TMVA_SOFIE_Models.py File Reference

Detailed Description

Example of inference with SOFIE using a set of models trained with Keras.

This tutorial shows how to store several models in a single header file and the weights in a ROOT binary file. The models are then evaluated using the RDataFrame First, generate the input model by running TMVA_Higgs_Classification.C.

This tutorial parses the input model and runs the inference using ROOT's JITing capability.

 
import contextlib
import os
import warnings
 
import numpy as np
import ROOT
from sklearn.model_selection import train_test_split
from tensorflow.keras.layers import Dense, Input
from tensorflow.keras.models import Sequential
from tensorflow.keras.optimizers import Adam
 
 
@contextlib.contextmanager
def expect_warning(category, message):
    """Silence a known third-party warning. Raise if it stops firing.
 
    Notifies us to drop the workaround once the upstream library is fixed.
    """
    with warnings.catch_warnings(record=True) as caught:
        warnings.simplefilter("always")
        yield
    seen = False
    for w in caught:
        if issubclass(w.category, category) and message in str(w.message):
            seen = True
        else:
            warnings.warn_explicit(w.message, w.category, w.filename, w.lineno)
    if not seen:
        raise RuntimeError(
            f"Expected {category.__name__} containing {message!r} was not "
            "emitted. This tutorial's workaround can probably be removed."
        )
 
 
## generate and train Keras models with different architectures
 
 
def CreateModel(nlayers=4, nunits=64):
    model = Sequential()
    model.add(Input(shape=(7,)))
    model.add(Dense(nunits, activation="relu"))
    for i in range(1, nlayers):
        model.add(Dense(nunits, activation="relu"))
 
    model.add(Dense(1, activation="sigmoid"))
    model.compile(loss="binary_crossentropy", optimizer=Adam(learning_rate=0.001), weighted_metrics=["accuracy"])
    model.summary()
    return model
 
 
def PrepareData():
    # get the input data
    inputFile = str(ROOT.gROOT.GetTutorialDir()) + "/machine_learning/data/Higgs_data.root"
 
    df1 = ROOT.RDataFrame("sig_tree", inputFile)
    sigData = df1.AsNumpy(columns=["m_jj", "m_jjj", "m_lv", "m_jlv", "m_bb", "m_wbb", "m_wwbb"])
    # print(sigData)
 
    # stack all the 7 numpy array in a single array (nevents x nvars)
    xsig = np.column_stack(list(sigData.values()))
    data_sig_size = xsig.shape[0]
    print("size of data", data_sig_size)
 
    # make SOFIE inference on background data
    df2 = ROOT.RDataFrame("bkg_tree", inputFile)
    bkgData = df2.AsNumpy(columns=["m_jj", "m_jjj", "m_lv", "m_jlv", "m_bb", "m_wbb", "m_wwbb"])
    xbkg = np.column_stack(list(bkgData.values()))
    data_bkg_size = xbkg.shape[0]
 
    ysig = np.ones(data_sig_size)
    ybkg = np.zeros(data_bkg_size)
    inputs_data = np.concatenate((xsig, xbkg), axis=0)
    inputs_targets = np.concatenate((ysig, ybkg), axis=0)
 
    # split data in training and test data
 
    x_train, x_test, y_train, y_test = train_test_split(inputs_data, inputs_targets, test_size=0.50, random_state=1234)
 
    return x_train, y_train, x_test, y_test
 
 
def TrainModel(model, x, y, name):
    model.fit(x, y, epochs=5, batch_size=50)
    modelFile = name + ".keras"
    # Keras' internal ``np.array(x)`` (TensorFlow backend) does not yet
    # implement the NumPy 2.0 ``__array__(copy=...)`` signature, so saving
    # emits a DeprecationWarning that we cannot fix from user code.
    if tuple(int(p) for p in np.__version__.split(".")[:2]) >= (2, 0):
        ctx = expect_warning(DeprecationWarning, "__array__ implementation doesn't accept a copy keyword")
    else:
        ctx = contextlib.nullcontext()
 
    with ctx:
        model.save(modelFile)
    return modelFile
 
 
### run the models
 
x_train, y_train, x_test, y_test = PrepareData()
 
## create models and train them
 
model1 = TrainModel(CreateModel(4, 64), x_train, y_train, "Higgs_Model_4L_50")
model2 = TrainModel(CreateModel(4, 64), x_train, y_train, "Higgs_Model_4L_200")
model3 = TrainModel(CreateModel(4, 64), x_train, y_train, "Higgs_Model_2L_500")
 
# evaluate with SOFIE the 3 trained models
 
 
def GenerateModelCode(modelFile, generatedHeaderFile):
    model = ROOT.TMVA.Experimental.SOFIE.PyKeras.Parse(modelFile)
 
    print("Generating inference code for the Keras model from ", modelFile, "in the header ", generatedHeaderFile)
    # Generating inference code using a ROOT binary file
    model.Generate(ROOT.TMVA.Experimental.SOFIE.Options.kRootBinaryWeightFile)
    # add option to append to the same file the generated headers (pass True for append flag)
    model.OutputGenerated(generatedHeaderFile, True)
    # model.PrintGenerated()
    return generatedHeaderFile
 
 
generatedHeaderFile = "Higgs_Model.hxx"
# need to remove existing header file since we are appending on same one
if os.path.exists(generatedHeaderFile):
    print("removing existing file", generatedHeaderFile)
    os.remove(generatedHeaderFile)
 
weightFile = "Higgs_Model.root"
if os.path.exists(weightFile):
    print("removing existing file", weightFile)
    os.remove(weightFile)
 
GenerateModelCode(model1, generatedHeaderFile)
GenerateModelCode(model2, generatedHeaderFile)
GenerateModelCode(model3, generatedHeaderFile)
 
# compile the generated code
 
ROOT.gInterpreter.Declare('#include "' + generatedHeaderFile + '"')
 
 
# run the inference on the test data
session1 = ROOT.TMVA_SOFIE_Higgs_Model_4L_50.Session("Higgs_Model.root")
session2 = ROOT.TMVA_SOFIE_Higgs_Model_4L_200.Session("Higgs_Model.root")
session3 = ROOT.TMVA_SOFIE_Higgs_Model_2L_500.Session("Higgs_Model.root")
 
hs1 = ROOT.TH1D("hs1", "Signal result 4L 50", 100, 0, 1)
hs2 = ROOT.TH1D("hs2", "Signal result 4L 200", 100, 0, 1)
hs3 = ROOT.TH1D("hs3", "Signal result 2L 500", 100, 0, 1)
 
hb1 = ROOT.TH1D("hb1", "Background result 4L 50", 100, 0, 1)
hb2 = ROOT.TH1D("hb2", "Background result 4L 200", 100, 0, 1)
hb3 = ROOT.TH1D("hb3", "Background result 2L 500", 100, 0, 1)
 
 
def EvalModel(session, x):
    result = session.infer(x)
    return result[0]
 
 
for i in range(0, x_test.shape[0]):
    result1 = EvalModel(session1, x_test[i, :])
    result2 = EvalModel(session2, x_test[i, :])
    result3 = EvalModel(session3, x_test[i, :])
    if y_test[i] == 1:
        hs1.Fill(result1)
        hs2.Fill(result2)
        hs3.Fill(result3)
    else:
        hb1.Fill(result1)
        hb2.Fill(result2)
        hb3.Fill(result3)
 
 
def PlotHistos(hs, hb):
    hs.SetLineColor("kRed")
    hb.SetLineColor("kBlue")
    hs.Draw()
    hb.Draw("same")
 
 
c1 = ROOT.TCanvas()
c1.Divide(1, 3)
c1.cd(1)
PlotHistos(hs1, hb1)
c1.cd(2)
PlotHistos(hs2, hb2)
c1.cd(3)
PlotHistos(hs3, hb3)
c1.Draw()
 
## draw also ROC curves
 
 
def GetContent(h):
    n = h.GetNbinsX()
    x = ROOT.std.vector["float"](n)
    w = ROOT.std.vector["float"](n)
    for i in range(0, n):
        x[i] = h.GetBinCenter(i + 1)
        w[i] = h.GetBinContent(i + 1)
    return x, w
 
 
def MakeROCCurve(hs, hb):
    xs, ws = GetContent(hs)
    xb, wb = GetContent(hb)
    roc = ROOT.TMVA.ROCCurve(xs, xb, ws, wb)
    print("ROC integral for ", hs.GetName(), roc.GetROCIntegral())
    curve = roc.GetROCCurve()
    curve.SetName(hs.GetName())
    return roc, curve
 
 
c2 = ROOT.TCanvas()
 
r1, curve1 = MakeROCCurve(hs1, hb1)
curve1.SetLineColor("kRed")
curve1.Draw("AC")
 
r2, curve2 = MakeROCCurve(hs2, hb2)
curve2.SetLineColor("kBlue")
curve2.Draw("C")
 
r3, curve3 = MakeROCCurve(hs3, hb3)
curve3.SetLineColor("kGreen")
curve3.Draw("C")
 
c2.Draw()

size of data 10000
Model: "sequential"
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Layer (type)                    ┃ Output Shape           ┃       Param # ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ dense (Dense)                   │ (None, 64)             │           512 │
├─────────────────────────────────┼────────────────────────┼───────────────┤
│ dense_1 (Dense)                 │ (None, 64)             │         4,160 │
├─────────────────────────────────┼────────────────────────┼───────────────┤
│ dense_2 (Dense)                 │ (None, 64)             │         4,160 │
├─────────────────────────────────┼────────────────────────┼───────────────┤
│ dense_3 (Dense)                 │ (None, 64)             │         4,160 │
├─────────────────────────────────┼────────────────────────┼───────────────┤
│ dense_4 (Dense)                 │ (None, 1)              │            65 │
└─────────────────────────────────┴────────────────────────┴───────────────┘
 Total params: 13,057 (51.00 KB)
 Trainable params: 13,057 (51.00 KB)
 Non-trainable params: 0 (0.00 B)
Epoch 1/5
 
␛[1m  1/200␛[0m ␛[37m━━━━━━━━━━━━━━━━━━━━␛[0m ␛[1m2:11␛[0m 660ms/step - accuracy: 0.4600 - loss: 0.7007␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m 54/200␛[0m ␛[32m━━━━━␛[0m␛[37m━━━━━━━━━━━━━━━␛[0m ␛[1m0s␛[0m 951us/step - accuracy: 0.5526 - loss: 0.6814  ␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m111/200␛[0m ␛[32m━━━━━━━━━━━␛[0m␛[37m━━━━━━━━━␛[0m ␛[1m0s␛[0m 914us/step - accuracy: 0.5741 - loss: 0.6719␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m175/200␛[0m ␛[32m━━━━━━━━━━━━━━━━━␛[0m␛[37m━━━␛[0m ␛[1m0s␛[0m 867us/step - accuracy: 0.5831 - loss: 0.6674␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m200/200␛[0m ␛[32m━━━━━━━━━━━━━━━━━━━━␛[0m␛[37m␛[0m ␛[1m1s␛[0m 911us/step - accuracy: 0.5874 - loss: 0.6663
Epoch 2/5
 
␛[1m  1/200␛[0m ␛[37m━━━━━━━━━━━━━━━━━━━━␛[0m ␛[1m2s␛[0m 12ms/step - accuracy: 0.6400 - loss: 0.6203␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m 64/200␛[0m ␛[32m━━━━━━␛[0m␛[37m━━━━━━━━━━━━━━␛[0m ␛[1m0s␛[0m 803us/step - accuracy: 0.6366 - loss: 0.6408␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m128/200␛[0m ␛[32m━━━━━━━━━━━━␛[0m␛[37m━━━━━━━━␛[0m ␛[1m0s␛[0m 792us/step - accuracy: 0.6309 - loss: 0.6460␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m190/200␛[0m ␛[32m━━━━━━━━━━━━━━━━━━━␛[0m␛[37m━␛[0m ␛[1m0s␛[0m 797us/step - accuracy: 0.6352 - loss: 0.6422␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m200/200␛[0m ␛[32m━━━━━━━━━━━━━━━━━━━━␛[0m␛[37m␛[0m ␛[1m0s␛[0m 844us/step - accuracy: 0.6363 - loss: 0.6412
Epoch 3/5
 
␛[1m  1/200␛[0m ␛[37m━━━━━━━━━━━━━━━━━━━━␛[0m ␛[1m2s␛[0m 11ms/step - accuracy: 0.7200 - loss: 0.6064␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m 67/200␛[0m ␛[32m━━━━━━␛[0m␛[37m━━━━━━━━━━━━━━␛[0m ␛[1m0s␛[0m 759us/step - accuracy: 0.6406 - loss: 0.6355␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m133/200␛[0m ␛[32m━━━━━━━━━━━━━␛[0m␛[37m━━━━━━━␛[0m ␛[1m0s␛[0m 759us/step - accuracy: 0.6417 - loss: 0.6333␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m200/200␛[0m ␛[32m━━━━━━━━━━━━━━━━━━━━␛[0m␛[37m␛[0m ␛[1m0s␛[0m 801us/step - accuracy: 0.6461 - loss: 0.6287
Epoch 4/5
 
␛[1m  1/200␛[0m ␛[37m━━━━━━━━━━━━━━━━━━━━␛[0m ␛[1m2s␛[0m 11ms/step - accuracy: 0.6800 - loss: 0.6063␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m 55/200␛[0m ␛[32m━━━━━␛[0m␛[37m━━━━━━━━━━━━━━━␛[0m ␛[1m0s␛[0m 932us/step - accuracy: 0.6516 - loss: 0.6262␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m123/200␛[0m ␛[32m━━━━━━━━━━━━␛[0m␛[37m━━━━━━━━␛[0m ␛[1m0s␛[0m 822us/step - accuracy: 0.6528 - loss: 0.6187␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m193/200␛[0m ␛[32m━━━━━━━━━━━━━━━━━━━␛[0m␛[37m━␛[0m ␛[1m0s␛[0m 783us/step - accuracy: 0.6516 - loss: 0.6191␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m200/200␛[0m ␛[32m━━━━━━━━━━━━━━━━━━━━␛[0m␛[37m␛[0m ␛[1m0s␛[0m 830us/step - accuracy: 0.6517 - loss: 0.6186
Epoch 5/5
 
␛[1m  1/200␛[0m ␛[37m━━━━━━━━━━━━━━━━━━━━␛[0m ␛[1m2s␛[0m 11ms/step - accuracy: 0.6400 - loss: 0.6010␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m 70/200␛[0m ␛[32m━━━━━━━␛[0m␛[37m━━━━━━━━━━━━━␛[0m ␛[1m0s␛[0m 736us/step - accuracy: 0.6671 - loss: 0.6114␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m138/200␛[0m ␛[32m━━━━━━━━━━━━━␛[0m␛[37m━━━━━━━␛[0m ␛[1m0s␛[0m 738us/step - accuracy: 0.6630 - loss: 0.6125␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m200/200␛[0m ␛[32m━━━━━━━━━━━━━━━━━━━━␛[0m␛[37m␛[0m ␛[1m0s␛[0m 785us/step - accuracy: 0.6665 - loss: 0.6093
Model: "sequential_1"
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Layer (type)                    ┃ Output Shape           ┃       Param # ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ dense_5 (Dense)                 │ (None, 64)             │           512 │
├─────────────────────────────────┼────────────────────────┼───────────────┤
│ dense_6 (Dense)                 │ (None, 64)             │         4,160 │
├─────────────────────────────────┼────────────────────────┼───────────────┤
│ dense_7 (Dense)                 │ (None, 64)             │         4,160 │
├─────────────────────────────────┼────────────────────────┼───────────────┤
│ dense_8 (Dense)                 │ (None, 64)             │         4,160 │
├─────────────────────────────────┼────────────────────────┼───────────────┤
│ dense_9 (Dense)                 │ (None, 1)              │            65 │
└─────────────────────────────────┴────────────────────────┴───────────────┘
 Total params: 13,057 (51.00 KB)
 Trainable params: 13,057 (51.00 KB)
 Non-trainable params: 0 (0.00 B)
Epoch 1/5
 
␛[1m  1/200␛[0m ␛[37m━━━━━━━━━━━━━━━━━━━━␛[0m ␛[1m1:58␛[0m 598ms/step - accuracy: 0.5200 - loss: 0.6942␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m 62/200␛[0m ␛[32m━━━━━━␛[0m␛[37m━━━━━━━━━━━━━━␛[0m ␛[1m0s␛[0m 831us/step - accuracy: 0.5635 - loss: 0.6801  ␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m126/200␛[0m ␛[32m━━━━━━━━━━━━␛[0m␛[37m━━━━━━━━␛[0m ␛[1m0s␛[0m 810us/step - accuracy: 0.5757 - loss: 0.6737␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m192/200␛[0m ␛[32m━━━━━━━━━━━━━━━━━━━␛[0m␛[37m━␛[0m ␛[1m0s␛[0m 795us/step - accuracy: 0.5884 - loss: 0.6675␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m200/200␛[0m ␛[32m━━━━━━━━━━━━━━━━━━━━␛[0m␛[37m␛[0m ␛[1m1s␛[0m 840us/step - accuracy: 0.5909 - loss: 0.6656
Epoch 2/5
 
␛[1m  1/200␛[0m ␛[37m━━━━━━━━━━━━━━━━━━━━␛[0m ␛[1m2s␛[0m 11ms/step - accuracy: 0.6200 - loss: 0.6648␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m 66/200␛[0m ␛[32m━━━━━━␛[0m␛[37m━━━━━━━━━━━━━━␛[0m ␛[1m0s␛[0m 777us/step - accuracy: 0.6379 - loss: 0.6395␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m131/200␛[0m ␛[32m━━━━━━━━━━━━━␛[0m␛[37m━━━━━━━␛[0m ␛[1m0s␛[0m 779us/step - accuracy: 0.6366 - loss: 0.6388␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m197/200␛[0m ␛[32m━━━━━━━━━━━━━━━━━━━␛[0m␛[37m━␛[0m ␛[1m0s␛[0m 773us/step - accuracy: 0.6363 - loss: 0.6375␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m200/200␛[0m ␛[32m━━━━━━━━━━━━━━━━━━━━␛[0m␛[37m␛[0m ␛[1m0s␛[0m 822us/step - accuracy: 0.6367 - loss: 0.6378
Epoch 3/5
 
␛[1m  1/200␛[0m ␛[37m━━━━━━━━━━━━━━━━━━━━␛[0m ␛[1m2s␛[0m 11ms/step - accuracy: 0.6000 - loss: 0.6332␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m 68/200␛[0m ␛[32m━━━━━━␛[0m␛[37m━━━━━━━━━━━━━━␛[0m ␛[1m0s␛[0m 756us/step - accuracy: 0.6376 - loss: 0.6380␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m136/200␛[0m ␛[32m━━━━━━━━━━━━━␛[0m␛[37m━━━━━━━␛[0m ␛[1m0s␛[0m 751us/step - accuracy: 0.6493 - loss: 0.6315␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m200/200␛[0m ␛[32m━━━━━━━━━━━━━━━━━━━━␛[0m␛[37m␛[0m ␛[1m0s␛[0m 797us/step - accuracy: 0.6500 - loss: 0.6270
Epoch 4/5
 
␛[1m  1/200␛[0m ␛[37m━━━━━━━━━━━━━━━━━━━━␛[0m ␛[1m2s␛[0m 11ms/step - accuracy: 0.6400 - loss: 0.6052␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m 70/200␛[0m ␛[32m━━━━━━━␛[0m␛[37m━━━━━━━━━━━━━␛[0m ␛[1m0s␛[0m 726us/step - accuracy: 0.6606 - loss: 0.6147␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m138/200␛[0m ␛[32m━━━━━━━━━━━━━␛[0m␛[37m━━━━━━━␛[0m ␛[1m0s␛[0m 732us/step - accuracy: 0.6633 - loss: 0.6147␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m200/200␛[0m ␛[32m━━━━━━━━━━━━━━━━━━━━␛[0m␛[37m␛[0m ␛[1m0s␛[0m 781us/step - accuracy: 0.6572 - loss: 0.6200
Epoch 5/5
 
␛[1m  1/200␛[0m ␛[37m━━━━━━━━━━━━━━━━━━━━␛[0m ␛[1m2s␛[0m 11ms/step - accuracy: 0.6400 - loss: 0.5676␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m 70/200␛[0m ␛[32m━━━━━━━␛[0m␛[37m━━━━━━━━━━━━━␛[0m ␛[1m0s␛[0m 729us/step - accuracy: 0.6840 - loss: 0.6023␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m137/200␛[0m ␛[32m━━━━━━━━━━━━━␛[0m␛[37m━━━━━━━␛[0m ␛[1m0s␛[0m 743us/step - accuracy: 0.6771 - loss: 0.6092␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m200/200␛[0m ␛[32m━━━━━━━━━━━━━━━━━━━━␛[0m␛[37m␛[0m ␛[1m0s␛[0m 798us/step - accuracy: 0.6684 - loss: 0.6124
Model: "sequential_2"
┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓
┃ Layer (type)                    ┃ Output Shape           ┃       Param # ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩
│ dense_10 (Dense)                │ (None, 64)             │           512 │
├─────────────────────────────────┼────────────────────────┼───────────────┤
│ dense_11 (Dense)                │ (None, 64)             │         4,160 │
├─────────────────────────────────┼────────────────────────┼───────────────┤
│ dense_12 (Dense)                │ (None, 64)             │         4,160 │
├─────────────────────────────────┼────────────────────────┼───────────────┤
│ dense_13 (Dense)                │ (None, 64)             │         4,160 │
├─────────────────────────────────┼────────────────────────┼───────────────┤
│ dense_14 (Dense)                │ (None, 1)              │            65 │
└─────────────────────────────────┴────────────────────────┴───────────────┘
 Total params: 13,057 (51.00 KB)
 Trainable params: 13,057 (51.00 KB)
 Non-trainable params: 0 (0.00 B)
Epoch 1/5
 
␛[1m  1/200␛[0m ␛[37m━━━━━━━━━━━━━━━━━━━━␛[0m ␛[1m1:59␛[0m 600ms/step - accuracy: 0.5800 - loss: 0.6941␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m 60/200␛[0m ␛[32m━━━━━━␛[0m␛[37m━━━━━━━━━━━━━━␛[0m ␛[1m0s␛[0m 849us/step - accuracy: 0.5507 - loss: 0.6840  ␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m123/200␛[0m ␛[32m━━━━━━━━━━━━␛[0m␛[37m━━━━━━━━␛[0m ␛[1m0s␛[0m 825us/step - accuracy: 0.5633 - loss: 0.6767␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m189/200␛[0m ␛[32m━━━━━━━━━━━━━━━━━━␛[0m␛[37m━━␛[0m ␛[1m0s␛[0m 804us/step - accuracy: 0.5794 - loss: 0.6703␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m200/200␛[0m ␛[32m━━━━━━━━━━━━━━━━━━━━␛[0m␛[37m␛[0m ␛[1m1s␛[0m 850us/step - accuracy: 0.5826 - loss: 0.6684
Epoch 2/5
 
␛[1m  1/200␛[0m ␛[37m━━━━━━━━━━━━━━━━━━━━␛[0m ␛[1m2s␛[0m 11ms/step - accuracy: 0.6800 - loss: 0.6438␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m 66/200␛[0m ␛[32m━━━━━━␛[0m␛[37m━━━━━━━━━━━━━━␛[0m ␛[1m0s␛[0m 776us/step - accuracy: 0.6185 - loss: 0.6513␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m130/200␛[0m ␛[32m━━━━━━━━━━━━━␛[0m␛[37m━━━━━━━␛[0m ␛[1m0s␛[0m 783us/step - accuracy: 0.6298 - loss: 0.6406␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m195/200␛[0m ␛[32m━━━━━━━━━━━━━━━━━━━␛[0m␛[37m━␛[0m ␛[1m0s␛[0m 781us/step - accuracy: 0.6355 - loss: 0.6382␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m200/200␛[0m ␛[32m━━━━━━━━━━━━━━━━━━━━␛[0m␛[37m␛[0m ␛[1m0s␛[0m 830us/step - accuracy: 0.6350 - loss: 0.6388
Epoch 3/5
 
␛[1m  1/200␛[0m ␛[37m━━━━━━━━━━━━━━━━━━━━␛[0m ␛[1m2s␛[0m 11ms/step - accuracy: 0.5400 - loss: 0.6886␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m 66/200␛[0m ␛[32m━━━━━━␛[0m␛[37m━━━━━━━━━━━━━━␛[0m ␛[1m0s␛[0m 777us/step - accuracy: 0.6558 - loss: 0.6239␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m133/200␛[0m ␛[32m━━━━━━━━━━━━━␛[0m␛[37m━━━━━━━␛[0m ␛[1m0s␛[0m 767us/step - accuracy: 0.6511 - loss: 0.6270␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m200/200␛[0m ␛[32m━━━━━━━━━━━━━━━━━━━━␛[0m␛[37m␛[0m ␛[1m0s␛[0m 806us/step - accuracy: 0.6502 - loss: 0.6258
Epoch 4/5
 
␛[1m  1/200␛[0m ␛[37m━━━━━━━━━━━━━━━━━━━━␛[0m ␛[1m2s␛[0m 11ms/step - accuracy: 0.6000 - loss: 0.6271␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m 68/200␛[0m ␛[32m━━━━━━␛[0m␛[37m━━━━━━━━━━━━━━␛[0m ␛[1m0s␛[0m 753us/step - accuracy: 0.6571 - loss: 0.6236␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m137/200␛[0m ␛[32m━━━━━━━━━━━━━␛[0m␛[37m━━━━━━━␛[0m ␛[1m0s␛[0m 740us/step - accuracy: 0.6562 - loss: 0.6219␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m200/200␛[0m ␛[32m━━━━━━━━━━━━━━━━━━━━␛[0m␛[37m␛[0m ␛[1m0s␛[0m 790us/step - accuracy: 0.6564 - loss: 0.6183
Epoch 5/5
 
␛[1m  1/200␛[0m ␛[37m━━━━━━━━━━━━━━━━━━━━␛[0m ␛[1m2s␛[0m 11ms/step - accuracy: 0.5200 - loss: 0.7222␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m 70/200␛[0m ␛[32m━━━━━━━␛[0m␛[37m━━━━━━━━━━━━━␛[0m ␛[1m0s␛[0m 730us/step - accuracy: 0.6769 - loss: 0.6053␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m139/200␛[0m ␛[32m━━━━━━━━━━━━━␛[0m␛[37m━━━━━━━␛[0m ␛[1m0s␛[0m 731us/step - accuracy: 0.6663 - loss: 0.6110␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈␈
␛[1m200/200␛[0m ␛[32m━━━━━━━━━━━━━━━━━━━━␛[0m␛[37m␛[0m ␛[1m0s␛[0m 782us/step - accuracy: 0.6670 - loss: 0.6105
PyKeras: parsing model  Higgs_Model_4L_50.keras
Generating inference code for the Keras model from  Higgs_Model_4L_50.keras in the header  Higgs_Model.hxx
PyKeras: parsing model  Higgs_Model_4L_200.keras
Generating inference code for the Keras model from  Higgs_Model_4L_200.keras in the header  Higgs_Model.hxx
PyKeras: parsing model  Higgs_Model_2L_500.keras
Generating inference code for the Keras model from  Higgs_Model_2L_500.keras in the header  Higgs_Model.hxx
ROC integral for  hs1 0.7375013680429018
ROC integral for  hs2 0.7377051544336427
ROC integral for  hs3 0.731766208188289

Author

Lorenzo Moneta

Definition in file TMVA_SOFIE_Models.py.