/** metacode_derived_features_streaming.txt Script to calculate streaming data DolphinDB Inc. DolphinDB server version: 2.00.6 2022.05.09 Last modification time: 2022.08.31 */ /** Attention: 1. The developer need to import level2 snapshot data into the database in advance 2. The developer need to install Xgboost plugin and save model into specified path in advance 3. There are two places in the script that need to be modified according to the environment */ //clean environment def cleanEnvironment(){ try{ unsubscribeTable(tableName=`snapshotStream, actionName="aggrFeatures10min") } catch(ex){ print(ex) } try{ unsubscribeTable(tableName=`aggrFeatures10min, actionName="predictRV") } catch(ex){ print(ex) } try{ dropStreamEngine("aggrFeatures10min") } catch(ex){ print(ex) } try{ dropStreamTable(`snapshotStream) } catch(ex){ print(ex) } try{ dropStreamTable(`aggrFeatures10min) } catch(ex){ print(ex) } try{ dropStreamTable(`result10min) } catch(ex){ print(ex) } undef all } cleanEnvironment() go //login account login("admin", "123456") /** part1: define functions */ def logReturn(s){ return log(s)-log(prev(s)) } def realizedVolatility(s){ return sqrt(sum2(s)) } //create meta code def createAggMetaCode(aggDict){ metaCode = [] metaCodeColName = [] for(colName in aggDict.keys()){ for(funcName in aggDict[colName]) { metaCode.append!(sqlCol(colName, funcByName(funcName), colName + `_ + funcName$STRING)) metaCodeColName.append!(colName + `_ + funcName$STRING) } } return metaCode, metaCodeColName$STRING } /** part2: feature engineering */ features = { "DateTime":[`count] } for( i in 0..9) { features["Wap"+i] = [`sum, `mean, `std] features["LogReturn"+i] = [`sum, `realizedVolatility, `mean, `std] features["LogReturnOffer"+i] = [`sum, `realizedVolatility, `mean, `std] features["LogReturnBid"+i] = [`sum, `realizedVolatility, `mean, `std] } features["WapBalance"] = [`sum, `mean, `std] features["PriceSpread"] = [`sum, `mean, `std] features["BidSpread"] = [`sum, `mean, `std] features["OfferSpread"] = [`sum, `mean, `std] features["TotalVolume"] = [`sum, `mean, `std] features["VolumeImbalance"] = [`sum, `mean, `std] aggMetaCode, metaCodeColName = createAggMetaCode(features) /** part3: import plugin and load saved model modified location 1: Path to Plugin, modelSavePath */ try{ loadPlugin(getHomeDir()+"/plugins/xgboost/PluginXgboost.txt") } catch(ex){ print(ex) } modelSavePath = "/hdd/hdd9/machineLearning/model/001.model" model = xgboost::loadModel(modelSavePath) /** part4: define stream table */ name = `SecurityID`DateTime`PreClosePx`OpenPx`HighPx`LowPx`LastPx`TotalVolumeTrade`TotalValueTrade`BidPrice0`BidPrice1`BidPrice2`BidPrice3`BidPrice4`BidPrice5`BidPrice6`BidPrice7`BidPrice8`BidPrice9`BidOrderQty0`BidOrderQty1`BidOrderQty2`BidOrderQty3`BidOrderQty4`BidOrderQty5`BidOrderQty6`BidOrderQty7`BidOrderQty8`BidOrderQty9`OfferPrice0`OfferPrice1`OfferPrice2`OfferPrice3`OfferPrice4`OfferPrice5`OfferPrice6`OfferPrice7`OfferPrice8`OfferPrice9`OfferOrderQty0`OfferOrderQty1`OfferOrderQty2`OfferOrderQty3`OfferOrderQty4`OfferOrderQty5`OfferOrderQty6`OfferOrderQty7`OfferOrderQty8`OfferOrderQty9 type =`SYMBOL`TIMESTAMP`DOUBLE`DOUBLE`DOUBLE`DOUBLE`DOUBLE`INT`DOUBLE`DOUBLE`DOUBLE`DOUBLE`DOUBLE`DOUBLE`DOUBLE`DOUBLE`DOUBLE`DOUBLE`DOUBLE`INT`INT`INT`INT`INT`INT`INT`INT`INT`INT`DOUBLE`DOUBLE`DOUBLE`DOUBLE`DOUBLE`DOUBLE`DOUBLE`DOUBLE`DOUBLE`DOUBLE`INT`INT`INT`INT`INT`INT`INT`INT`INT`INT share streamTable(100000:0, name, type) as snapshotStream share streamTable(100000:0 , `DateTime`SecurityID <- metaCodeColName <- (metaCodeColName+"_150") <- (metaCodeColName+"_300") <- (metaCodeColName+"_450"),`TIMESTAMP`SYMBOL <- take(`DOUBLE, 676)) as aggrFeatures10min share streamTable(100000:0 , `Predicted`SecurityID`DateTime, `FLOAT`SYMBOL`TIMESTAMP) as result10min /** part5: define aggregate function ,calculate features metrics */ defg featureEngineering(DateTime, BidPrice, BidOrderQty, OfferPrice, OfferOrderQty, aggMetaCode){ wap = (BidPrice * OfferOrderQty + BidOrderQty * OfferPrice) \ (BidOrderQty + OfferOrderQty) wapBalance = abs(wap[0] - wap[1]) priceSpread = (OfferPrice[0] - BidPrice[0]) \ ((OfferPrice[0] + BidPrice[0]) \ 2) BidSpread = BidPrice[0] - BidPrice[1] OfferSpread = OfferPrice[0] - OfferPrice[1] totalVolume = OfferOrderQty.rowSum() + BidOrderQty.rowSum() volumeImbalance = abs(OfferOrderQty.rowSum() - BidOrderQty.rowSum()) LogReturnWap = logReturn(wap) LogReturnOffer = logReturn(OfferPrice) LogReturnBid = logReturn(BidPrice) subTable = table(DateTime as `DateTime, BidPrice, BidOrderQty, OfferPrice, OfferOrderQty, wap, wapBalance, priceSpread, BidSpread, OfferSpread, totalVolume, volumeImbalance, LogReturnWap, LogReturnOffer, LogReturnBid) colNum = 0..9$STRING colName = `DateTime <- (`BidPrice + colNum) <- (`BidOrderQty + colNum) <- (`OfferPrice + colNum) <- (`OfferOrderQty + colNum) <- (`Wap + colNum) <- `WapBalance`PriceSpread`BidSpread`OfferSpread`TotalVolume`VolumeImbalance <- (`LogReturn + colNum) <- (`LogReturnOffer + colNum) <- (`LogReturnBid + colNum) subTable.rename!(colName) subTable['BarDateTime'] = bar(subTable['DateTime'], 10m) result = sql(select = aggMetaCode, from = subTable).eval().matrix() result150 = sql(select = aggMetaCode, from = subTable, where = = (time(BarDateTime) + 150*1000) >).eval().matrix() result300 = sql(select = aggMetaCode, from = subTable, where = = (time(BarDateTime) + 300*1000) >).eval().matrix() result450 = sql(select = aggMetaCode, from = subTable, where = = (time(BarDateTime) + 450*1000) >).eval().matrix() return concatMatrix([result, result150, result300, result450]) } metrics=sqlColAlias(, metaCodeColName <- (metaCodeColName+"_150") <- (metaCodeColName+"_300") <- (metaCodeColName+"_450")) /** part6: register stream computing engine and define two subscribe tables */ createTimeSeriesEngine(name="aggrFeatures10min", windowSize=600000, step=600000, metrics=metrics, dummyTable=snapshotStream, outputTable=aggrFeatures10min, timeColumn=`DateTime, useWindowStartTime=true, keyColumn=`SecurityID, forceTriggerTime = 5) subscribeTable(tableName="snapshotStream", actionName="aggrFeatures10min", offset=-1, handler=getStreamEngine("aggrFeatures10min"), msgAsTable=true, batchSize=2000, throttle=1, hash=0, reconnect=true) def predictRV(mutable result10min, model, mutable msg){ temp_table = select SecurityID, DateTime from msg msg.update!(`SecurityID_int, int(msg[`SecurityID])).dropColumns!(`SecurityID`DateTime`LogReturn0_realizedVolatility) Predicted = xgboost::predict(model , msg) temp_table_2 = table(Predicted, temp_table) result10min.append!(temp_table_2) } subscribeTable(tableName="aggrFeatures10min", actionName="predictRV", offset=-1, handler=predictRV{result10min, model}, msgAsTable=true, hash=1, reconnect=true) go /** part7: replay history data modified location 2: csvFilePath */ csvFilePath = "/hdd/hdd9/machineLearning/testSnapshot.csv" testSnapshot = loadText(filename=csvFilePath, schema=table(snapshotStream.schema().colDefs.name, snapshotStream.schema().colDefs.typeString)) submitJob("replay", "replay", replay{testSnapshot, snapshotStream, `DateTime, `DateTime, 20000, true, 1}) /** part8: check result */ sleep(1000) select * from result10min