简介

提供了常用的监督分类分类器。

dTrees

决策树分类分类器。

函数	返回值
dTrees()	PIEDT分类器

示例

import pie
from pie import *
Map = pie.Map()
image = pie.Image('user/17090142114/PGDB001/World84').select(["B1","B2","B3"])
featureCollection = pie.FeatureCollection('user/17090142114/PGDB001/WorldROI')
training = image.sampleRegions(featureCollection,["type"],50000)
classifer = pie.dTrees().train(training,"type",["B1","B2","B3"]).getInfo()
resultImage = image.classify(classifer,"classifyA")
visParam = {
  'opacity':1,
  'uniqueValue':'1,2,3,4',
  'palette': 'EAF2F5,000032,1F3600,FAFFC8'
}
Map.addLayer(resultImage,visParam,'DT_result')
Map.setCenter(0,0,0)
Map

knn

K最近邻分类分类器。

函数	返回值
knn()	PIEKNN分类器

示例

import pie
from pie import *
Map = pie.Map()
image = pie.Image('user/17090142114/PGDB001/World84').select(["B1","B2","B3"])
featureCollection = pie.FeatureCollection('user/17090142114/PGDB001/WorldROI')
training = image.sampleRegions(featureCollection,["type"],50000)
classifer = pie.knn().train(training,"type",["B1","B2","B3"]).getInfo()
resultImage = image.classify(classifer,"classifyA")
visParam = {
  'opacity':1,
  'uniqueValue':'1,2,3,4',
  'palette': 'EAF2F5,000032,1F3600,FAFFC8'
}
Map.addLayer(resultImage,visParam,'Knn_result')
Map.setCenter(0,0,0)
Map

normalBayes

正态贝叶斯分类分类器。

函数	返回值
normalBayes()	PIENormalBayes分类器

示例

import pie
from pie import *
Map = pie.Map()
image= pie.Image('user/17090142114/PGDB001/World84').select(["B1","B2","B3"])
featureCollection= pie.FeatureCollection('user/17090142114/PGDB001/WorldROI')
training= image.sampleRegions(featureCollection,["type"],50000)
classifer= pie.normalBayes().train(training,"type",["B1","B2","B3"], 1, 1)
resultImage = image.classify(classifer,"classifyA")
visParam = {
    'opacity':1,
    'uniqueValue':'1,2,3,4',
    'palette': 'EAF2F5,000032,1F3600,FAFFC8'}
Map.addLayer(resultImage,visParam, "NormalBayes")
Map.setCenter(0,0,0)
Map

rTrees

随机森林分类分类器。

函数	返回值
rTrees()	PIERTrees分类器

示例

import pie
from pie import *
Map = pie.Map()
image = pie.Image('user/17090142114/PGDB001/World84').select(["B1","B2","B3"])
featureCollection = pie.FeatureCollection('user/17090142114/PGDB001/WorldROI')
training = image.sampleRegions(featureCollection,["type"],50000)
classifer = pie.rTrees().train(training, "type", ["B1", "B2", "B3"], 100, 1)
resultImage = image.classify(classifer,"classifyA")
visParam = {
    'opacity':1,
    'uniqueValue':'1,2,3,4',
    'palette': 'EAF2F5,000032,1F3600,FAFFC8'
}
Map.addLayer(resultImage,visParam, "RTees")
Map.setCenter(0,0,0)
Map

svm

SVM（支持向量机）监督分类分类器。

函数	返回值
svm()	PIESVM分类器

示例

import pie
from pie import *
Map = pie.Map()
image = pie.Image('user/17090142114/PGDB001/World84').select(["B1", "B2", "B3"])
featureCollection = pie.FeatureCollection('user/17090142114/PGDB001/WorldROI')
training = image.sampleRegions(featureCollection, ["type"], 50000)
classifer = pie.svm().train(training, "type", ["B1", "B2", "B3"], 100, 10)
resultImage = image.classify(classifer, "classifyA")
visParam = {
    'opacity': 1,
    'uniqueValue': '1,2,3,4',
    'palette': 'EAF2F5,000032,1F3600,FAFFC8'
}
Map.addLayer(resultImage, visParam, "SVM")
Map.setCenter(0, 0, 0)
Map

train

监督分类分类器训练。

函数	返回值
train(features,classProperty,inputProperties,subsampling,subsamplingSeed)	监督分类器训练结果

参数	类型	说明
features	FeatureCollection	样本点
classProperty	String	分类类别字段
inputProperties	List	分类计算字段
subsampling	Float	未启用
subsamplingSeed	Int	未启用

示例

import pie
from pie import *
Map = pie.Map()
image= pie.Image('user/17090142114/PGDB001/World84').select(["B1","B2","B3"])
featureCollection= pie.FeatureCollection('user/17090142114/PGDB001/WorldROI')
training= image.sampleRegions(featureCollection,["type"],50000)
classifer= pie.normalBayes().train(training,"type",["B1","B2","B3"], 1, 1)
resultImage = image.classify(classifer,"classifyA")
visParam = {
    'opacity':1,
    'uniqueValue':'1,2,3,4',
    'palette': 'EAF2F5,000032,1F3600,FAFFC8'}
Map.addLayer(resultImage,visParam, "NormalBayes")
Map.setCenter(0,0,0)
Map