笛卡尔质心聚合
编辑笛卡尔质心聚合
编辑一种度量聚合，用于计算点和形状字段所有坐标值的加权质心。
示例
resp = client.indices.create(
index="museums",
mappings={
"properties": {
"location": {
"type": "point"
}
}
},
)
print(resp)
resp1 = client.bulk(
index="museums",
refresh=True,
operations=[
{
"index": {
"_id": 1
}
},
{
"location": "POINT (491.2350 5237.4081)",
"city": "Amsterdam",
"name": "NEMO Science Museum"
},
{
"index": {
"_id": 2
}
},
{
"location": "POINT (490.1618 5236.9219)",
"city": "Amsterdam",
"name": "Museum Het Rembrandthuis"
},
{
"index": {
"_id": 3
}
},
{
"location": "POINT (491.4722 5237.1667)",
"city": "Amsterdam",
"name": "Nederlands Scheepvaartmuseum"
},
{
"index": {
"_id": 4
}
},
{
"location": "POINT (440.5200 5122.2900)",
"city": "Antwerp",
"name": "Letterenhuis"
},
{
"index": {
"_id": 5
}
},
{
"location": "POINT (233.6389 4886.1111)",
"city": "Paris",
"name": "Musée du Louvre"
},
{
"index": {
"_id": 6
}
},
{
"location": "POINT (232.7000 4886.0000)",
"city": "Paris",
"name": "Musée d'Orsay"
}
],
)
print(resp1)
resp2 = client.search(
index="museums",
size="0",
aggs={
"centroid": {
"cartesian_centroid": {
"field": "location"
}
}
},
)
print(resp2)
response = client.indices.create(
index: 'museums',
body: {
mappings: {
properties: {
location: {
type: 'point'
}
}
}
}
)
puts response
response = client.bulk(
index: 'museums',
refresh: true,
body: [
{
index: {
_id: 1
}
},
{
location: 'POINT (491.2350 5237.4081)',
city: 'Amsterdam',
name: 'NEMO Science Museum'
},
{
index: {
_id: 2
}
},
{
location: 'POINT (490.1618 5236.9219)',
city: 'Amsterdam',
name: 'Museum Het Rembrandthuis'
},
{
index: {
_id: 3
}
},
{
location: 'POINT (491.4722 5237.1667)',
city: 'Amsterdam',
name: 'Nederlands Scheepvaartmuseum'
},
{
index: {
_id: 4
}
},
{
location: 'POINT (440.5200 5122.2900)',
city: 'Antwerp',
name: 'Letterenhuis'
},
{
index: {
_id: 5
}
},
{
location: 'POINT (233.6389 4886.1111)',
city: 'Paris',
name: 'Musée du Louvre'
},
{
index: {
_id: 6
}
},
{
location: 'POINT (232.7000 4886.0000)',
city: 'Paris',
name: "Musée d'Orsay"
}
]
)
puts response
response = client.search(
index: 'museums',
size: 0,
body: {
aggregations: {
centroid: {
cartesian_centroid: {
field: 'location'
}
}
}
}
)
puts response
const response = await client.indices.create({
index: "museums",
mappings: {
properties: {
location: {
type: "point",
},
},
},
});
console.log(response);
const response1 = await client.bulk({
index: "museums",
refresh: "true",
operations: [
{
index: {
_id: 1,
},
},
{
location: "POINT (491.2350 5237.4081)",
city: "Amsterdam",
name: "NEMO Science Museum",
},
{
index: {
_id: 2,
},
},
{
location: "POINT (490.1618 5236.9219)",
city: "Amsterdam",
name: "Museum Het Rembrandthuis",
},
{
index: {
_id: 3,
},
},
{
location: "POINT (491.4722 5237.1667)",
city: "Amsterdam",
name: "Nederlands Scheepvaartmuseum",
},
{
index: {
_id: 4,
},
},
{
location: "POINT (440.5200 5122.2900)",
city: "Antwerp",
name: "Letterenhuis",
},
{
index: {
_id: 5,
},
},
{
location: "POINT (233.6389 4886.1111)",
city: "Paris",
name: "Musée du Louvre",
},
{
index: {
_id: 6,
},
},
{
location: "POINT (232.7000 4886.0000)",
city: "Paris",
name: "Musée d'Orsay",
},
],
});
console.log(response1);
const response2 = await client.search({
index: "museums",
size: 0,
aggs: {
centroid: {
cartesian_centroid: {
field: "location",
},
},
},
});
console.log(response2);
PUT /museums
{
"mappings": {
"properties": {
"location": {
"type": "point"
}
}
}
}
POST /museums/_bulk?refresh
{"index":{"_id":1}}
{"location": "POINT (491.2350 5237.4081)", "city": "Amsterdam", "name": "NEMO Science Museum"}
{"index":{"_id":2}}
{"location": "POINT (490.1618 5236.9219)", "city": "Amsterdam", "name": "Museum Het Rembrandthuis"}
{"index":{"_id":3}}
{"location": "POINT (491.4722 5237.1667)", "city": "Amsterdam", "name": "Nederlands Scheepvaartmuseum"}
{"index":{"_id":4}}
{"location": "POINT (440.5200 5122.2900)", "city": "Antwerp", "name": "Letterenhuis"}
{"index":{"_id":5}}
{"location": "POINT (233.6389 4886.1111)", "city": "Paris", "name": "Musée du Louvre"}
{"index":{"_id":6}}
{"location": "POINT (232.7000 4886.0000)", "city": "Paris", "name": "Musée d'Orsay"}
POST /museums/_search?size=0
{
"aggs": {
"centroid": {
"cartesian_centroid": {
"field": "location"
}
}
}
}
上述聚合演示了如何计算所有博物馆文档的 location 字段的质心。
上述聚合的响应
{
...
"aggregations": {
"centroid": {
"location": {
"x": 396.6213124593099,
"y": 5100.982991536458
},
"count": 6
}
}
}
cartesian_centroid 聚合与其他桶聚合组合使用时更有趣。
示例
resp = client.search(
index="museums",
size="0",
aggs={
"cities": {
"terms": {
"field": "city.keyword"
},
"aggs": {
"centroid": {
"cartesian_centroid": {
"field": "location"
}
}
}
}
},
)
print(resp)
const response = await client.search({
index: "museums",
size: 0,
aggs: {
cities: {
terms: {
field: "city.keyword",
},
aggs: {
centroid: {
cartesian_centroid: {
field: "location",
},
},
},
},
},
});
console.log(response);
POST /museums/_search?size=0
{
"aggs": {
"cities": {
"terms": { "field": "city.keyword" },
"aggs": {
"centroid": {
"cartesian_centroid": { "field": "location" }
}
}
}
}
}
以上示例使用cartesian_centroid 作为术语桶聚合的子聚合，以查找每个城市博物馆的中心位置。
上述聚合的响应
{
...
"aggregations": {
"cities": {
"sum_other_doc_count": 0,
"doc_count_error_upper_bound": 0,
"buckets": [
{
"key": "Amsterdam",
"doc_count": 3,
"centroid": {
"location": {
"x": 490.9563293457031,
"y": 5237.16552734375
},
"count": 3
}
},
{
"key": "Paris",
"doc_count": 2,
"centroid": {
"location": {
"x": 233.16944885253906,
"y": 4886.0556640625
},
"count": 2
}
},
{
"key": "Antwerp",
"doc_count": 1,
"centroid": {
"location": {
"x": 440.5199890136719,
"y": 5122.2900390625
},
"count": 1
}
}
]
}
}
}
在shape字段上使用笛卡尔质心聚合
编辑形状的质心度量比点的质心度量更细致。包含形状的特定聚合桶的质心是该桶中最高维度形状类型的质心。例如，如果一个桶包含由多边形和线组成的形状，则这些线不会对质心度量做出贡献。每种形状类型的质心计算方式不同。通过圆形处理器摄取的包络线和圆形被视为多边形。
| 几何类型 | 质心计算 |
|---|---|
[多]点 |
所有坐标的等权平均值 |
[多]线串 |
每段的质心的加权平均值，其中每段的权重是其长度（单位与坐标相同） |
[多]多边形 |
所有多边形三角形的质心的加权平均值，其中三角形由每两个连续顶点和起点形成。孔具有负权重。权重表示三角形的面积（单位为坐标单位的平方） |
几何集合 |
所有具有最高维度的底层几何图形的质心。如果存在多边形和线和/或点，则忽略线和/或点。如果存在线和点，则忽略点 |
示例
resp = client.indices.create(
index="places",
mappings={
"properties": {
"geometry": {
"type": "shape"
}
}
},
)
print(resp)
resp1 = client.bulk(
index="places",
refresh=True,
operations=[
{
"index": {
"_id": 1
}
},
{
"name": "NEMO Science Museum",
"geometry": "POINT(491.2350 5237.4081)"
},
{
"index": {
"_id": 2
}
},
{
"name": "Sportpark De Weeren",
"geometry": {
"type": "Polygon",
"coordinates": [
[
[
496.5305328369141,
5239.347642069457
],
[
496.6979026794433,
5239.172175893484
],
[
496.9425201416015,
5239.238958618537
],
[
496.7944622039794,
5239.420969150824
],
[
496.5305328369141,
5239.347642069457
]
]
]
}
}
],
)
print(resp1)
resp2 = client.search(
index="places",
size="0",
aggs={
"centroid": {
"cartesian_centroid": {
"field": "geometry"
}
}
},
)
print(resp2)
response = client.indices.create(
index: 'places',
body: {
mappings: {
properties: {
geometry: {
type: 'shape'
}
}
}
}
)
puts response
response = client.bulk(
index: 'places',
refresh: true,
body: [
{
index: {
_id: 1
}
},
{
name: 'NEMO Science Museum',
geometry: 'POINT(491.2350 5237.4081)'
},
{
index: {
_id: 2
}
},
{
name: 'Sportpark De Weeren',
geometry: {
type: 'Polygon',
coordinates: [
[
[
496.5305328369141,
5239.347642069457
],
[
496.6979026794433,
5239.172175893484
],
[
496.9425201416015,
5239.238958618537
],
[
496.7944622039794,
5239.420969150824
],
[
496.5305328369141,
5239.347642069457
]
]
]
}
}
]
)
puts response
response = client.search(
index: 'places',
size: 0,
body: {
aggregations: {
centroid: {
cartesian_centroid: {
field: 'geometry'
}
}
}
}
)
puts response
const response = await client.indices.create({
index: "places",
mappings: {
properties: {
geometry: {
type: "shape",
},
},
},
});
console.log(response);
const response1 = await client.bulk({
index: "places",
refresh: "true",
operations: [
{
index: {
_id: 1,
},
},
{
name: "NEMO Science Museum",
geometry: "POINT(491.2350 5237.4081)",
},
{
index: {
_id: 2,
},
},
{
name: "Sportpark De Weeren",
geometry: {
type: "Polygon",
coordinates: [
[
[496.5305328369141, 5239.347642069457],
[496.6979026794433, 5239.172175893484],
[496.9425201416015, 5239.238958618537],
[496.7944622039794, 5239.420969150824],
[496.5305328369141, 5239.347642069457],
],
],
},
},
],
});
console.log(response1);
const response2 = await client.search({
index: "places",
size: 0,
aggs: {
centroid: {
cartesian_centroid: {
field: "geometry",
},
},
},
});
console.log(response2);
PUT /places
{
"mappings": {
"properties": {
"geometry": {
"type": "shape"
}
}
}
}
POST /places/_bulk?refresh
{"index":{"_id":1}}
{"name": "NEMO Science Museum", "geometry": "POINT(491.2350 5237.4081)" }
{"index":{"_id":2}}
{"name": "Sportpark De Weeren", "geometry": { "type": "Polygon", "coordinates": [ [ [ 496.5305328369141, 5239.347642069457 ], [ 496.6979026794433, 5239.1721758934835 ], [ 496.9425201416015, 5239.238958618537 ], [ 496.7944622039794, 5239.420969150824 ], [ 496.5305328369141, 5239.347642069457 ] ] ] } }
POST /places/_search?size=0
{
"aggs": {
"centroid": {
"cartesian_centroid": {
"field": "geometry"
}
}
}
}
{
...
"aggregations": {
"centroid": {
"location": {
"x": 496.74041748046875,
"y": 5239.29638671875
},
"count": 2
}
}
}