文章目录世界上并没有完美的程序，但是我们并不因此而沮丧，因为写程序就是一个不断追求完美的过程。
- normalize
- 场景
- 作用
- 配置项
- 使用
- 索引
- method : rescale_0_1
- method : rescale_0_100
- method : percent_of_sum
- method : mean
- method : softmax
- 父管道聚合
- bucket值的规范化
-
buckets_path
- 要规范化的bucket的路径
-
method
-
规范化方法
-
rescale_0_1
- (x - min) / (max - min)
-
rescale_0_100
- 100 * (x - min) / (max - min)
-
percent_of_sum
- x / sum
-
mean
- (x - mean) / (max - min)
-
zscore(实测时invalid)
- (x - mean) / stdev
- stdev是标准差
-
softmax
- e^x / sum_e_x
- sum_e_x是原始值的指数和
-
-
format
- 返回值的格式
- 如，00.00%，00.00
DELETE /normalize_test
PUT /normalize_test
{
"mappings" : {
"properties" : {
"type" : {"type" : "integer"},
"num" : {"type" : "integer"},
"date" : {"type" : "date"}
}
}
}
POST /normalize_test/_bulk
{"index" : {"_id" : 1}}
{"type" : 1, "num" : 400, "date" : "2001-01-10"}
{"index" : {"_id" : 2}}
{"type" : 2, "num" : 450, "date" : "2001-01-20"}
{"index" : {"_id" : 3}}
{"type" : 1, "num" : 580, "date" : "2001-02-10"}
{"index" : {"_id" : 4}}
{"type" : 2, "num" : 990, "date" : "2001-03-20"}
{"index" : {"_id" : 5}}
{"type" : 1, "num" : 660, "date" : "2001-04-21"}
{"index" : {"_id" : 6}}
{"type" : 1, "num" : 680, "date" : "2001-05-21"}
method : rescale_0_1
GET /normalize_test/_search
{
"size" : 0,
"aggs" : {
"month_aggs" : {
"date_histogram": {
"field" : "date",
"calendar_interval" : "month"
},
"aggs" : {
"sum_aggs" : {
"sum": {
"field": "num"
}
},
"normalize_aggs" : {
"normalize" : {
"buckets_path" : "sum_aggs",
"method" : "rescale_0_1",
"format" : "00.00"
}
}
}
}
}
}
- 返回结果
{
"took" : 2,
"timed_out" : false,
"_shards" : {
"total" : 1,
"successful" : 1,
"skipped" : 0,
"failed" : 0
},
"hits" : {
"total" : {
"value" : 6,
"relation" : "eq"
},
"max_score" : null,
"hits" : [ ]
},
"aggregations" : {
"month_aggs" : {
"buckets" : [
{
"key_as_string" : "2001-01-01T00:00:00.000Z",
"key" : 978307200000,
"doc_count" : 2,
"sum_aggs" : {
"value" : 850.0
},
"normalize_aggs" : {
"value" : 0.6585365853658537,
"value_as_string" : "00.66"
}
},
{
"key_as_string" : "2001-02-01T00:00:00.000Z",
"key" : 980985600000,
"doc_count" : 1,
"sum_aggs" : {
"value" : 580.0
},
"normalize_aggs" : {
"value" : 0.0,
"value_as_string" : "00.00"
}
},
{
"key_as_string" : "2001-03-01T00:00:00.000Z",
"key" : 983404800000,
"doc_count" : 1,
"sum_aggs" : {
"value" : 990.0
},
"normalize_aggs" : {
"value" : 1.0,
"value_as_string" : "01.00"
}
},
{
"key_as_string" : "2001-04-01T00:00:00.000Z",
"key" : 986083200000,
"doc_count" : 1,
"sum_aggs" : {
"value" : 660.0
},
"normalize_aggs" : {
"value" : 0.1951219512195122,
"value_as_string" : "00.20"
}
},
{
"key_as_string" : "2001-05-01T00:00:00.000Z",
"key" : 988675200000,
"doc_count" : 1,
"sum_aggs" : {
"value" : 680.0
},
"normalize_aggs" : {
"value" : 0.24390243902439024,
"value_as_string" : "00.24"
}
}
]
}
}
}
method : rescale_0_100
GET /normalize_test/_search
{
"size" : 0,
"aggs" : {
"month_aggs" : {
"date_histogram": {
"field" : "date",
"calendar_interval" : "month"
},
"aggs" : {
"sum_aggs" : {
"sum": {
"field": "num"
}
},
"normalize_aggs" : {
"normalize" : {
"buckets_path" : "sum_aggs",
"method" : "rescale_0_100",
"format" : "00"
}
}
}
}
}
}
- 返回结果
{
"took" : 34,
"timed_out" : false,
"_shards" : {
"total" : 1,
"successful" : 1,
"skipped" : 0,
"failed" : 0
},
"hits" : {
"total" : {
"value" : 6,
"relation" : "eq"
},
"max_score" : null,
"hits" : [ ]
},
"aggregations" : {
"month_aggs" : {
"buckets" : [
{
"key_as_string" : "2001-01-01T00:00:00.000Z",
"key" : 978307200000,
"doc_count" : 2,
"sum_aggs" : {
"value" : 850.0
},
"normalize_aggs" : {
"value" : 65.85365853658537,
"value_as_string" : "66"
}
},
{
"key_as_string" : "2001-02-01T00:00:00.000Z",
"key" : 980985600000,
"doc_count" : 1,
"sum_aggs" : {
"value" : 580.0
},
"normalize_aggs" : {
"value" : 0.0,
"value_as_string" : "00"
}
},
{
"key_as_string" : "2001-03-01T00:00:00.000Z",
"key" : 983404800000,
"doc_count" : 1,
"sum_aggs" : {
"value" : 990.0
},
"normalize_aggs" : {
"value" : 100.0,
"value_as_string" : "100"
}
},
{
"key_as_string" : "2001-04-01T00:00:00.000Z",
"key" : 986083200000,
"doc_count" : 1,
"sum_aggs" : {
"value" : 660.0
},
"normalize_aggs" : {
"value" : 19.51219512195122,
"value_as_string" : "20"
}
},
{
"key_as_string" : "2001-05-01T00:00:00.000Z",
"key" : 988675200000,
"doc_count" : 1,
"sum_aggs" : {
"value" : 680.0
},
"normalize_aggs" : {
"value" : 24.390243902439025,
"value_as_string" : "24"
}
}
]
}
}
}
method : percent_of_sum
GET /normalize_test/_search
{
"size" : 0,
"aggs" : {
"month_aggs" : {
"date_histogram": {
"field" : "date",
"calendar_interval" : "month"
},
"aggs" : {
"sum_aggs" : {
"sum": {
"field": "num"
}
},
"normalize_aggs" : {
"normalize" : {
"buckets_path" : "sum_aggs",
"method" : "percent_of_sum",
"format" : "00.00%"
}
}
}
}
}
}
- 返回结果
{
"took" : 3,
"timed_out" : false,
"_shards" : {
"total" : 1,
"successful" : 1,
"skipped" : 0,
"failed" : 0
},
"hits" : {
"total" : {
"value" : 6,
"relation" : "eq"
},
"max_score" : null,
"hits" : [ ]
},
"aggregations" : {
"month_aggs" : {
"buckets" : [
{
"key_as_string" : "2001-01-01T00:00:00.000Z",
"key" : 978307200000,
"doc_count" : 2,
"sum_aggs" : {
"value" : 850.0
},
"normalize_aggs" : {
"value" : 0.22606382978723405,
"value_as_string" : "22.61%"
}
},
{
"key_as_string" : "2001-02-01T00:00:00.000Z",
"key" : 980985600000,
"doc_count" : 1,
"sum_aggs" : {
"value" : 580.0
},
"normalize_aggs" : {
"value" : 0.15425531914893617,
"value_as_string" : "15.43%"
}
},
{
"key_as_string" : "2001-03-01T00:00:00.000Z",
"key" : 983404800000,
"doc_count" : 1,
"sum_aggs" : {
"value" : 990.0
},
"normalize_aggs" : {
"value" : 0.2632978723404255,
"value_as_string" : "26.33%"
}
},
{
"key_as_string" : "2001-04-01T00:00:00.000Z",
"key" : 986083200000,
"doc_count" : 1,
"sum_aggs" : {
"value" : 660.0
},
"normalize_aggs" : {
"value" : 0.17553191489361702,
"value_as_string" : "17.55%"
}
},
{
"key_as_string" : "2001-05-01T00:00:00.000Z",
"key" : 988675200000,
"doc_count" : 1,
"sum_aggs" : {
"value" : 680.0
},
"normalize_aggs" : {
"value" : 0.18085106382978725,
"value_as_string" : "18.09%"
}
}
]
}
}
}
method : mean
GET /normalize_test/_search
{
"size" : 0,
"aggs" : {
"month_aggs" : {
"date_histogram": {
"field" : "date",
"calendar_interval" : "month"
},
"aggs" : {
"sum_aggs" : {
"sum": {
"field": "num"
}
},
"normalize_aggs" : {
"normalize" : {
"buckets_path" : "sum_aggs",
"method" : "mean",
"format" : "0.00"
}
}
}
}
}
}
- 返回结果
{
"took" : 5,
"timed_out" : false,
"_shards" : {
"total" : 1,
"successful" : 1,
"skipped" : 0,
"failed" : 0
},
"hits" : {
"total" : {
"value" : 6,
"relation" : "eq"
},
"max_score" : null,
"hits" : [ ]
},
"aggregations" : {
"month_aggs" : {
"buckets" : [
{
"key_as_string" : "2001-01-01T00:00:00.000Z",
"key" : 978307200000,
"doc_count" : 2,
"sum_aggs" : {
"value" : 850.0
},
"normalize_aggs" : {
"value" : 0.23902439024390243,
"value_as_string" : "0.24"
}
},
{
"key_as_string" : "2001-02-01T00:00:00.000Z",
"key" : 980985600000,
"doc_count" : 1,
"sum_aggs" : {
"value" : 580.0
},
"normalize_aggs" : {
"value" : -0.4195121951219512,
"value_as_string" : "-0.42"
}
},
{
"key_as_string" : "2001-03-01T00:00:00.000Z",
"key" : 983404800000,
"doc_count" : 1,
"sum_aggs" : {
"value" : 990.0
},
"normalize_aggs" : {
"value" : 0.5804878048780487,
"value_as_string" : "0.58"
}
},
{
"key_as_string" : "2001-04-01T00:00:00.000Z",
"key" : 986083200000,
"doc_count" : 1,
"sum_aggs" : {
"value" : 660.0
},
"normalize_aggs" : {
"value" : -0.22439024390243903,
"value_as_string" : "-0.22"
}
},
{
"key_as_string" : "2001-05-01T00:00:00.000Z",
"key" : 988675200000,
"doc_count" : 1,
"sum_aggs" : {
"value" : 680.0
},
"normalize_aggs" : {
"value" : -0.17560975609756097,
"value_as_string" : "-0.18"
}
}
]
}
}
}
method : softmax
- 由于是指数，所以测试本方法时索引中的num都去掉一个0
GET /normalize_test/_search
{
"size" : 0,
"aggs" : {
"month_aggs" : {
"date_histogram": {
"field" : "date",
"calendar_interval" : "month"
},
"aggs" : {
"sum_aggs" : {
"sum": {
"field": "num"
}
},
"normalize_aggs" : {
"normalize" : {
"buckets_path" : "sum_aggs",
"method" : "softmax"
}
}
}
}
}
}
- 返回结果
{
"took" : 0,
"timed_out" : false,
"_shards" : {
"total" : 1,
"successful" : 1,
"skipped" : 0,
"failed" : 0
},
"hits" : {
"total" : {
"value" : 6,
"relation" : "eq"
},
"max_score" : null,
"hits" : [ ]
},
"aggregations" : {
"month_aggs" : {
"buckets" : [
{
"key_as_string" : "2001-01-01T00:00:00.000Z",
"key" : 978307200000,
"doc_count" : 2,
"sum_aggs" : {
"value" : 85.0
},
"normalize_aggs" : {
"value" : 8.315280276640997E-7
}
},
{
"key_as_string" : "2001-02-01T00:00:00.000Z",
"key" : 980985600000,
"doc_count" : 1,
"sum_aggs" : {
"value" : 58.0
},
"normalize_aggs" : {
"value" : 1.5628808897545835E-18
}
},
{
"key_as_string" : "2001-03-01T00:00:00.000Z",
"key" : 983404800000,
"doc_count" : 1,
"sum_aggs" : {
"value" : 99.0
},
"normalize_aggs" : {
"value" : 0.9999991684719333
}
},
{
"key_as_string" : "2001-04-01T00:00:00.000Z",
"key" : 986083200000,
"doc_count" : 1,
"sum_aggs" : {
"value" : 66.0
},
"normalize_aggs" : {
"value" : 4.658882271108809E-15
}
},
{
"key_as_string" : "2001-05-01T00:00:00.000Z",
"key" : 988675200000,
"doc_count" : 1,
"sum_aggs" : {
"value" : 68.0
},
"normalize_aggs" : {
"value" : 3.442474245953642E-14
}
}
]
}
}
}
欢迎分享，转载请注明来源：内存溢出
微信扫一扫
支付宝扫一扫
评论列表（0条）