How to serialize & deserialize SVM model object through Python API?

[Matt_P_1]

Hi,

I'm working through the Pydaal API but can't figure out how to serialize and deserialize a trained (SVM) model; every example script trains and tests the model in one shot, but I need to serialize it for later to predict on new data as it becomes available. How can I do this through the Python API?

Thanks,

Matt

[Preethi_V_Intel]

Hi Matt,

You can use the Serialization and Deserialization interfaces available in DAAL to serialize model results into buffer(or even save to disk) and de-serialize at a later point to reconstruct the trained model. Note that, while de-serializing, an empty object must be constructed with the same type of serialized object.

Below code snippet is an extended functionality to the existing SVM code. Also, I have attached the complete code for your reference

import numpy as np
from daal.data_management import InputDataArchive, OutputDataArchive

def Serialize(model):

    # Construct input data archive Object
    # Serialize model contents into data archive Object
    # Copy data archive contents to numpy array
    dataArch = InputDataArchive ()
    model.serialize (dataArch)
    length = dataArch.getSizeOfArchive ()
    buffer_array = np.zeros (length, dtype=np.ubyte)
    dataArch.copyArchiveToArray (buffer_array)
    return buffer_array

if __name__ == "__main__":

    trainingResult = trainModel()
    buffer = Serialize(trainingResult)  # you can save this serialized object (which is a numpy array)to your disk
    #np.save(path, buffer)

    #Deserialization starts here
    #buffer = np.load(path) #Load numpy array
    dataArch = OutputDataArchive (buffer)
    trainedModel = training.Result()  # Construct an empty training result object
    trainedModel.deserialize (dataArch)  # Deserialize into trainedModel
    testModel(trainedModel)