Skip to main content
Weaviate Docs (migrated from docs.weaviate.io) Docs

Search documentation

Type to search this documentation.

On this pageOverview

Hybrid search

Hybrid search combines the results of a vector search and a keyword (BM25F) search by fusing the two result sets.

The fusion method and the relative weights are configurable.

Section titled “Basic hybrid search”

Combine the results of a vector search and a keyword search. The search uses a single query string.

Python
jeopardy = client.collections.use("JeopardyQuestion")response = jeopardy.query.hybrid(query="food", limit=3)for o in response.objects:    print(o.properties)
JavaScript/TypeScript
const jeopardy = client.collections.use('JeopardyQuestion');
Go
ctx := context.Background()
className := "JeopardyQuestion"
query := "food"
limit := 3

q := client.GraphQL().Get().
  WithClassName(className).
  WithFields(graphql.Field{Name: "question"}, graphql.Field{Name: "answer"}).
  WithHybrid(client.GraphQL().HybridArgumentBuilder().WithQuery(query)).
  WithLimit(limit)

result, err := q.Do(ctx)
Java
CollectionHandle<Map<String, Object>> jeopardy =    client.collections.use("JeopardyQuestion");var response = jeopardy.query.hybrid(    "food", q -> q.limit(3));for (var o : response.objects()) {  System.out.println(o.properties());}
C#
var jeopardy = client.Collections.Use("JeopardyQuestion");var response = await jeopardy.Query.Hybrid(    "food",    limit: 3);foreach (var o in response.Objects){    Console.WriteLine(JsonSerializer.Serialize(o.Properties));}
GraphQL
{  Get {    JeopardyQuestion(      limit: 3      hybrid: {        query: "food"      }    ) {      question      answer    }  }}
Example response

The output is like this:

JSON
{
  "data": {
    "Get": {
      "JeopardyQuestion": [
        {
          "answer": "a closer grocer",
          "question": "A nearer food merchant"
        },
        {
          "answer": "Famine",
          "question": "From the Latin for \"hunger\", it's a period when food is extremely scarce"
        },
        {
          "answer": "Tofu",
          "question": "A popular health food, this soybean curd is used to make a variety of dishes & an ice cream substitute"
        }
      ]
    }
  }
}

Named vectors

Section titled “Named vectors”

A hybrid search on a collection that has named vectors must specify a target vector. Weaviate uses the query vector to search the target vector space.

Python
reviews = client.collections.use("WineReviewNV")response = reviews.query.hybrid(    query="A French Riesling",    target_vector="title_country",    limit=3)for o in response.objects:    print(o.properties)
JavaScript/TypeScript
const myNVCollection = client.collections.use('WineReviewNV');const result = await myNVCollection.query.hybrid('a sweet German white wine', {  targetVector: 'title_country',  limit: 2,})for (let object of result.objects) {  console.log(JSON.stringify(object.properties, null, 2));}
Java
C#
var reviews = client.Collections.Use("WineReviewNV");var response = await reviews.Query.Hybrid(    vectors: v => v.NearText(["A French Riesling"]).TargetVectorsMinimum("title_country"),    limit: 3);foreach (var o in response.Objects){    Console.WriteLine(JsonSerializer.Serialize(o.Properties));}
GraphQL
{  Get {    WineReviewNV(      limit: 2      hybrid: {        targetVectors: ["title_country"]        query: "A French Riesling"      }    ) {      title      review_body      country    }  }}
Example response

The output is like this:

JSON

Explain the search results

Section titled “Explain the search results”

To see the object rankings, set the explain score field in your query. The search rankings are part of the object metadata. Weaviate uses the score to order the search results.

Python
from weaviate.classes.query import MetadataQueryjeopardy = client.collections.use("JeopardyQuestion")response = jeopardy.query.hybrid(    query="food",    alpha=0.5,    return_metadata=MetadataQuery(score=True, explain_score=True),    limit=3,)for o in response.objects:    print(o.properties)    print(o.metadata.score, o.metadata.explain_score)
JavaScript/TypeScript
const jeopardy = client.collections.use('JeopardyQuestion');
Go
ctx := context.Background()
className := "JeopardyQuestion"
query := "food"
limit := 3

result, err := client.GraphQL().Get().
  WithClassName(className).
  WithFields(
    graphql.Field{Name: "question"},
    graphql.Field{Name: "answer"},
    graphql.Field{Name: "_additional", Fields: []graphql.Field{{Name: "score"}, {Name: "explainScore"}}},
  ).
  WithHybrid(client.GraphQL().HybridArgumentBuilder().WithQuery(query)).
  WithLimit(limit).
  Do(ctx)
Java
CollectionHandle<Map<String, Object>> jeopardy =    client.collections.use("JeopardyQuestion");var response = jeopardy.query.hybrid("food", q -> q.alpha(0.5f)    .returnMetadata(Metadata.SCORE, Metadata.EXPLAIN_SCORE)    .limit(3));for (var o : response.objects()) {  System.out.println(o.properties());  System.out      .println(o.queryMetadata().score() + " " + o.queryMetadata().explainScore());}
C#
var jeopardy = client.Collections.Use("JeopardyQuestion");var response = await jeopardy.Query.Hybrid(    "food",    alpha: 0.5f,    returnMetadata: MetadataOptions.Score | MetadataOptions.ExplainScore,    limit: 3);foreach (var o in response.Objects){    Console.WriteLine(JsonSerializer.Serialize(o.Properties));    Console.WriteLine(        $"Score: {o.Metadata.Score}, Explain Score: {o.Metadata.ExplainScore}"    );}
GraphQL
{  Get {    JeopardyQuestion(      limit: 3      hybrid: {        query: "food"      }    ) {      question      answer      _additional {        score        explainScore      }    }  }}
Example response

The output is like this:

JSON
{
  "data": {
    "Get": {
      "JeopardyQuestion": [
        {
          "_additional": {
            "explainScore": "(bm25)\n(hybrid) Document df958a90-c3ad-5fde-9122-cd777c22da6c contributed 0.003968253968253968 to the score\n(hybrid) Document df958a90-c3ad-5fde-9122-cd777c22da6c contributed 0.012295081967213115 to the score",
            "score": "0.016263336"
          },
          "answer": "a closer grocer",
          "question": "A nearer food merchant"
        },
        {
          "_additional": {
            "explainScore": "(vector) [0.0223698 -0.02752683 -0.0061537363 0.0023812135 -0.00036100898 -0.0078375945 -0.018505432 -0.037500713 -0.0042215516 -0.012620432]...  \n(hybrid) Document ec776112-e651-519d-afd1-b48e6237bbcb contributed 0.012096774193548387 to the score",
            "score": "0.012096774"
          },
          "answer": "Famine",
          "question": "From the Latin for \"hunger\", it's a period when food is extremely scarce"
        },
        {
          "_additional": {
            "explainScore": "(vector) [0.0223698 -0.02752683 -0.0061537363 0.0023812135 -0.00036100898 -0.0078375945 -0.018505432 -0.037500713 -0.0042215516 -0.012620432]...  \n(hybrid) Document 98807640-cd16-507d-86a1-801902d784de contributed 0.011904761904761904 to the score",
            "score": "0.011904762"
          },
          "answer": "Tofu",
          "question": "A popular health food, this soybean curd is used to make a variety of dishes & an ice cream substitute"
        }
      ]
    }
  }
}
Section titled “Balance keyword and vector search”

Hybrid search results can favor the keyword component or the vector component. To change the relative weights of the keyword and vector components, set the alpha value in your query.

  • An alpha of 1 is a pure vector search.
  • An alpha of 0 is a pure keyword search.

If you do not set alpha, the effective weighting depends on your client. See Alpha parameter.

Python
jeopardy = client.collections.use("JeopardyQuestion")response = jeopardy.query.hybrid(    query="food",    alpha=0.25,    limit=3,)for o in response.objects:    print(o.properties)
JavaScript/TypeScript
const jeopardy = client.collections.use('JeopardyQuestion');
Go
ctx := context.Background()
className := "JeopardyQuestion"
query := "food"
limit := 3
alpha := float32(0.25)

result, err := client.GraphQL().Get().
  WithClassName(className).
  WithFields(graphql.Field{Name: "question"}, graphql.Field{Name: "answer"}).
  WithHybrid(client.GraphQL().HybridArgumentBuilder().
    WithQuery(query).
    WithAlpha(alpha),
  ).
  WithLimit(limit).
  Do(ctx)
Java
CollectionHandle<Map<String, Object>> jeopardy =    client.collections.use("JeopardyQuestion");var response = jeopardy.query.hybrid("food", q -> q    .alpha(0.25f)    .limit(3));for (var o : response.objects()) {  System.out.println(o.properties());}
C#
var jeopardy = client.Collections.Use("JeopardyQuestion");var response = await jeopardy.Query.Hybrid(    "food",    alpha: 0.25f,    limit: 3);foreach (var o in response.Objects){    Console.WriteLine(JsonSerializer.Serialize(o.Properties));}
GraphQL
{  Get {    JeopardyQuestion(      limit: 3      hybrid: {        query: "food"        alpha: 0.25      }    ) {      question      answer    }  }}
Example response

The output is like this:

JSON
{
  "data": {
    "Get": {
      "JeopardyQuestion": [
        {
          "answer": "a closer grocer",
          "question": "A nearer food merchant"
        },
        {
          "answer": "food stores (supermarkets)",
          "question": "This type of retail store sells more shampoo & makeup than any other"
        },
        {
          "answer": "cake",
          "question": "Devil's food & angel food are types of this dessert"
        }
      ]
    }
  }
}

Change the fusion method

Section titled “Change the fusion method”

Relative Score Fusion is the default fusion method starting in v1.24.

  • To use the keyword and vector search relative scores instead of the search rankings, use Relative Score Fusion.
  • To use autocut with the hybrid operator, use Relative Score Fusion.
Python
from weaviate.classes.query import HybridFusionjeopardy = client.collections.use("JeopardyQuestion")response = jeopardy.query.hybrid(    query="food",    fusion_type=HybridFusion.RELATIVE_SCORE,    limit=3,)for o in response.objects:    print(o.properties)
JavaScript/TypeScript
const jeopardy = client.collections.use('JeopardyQuestion');
Go
ctx := context.Background()
className := "JeopardyQuestion"
query := "food"
limit := 3
fusionType := "relativeScoreFusion"

result, err := client.GraphQL().Get().
  WithClassName(className).
  WithFields(graphql.Field{Name: "question"}, graphql.Field{Name: "answer"}).
  WithHybrid(client.GraphQL().HybridArgumentBuilder().
    WithQuery(query).
    WithFusionType(graphql.FusionType(fusionType)),
  ).
  WithLimit(limit).
  Do(ctx)
Java
CollectionHandle<Map<String, Object>> jeopardy =    client.collections.use("JeopardyQuestion");var response = jeopardy.query.hybrid("food", q -> q    .fusionType(FusionType.RELATIVE_SCORE)    .limit(3));for (var o : response.objects()) {  System.out.println(o.properties());}
C#
var jeopardy = client.Collections.Use("JeopardyQuestion");var response = await jeopardy.Query.Hybrid(    "food",    fusionType: HybridFusion.RelativeScore,    limit: 3);foreach (var o in response.Objects){    Console.WriteLine(JsonSerializer.Serialize(o.Properties));}
GraphQL
{  Get {    JeopardyQuestion(      limit: 3      hybrid: {        query: "food"        fusionType: relativeScoreFusion      }    ) {      question      answer    }  }}
Example response

The output is like this:

JSON
{
  "data": {
    "Get": {
      "JeopardyQuestion": [
        {
          "answer": "a closer grocer",
          "question": "A nearer food merchant"
        },
        {
          "answer": "food stores (supermarkets)",
          "question": "This type of retail store sells more shampoo & makeup than any other"
        },
        {
          "answer": "cake",
          "question": "Devil's food & angel food are types of this dessert"
        }
      ]
    }
  }
}
Additional information

For a discussion of fusion methods, see this blog post and this reference page.

Keyword search operators

Section titled “Keyword search operators”

Keyword (BM25) search operators define how many of the query tokens must match, and whether they must all match within a single searched property. The options are or (default), and, and and_cross (available from v1.38.8).

The keyword leg of a hybrid query accepts the same operators as a standalone keyword search. For and_cross, which matches every token across the searched properties combined, see BM25 search: and_cross.

or

Section titled “or”

With the or operator, the search returns objects that contain at least minimumOrTokensMatch of the tokens in the search string.

Python
from weaviate.classes.query import BM25Operatorjeopardy = client.collections.use("JeopardyQuestion")response = jeopardy.query.hybrid(    query="Australian mammal cute",    bm25_operator=BM25Operator.or_(minimum_match=2),    limit=3,)for o in response.objects:    print(o.properties)
JS/TS
const jeopardy = client.collections.use('JeopardyQuestion');
Java
CollectionHandle<Map<String, Object>> jeopardy =    client.collections.use("JeopardyQuestion");var response = jeopardy.query.hybrid(    "Australian mammal cute", c -> c.searchOperator(SearchOperator.or(2))        .limit(3));for (var o : response.objects()) {  System.out.println(o.properties());}
C#
var jeopardy = client.Collections.Use("JeopardyQuestion");var response = await jeopardy.Query.Hybrid(    "Australian mammal cute",    bm25Operator: new BM25Operator.Or(MinimumMatch: 1),    limit: 3);foreach (var o in response.Objects){    Console.WriteLine(JsonSerializer.Serialize(o.Properties));}
GraphQL
{  Get {    JeopardyQuestion(      limit: 3      hybrid: {        query: "Australian mammal cute"        bm25SearchOperator: {          operator: Or,          minimumOrTokensMatch: 2        }      }    ) {      question      answer    }  }}

and

Section titled “and”

With the and operator, the search returns objects where all tokens in the search string appear together within a single searched property.

Python
from weaviate.classes.query import BM25Operatorjeopardy = client.collections.use("JeopardyQuestion")response = jeopardy.query.hybrid(    query="Australian mammal cute",    bm25_operator=BM25Operator.and_(),  # Each result must include all tokens (e.g. "australian", "mammal", "cute")    limit=3,)for o in response.objects:    print(o.properties)
JS/TS
const jeopardy = client.collections.use('JeopardyQuestion');
Java
CollectionHandle<Map<String, Object>> jeopardy =    client.collections.use("JeopardyQuestion");var response = jeopardy.query.hybrid(    "Australian mammal cute"// .bm25Operator(BM25Operator.and()) // Each result must include all tokens// (e.g. "australian", "mammal", "cute")// .limit(3));for (var o : response.objects()) {  System.out.println(o.properties());}
C#
var jeopardy = client.Collections.Use("JeopardyQuestion");var response = await jeopardy.Query.Hybrid(    "Australian mammal cute",    bm25Operator: new BM25Operator.And(), // Each result must include all tokens    limit: 3);foreach (var o in response.Objects){    Console.WriteLine(JsonSerializer.Serialize(o.Properties));}
GraphQL
{  Get {    JeopardyQuestion(      limit: 3      hybrid: {        query: "Australian mammal cute"        bm25SearchOperator: {          operator: And,        }      }    ) {      question      answer    }  }}

Specify keyword search properties

Section titled “Specify keyword search properties”

The keyword search portion of hybrid search can be directed to only search a subset of object properties. This does not affect the vector search portion.

Python
jeopardy = client.collections.use("JeopardyQuestion")response = jeopardy.query.hybrid(    query="food",    query_properties=["question"],    alpha=0.25,    limit=3,)for o in response.objects:    print(o.properties)
JavaScript/TypeScript
const jeopardy = client.collections.use('JeopardyQuestion');
Go
ctx := context.Background()
className := "JeopardyQuestion"
query := "food"
limit := 3
alpha := float32(0.25)
properties := []string{"question"}

result, err := client.GraphQL().Get().
  WithClassName(className).
  WithFields(graphql.Field{Name: "question"}, graphql.Field{Name: "answer"}).
  WithHybrid(client.GraphQL().HybridArgumentBuilder().
    WithQuery(query).
    WithAlpha(alpha).
    WithProperties(properties),
  ).
  WithLimit(limit).
  Do(ctx)
Java
CollectionHandle<Map<String, Object>> jeopardy =    client.collections.use("JeopardyQuestion");var response = jeopardy.query.hybrid("food", q -> q    .queryProperties("question")    .alpha(0.25f)    .limit(3));for (var o : response.objects()) {  System.out.println(o.properties());}
C#
var jeopardy = client.Collections.Use("JeopardyQuestion");var response = await jeopardy.Query.Hybrid(    "food",    queryProperties: ["question"],    alpha: 0.25f,    limit: 3);foreach (var o in response.Objects){    Console.WriteLine(JsonSerializer.Serialize(o.Properties));}
GraphQL
{  Get {    JeopardyQuestion(      limit: 3      hybrid: {        query: "food"        properties: ["question"]        alpha: 0.25      }    ) {      question      answer    }  }}
Example response

The output is like this:

JSON
{
  "data": {
    "Get": {
      "JeopardyQuestion": [
        {
          "answer": "a closer grocer",
          "question": "A nearer food merchant"
        },
        {
          "answer": "cake",
          "question": "Devil's food & angel food are types of this dessert"
        },
        {
          "answer": "honey",
          "question": "The primary source of this food is the Apis mellifera"
        }
      ]
    }
  }
}

Set weights on property values

Section titled “Set weights on property values”

Specify the relative value of an object's properties in the keyword search. Higher values increase the property's contribution to the search score.

Python
jeopardy = client.collections.use("JeopardyQuestion")response = jeopardy.query.hybrid(    query="food",    query_properties=["question^2", "answer"],    alpha=0.25,    limit=3,)for o in response.objects:    print(o.properties)
JavaScript/TypeScript
const jeopardy = client.collections.use('JeopardyQuestion');
Go
ctx := context.Background()
className := "JeopardyQuestion"
query := "food"
limit := 3
alpha := float32(0.25)
properties := []string{"question^2", "answer"}

result, err := client.GraphQL().Get().
  WithClassName(className).
  WithFields(graphql.Field{Name: "question"}, graphql.Field{Name: "answer"}).
  WithHybrid(client.GraphQL().HybridArgumentBuilder().
    WithQuery(query).
    WithAlpha(alpha).
    WithProperties(properties),
  ).
  WithLimit(limit).
  Do(ctx)
Java
CollectionHandle<Map<String, Object>> jeopardy =    client.collections.use("JeopardyQuestion");var response = jeopardy.query.hybrid("food", q -> q    .queryProperties("question^2", "answer")    .alpha(0.25f)    .limit(3));for (var o : response.objects()) {  System.out.println(o.properties());}
C#
var jeopardy = client.Collections.Use("JeopardyQuestion");var response = await jeopardy.Query.Hybrid(    "food",    queryProperties: ["question^2", "answer"],    alpha: 0.25f,    limit: 3);foreach (var o in response.Objects){    Console.WriteLine(JsonSerializer.Serialize(o.Properties));}
GraphQL
{  Get {    JeopardyQuestion(      limit: 3      hybrid: {        query: "food"        properties: ["question^2", "answer"]        alpha: 0.25      }    ) {      question      answer    }  }}
Example response

The output is like this:

JSON
{
  "data": {
    "Get": {
      "JeopardyQuestion": [
        {
          "answer": "a closer grocer",
          "question": "A nearer food merchant"
        },
        {
          "answer": "cake",
          "question": "Devil's food & angel food are types of this dessert"
        },
        {
          "answer": "food stores (supermarkets)",
          "question": "This type of retail store sells more shampoo & makeup than any other"
        }
      ]
    }
  }
}

Specify a search vector

Section titled “Specify a search vector”

The vector component of hybrid search can use a query string or a query vector. To specify a query vector instead of a query string, provide a query vector (for the vector search) and a query string (for the keyword search) in your query.

Python
query_vector = [-0.02] * 1536  # Some vector that is compatible with object vectorsjeopardy = client.collections.use("JeopardyQuestion")response = jeopardy.query.hybrid(    query="food",    vector=query_vector,    alpha=0.25,    limit=3,)for o in response.objects:    print(o.properties)
JavaScript/TypeScript
const jeopardy = client.collections.use('JeopardyQuestion');
Go
ctx := context.Background()
className := "JeopardyQuestion"
query := "food"
limit := 3
//Create a vector 384 dimensions long

// Define the length of the slice
length := 384

// Initialize the slice with the specified length
values := make([]float32, length)

// Fill the slice with values
for i := 0; i < length; i++ {
  values[i] = 0.1 * float32(i+1)
}

vector := values

result, err := client.GraphQL().Get().
  WithClassName(className).
  WithFields(graphql.Field{Name: "question"}, graphql.Field{Name: "answer"}).
  WithHybrid(client.GraphQL().HybridArgumentBuilder().
    WithQuery(query).
    WithVector(vector),
  ).
  WithLimit(limit).
  Do(ctx)
Java
float[] queryVector = new float[1536]; // Some vector that is compatible with object vectorsfor (int i = 0; i < queryVector.length; i++) {  queryVector[i] = -0.02f;}CollectionHandle<Map<String, Object>> jeopardy =    client.collections.use("JeopardyQuestion");var response = jeopardy.query.hybrid("food", q -> q    // .nearVector(NearVector.of(queryVector))    .alpha(0.25f)    .limit(3));for (var o : response.objects()) {  System.out.println(o.properties());}
C#
var queryVector = Enumerable.Repeat(-0.02f, 1536).ToArray();var jeopardy = client.Collections.Use("JeopardyQuestion");var response = await jeopardy.Query.Hybrid(    "food",    vectors: queryVector,    alpha: 0.25f,    limit: 3);foreach (var o in response.Objects){    Console.WriteLine(JsonSerializer.Serialize(o.Properties));}
GraphQL
{  Get {    JeopardyQuestion(      limit: 3      hybrid: {        query: "food"        vector: [0.013085687533020973, -0.00777443777769804, 0.005439540836960077, -0.021052561700344086, -0.02164270170032978, -0.006985447835177183, -0.018974246457219124, -0.025260508060455322, -0.0013630924513563514, -0.03597281128168106, 0.027993107214570045, 0.01635710895061493, -0.02948128432035446, 0.009159981273114681, -0.0026780758053064346, 0.001033544773235917, 0.04051431640982628, 0.001919554895721376, 0.024952609091997147, -0.01960287243127823, 0.0001997531799133867, 0.031405650079250336, 0.021142365410923958, -0.007954045198857784, -0.0008338918560184538, -0.0040572043508291245, 7.381747127510607e-05, -0.019051222130656242, 0.004942412953823805, -0.01888444274663925, 0.028121398761868477, 0.004631306976079941, -0.031559597700834274, -0.003143130801618099, 0.01867917738854885, -0.024208521470427513, 0.0056351847015321255, -0.019333461299538612, -0.0012756941141560674, 0.01737060956656933, 0.020128866657614708, -0.007755194325000048, -0.0071329823695123196, -0.007761608809232712, -0.0074986121617257595, 0.00579554820433259, 0.002782312221825123, 0.01349621918052435, 0.003954571671783924, 0.003170392708852887, 0.017678506672382355, 0.008980373851954937, -0.027454284951090813, -0.004377932287752628, 0.013547535054385662, 0.028506271541118622, -0.011225467547774315, -0.003855146234855056, -8.654634439153597e-05, -0.021770991384983063, -0.004724318161606789, 0.037281379103660583, -0.0543954074382782, 0.01912819594144821, 0.009916898794472218, -0.007806510664522648, 0.0035921495873481035, 0.011757874861359596, -0.004980900324881077, -0.014381427317857742, 0.005952704697847366, 0.009839924052357674, -0.02256639674305916, 0.014561034739017487, 0.01888444274663925, 0.0006859562126919627, 0.0024984683841466904, 0.0033355674240738153, -0.007633317727595568, -0.015330781228840351, 0.02297692745923996, -0.02996879070997238, 0.017473241314291954, 0.01232235599309206, 0.019949259236454964, 0.009769363328814507, -0.038307707756757736, 0.0278134997934103, 0.012264625169336796, -0.007062422577291727, -0.013316611759364605, -0.00465055089443922, 0.013188320212066174, 0.008210627362132072, -0.023862136527895927, 0.006998276803642511, -0.005157300271093845, 0.036024127155542374, -0.0211167074739933, -0.0013109742430970073, -0.0014577071415260434, 0.00858267117291689, -0.004637721460312605, -0.0176656786352396, -0.0035793203860521317, -0.004618478007614613, -0.0015691599110141397, -0.018692007288336754, 0.016536716371774673, -0.012476304545998573, -0.03150828182697296, 0.029789183288812637, 0.01989794336259365, -0.03304777666926384, -0.025234850123524666, 0.0034093346912413836, 0.04154064506292343, 0.013188320212066174, 4.1249832065659575e-06, -0.015279464423656464, -0.01488176267594099, 0.016498230397701263, 0.03025102987885475, -0.015574533492326736, 0.033304356038570404, 0.014445573091506958, 0.018114697188138962, -0.011077932082116604, -0.020719004794955254, -0.008069506846368313, 0.0259789377450943, 0.013079272583127022, 0.014920249581336975, -0.005500479135662317, -0.025119388476014137, 0.017062710598111153, -0.0018938967259600759, -0.003059741575270891, -0.005179751198738813, -0.0017062709666788578, 0.02025715820491314, 0.039077453315258026, -0.01520248968154192, 0.012912494130432606, -0.018730493262410164, 0.022669028490781784, 0.014458402059972286, 0.01916668377816677, -0.013406415469944477, 0.0027758977375924587, 0.02151441015303135, -0.025080900639295578, 0.017999235540628433, -0.01716534234583378, 0.014445573091506958, 0.014868932776153088, -0.008326089009642601, 0.021540068089962006, -0.009820680133998394, 0.0033516038674861193, -0.006863571237772703, 0.0016012326814234257, 0.00046625779941678047, -0.014419914223253727, 0.009442221373319626, 0.021976256743073463, 0.01558736339211464, 0.005356151610612869, 0.013104931451380253, -0.0007625299622304738, -0.0013294160598888993, 0.024580566212534904, -0.011308856308460236, 0.03032800555229187, -0.006594160106033087, 0.023041073232889175, -0.0036466731689870358, 0.004679416306316853, -0.01162958424538374, 0.031764864921569824, -0.01007084734737873, -0.004990521818399429, 0.049135472625494, 0.016062039881944656, -0.005660842638462782, -0.019217999652028084, 0.0008852082537487149, 0.011879751458764076, -0.00190031121019274, -0.03250895440578461, 0.006546051241457462, 0.04238736256957054, 0.014009383507072926, -0.012232552282512188, -0.6835347414016724, -0.006443418096750975, 0.003932120744138956, -0.02776218391954899, 0.027839159592986107, 0.035998471081256866, 0.0012444232124835253, 0.0010808521183207631, 0.005522930063307285, -0.01676764152944088, -0.012976639904081821, 0.010468550026416779, -0.003197654616087675, 0.007274102885276079, -0.00998745858669281, -0.025119388476014137, 0.030302347615361214, -0.01453537680208683, -0.04323408380150795, 0.023002585396170616, -0.02207889035344124, 0.02712072804570198, -0.007633317727595568, -0.0038294880650937557, 0.025991767644882202, -0.002905792323872447, 0.004855816252529621, -0.019949259236454964, 0.00048149237409234047, -0.006250981707125902, -0.017691336572170258, -0.005445955321192741, 0.0037910006940364838, -0.01242498867213726, 0.0669679269194603, -0.00747936824336648, -0.015574533492326736, 0.027505602687597275, 0.024336813017725945, 0.0389748215675354, -0.006587745621800423, -0.012655912898480892, 0.022014744579792023, 0.017524557188153267, -0.009442221373319626, -0.015536046586930752, 0.038102444261312485, 0.016613692045211792, 0.021873624995350838, -0.013098516501486301, 0.025080900639295578, 0.019102538004517555, -0.01688310317695141, 0.014137674123048782, 0.012752130627632141, -0.004724318161606789, 0.031610917299985886, -0.024349642917513847, 0.02603025548160076, 0.010269698686897755, -0.007941216230392456, 0.00834533292800188, 0.0005380206275731325, -0.03150828182697296, 0.0002830421435646713, -0.021629871800541878, -0.012681570835411549, 0.01114207785576582, 0.030995119363069534, 0.0018008856568485498, -0.0017800383502617478, 0.021950598806142807, -0.003095021704211831, -0.00533690769225359, 0.018935760483145714, 0.015279464423656464, 0.019384779036045074, 0.009589755907654762, 0.005211824085563421, 0.0006318334490060806, -0.010859837755560875, 0.00035219904384575784, -0.030481955036520958, 0.0026139302644878626, 0.006504356395453215, 0.008229871280491352, -0.03661426529288292, 0.005215031560510397, 0.0012764959828928113, 0.013175491243600845, 0.02907075360417366, 0.0025481809861958027, -0.004807707387953997, -0.012739301659166813, 0.0014464816777035594, -0.0017463619587942958, -0.03435634449124336, 0.004156630020588636, -0.0032634036615490913, -0.016818957403302193, -0.018127525225281715, 0.015061370097100735, 0.00990406982600689, -0.0008860100642777979, 0.01276495959609747, 0.0259789377450943, 0.01045572105795145, 0.034407660365104675, 0.02747994288802147, -0.042566969990730286, 0.014407085254788399, -0.00013610879250336438, -0.013970895670354366, -0.01984662562608719, 0.003755720565095544, -0.020680518820881844, 0.01708836853504181, -0.010648157447576523, 0.005221446044743061, -0.035177405923604965, -0.0045286742970347404, -0.01708836853504181, 0.013034370727837086, -0.0012845142045989633, -0.0063087125308811665, 0.03338133171200752, 0.012290283106267452, -0.01463800948113203, -0.021296314895153046, -0.010789277963340282, 0.015766970813274384, 0.010949641466140747, 0.011308856308460236, -0.019833797588944435, 0.012848349288105965, 0.007152226287871599, 0.020205840468406677, -0.016331451013684273, 0.013855434022843838, -0.01640842668712139, -0.008107994683086872, -0.010872666724026203, -0.007242029998451471, -0.006908473093062639, 0.004240019246935844, -0.025119388476014137, 0.00765256118029356, 0.0022595261689275503, -0.010179894976317883, -0.009474294260144234, 0.002902585081756115, -0.00389684084802866, -0.021873624995350838, -0.0032698181457817554, 0.023130876943469048, -0.019577214494347572, -0.004086070228368044, -0.006052130367606878, 0.009134323336184025, -0.015125514939427376, -0.0040604118257761, 0.020808808505535126, -0.010808520950376987, 0.016652178019285202, -0.021206511184573174, -0.014625180512666702, 0.002657228382304311, -0.002690904773771763, 0.007165055256336927, -0.03312474861741066, 0.024208521470427513, -0.013060029596090317, 0.010077262297272682, 0.019012734293937683, -0.014304452575743198, 0.033791862428188324, -0.02046242356300354, 0.009673145599663258, 0.019679848104715347, 0.004188702907413244, 0.0067865969613194466, -0.028480613604187965, -0.029686549678444862, -0.0015900072176009417, 0.013791288249194622, -0.0070880805142223835, 0.03576754406094551, 0.022964099422097206, -0.026402298361063004, 0.02801876701414585, -0.022348301485180855, -0.000972606532741338, 0.0023493298795074224, 0.0028400432784110308, 0.004807707387953997, -0.0026235519908368587, 0.02513221837580204, 0.02727467752993107, 0.007575586903840303, -0.018178842961788177, 0.015818286687135696, -0.021142365410923958, 0.018538057804107666, -0.02119368128478527, 0.027428627014160156, -0.016626520082354546, -0.004506223369389772, -0.014445573091506958, 0.018704835325479507, 0.020411105826497078, 0.035536620765924454, -0.026761513203382492, -0.032072763890028, 0.005173336714506149, 0.004531881306320429, 0.036793872714042664, 3.102037589997053e-05, 0.0017367401160299778, 0.006250981707125902, 5.0188864406663924e-05, 0.023733844980597496, -0.015497559681534767, -0.0013462542556226254, -0.008255529217422009, -0.039077453315258026, 0.022438105195760727, -0.013380756601691246, 0.010487793944776058, 0.002185758901759982, -0.026864146813750267, -0.0071714697405695915, 0.02582498826086521, -0.006501149386167526, 0.0024663954973220825, 0.02524768002331257, -0.0340997613966465, -0.005564624443650246, -0.022835807874798775, 0.008512111380696297, 0.005567831918597221, -0.007004691753536463, 0.028711538761854172, 0.022104548290371895, 0.014253135770559311, 0.012604596093297005, -0.01599789410829544, 0.027454284951090813, 0.003919291775673628, -0.026735855266451836, 0.00616117799654603, 0.0073767355643212795, 0.0033099092543125153, -0.027556918561458588, 0.006097032222896814, 0.020244328305125237, 0.006318334490060806, 0.008993202820420265, -0.0174604132771492, 0.014676496386528015, 0.017075538635253906, 0.017203830182552338, 0.007190713658928871, 0.015895262360572815, 0.0021151986438781023, 0.005086740478873253, -0.0014793562004342675, -0.013675826601684093, -0.024939781054854393, -0.019949259236454964, -0.0022739588748663664, 0.0009325155988335609, -0.026043083518743515, 0.00585648650303483, 0.007030349690467119, -0.0026059120427817106, -0.010385161265730858, -0.015689995139837265, 0.01193748228251934, -0.0043073720298707485, -0.01148846372961998, -0.033920153975486755, -0.0259789377450943, 0.02991747297346592, 0.011841264553368092, -0.006587745621800423, 0.0064370036125183105, -0.01544624287635088, 0.022361131384968758, -0.011090761981904507, -0.008403063751757145, 0.02554274909198284, 0.014445573091506958, 0.0013510651187971234, 0.010301771573722363, -0.0035504549741744995, 0.026761513203382492, 0.003954571671783924, -0.02329765446484089, 0.014137674123048782, 0.005426711402833462, 0.012245381250977516, -0.025966109707951546, -0.01708836853504181, -0.008871326223015785, 0.03145696595311165, 0.009756534360349178, -0.007011106237769127, 0.00013781266170553863, -0.00096539017977193, 0.005487649701535702, 0.003014839719980955, 0.005817999131977558, -0.0025032791309058666, 0.004454907029867172, 0.030225371941924095, -0.012816276401281357, -0.015330781228840351, -0.004721110686659813, 0.009102250449359417, 0.0011610339861363173, -0.013714313507080078, -0.05208616703748703, -0.04315711185336113, 0.01729363389313221, 0.05216314271092415, 0.01325246598571539, 0.016947248950600624, 0.008524940349161625, 0.0024663954973220825, -0.024182863533496857, -0.03697348013520241, 0.003208880079910159, 8.890168828656897e-05, -0.021129537373781204, 0.012534036301076412, -0.009121494367718697, 0.01232235599309206, -0.005317664239555597, 0.012976639904081821, -0.019949259236454964, 0.012957395985722542, -0.017588702961802483, 0.027916133403778076, -0.019102538004517555, 0.014920249581336975, 0.012636668980121613, 0.02127065695822239, 0.019551556557416916, 0.0340997613966465, 0.02363121137022972, 0.04351632669568062, 0.025504261255264282, 0.010000287555158138, -0.02642795629799366, -0.011828435584902763, 0.006218908820301294, -0.0006951771210879087, 0.0045126378536224365, -0.02034696191549301, -0.006462662015110254, 0.02955825813114643, 0.006122690625488758, 0.01676764152944088, -0.011828435584902763, 0.005975155625492334, 0.021052561700344086, -0.009467880241572857, -0.01712685637176037, 0.017896601930260658, -0.026966780424118042, -0.017139684408903122, 0.04136103764176369, -0.01558736339211464, -0.023887794464826584, 0.0255812369287014, 0.0028288178145885468, 0.006857156753540039, -0.03340699151158333, 0.021989086642861366, 0.009224127046763897, 0.009467880241572857, -0.0012612614082172513, -0.013027956709265709, -0.017601532861590385, -0.00846720952540636, -0.02196342870593071, -0.008332503959536552, -0.009089421480894089, -0.01669066585600376, -0.013342269696295261, -0.0176528487354517, -0.011475634761154652, -0.022540738806128502, 0.010173480957746506, 0.010847008787095547, -0.012829105369746685, 0.007395979482680559, -0.0013582815881818533, 0.01985945552587509, 0.010616084560751915, -0.0229384396225214, -0.008396649733185768, 0.020116036757826805, 0.022964099422097206, -0.016793299466371536, -0.015600192360579967, 0.002857683226466179, -0.00562235526740551, -0.010096506215631962, 0.010442892089486122, 0.009384490549564362, -0.035741887986660004, 0.020372619852423668, 0.0012748923618346453, -0.016536716371774673, 0.024144375696778297, 0.0035312112886458635, -0.006420967169106007, -0.006241359747946262, -0.007274102885276079, -0.005288798827677965, -0.0012805050937458873, -0.01652388833463192, -0.03604978695511818, -0.014830445870757103, 0.015317952260375023, -0.01439425628632307, -0.007062422577291727, -0.003717233194038272, 0.01530512236058712, 0.0011818812927231193, -0.011007372289896011, 0.0003121080808341503, 0.007594830356538296, 0.019577214494347572, -0.01007084734737873, 0.005054667592048645, 0.022155864164233208, -0.02464471198618412, 0.019782479852437973, -0.008415892720222473, 0.023810818791389465, -0.0044516995549201965, -0.013483390212059021, 0.0035344185307621956, -0.02383647859096527, 0.008184969425201416, -0.0033580183517187834, 0.012950981967151165, -0.0018522021127864718, 0.0011353758163750172, 0.0035023458767682314, -0.006869985722005367, -0.010686644352972507, -0.010288942605257034, 0.02209172025322914, -0.0031655817292630672, -0.02996879070997238, -0.008184969425201416, -0.016049209982156754, 0.01286759227514267, 0.006882815156131983, -0.01534361019730568, -0.02663322351872921, -0.0021617042366415262, -0.008178554475307465, 3.533015478751622e-05, 0.0031046434305608273, -0.0066198185086250305, -0.02991747297346592, 0.0032185018062591553, -0.0324832946062088, 0.004830158315598965, 0.03543398901820183, -0.006206079851835966, -0.014817616902291775, -0.016049209982156754, 0.006026472430676222, -0.016331451013684273, -0.03153394162654877, -0.026607563719153404, -0.015882432460784912, -0.005872522946447134, 0.042361706495285034, 0.023246338590979576, 0.0025674246717244387, 0.023438775911927223, -0.013829775154590607, -0.022874295711517334, 0.006241359747946262, -0.004833365324884653, 0.005878937430679798, -0.013662997633218765, 0.018845954909920692, 0.045645955950021744, 0.015933748334646225, -0.018050551414489746, -0.0034510293044149876, -0.010115749202668667, 0.031200384721159935, -0.009166396223008633, -0.022296985611319542, -0.01138583105057478, -0.022399617359042168, 0.022258497774600983, 0.00609061773866415, -0.02034696191549301, 0.025029584765434265, 0.04243868216872215, -0.013290952891111374, 0.01660086214542389, -0.009269028902053833, 0.018499569967389107, 0.016780469566583633, 0.009467880241572857, -0.023477263748645782, 0.01216199155896902, -0.022887123748660088, -0.0064754909835755825, -0.014868932776153088, -0.012918909080326557, 0.01221972331404686, -0.02399042621254921, -0.0015306726563721895, 0.008993202820420265, 0.009262613952159882, -0.006145141553133726, -0.02159138396382332, -0.01359885185956955, 0.010808520950376987, -0.0158311165869236, 0.0031944471411406994, 0.005423504393547773, -0.010641742497682571, -0.019230829551815987, -0.01965419016778469, -0.020808808505535126, -0.022553566843271255, -0.00031290989136323333, 0.008685304783284664, -0.03953930363059044, 0.02529899589717388, -0.02062920108437538, -0.027736525982618332, 0.007447295822203159, -0.0034991384018212557, 0.02927601896226406, -0.002690904773771763, 0.022630542516708374, -0.0075820013880729675, 0.010628913529217243, -0.03130301833152771, 0.0015370871406048536, -0.0008515318622812629, 0.004493393935263157, 0.0011746649397537112, 0.006767353042960167, 0.030789852142333984, -0.022669028490781784, 0.007966874167323112, 0.017922259867191315, -0.012258210219442844, -0.028121398761868477, 0.020616373047232628, 0.0074152229353785515, 0.017486071214079857, -0.026530589908361435, -0.022887123748660088, -0.015779798850417137, 0.005946290213614702, -0.010134993121027946, 0.015600192360579967, -0.008370990864932537, -0.003912877291440964, -0.01920517161488533, 0.030174056068062782, -0.02866022102534771, 0.016960076987743378, -0.0009373265202157199, -0.011693730019032955, 0.0024744137190282345, -0.03438200429081917, -0.036845192313194275, 0.013778459280729294, 0.010423648171126842, 0.03720440715551376, -0.009435807354748249, 0.018730493262410164, 0.0051861656829714775, 0.014663667418062687, -0.014779129065573215, -0.003261800156906247, 0.012931738048791885, 0.020321302115917206, -0.02894246205687523, -0.03982154279947281, -0.013027956709265709, -0.012348013930022717, 0.008787937462329865, -0.03263724595308304, -0.016434084624052048, 0.023605553433299065, -0.00012598582543432713, -0.01669066585600376, 0.00453829625621438, 0.006190043408423662, 0.004076448269188404, 0.009346003644168377, 0.004714696202427149, -0.009269028902053833, -0.028121398761868477, 0.00096539017977193, 0.006196457892656326, -0.012091431766748428, -0.017062710598111153, -0.028403639793395996, 0.0158311165869236, -0.00407965574413538, 0.012213308364152908, -0.013650167733430862, 8.679691381985322e-05, 0.00011275580618530512, -0.021706845611333847, 0.02098841592669487, -0.03250895440578461, 0.013509048148989677, 0.004486979451030493, 0.006818669382482767, -0.015189660713076591, 0.004679416306316853, 0.018050551414489746, 0.003131905337795615, -0.005478028208017349, -0.018807468935847282, 0.0045511252246797085, 0.017383437603712082, 0.01033384446054697, -0.0013133796164765954, 0.03284250944852829, 0.007120153401046991, -0.0002808371209539473, -0.0225279089063406, -0.018910100683569908, 0.022681858390569687, -0.007857827469706535, -0.0041854954324662685, -0.008326089009642601, -0.023900622501969337, 0.016472570598125458, 0.01903839223086834, -0.0012067377101629972, 0.005118812900036573, -0.010891910642385483, 0.01415050309151411, -0.016536716371774673, 0.020885784178972244, -0.000979822943918407, -0.00815931148827076, -0.005192580632865429, 0.010693059302866459, -0.015433413907885551, 0.011905409395694733, -0.01737060956656933, -0.0032521781977266073, 0.0033259454648941755, 0.058603353798389435, 0.004692245274782181, 0.008236285299062729, 0.0031463380437344313, -0.019833797588944435, -0.02301541529595852, -0.022309813648462296, -0.0017174964305013418, -0.02359272539615631, -0.010077262297272682, 0.01162958424538374, 0.009878410957753658, -0.013868262991309166, -0.02354140765964985, 0.03019971400499344, -0.015972236171364784, -0.0067032077349722385, -0.015215318650007248, 0.004316993989050388, 0.02472168579697609, 0.006780182011425495, -0.03345830738544464, 0.012277454137802124, 0.01171297300606966, 0.011180565692484379, 0.008447965607047081, 0.004608856048434973, 0.014137674123048782, -0.0196926761418581, 0.018858784809708595, -0.002015773206949234, -0.027659550309181213, -0.02025715820491314, 0.008499282412230968, 0.007973289117217064, 0.025837818160653114, 0.0005063487333245575, -0.013534706085920334, -0.02408022992312908, -0.044722262769937515, -0.01232235599309206, 0.012617425061762333, 0.005747439339756966, 0.0075820013880729675, -0.017396267503499985, -0.014830445870757103, 0.0022771661169826984, 0.00034357947879470885, -0.028249690309166908, -0.005426711402833462, 0.008909814059734344, 0.007049593608826399, 0.021052561700344086, -0.0023284826893359423, -0.0052920058369636536, 0.00941656343638897, -0.026094401255249977, 0.00616117799654603, 0.003858353476971388, -0.01684461534023285, 0.008961129933595657, -0.009673145599663258, -0.011963141150772572, -0.014009383507072926, -0.019153853878378868, -0.01697290688753128, -0.007594830356538296, 0.02712072804570198, -0.00564159918576479, -0.00581479212269187, -0.021527238190174103, 0.006728865671902895, -0.0043073720298707485, 0.020552227273583412, -0.003710818709805608, 0.012918909080326557, 0.015882432460784912, 0.005898181349039078, 0.014432743191719055, -0.019461752846837044, 0.00024976665736176074, 0.0008150491048581898, 0.0021264241077005863, -0.008306846022605896, -0.005381809547543526, 0.010410819202661514, -0.010173480957746506, -0.002830421319231391, 0.019436094909906387, -0.017563045024871826, -0.020770322531461716, 0.004037960898131132, -0.018730493262410164, -0.00880718044936657, 0.011745045892894268, 0.18124960362911224, 0.010731546208262444, -0.0013294160598888993, 0.03794849291443825, 0.010956056416034698, 0.03723006322979927, 0.02935299277305603, 0.005340115167200565, 0.01033384446054697, 0.030558928847312927, -0.001111321267671883, 0.016754811629652977, -0.010641742497682571, 0.0006975826108828187, 0.00545878428965807, -0.01688310317695141, -0.014868932776153088, -0.025183534249663353, -0.009166396223008633, -0.013534706085920334, -0.010507036931812763, -0.019820967689156532, -0.013444902375340462, -0.0038294880650937557, 0.029583917930722237, 0.0178581140935421, -0.017973575741052628, -0.01335509866476059, 0.034895166754722595, 0.019346291199326515, 0.008088750764727592, -0.00846079457551241, 0.0025722356513142586, -0.00622211629524827, -0.02042393572628498, 0.014843274839222431, -0.005320871248841286, -0.018974246457219124, 0.009711632505059242, -0.012822690419852734, -0.007197128143161535, -0.00787707045674324, 0.008358161896467209, -0.03320172429084778, -0.020475251600146294, 0.0036113932728767395, -0.006940545979887247, 0.0025129010900855064, 0.005933461245149374, 0.008499282412230968, -0.019051222130656242, -0.026684539392590523, 0.0005632779211737216, 0.0278134997934103, 0.0119759701192379, 0.00838382076472044, 0.020924270153045654, 0.008916228078305721, 0.0006194052402861416, 0.005465198773890734, 0.003152752760797739, 0.03515174984931946, -0.013239637017250061, 0.029455626383423805, -0.017357779666781425, 0.016472570598125458, -0.01102020125836134, -0.019820967689156532, -0.0022017951123416424, -0.01359885185956955, 0.035741887986660004, -0.022476593032479286, -0.010705888271331787, -0.011110004968941212, -0.009179225191473961, -0.01716534234583378, 0.00860833004117012, 0.02606874145567417, -0.0032970800530165434, 0.019872283563017845, -0.014714984223246574, 0.007575586903840303, -0.01680612750351429, -0.0025658211670815945, -0.01439425628632307, -0.0373583547770977, 0.014843274839222431, -0.007421637419611216, -0.01692158915102482, -0.02561972290277481, -0.0077038779854774475, -0.0021151986438781023, -0.008281187154352665, 0.012809861451387405, 0.01684461534023285, -0.005240689497441053, -0.03835902363061905, 0.013765630312263966, 0.014265964739024639, -0.008075921796262264, -0.01684461534023285, -0.03864126652479172, 0.016010724008083344, 0.0005817197379656136, -0.025324653834104538, -0.016062039881944656, -0.013996553607285023, 0.012232552282512188, 0.008967544883489609, -0.007344662677496672, 0.014342939481139183, 0.004150215536355972, -0.004124557599425316, -0.005811584647744894, -0.006940545979887247, -0.008800766430795193, 0.008653231896460056, -0.010276113636791706, 0.021976256743073463, -0.012745716609060764, 0.03807678446173668, -0.003932120744138956, 0.0017559838015586138, 0.020885784178972244, -0.04903284087777138, -0.02650493197143078, -0.02160421386361122, 0.004836572799831629, 0.015420584939420223, -0.03271421790122986, 0.014330110512673855, -0.024696027860045433, -0.019872283563017845, 0.0034510293044149876, -0.006882815156131983, 0.000971804722212255, 0.006084203254431486, 0.01888444274663925, -0.029378650709986687, 0.001085663097910583, -0.007081666029989719, -0.0034927239175885916, 0.014253135770559311, -0.014330110512673855, -0.004567161668092012, -0.018409766256809235, 8.809987048152834e-05, -0.0016373145626857877, -0.0082234563305974, -0.010045189410448074, -0.0034959311597049236, -0.020359789952635765, -0.010276113636791706, -0.005295213311910629, 0.016087697818875313, -0.040411680936813354, -0.03256027027964592, -0.004791670944541693, -0.02062920108437538, 0.005747439339756966, -0.014009383507072926, -0.014060699380934238, 0.024991096928715706, -0.006174006965011358, -0.007434466388076544, -0.022515080869197845, -0.16339148581027985, 0.029609575867652893, 0.006722451187670231, -0.037332694977521896, 0.021976256743073463, -0.022361131384968758, -0.0043458594009280205, -0.005083533003926277, 0.009403734467923641, 0.0114692198112607, 0.020475251600146294, -0.01242498867213726, -0.04600517079234123, -0.031610917299985886, 0.02942996844649315, -0.007530685048550367, -0.0010920775821432471, 0.019667018204927444, 0.007312590256333351, 0.022194352000951767, 0.013983724638819695, -0.006613404024392366, 0.018358450382947922, -0.01757587492465973, -0.0015042126178741455, 0.021334802731871605, 0.004499808885157108, 0.01993642933666706, -0.014368598349392414, 0.008563428185880184, 0.012418573722243309, 0.0027261849027127028, 0.014766300097107887, 0.03820507600903511, 0.0012227741535753012, -0.01509985700249672, -0.01033384446054697, -0.0005817197379656136, 0.0033291529398411512, 0.016023552045226097, 0.010147822089493275, 0.027736525982618332, -0.002987577812746167, -0.0002890557807404548, -0.01903839223086834, 0.007235615514218807, 0.02907075360417366, -0.006546051241457462, 0.011000958271324635, -0.0002021586406044662, 0.0162416473031044, -0.013393585570156574, 0.0011882958933711052, 0.008043848909437656, 0.020282816141843796, 0.012328770011663437, -0.0011650430969893932, -0.0016252873465418816, 0.005708951968699694, 0.005198995117098093, -0.003941742703318596, 0.0032024653628468513, 0.014458402059972286, -0.00858267117291689, -0.00953844003379345, -0.02184796705842018, -0.02297692745923996, 0.010122164152562618, 0.0176656786352396, 0.005939875729382038, -0.004679416306316853, -0.002729392144829035, 0.0031607707496732473, -0.010494207963347435, 0.009461465291678905, 0.009929727762937546, -0.0269924383610487, -0.00016016335575841367, 0.009711632505059242, -0.01721666008234024, -0.0063087125308811665, -0.008204213343560696, -0.0033355674240738153, 0.012687984853982925, 0.004515845328569412, 0.005776304751634598, -0.009666730649769306, 0.0033387746661901474, -0.007780852261930704, -0.010115749202668667, 0.028506271541118622, -0.0077038779854774475, -0.007492197677493095, -0.02209172025322914, -0.011360173113644123, 0.0118733374401927, 0.012630254030227661, 0.011918239295482635, 0.007973289117217064, -0.009647487662732601, 0.01834562048316002, -0.002833628561347723, -0.019782479852437973, 0.011340929195284843, 0.019089708104729652, 0.014201819896697998, 0.0023236717097461224, 0.010186309926211834, 0.023977598175406456, 0.0018618239555507898, 0.005205409601330757, 0.033637914806604385, 0.013662997633218765, 0.00544916233047843, -0.015510388650000095, 0.03186749666929245, -0.005378602538257837, -0.03235500305891037, 0.013752801343798637, 0.0006743298727087677, -0.004426041152328253, -0.012495548464357853, 0.010256869718432426, 0.008531355299055576, 0.0034317856188863516, -0.0025112973526120186, -0.07861676067113876, -0.007062422577291727, 0.0063311634585261345, 0.02537596970796585, -0.02704375423491001, 0.027864817529916763, 0.004207946360111237, 0.013329440727829933, -0.01579262875020504, 0.01815318502485752, 0.007120153401046991, -0.02359272539615631, -0.02627400867640972, -0.013752801343798637, 0.015459071844816208, 0.029301676899194717, 0.0025161083322018385, -0.003045308869332075, -0.017601532861590385, -0.0025818573776632547, -0.009910483844578266, -0.010410819202661514, -0.011167735792696476, 0.0014865725534036756, -0.009756534360349178, 0.009429392404854298, -0.014278794638812542, 0.007960460148751736, 0.007152226287871599, -0.001069626654498279, 0.018576543778181076, -0.012874007225036621, 0.002876926911994815, -0.04015510156750679, -4.9462214519735426e-05, -0.0053080422803759575, 0.007819339632987976, -0.001986907795071602, 0.012354428879916668, -0.01244423259049654, 0.01266874186694622, 0.0017367401160299778, -0.011693730019032955, -0.047108475118875504, 0.020513739436864853, -0.013855434022843838, -0.007004691753536463, 0.005785926710814238, 0.029455626383423805, -0.03610110282897949, -0.005981570575386286, 0.00186984206084162, -0.013791288249194622, -0.0074537103064358234, 0.014933078549802303, -0.02078315056860447, 0.012315941043198109, -0.025812160223722458, -0.020026233047246933, 0.007883485406637192, -0.011270369403064251, -0.020885784178972244, -0.023772332817316055, 0.0340997613966465, 0.007941216230392456, 0.013470560312271118, -0.025914791971445084, -0.050854574888944626, 0.0158311165869236, -0.01140507496893406, -0.01242498867213726, -0.007203542627394199, -0.014522546902298927, -0.006959789898246527, -0.015407755970954895, -0.01896141842007637, -0.024708857759833336, 0.0023733845446258783, 0.020847296342253685, -0.029045093804597855, -0.010167066007852554, -0.021283484995365143, 0.010391575284302235, -0.018640689551830292, 0.012104260735213757, 0.018012063577771187, 0.012072187848389149, 0.009878410957753658, -0.012392915785312653, -0.01875615119934082, 0.009038104675710201, -0.0017110819462686777, 0.03325304016470909, -0.020654859021306038, -0.02087295427918434, 0.02010320872068405, -0.007235615514218807, -0.01566433720290661, 0.013778459280729294, -0.023823648691177368, -0.013290952891111374, 0.008775108493864536, -0.044593971222639084, 0.0067032077349722385, -0.018089039251208305, -0.015369268134236336, 0.013329440727829933, -0.008563428185880184, -0.00937807559967041, 0.005975155625492334, 0.0062926760874688625, 0.015561704523861408, -0.012803447432816029, 0.009576926939189434, 0.010372331365942955, -0.001885878504253924, 0.0023894209880381823, -0.005628770217299461, 0.02927601896226406, 0.014214648865163326, 0.01672915369272232, -0.001007084734737873, 0.019795309752225876, 0.013675826601684093, -0.0036242222413420677, -0.0113665871322155, 0.003111058147624135, 0.008845668286085129, -0.005073911044746637, 0.028531929478049278, -0.010391575284302235, -0.0194874107837677, 0.019064050167798996, -0.019269315525889397, -0.012578938156366348, 0.032457634806632996, 0.0022819770965725183, -0.02359272539615631, -0.003890426130965352, 0.015266635455191135, -0.0046954527497291565, 0.013079272583127022, -0.0031190761364996433, -0.02655624784529209, 0.019679848104715347, -0.025914791971445084, -0.014817616902291775, -0.014022212475538254, -0.008685304783284664, 0.00998745858669281, -0.005138056818395853, 0.01599789410829544, 0.029096411541104317, 0.011430732905864716, 0.013470560312271118, -0.03779454529285431, 0.0014408689457923174, -0.02139894850552082, 0.010564768686890602, 0.02732599526643753, 0.008787937462329865, 0.001094483071938157, 0.02256639674305916, 0.006472283508628607, -0.025145046412944794, -0.016536716371774673, -0.024991096928715706, -0.010956056416034698, -0.010244040749967098, 0.0061643850058317184, 0.023310484364628792, -0.02789047546684742, -0.022438105195760727, 0.007934801280498505, 0.0033034945372492075, -0.007524270098656416, 0.018935760483145714, 0.003867975203320384, 0.007267688401043415, 0.001003877492621541, -0.0003912877000402659, 0.0442604124546051, 0.019923601299524307, 0.026812830939888954, -0.020885784178972244, 0.03758927807211876, 0.0017784347292035818, -0.0034702729899436235, 0.0004931187140755355, 0.0014713379787281156, 0.01404787041246891, 0.0005945488810539246, 0.016139013692736626, 0.0022739588748663664, 0.0276082344353199, -0.0026524176355451345, 0.020924270153045654, 0.016190331429243088, -0.007299760822206736, -0.027556918561458588, 0.00996821466833353, 0.018076209351420403, 0.019743993878364563, -0.010442892089486122, -0.01708836853504181, -0.013483390212059021, -0.011905409395694733, -0.014060699380934238, -0.04857099428772926, -0.018858784809708595, -0.025606894865632057, 0.0043234084732830524, 0.025812160223722458, 0.010507036931812763, 0.016382766887545586, 0.030071422457695007, -0.01530512236058712, 0.004288128577172756, -0.02590196393430233, -0.002894566860049963, 0.00810157973319292, 0.0013117759954184294, 0.002196984365582466, 0.015292293392121792, 0.009564097970724106, -0.0024022499565035105, -0.007081666029989719, 0.002812781371176243, 0.002102369675412774, -0.03096945956349373, 0.015420584939420223, 0.009589755907654762, -0.0064754909835755825, -0.0073318337090313435, -0.002026998670771718, -0.015394926071166992, -0.02712072804570198, 0.00927544292062521, 0.015035711228847504, 0.020167354494333267, -0.019820967689156532, 0.04213078320026398, 0.01640842668712139, 0.010244040749967098, -0.002860890468582511, 0.021078219637274742, 0.01858937367796898, 0.0047179036773741245, -0.013162662275135517, -0.014894591644406319, -0.013053614646196365, 0.004884681664407253, 0.014407085254788399, 0.007594830356538296, -0.001111321267671883, -0.027582576498389244, -0.0024455483071506023, -0.021655529737472534, -0.007505026645958424, -0.009429392404854298, -0.016177501529455185, 0.0072484444826841354, 0.017139684408903122, 0.014561034739017487, 0.007075251545757055, -0.02289995364844799, 0.006411345209926367, 0.018768981099128723, 0.0026347774546593428, -0.008871326223015785, -0.03989851847290993, 0.008204213343560696, 0.013060029596090317, -0.03299646079540253, -0.015536046586930752, 0.02464471198618412, -0.02894246205687523, -0.006485112942755222, 0.008768693543970585, 0.016831785440444946, 0.035664912313222885, -0.0032938728109002113, 0.022425275295972824, -0.02050090953707695, -0.00025557982735335827, 0.013239637017250061, -0.016652178019285202, 0.004143801052123308, -0.0032072763424366713, -0.025273337960243225]      }    ) {      question      answer    }  }}
Example response

The output is like this:

JSON
{
  "data": {
    "Get": {
      "JeopardyQuestion": [
        {
          "answer": "Risotto",
          "question": "From the Italian word for rice, it's a rice dish cooked with broth & often grated cheese"
        },
        {
          "answer": "arrabiata",
          "question": "Italian for \"angry\", it describes a pasta sauce spiced up with plenty of chiles"
        },
        {
          "answer": "Fettucine Alfredo",
          "question": "Ribbon-shaped noodles, sweet butter, cream, parmesan cheese & black pepper make up this pasta dish"
        }
      ]
    }
  }
}

Vector search parameters

Section titled “Vector search parameters”

You can specify vector similarity search parameters similar to near text or near vector searches, such as group by and move to / move away. An equivalent distance threshold for vector search can be specified with the max vector distance parameter.

Python
from weaviate.classes.query import HybridVector, Move, HybridFusionjeopardy = client.collections.use("JeopardyQuestion")response = jeopardy.query.hybrid(    query="California",    max_vector_distance=0.4,  # Maximum threshold for the vector search component    vector=HybridVector.near_text(        query="large animal",        move_away=Move(force=0.5, concepts=["mammal", "terrestrial"]),    ),    alpha=0.75,    limit=5,)
JavaScript/TypeScript
const jeopardy = client.collections.use('JeopardyQuestion');
Java
CollectionHandle<Map<String, Object>> jeopardy =    client.collections.use("JeopardyQuestion");var intermediateResponse = jeopardy.query    .nearText("large animal",        c -> c.moveAway(0.5f,            from -> from.concepts("mammal", "terrestrial")))    .objects()    .get(0)    .vectors()    .getDefaultSingle();var response = jeopardy.query.hybrid("California", q -> q    .maxVectorDistance(0.4f)    .nearVector(NearVector.of(intermediateResponse))    .alpha(0.75f)    .limit(5));
C#
var jeopardy = client.Collections.Use("JeopardyQuestion");var response = await jeopardy.Query.Hybrid(    "California",    maxVectorDistance: 0.4f,    vectors: v =>        v.NearText(            "large animal",            moveAway: new Move(force: 0.5f, concepts: ["mammal", "terrestrial"])        ),    alpha: 0.75f,    limit: 5);
Example response

The output is like this:

JSON
{
  "data": {
    "Get": {
      "JeopardyQuestion": [
        {
          "answer": "Rhinoceros",
          "points": 400,
          "question": "The \"black\" species of this large horned mammal can grasp twigs with its upper lip"
        },
        {
          "answer": "the hippopotamus",
          "points": 400,
          "question": "Close relative of the pig, though its name means \"river horse\""
        },
        {
          "answer": "buffalo",
          "points": 400,
          "question": "Animal that was the main staple of the Plains Indians economy"
        },
        {
          "answer": "California",
          "points": 200,
          "question": "Its state animal is the grizzly bear, & the state tree is a type of redwood"
        },
        {
          "answer": "California",
          "points": 200,
          "question": "This western state sent its first refrigerated trainload of oranges back east February 14, 1886"
        }
      ]
    }
  }
}

Hybrid search thresholds

Section titled “Hybrid search thresholds”

The only available search threshold is max vector distance, which will set the maximum allowable distance for the vector search component.

Python
from weaviate.classes.query import HybridVector, Move, HybridFusionjeopardy = client.collections.use("JeopardyQuestion")response = jeopardy.query.hybrid(    query="California",    max_vector_distance=0.4,  # Maximum threshold for the vector search component    alpha=0.75,    limit=5,)
JavaScript/TypeScript
const jeopardy = client.collections.use('JeopardyQuestion');
Java
CollectionHandle<Map<String, Object>> jeopardy =    client.collections.use("JeopardyQuestion");var response = jeopardy.query.hybrid("California", q -> q    .maxVectorDistance(0.4f) // Maximum threshold for the vector search component    .alpha(0.75f)    .limit(5));
C#
var jeopardy = client.Collections.Use("JeopardyQuestion");var response = await jeopardy.Query.Hybrid(    "California",    maxVectorDistance: 0.4f, // Maximum threshold for the vector search component    alpha: 0.75f,    limit: 5);

Group results

Section titled “Group results”

Define criteria to group search results.

Python
# Grouping parameters
group_by = GroupBy(
    prop="round",  # group by this property
    objects_per_group=3,  # maximum objects per group
    number_of_groups=2,  # maximum number of groups
)

# Query
jeopardy = client.collections.use("JeopardyQuestion")
response = jeopardy.query.hybrid(
    alpha=0.75,
    query="California",
    group_by=group_by
)

for grp_name, grp_content in response.groups.items():
    print(grp_name, grp_content.objects)
JavaScript/TypeScript
const jeopardy = client.collections.use('JeopardyQuestion');
Java
// Query
CollectionHandle<Map<String, Object>> jeopardy =
    client.collections.use("JeopardyQuestion");
var response = jeopardy.query.hybrid("California", q -> q.alpha(0.75f),
    GroupBy.property("round", // group by this property
        2, // maximum number of groups
        3 // maximum objects per group
    ));

response.groups().forEach((groupName, group) -> {
  System.out.println(group.name() + " " + group.objects());
});
C#
// Query
var jeopardy = client.Collections.Use("JeopardyQuestion");
var response = await jeopardy.Query.Hybrid(
    "California",
    alpha: 0.75f,
    groupBy: new GroupByRequest("round") // group by this property
    {
        NumberOfGroups = 2, // maximum number of groups
        ObjectsPerGroup = 3, // maximum objects per group
    }
);

foreach (var group in response.Groups.Values)
{
    Console.WriteLine($"{group.Name} {JsonSerializer.Serialize(group.Objects)}");
}
Example response

The response is like this:

'Jeopardy!'
'Double Jeopardy!'

limit & offset

Section titled “limit & offset”

Use limit to set a fixed maximum number of objects to return.

Optionally, use offset to paginate the results.

Python
jeopardy = client.collections.use("JeopardyQuestion")response = jeopardy.query.hybrid(    query="food",    limit=3,    offset=1)for o in response.objects:    print(o.properties)
JavaScript/TypeScript
const jeopardy = client.collections.use('JeopardyQuestion');
Go
ctx := context.Background()
className := "JeopardyQuestion"
query := "safety"
limit := 3

result, err := client.GraphQL().Get().
  WithClassName(className).
  WithFields(
    graphql.Field{Name: "question"},
    graphql.Field{Name: "answer"},
    graphql.Field{Name: "_additional", Fields: []graphql.Field{{Name: "score"}}},
  ).
  WithHybrid(client.GraphQL().HybridArgumentBuilder().WithQuery(query)).
  WithLimit(limit).
  Do(ctx)
Java
CollectionHandle<Map<String, Object>> jeopardy =    client.collections.use("JeopardyQuestion");var response = jeopardy.query.hybrid("food", q -> q    .limit(3)    .offset(1));for (var o : response.objects()) {  System.out.println(o.properties());}
C#
var jeopardy = client.Collections.Use("JeopardyQuestion");var response = await jeopardy.Query.Hybrid(    "food",    limit: 3,    offset: 1);foreach (var o in response.Objects){    Console.WriteLine(JsonSerializer.Serialize(o.Properties));}
GraphQL
{  Get {    JeopardyQuestion(      hybrid: {        query: "safety"      }      limit: 3    ) {      question      answer      _additional {        score      }    }  }}

Limit result groups

Section titled “Limit result groups”

To limit results to groups with similar distances from the query, use the autocut filter. Specify the Relative Score Fusion ranking method when you use autocut with hybrid search.

Python
from weaviate.classes.query import HybridFusionjeopardy = client.collections.use("JeopardyQuestion")response = jeopardy.query.hybrid(    query="food",    fusion_type=HybridFusion.RELATIVE_SCORE,    auto_limit=1)for o in response.objects:    print(o.properties)
JavaScript/TypeScript
const jeopardy = client.collections.use('JeopardyQuestion');
Go
ctx := context.Background()
className := "JeopardyQuestion"
query := "safety"
autocut := 1

result, err := client.GraphQL().Get().
  WithClassName(className).
  WithFields(
    graphql.Field{Name: "question"},
    graphql.Field{Name: "answer"},
    graphql.Field{Name: "_additional", Fields: []graphql.Field{{Name: "score"}}},
  ).
  WithHybrid(client.GraphQL().HybridArgumentBuilder().WithQuery(query)).
  WithAutocut(autocut).
  Do(ctx)
Java
C#
var jeopardy = client.Collections.Use("JeopardyQuestion");var response = await jeopardy.Query.Hybrid(    "food",    fusionType: HybridFusion.RelativeScore,    autoLimit: 1);foreach (var o in response.Objects){    Console.WriteLine(JsonSerializer.Serialize(o.Properties));}
GraphQL
{  Get {    JeopardyQuestion(      hybrid: {        query: "safety"      }      autocut: 1    ) {      question      answer      _additional {        score      }    }  }}
Example response

The output is like this:

JSON
{
  "data": {
    "Get": {
      "JeopardyQuestion": [
        {
          "answer": "Guards",
          "question": "Life, Security, Shin",
          "_additional": {
            "score": "0.75"
          },
        },
        # ... trimmed for brevity
      ]
    }
  }
}

Filter results

Section titled “Filter results”

To narrow your search results, use a filter.

Python
from weaviate.classes.query import Filterjeopardy = client.collections.use("JeopardyQuestion")response = jeopardy.query.hybrid(    query="food",    filters=Filter.by_property("round").equal("Double Jeopardy!"),    limit=3,)for o in response.objects:    print(o.properties)
JavaScript/TypeScript
const jeopardy = client.collections.use('JeopardyQuestion');
Go
ctx := context.Background()
className := "JeopardyQuestion"
query := "food"
limit := 3

filter := filters.Where().
  WithPath([]string{"round"}).
  WithOperator(filters.Equal).
  WithValueString("Double Jeopardy!")

result, err := client.GraphQL().Get().
  WithClassName(className).
  WithFields(graphql.Field{Name: "question"}, graphql.Field{Name: "answer"}, graphql.Field{Name: "round"}).
  WithHybrid(client.GraphQL().HybridArgumentBuilder().WithQuery(query)).
  WithWhere(filter).
  WithLimit(limit).
  Do(ctx)
Java
CollectionHandle<Map<String, Object>> jeopardy =    client.collections.use("JeopardyQuestion");var response = jeopardy.query.hybrid("food", q -> q    .filters(Filter.property("round").eq("Double Jeopardy!"))    .limit(3));for (var o : response.objects()) {  System.out.println(o.properties());}
C#
var jeopardy = client.Collections.Use("JeopardyQuestion");var response = await jeopardy.Query.Hybrid(    "food",    filters: Filter.Property("round").IsEqual("Double Jeopardy!"),    limit: 3);foreach (var o in response.Objects){    Console.WriteLine(JsonSerializer.Serialize(o.Properties));}
GraphQL
{  Get {    JeopardyQuestion(      limit: 3      hybrid: {        query: "food"      }      where: {        path: ["round"]        operator: Equal        valueText: "Double Jeopardy!"      }    ) {      question      answer      round    }  }}
Example response

The output is like this:

JSON
{
  "data": {
    "Get": {
      "JeopardyQuestion": [
        {
          "answer": "food stores (supermarkets)",
          "question": "This type of retail store sells more shampoo & makeup than any other",
          "round": "Double Jeopardy!"
        },
        {
          "answer": "Tofu",
          "question": "A popular health food, this soybean curd is used to make a variety of dishes & an ice cream substitute",
          "round": "Double Jeopardy!"
        },
        {
          "answer": "gastronomy",
          "question": "This word for the art & science of good eating goes back to Greek for \"belly\"",
          "round": "Double Jeopardy!"
        }
      ]
    }
  }
}

Tokenization

Section titled “Tokenization”

Weaviate converts filter terms into tokens. The default tokenization is word. The word tokenizer keeps alphanumeric characters, lowercase them and splits on whitespace. It converts a string like "Test_domain_weaviate" into "test", "domain", and "weaviate".

For details and additional tokenization methods, see Tokenization.

Soft-rank with Boost

Section titled “Soft-rank with Boost”

Hybrid queries accept an optional boost argument that promotes or demotes matching documents without removing them. This is useful for biasing results by recency, popularity, a soft filter, or another property.

The boost runs once over the fused hybrid result. The BM25 and vector sub-search legs do not see the boost themselves. Hybrid's own alpha blend runs first, and the boost rescores the fused candidate pool on top.

See Boost for the supported condition types (filter, property value, time decay, numeric decay), curve choices, blending semantics, and depth tuning.

Diversity selection (MMR)

Section titled “Diversity selection (MMR)”

Hybrid search fuses a keyword result set and a vector result set, which often means the top of the fused list is a cluster of near-duplicates. Maximal Marginal Relevance (MMR) reranks that list to balance relevance with diversity, so that each selected object adds something new to the result set.

Diversity selection runs after fusion. Both search legs run first, their results are fused with the configured alpha and fusion method, and the diversity pass then picks a diverse subset of the fused candidates.

Python
from weaviate.classes.query import Diversitycollection = client.collections.get("MMRDemo")# Fuse the keyword and vector results into 20 candidates, then select 5 diverse resultsresponse = collection.query.hybrid(    query="Question",    vector=base_vec,    limit=20,    diversity_selection=Diversity.mmr(        limit=5,        balance=0.5,    ),)for o in response.objects:    print(o.properties["question"])

Important notes:

  • Top-level only: set diversity selection on the hybrid query itself. Setting it on a sub-search is rejected with an error.
  • Two limits: the query's top-level limit is the candidate window that gets diversified, and the diversity limit is the number of results returned. The diversity limit must be at least 1 and no larger than the query limit.
  • Ordering: results come back in MMR order, not fused-score order.
  • Pagination: offset moves the candidate window, so it must advance by the query limit, not by the number of returned objects. Weaviate does not validate this, and getting it wrong silently repeats some objects across pages while skipping others. See Pagination.
  • Not supported: multi-vector collections. Weaviate rejects these queries with an error.

For the parameters, the relevance and diversity trade-off, and vector search examples, see Diversity selection (MMR).

Section titled “Related pages”

Questions and feedback

Section titled “Questions and feedback”

Have a question or feedback? Here's how to reach us.

Community Forum

Ask questions and connect with other developers on our Community forum.

Support

Weaviate Cloud user or customer? Find the right channel on the Support page.

Suggest an edit

Propose a replacement for this page. The site team reviews it before applying any changes.

Export
Documentation menu