def test_dnn_trainModel_exceptions():
dnn = DNN("Name", 10, [40])
testDataArray = np.ndarray(shape=(20,10))
testDataArrayLabels = np.ndarray(shape=(20,1))
testDataArrayWeight = np.ndarray(shape=(20,1))
with pytest.raises(RuntimeError):
dnn.trainModel(testDataArray, testDataArrayLabels,testDataArrayWeight, "someFolder")
dnn.modelCompiled = True
with pytest.raises(TypeError):
dnn.trainModel("NotAnArray", testDataArrayLabels,testDataArrayWeight, "someFolder")
with pytest.raises(TypeError):
dnn.trainModel(testDataArray, testDataArrayLabels,"NotAnArray", "someFolder")
def setupDNN(modelDefinitions):
thisDNN = DNN(
identifier = modelDefinitions.config.net.name,
inputDim = modelDefinitions.config.net.inputDimention,
layerDims = modelDefinitions.config.net.layerDimentions,
weightDecay = modelDefinitions.config.net.useWeightDecay,
activation = modelDefinitions.config.net.activation,
outputActivation = modelDefinitions.config.net.outputActivation,
loss = modelDefinitions.config.net.loss,
metric = ["acc"],
batchSize = modelDefinitions.config.net.batchSize
)
thisDNN.loadModel(modelDefinitions.trainingOutput
)
return thisDNN
def test_dnn_trainModel(mocker):
testDataArray = np.ndarray(shape=(20,10))
testDataArrayLabels = np.ndarray(shape=(20,1))
testDataArrayWeight = np.ndarray(shape=(20,1))
dnn = DNN("Name", 10, [40])
dnn.optimizer = "adam"
dnn.buildModel()
dnn.compileModel()
dnn.isBinary = True
m = mocker.MagicMock
m.epoch = 100*[0]
m.history = {"loss" : 99*[0]+[1]}
mocker.patch.object(keras.models.Model , "fit", return_value = m)
print(m)
dnn.trainModel(testDataArray, testDataArrayLabels,testDataArrayWeight, "someFolder")
assert dnn.StopValues == (100, {"loss" : 1})
def test_dnn_compileModel_exceptions():
dnn = DNN("Name", 10, [40])
with pytest.raises(RuntimeError):
dnn.compileModel()
dnn.optimizer = "adagrad"
with pytest.raises(RuntimeError):
dnn.compileModel()
def test_dnn_init():
dnn = DNN("Name", 10, [40])
assert dnn.name == "Name"
assert dnn.inputDimention == 10
assert dnn.nLayer == 1
assert isinstance(dnn.layerDimention, list)
assert dnn.layerDimention[0] == 40
assert dnn.activation == "relu"
assert dnn.outputActivation == "softmax"
def trainDNN(config, batch=False, addMetrics=["MEM"]):
logging.debug("Output folder")
checkNcreateFolder(config.output, onlyFolder=True)
logging.debug("Copying used config to outputfolder")
shutil.copy2(config.path, config.output + "/usedConfig.cfg")
logging.info("Initializing samples and data")
allSample, data = initialize(config, incGenWeights=config.includeGenWeight)
logging.info("Initializing DNN")
thisDNN = DNN(identifier=config.net.name,
inputDim=config.net.inputDimention,
layerDims=config.net.layerDimentions,
weightDecay=config.net.useWeightDecay,
weightDecayLambda=config.net.weightDecayLambda,
activation=config.net.activation,
outputActivation=config.net.outputActivation,
loss=config.net.loss,
metric=["acc"],
batchSize=config.net.batchSize)
logging.info("Setting optimizer")
logging.debug("In config: %s", config.net.optimizer)
thisDNN.optimizer = config.net.optimizer
logging.info("Building model")
if config.net.loss == "binary_crossentropy":
thisDNN.buildModel(nClasses=1,
dropoutAll=config.net.dropoutAll,
dropoutOutput=config.net.dropoutOutput,
dropoutPercent=config.net.dropoutPercent)
else:
thisDNN.buildModel(nClasses=len(data.outputClasses),
dropoutAll=config.net.dropoutAll,
dropoutOutput=config.net.dropoutOutput,
dropoutPercent=config.net.dropoutPercent)
logging.info("Compiling model")
thisDNN.compileModel()
thisDNN.network.summary(print_fn=logging.warning)
if not batch:
input("Press ret")
trainData = data.getTrainData()
trainLabels = data.trainLabels
trainWeights = data.trainTrainingWeights
logging.info("Training DNN")
thisDNN.trainModel(trainData,
trainLabels,
trainWeights,
config.output,
epochs=config.net.trainEpochs,
valSplit=config.net.validationSplit,
earlyStopping=config.net.doEarlyStopping,
patience=config.net.StoppingPatience)
testData = data.getTestData()
testLabels = data.testLabels
testWeights = data.testTrainingWeights
#TODO: Make this configurable
ROCMetrics = []
if (len(addMetrics) == 1 and addMetrics[0] == ""):
addMetrics = []
for metric in addMetrics:
ROCMetrics.append(
(metric, data.getTestData(asMatrix=False)[metric].values))
logging.info("Model evaluation")
thisDNN.evalModel(testData,
testWeights,
testLabels,
trainData,
trainWeights,
trainLabels,
config.trainingVariables,
config.output,
data.outputClasses,
plotMetics=True,
saveData=True,
addROCMetrics=ROCMetrics,
forceColors=config.forceColors)
logging.info("Saving model")
thisDNN.saveModel(config.output, data.transformations)
def test_dnn_compileModel():
dnn = DNN("Name", 10, [40])
dnn.optimizer = "adagrad"
dnn.buildModel()
assert dnn.compileModel(writeyml=True)
def test_dnn_buildModel():
dnn = DNN("Name", 10, [40])
assert dnn.buildModel(plot=True)
def test_dnn_init_exceptions():
with pytest.raises(TypeError):
DNN("Name", 10, "Blubb")
with pytest.raises(RuntimeError):
DNN("Name", 10, [])