + if (nextBatch > leftCount) { |
| 153 | + nextBatch = leftCount; |
| 154 | + } |
| 155 | + List<JsonObject> rows = new ArrayList<>(); |
| 156 | + for (int i = 0; i < nextBatch; i++) { |
| 157 | + JsonObject row = new JsonObject(); |
| 158 | + int id = insertedCount + i; |
| 159 | + row.addProperty(ID_FIELD, id); |
| 160 | + row.addProperty(NAME_FIELD, "film_" + id); |
| 161 | + JsonArray structArr = new JsonArray(); |
| 162 | + for (int k = 0; k < 5; k++) { |
| 163 | + JsonObject struct = new JsonObject(); |
| 164 | + struct.addProperty(FRAME_FIELD, ran.nextInt(1000000)); |
| 165 | + struct.add(CLIP_VECTOR_FIELD, JsonUtils.toJsonTree(CommonUtils.generateFloatVector(VECTOR_DIM))); |
| 166 | + struct.addProperty(DESC_FIELD, "clip_description_" + id); |
| 167 | + struct.add(DESC_VECTOR_FIELD, JsonUtils.toJsonTree(CommonUtils.generateFloatVector(VECTOR_DIM))); |
| 168 | + structArr.add(struct); |
| 169 | + } |
| 170 | + row.add(STRUCT_FIELD, structArr); |
| 171 | + rows.add(row); |
| 172 | + } |
| 173 | + |
| 174 | + InsertResp insertResp = client.insert(InsertReq.builder() |
| 175 | + .collectionName(COLLECTION_NAME) |
| 176 | + .data(rows) |
| 177 | + .build()); |
| 178 | + insertedCount += (int) insertResp.getInsertCnt(); |
| 179 | + System.out.println("Inserted row count: " + insertResp.getInsertCnt()); |
| 180 | + } |
| 181 | + |
| 182 | + QueryResp countR = client.query(QueryReq.builder() |
| 183 | + .collectionName(COLLECTION_NAME) |
| 184 | + .outputFields(Collections.singletonList("count(*)")) |
| 185 | + .consistencyLevel(ConsistencyLevel.STRONG) |
| 186 | + .build()); |
| 187 | + System.out.printf("%d rows persisted\n", (long)countR.getQueryResults().get(0).getEntity().get("count(*)")); |
| 188 | + |
| 189 | + } |
| 190 | + private static void query(String filter) { |
| 191 | + System.out.println("==================================================="); |
| 192 | + System.out.println("Query with filter expression: " + filter); |
| 193 | + QueryResp queryResp = client.query(QueryReq.builder() |
| 194 | + .collectionName(COLLECTION_NAME) |
| 195 | + .filter(filter) |
| 196 | + .consistencyLevel(ConsistencyLevel.BOUNDED) |
| 197 | + .outputFields(Collections.singletonList(STRUCT_FIELD)) |
| 198 | + .build()); |
| 199 | + List<QueryResp.QueryResult> queryResults = queryResp.getQueryResults(); |
| 200 | + for (QueryResp.QueryResult result : queryResults) { |
| 201 | + System.out.println(result.getEntity()); |
| 202 | + } |
| 203 | + } |
| 204 | + |
| 205 | + private static void search(String annsField, int nq, int targetVectorsPerNQ) { |
| 206 | + System.out.println("==================================================="); |
| 207 | + String msg = String.format("Search on field '%s' with nq=%d and vectors_per_nq=%d", annsField, nq, targetVectorsPerNQ); |
| 208 | + System.out.println(msg); |
| 209 | + List<BaseVector> searchData = new ArrayList<>(); |
| 210 | + for (int i = 0; i < nq; i++) { |
| 211 | + EmbeddingList embList = new EmbeddingList(); |
| 212 | + for (int k = 0; k < targetVectorsPerNQ; k++) { |
| 213 | + embList.add(new FloatVec(CommonUtils.generateFloatVector(VECTOR_DIM))); |
| 214 | + } |
| 215 | + searchData.add(embList); |
| 216 | + } |
| 217 | + |
| 218 | + int topK = 5; |
| 219 | + SearchResp searchResp = client.search(SearchReq.builder() |
| 220 | + .collectionName(COLLECTION_NAME) |
| 221 | + .annsField(annsField) |
| 222 | + .data(searchData) |
| 223 | + .limit(topK) |
| 224 | + .consistencyLevel(ConsistencyLevel.BOUNDED) |
| 225 | + .outputFields(Arrays.asList(NAME_FIELD, FRAME_FIELD, DESC_FIELD)) |
| 226 | + .build()); |
| 227 | + List<List<SearchResp.SearchResult>> searchResults = searchResp.getSearchResults(); |
| 228 | + for (int i = 0; i < searchResults.size(); i++) { |
| 229 | + System.out.println("Results of No." + i + " embedding list"); |
| 230 | + List<SearchResp.SearchResult> results = searchResults.get(i); |
| 231 | + for (SearchResp.SearchResult result : results) { |
| 232 | + System.out.println(result); |
| 233 | + } |
| 234 | + } |
| 235 | + } |
| 236 | + |
| 237 | + public static void main(String[] args) { |
| 238 | + createCollection(); |
| 239 | + insertData(2000); |
| 240 | + query(ID_FIELD + " <= 5"); |
| 241 | + search(CLIP_VECTOR_FIELD, 2, 3); |
| 242 | + search(DESC_VECTOR_FIELD, 1, 5); |
| 243 | + } |
| 244 | +} |
0 commit comments