Avg 聚合
编辑Avg 聚合
编辑一个 单值 指标聚合，用于计算从聚合文档中提取的数值的平均值。这些值可以从文档中特定的数值或 直方图 字段中提取。
假设数据由代表学生考试成绩（介于 0 和 100 之间）的文档组成，我们可以使用以下方式计算他们的平均分：
resp = client.search(
index="exams",
size="0",
aggs={
"avg_grade": {
"avg": {
"field": "grade"
}
}
},
)
print(resp)
response = client.search(
index: 'exams',
size: 0,
body: {
aggregations: {
avg_grade: {
avg: {
field: 'grade'
}
}
}
}
)
puts response
const response = await client.search({
index: "exams",
size: 0,
aggs: {
avg_grade: {
avg: {
field: "grade",
},
},
},
});
console.log(response);
POST /exams/_search?size=0
{
"aggs": {
"avg_grade": { "avg": { "field": "grade" } }
}
}
上述聚合计算所有文档的平均成绩。聚合类型为 avg，field 设置定义了文档中计算平均值的数值字段。上述操作将返回以下内容：
{
...
"aggregations": {
"avg_grade": {
"value": 75.0
}
}
}
聚合的名称（上面的 avg_grade）也用作从返回的响应中检索聚合结果的键。
脚本
编辑假设考试非常困难，你需要应用成绩更正。对 运行时字段 求平均值以获得更正后的平均值：
resp = client.search(
index="exams",
size="0",
runtime_mappings={
"grade.corrected": {
"type": "double",
"script": {
"source": "emit(Math.min(100, doc['grade'].value * params.correction))",
"params": {
"correction": 1.2
}
}
}
},
aggs={
"avg_corrected_grade": {
"avg": {
"field": "grade.corrected"
}
}
},
)
print(resp)
response = client.search(
index: 'exams',
size: 0,
body: {
runtime_mappings: {
'grade.corrected' => {
type: 'double',
script: {
source: "emit(Math.min(100, doc['grade'].value * params.correction))",
params: {
correction: 1.2
}
}
}
},
aggregations: {
avg_corrected_grade: {
avg: {
field: 'grade.corrected'
}
}
}
}
)
puts response
const response = await client.search({
index: "exams",
size: 0,
runtime_mappings: {
"grade.corrected": {
type: "double",
script: {
source: "emit(Math.min(100, doc['grade'].value * params.correction))",
params: {
correction: 1.2,
},
},
},
},
aggs: {
avg_corrected_grade: {
avg: {
field: "grade.corrected",
},
},
},
});
console.log(response);
POST /exams/_search?size=0
{
"runtime_mappings": {
"grade.corrected": {
"type": "double",
"script": {
"source": "emit(Math.min(100, doc['grade'].value * params.correction))",
"params": {
"correction": 1.2
}
}
}
},
"aggs": {
"avg_corrected_grade": {
"avg": {
"field": "grade.corrected"
}
}
}
}
缺失值
编辑missing 参数定义了应如何处理缺少值的文档。默认情况下，它们将被忽略，但也可以将它们视为具有一个值。
resp = client.search(
index="exams",
size="0",
aggs={
"grade_avg": {
"avg": {
"field": "grade",
"missing": 10
}
}
},
)
print(resp)
response = client.search(
index: 'exams',
size: 0,
body: {
aggregations: {
grade_avg: {
avg: {
field: 'grade',
missing: 10
}
}
}
}
)
puts response
const response = await client.search({
index: "exams",
size: 0,
aggs: {
grade_avg: {
avg: {
field: "grade",
missing: 10,
},
},
},
});
console.log(response);
直方图字段
编辑当在 直方图字段 上计算平均值时，聚合的结果是 values 数组中所有元素的加权平均值，同时考虑了 counts 数组中相同位置的数字。
例如，对于以下索引，该索引存储了不同网络的延迟指标的预聚合直方图：
resp = client.index(
index="metrics_index",
id="1",
document={
"network.name": "net-1",
"latency_histo": {
"values": [
0.1,
0.2,
0.3,
0.4,
0.5
],
"counts": [
3,
7,
23,
12,
6
]
}
},
)
print(resp)
resp1 = client.index(
index="metrics_index",
id="2",
document={
"network.name": "net-2",
"latency_histo": {
"values": [
0.1,
0.2,
0.3,
0.4,
0.5
],
"counts": [
8,
17,
8,
7,
6
]
}
},
)
print(resp1)
resp2 = client.search(
index="metrics_index",
size="0",
aggs={
"avg_latency": {
"avg": {
"field": "latency_histo"
}
}
},
)
print(resp2)
response = client.index(
index: 'metrics_index',
id: 1,
body: {
'network.name' => 'net-1',
latency_histo: {
values: [
0.1,
0.2,
0.3,
0.4,
0.5
],
counts: [
3,
7,
23,
12,
6
]
}
}
)
puts response
response = client.index(
index: 'metrics_index',
id: 2,
body: {
'network.name' => 'net-2',
latency_histo: {
values: [
0.1,
0.2,
0.3,
0.4,
0.5
],
counts: [
8,
17,
8,
7,
6
]
}
}
)
puts response
response = client.search(
index: 'metrics_index',
size: 0,
body: {
aggregations: {
avg_latency: {
avg: {
field: 'latency_histo'
}
}
}
}
)
puts response
const response = await client.index({
index: "metrics_index",
id: 1,
document: {
"network.name": "net-1",
latency_histo: {
values: [0.1, 0.2, 0.3, 0.4, 0.5],
counts: [3, 7, 23, 12, 6],
},
},
});
console.log(response);
const response1 = await client.index({
index: "metrics_index",
id: 2,
document: {
"network.name": "net-2",
latency_histo: {
values: [0.1, 0.2, 0.3, 0.4, 0.5],
counts: [8, 17, 8, 7, 6],
},
},
});
console.log(response1);
const response2 = await client.search({
index: "metrics_index",
size: 0,
aggs: {
avg_latency: {
avg: {
field: "latency_histo",
},
},
},
});
console.log(response2);
PUT metrics_index/_doc/1
{
"network.name" : "net-1",
"latency_histo" : {
"values" : [0.1, 0.2, 0.3, 0.4, 0.5],
"counts" : [3, 7, 23, 12, 6]
}
}
PUT metrics_index/_doc/2
{
"network.name" : "net-2",
"latency_histo" : {
"values" : [0.1, 0.2, 0.3, 0.4, 0.5],
"counts" : [8, 17, 8, 7, 6]
}
}
POST /metrics_index/_search?size=0
{
"aggs": {
"avg_latency":
{ "avg": { "field": "latency_histo" }
}
}
}
对于每个直方图字段，avg 聚合会将 values 数组 <1> 中的每个数字乘以其在 counts 数组 <2> 中的关联计数。最终，它将计算所有直方图的这些值的平均值并返回以下结果：
{
...
"aggregations": {
"avg_latency": {
"value": 0.29690721649
}
}
}