d3/d13/full-example_8cpp-example.html

#include "boonnano.h"

#include "boonconfig.h"

#include "clusterstats.h"

#include "autotune.h"


#include <fstream>

#include <sstream>

// Full pipeline for nano core usage


int main() {


    CBoonNano Nano;

//  Set up configuration (boonconfig.h)

    CNanoConfigParameters NCP;


//  init parameters

    int SWS = 1;

    int NUM_FEATURES = 15;

    float MinVal = -10.0;

    float MaxVal = 10.0;

    int Weight = 1;

    float PV = 0.05;


//  set parameters in config object

    NCP.SetParameters(NF_FLOAT, PV, SWS);

    for (uint16_t i = 0; i < NUM_FEATURES; i++) {

        NCP.AddFeature(MinVal, MaxVal, Weight, "column-");

    }


//  set config

    Nano.Setup(NCP);


//  EXAMPLE OUTPUT

    std::cout << NCP.GetStatusString() << std::endl;

// CNanoConfigParameters

//     Numeric Type: Float  PV: 0.050000  SWS: 1

//     Feature 0:

//         Label: column- MinVal: -10.000000 MaxVal: 10.000000 Weight: 1

//     Feature 1:

//         Label: column- MinVal: -10.000000 MaxVal: 10.000000 Weight: 1

//     Feature 2:

//         Label: column- MinVal: -10.000000 MaxVal: 10.000000 Weight: 1

//     Feature 3:

//         Label: column- MinVal: -10.000000 MaxVal: 10.000000 Weight: 1

//     Feature 4:

//         Label: column- MinVal: -10.000000 MaxVal: 10.000000 Weight: 1

//     Feature 5:

//         Label: column- MinVal: -10.000000 MaxVal: 10.000000 Weight: 1

//     Feature 6:

//         Label: column- MinVal: -10.000000 MaxVal: 10.000000 Weight: 1

//     Feature 7:

//         Label: column- MinVal: -10.000000 MaxVal: 10.000000 Weight: 1

//     Feature 8:

//         Label: column- MinVal: -10.000000 MaxVal: 10.000000 Weight: 1

//     Feature 9:

//         Label: column- MinVal: -10.000000 MaxVal: 10.000000 Weight: 1

//     Feature 10:

//         Label: column- MinVal: -10.000000 MaxVal: 10.000000 Weight: 1

//     Feature 11:

//         Label: column- MinVal: -10.000000 MaxVal: 10.000000 Weight: 1

//     Feature 12:

//         Label: column- MinVal: -10.000000 MaxVal: 10.000000 Weight: 1

//     Feature 13:

//         Label: column- MinVal: -10.000000 MaxVal: 10.000000 Weight: 1

//     Feature 14:

//         Label: column- MinVal: -10.000000 MaxVal: 10.000000 Weight: 1

//     Error Message:


//  Autotune (autotune.h)


//  set up autotuning config

    CAutotuneNanoConfig ANC(&Nano);  // init with boonnano object

    ANC.LoadNanoConfig(NCP); // Get config values from boon config object; for information like numeric type


//  init parameters

    CAutotuningParameters AP;

    AP.m_AutotunePV = true;

    AP.m_AutotuneRange = true;

    AP.m_AutotuneByFeature = false;

    AP.m_MaxClusters = 1500;


//  set autotuning config

    ANC.LoadAutotuningConfig(AP);


//  EXAMPLE OUTPUT

    std::cout << AP.GetStatusString() << std::endl;

// CAutotuningParameters

//     AutotunePV: true  AutotuneRange: true  AutotuneByFeature: false  MaxClusters: 1500

//     Autotune over ALL features


//  Load autotuning data

    std::string line;

    std::ifstream myFile("./docs/examples/Planets.csv");

    int length = 14400;  // hardcoded for the specific file

    float *Data = new float[NUM_FEATURES * length];

    float val;

    int i = 0;

    while (std::getline(myFile, line)) {

      std::stringstream ss(line);


      while (ss >> val) {

        Data[i] = val;

        i++;


        if (ss.peek() == ',')

          ss.ignore();

      }

    }

    myFile.close();


// Main autotune step

    ANC.Autotune((uint8_t*)Data, NUM_FEATURES * length * sizeof(float)); // runs the synchronous autotuning function


//  Configure the nano object using the newly autotuned config

    NCP.CopyFrom(ANC.GetNanoConfig());

    Nano.Setup(NCP);


//  EXAMPLE OUTPUT

    std::cout << NCP.GetStatusString() << std::endl;

// CNanoConfigParameters

//     Numeric Type: Float  PV: 0.061000  SWS: 1

//     Feature 0:

//         Label: column- MinVal: -457.800262 MaxVal: 1211.813721 Weight: 1

//     Feature 1:

//         Label: column- MinVal: -457.800262 MaxVal: 1211.813721 Weight: 1

//     Feature 2:

//         Label: column- MinVal: -457.800262 MaxVal: 1211.813721 Weight: 1

//     Feature 3:

//         Label: column- MinVal: -457.800262 MaxVal: 1211.813721 Weight: 1

//     Feature 4:

//         Label: column- MinVal: -457.800262 MaxVal: 1211.813721 Weight: 1

//     Feature 5:

//         Label: column- MinVal: -457.800262 MaxVal: 1211.813721 Weight: 1

//     Feature 6:

//         Label: column- MinVal: -457.800262 MaxVal: 1211.813721 Weight: 1

//     Feature 7:

//         Label: column- MinVal: -457.800262 MaxVal: 1211.813721 Weight: 1

//     Feature 8:

//         Label: column- MinVal: -457.800262 MaxVal: 1211.813721 Weight: 1

//     Feature 9:

//         Label: column- MinVal: -457.800262 MaxVal: 1211.813721 Weight: 1

//     Feature 10:

//         Label: column- MinVal: -457.800262 MaxVal: 1211.813721 Weight: 1

//     Feature 11:

//         Label: column- MinVal: -457.800262 MaxVal: 1211.813721 Weight: 1

//     Feature 12:

//         Label: column- MinVal: -457.800262 MaxVal: 1211.813721 Weight: 1

//     Feature 13:

//         Label: column- MinVal: -457.800262 MaxVal: 1211.813721 Weight: 1

//     Feature 14:

//         Label: column- MinVal: -457.800262 MaxVal: 1211.813721 Weight: 1

//     Error Message:


//  Clustering (boonnano.h)

// Set backend statuses

    Nano.SetLearningMode(true); // required to be true to create new clusters

    Nano.SetNanoClippingDetection(false); // normal clipping of pattern range for clusters


//  Cluster

    int *ClusterResults = new int[length];

    Nano.ClusterData(Data, length, ClusterResults);


//  EXAMPLE OUTPUT

    std::cout << Nano.GetStatusString() << std::endl;

// CBoonNano

//     CNanoConfigParameters

//         Numeric Type: Float  PV: 0.061000  SWS: 1

//         Feature 0:

//             Label: column- MinVal: -457.800262 MaxVal: 1211.813721 Weight: 1

//         Feature 1:

//             Label: column- MinVal: -457.800262 MaxVal: 1211.813721 Weight: 1

//         Feature 2:

//             Label: column- MinVal: -457.800262 MaxVal: 1211.813721 Weight: 1

//         Feature 3:

//             Label: column- MinVal: -457.800262 MaxVal: 1211.813721 Weight: 1

//         Feature 4:

//             Label: column- MinVal: -457.800262 MaxVal: 1211.813721 Weight: 1

//         Feature 5:

//             Label: column- MinVal: -457.800262 MaxVal: 1211.813721 Weight: 1

//         Feature 6:

//             Label: column- MinVal: -457.800262 MaxVal: 1211.813721 Weight: 1

//         Feature 7:

//             Label: column- MinVal: -457.800262 MaxVal: 1211.813721 Weight: 1

//         Feature 8:

//             Label: column- MinVal: -457.800262 MaxVal: 1211.813721 Weight: 1

//         Feature 9:

//             Label: column- MinVal: -457.800262 MaxVal: 1211.813721 Weight: 1

//         Feature 10:

//             Label: column- MinVal: -457.800262 MaxVal: 1211.813721 Weight: 1

//         Feature 11:

//             Label: column- MinVal: -457.800262 MaxVal: 1211.813721 Weight: 1

//         Feature 12:

//             Label: column- MinVal: -457.800262 MaxVal: 1211.813721 Weight: 1

//         Feature 13:

//             Label: column- MinVal: -457.800262 MaxVal: 1211.813721 Weight: 1

//         Feature 14:

//             Label: column- MinVal: -457.800262 MaxVal: 1211.813721 Weight: 1

//         Error Message:

//     IsNanoBackend: true

//     BackendManager is Non-nullptr

//     CNanoBackendManager

//         SGXManager is nullptr

//            BackendInUse 0: true

//         NumKernels: 1  InitComplete: true

//         Error Message:

//         NanoBackend 0

//     NanoBackend is Non-nullptr

//     nano_sw_backend Ptr = 5234524160

//         InitComplete: true   PatternLength: 15   NumOfPatternsClustered: 14400   Learning is ON

//         Error Message:

//             Cluster 0:

//             Size: 0  Created at Inference: 0  RI: 1000  FI: 0  DI: 0  PI: 1000  NN: 0  PCA: (0, 0, 0) RCA: ()

//             Cluster 1:

//             Size: 1200  Created at Inference: 1  RI: 0  FI: 1011  DI: 435  PI: 0  NN: 0  PCA: (71, 669, 103) RCA: (0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000)

//             Cluster 2:

//             Size: 1200  Created at Inference: 1201  RI: 0  FI: 1011  DI: 394  PI: 0  NN: 0  PCA: (132, 732, 200) RCA: (0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000)

//             Cluster 3:

//             Size: 1200  Created at Inference: 2401  RI: 0  FI: 1011  DI: 429  PI: 0  NN: 0  PCA: (0, 1000, 7) RCA: (0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000)

//             Cluster 4:

//             Size: 1200  Created at Inference: 3601  RI: 0  FI: 1011  DI: 426  PI: 0  NN: 0  PCA: (67, 710, 108) RCA: (0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000)

//             Cluster 5:

//             Size: 1200  Created at Inference: 4801  RI: 0  FI: 1011  DI: 342  PI: 0  NN: 0  PCA: (702, 46, 248) RCA: (0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000)

//             Cluster 6:

//             Size: 1200  Created at Inference: 6001  RI: 0  FI: 1011  DI: 326  PI: 0  NN: 0  PCA: (679, 147, 294) RCA: (0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000)

//             Cluster 7:

//             Size: 1200  Created at Inference: 7201  RI: 0  FI: 1011  DI: 335  PI: 0  NN: 0  PCA: (1000, 0, 218) RCA: (0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000)

//             Cluster 8:

//             Size: 1200  Created at Inference: 8401  RI: 0  FI: 1011  DI: 330  PI: 0  NN: 0  PCA: (696, 46, 262) RCA: (0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000)

//             Cluster 9:

//             Size: 1200  Created at Inference: 9601  RI: 0  FI: 1011  DI: 343  PI: 0  NN: 0  PCA: (259, 95, 611) RCA: (0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000)

//             Cluster 10:

//             Size: 1200  Created at Inference: 10801  RI: 0  FI: 1011  DI: 345  PI: 0  NN: 0  PCA: (248, 20, 723) RCA: (0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000)

//             Cluster 11:

//             Size: 1200  Created at Inference: 12001  RI: 0  FI: 1011  DI: 364  PI: 0  NN: 0  PCA: (218, 7, 1000) RCA: (0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000)

//             Cluster 12:

//             Size: 1200  Created at Inference: 13201  RI: 0  FI: 1011  DI: 341  PI: 0  NN: 0  PCA: (277, 32, 668) RCA: (0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000, 0.000000)

//     Error Message:


//  Extra Results (clusterstats.h)

    Nano.SetLearningMode(false); // no new clusters can be created. patterns that dont fit in the model will create ghost clusters

//  Load extra data

    std::ifstream myFile2("./docs/examples/Meteors.csv");

    length = 5;

    float *ExtraData = new float[NUM_FEATURES * length];

    i = 0;

    while (std::getline(myFile2, line)) {

      std::stringstream ss(line);


      while (ss >> val) {

        ExtraData[i] = val;

        i++;


        if (ss.peek() == ',')

          ss.ignore();

      }

    }

    myFile2.close();


    CClusterStats *AdvancedAnalyticsBuffer = new CClusterStats[length];

    Nano.ClusterData(ExtraData, length, ClusterResults, AdvancedAnalyticsBuffer);


//  EXAMPLE OUTPUT

    for (int i = 0; i < length; i++) {

        std::cout << AdvancedAnalyticsBuffer[i].GetStatusString() << std::endl;

    }

// Ghost ID: -1 Size: 1  Created at Inference: 14401  RI: 1000  FI: 1000  DI: 1000  PI: 1000  NN: 1  PCA: (0, 0, 0) RCA: (0.027284, 0.110398, 0.055556, 0.037944, 0.022727, 0.019536, 0.094028, 0.197816, 0.158552, 0.034615, 0.057143, 0.041184, 0.007653, 0.128205, 0.016477)


// Ghost ID: -2 Size: 1  Created at Inference: 14402  RI: 1000  FI: 1000  DI: 1000  PI: 1000  NN: 1  PCA: (0, 0, 0) RCA: (0.022539, 0.030809, 0.170290, 0.122399, 0.093583, 0.115995, 0.019060, 0.150485, 0.024969, 0.007692, 0.055901, 0.037323, 0.042092, 0.061050, 0.126743)


// Ghost ID: -1 Size: 2  Created at Inference: 14401  RI: 1000  FI: 1000  DI: 1000  PI: 1000  NN: 1  PCA: (0, 0, 0) RCA: (0.027284, 0.110398, 0.055556, 0.037944, 0.022727, 0.019536, 0.094028, 0.197816, 0.158552, 0.034615, 0.057143, 0.041184, 0.007653, 0.128205, 0.016477)


// Ghost ID: -3 Size: 1  Created at Inference: 14404  RI: 1000  FI: 1000  DI: 1000  PI: 1000  NN: 1  PCA: (0, 0, 0) RCA: (0.090154, 0.084724, 0.000000, 0.078335, 0.076203, 0.014652, 0.133418, 0.154126, 0.093633, 0.151282, 0.026087, 0.006435, 0.049745, 0.074481, 0.032953)


// Ghost ID: -4 Size: 1  Created at Inference: 14405  RI: 1000  FI: 1000  DI: 1000  PI: 1000  NN: 1  PCA: (0, 0, 0) RCA: (0.102017, 0.156611, 0.032609, 0.154223, 0.012032, 0.001221, 0.017789, 0.165049, 0.057428, 0.012821, 0.024845, 0.086229, 0.042092, 0.103785, 0.102662)

//  Clean up

    Nano.Reset();

    delete[] ClusterResults;

    delete[] AdvancedAnalyticsBuffer;

    delete[] Data;


    return 0;

}

autotune.h

boonconfig.h

NF_FLOAT
const uint16_t NF_FLOAT
Numeric code for float data type.
Definition boonconfig.h:74

boonnano.h

CAutotuneNanoConfig::Autotune
bool Autotune(uint8_t *PatternBuffer, uint64_t PatternBufferLengthInBytes)
Main autotune function.

CAutotuneNanoConfig::GetNanoConfig
CNanoConfigParameters & GetNanoConfig()
Get autotuned nano config.

CAutotuneNanoConfig::LoadAutotuningConfig
bool LoadAutotuningConfig(CAutotuningParameters &AP)
Setup autotuning config.

CAutotuneNanoConfig::LoadNanoConfig
bool LoadNanoConfig(CNanoConfigParameters &NCP)
Setup config for autotuning.

CAutotuneNanoConfig
Main autotuning functionality.
Definition autotune.h:85

CAutotuningParameters::GetStatusString
std::string GetStatusString(int CurrentIndent=0, int AdditionalIndent=4)
Generate print string of object.

CAutotuningParameters
Object for autotuning specific parameters.
Definition autotune.h:23

CAutotuningParameters::m_MaxClusters
uint16_t m_MaxClusters
only allow this many clusters for each PV that is tested
Definition autotune.h:67

CAutotuningParameters::m_AutotuneRange
bool m_AutotuneRange
true means tune the range of the features, false uses the range that has been configured
Definition autotune.h:64

CAutotuningParameters::m_AutotuneByFeature
bool m_AutotuneByFeature
true means each feature has its range tuned separately, false uses a min and max found over all the f...
Definition autotune.h:65

CAutotuningParameters::m_AutotunePV
bool m_AutotunePV
true means tune the percent variation, false uses the percent variation that its been set to
Definition autotune.h:63

CBoonNano::Setup
bool Setup(CNanoConfigParameters &NanoConfig)
Set configuration parameters for model clustering.

CBoonNano::SetLearningMode
void SetLearningMode(bool LearningOn)
Enable/disable new cluster creation.

CBoonNano::GetStatusString
std::string GetStatusString(int CurrentIndent=0, int AdditionalIndent=4)
Print string of object variables.

CBoonNano::SetNanoClippingDetection
void SetNanoClippingDetection(bool ClippingOn)
Turn on/off value out of bounds anomalies.

CBoonNano::ClusterData
int ClusterData(float *Pattern, CClusterStats *AdvancedAnalytics=nullptr)
Cluster float pattern.

CBoonNano::Reset
void Reset(bool BackendReset=true)
Clear object variables.

CBoonNano
Main class of processing data.
Definition boonnano.h:110

CClusterStats::GetStatusString
std::string GetStatusString(int CurrentIndent=0, int AdditionalIndent=4) const
Generate print string for object values.

CClusterStats
Cluster specific statistics.
Definition clusterstats.h:17

CNanoConfigParameters::GetStatusString
std::string GetStatusString(int CurrentIndent=0, int AdditionalIndent=4) const
Print string of object.

CNanoConfigParameters::AddFeature
bool AddFeature(float MinVal, float MaxVal, uint16_t Weight, const char *Label=nullptr, uint16_t NumInstances=1)
Add feature values to overall config.

CNanoConfigParameters::SetParameters
bool SetParameters(uint16_t NumericFormat, float PercentVariation, uint16_t StreamingWindowSize)
Initialize base config values.

CNanoConfigParameters::CopyFrom
void CopyFrom(const CNanoConfigParameters &Other)
Copy parameters from given object.

CNanoConfigParameters
Main object for configuration.
Definition boonconfig.h:86

clusterstats.h