examples/daal4py/saga.py

# ==============================================================================
# Copyright 2014 Intel Corporation
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# ==============================================================================

# daal4py Saga example for shared memory systems

from pathlib import Path

import numpy as np
from readcsv import pd_read_csv

import daal4py as d4p


def main(readcsv=pd_read_csv):
    data_path = Path(__file__).parent / "data" / "batch"
    infile = data_path / "XM.csv"
    # Read the data, let's have 3 independent variables
    data = readcsv(infile, range(1))
    dep_data = readcsv(infile, range(1, 2))
    nVectors = data.shape[0]

    # configure a Logistic Loss object
    logloss_algo = d4p.optimization_solver_logistic_loss(
        numberOfTerms=nVectors,
        penaltyL1=0.3,
        penaltyL2=0,
        interceptFlag=True,
        resultsToCompute="gradient",
    )
    logloss_algo.setup(data, dep_data)

    # configure an Saga object
    lr = np.array([[0.01]], dtype=np.double)
    niters = 100000
    saga_algo = d4p.optimization_solver_saga(
        nIterations=niters,
        accuracyThreshold=1e-5,
        batchSize=1,
        function=logloss_algo,
        learningRateSequence=lr,
        optionalResultRequired=True,
    )

    # finally do the computation
    inp = np.zeros((2, 1), dtype=np.double)
    res = saga_algo.compute(inp, None)

    # The Saga result provides minimum and nIterations
    assert res.minimum.shape == inp.shape and res.nIterations[0][0] <= niters
    assert np.allclose(res.minimum, [[-0.17663868], [0.35893627]])

    return res


if __name__ == "__main__":
    res = main()
    print("\nMinimum:\n", res.minimum)
    print("\nNumber of iterations performed:\n", res.nIterations[0][0])
    print("All looks good!")