ImageHandsOn_Solution3.cntk
# CNTK Configuration File for training a simple CIFAR-10 convnet.
# During the hands-on tutorial, this will be fleshed out into a ResNet-20 model.
command = TrainConvNet:Eval
makeMode = false ; traceLevel = 0 ; deviceId = "auto"
rootDir = "." ; dataDir = "$rootDir$" ; modelDir = "$rootDir$/Models"
modelPath = "$modelDir$/cifar10.cmf"
# Training action for a convolutional network
TrainConvNet = {
action = "train"
BrainScriptNetworkBuilder = {
imageShape = 32:32:3
labelDim = 10
model (features) = {
MyLayer (x, depth, initValueScale) =
{
c = ConvolutionalLayer {depth, (5:5), pad = true, ##### no activation=ReLU
init = "gaussian", initValueScale = initValueScale} (x)
b = BatchNormalizationLayer {spatialRank = 2} (c)
r = ReLU (b) ##### now called explicitly
p = MaxPoolingLayer {(3:3), stride = (2:2)} (r)
}.p
featNorm = features - Constant (128)
p1 = MyLayer (featNorm, 32, 0.0043)
p2 = MyLayer (p1, 32, 1.414)
p3 = MyLayer (p2, 64, 1.414)
d1 = DenseLayer {64, init = "gaussian", initValueScale = 12} (p3)
d1_bnr = ReLU (BatchNormalizationLayer {normalizationTimeConstant = 4096} (d1)) ##### added BN and explicit ReLU
d1_d = Dropout (d1_bnr) ##### d1 -> d1_bnr
z = LinearLayer {10, init = "gaussian", initValueScale = 1.5} (d1_d)
}.z
# inputs
features = Input {imageShape}
labels = Input {labelDim}
# apply model to features
z = model (features)
# connect to system
ce = CrossEntropyWithSoftmax (labels, z)
errs = ClassificationError (labels, z)
featureNodes = (features)
labelNodes = (labels)
criterionNodes = (ce)
evaluationNodes = (errs)
outputNodes = (z)
}
SGD = {
epochSize = 50000
maxEpochs = 10 ; minibatchSize = 64
learningRatesPerSample = 0.00046875*7:0.00015625*10:0.000046875*10:0.000015625
momentumAsTimeConstant = 0
L2RegWeight = 0
dropoutRate = 0*1:0.5
firstMBsToShowResult = 10 ; numMBsToShowResult = 100
}
reader = {
verbosity = 0 ; randomize = true
deserializers = ({
type = "ImageDeserializer" ; module = "ImageReader"
file = "$dataDir$/cifar-10-batches-py/train_map.txt"
input = {
features = { transforms = (
{ type = "Crop" ; cropType = "random" ; cropRatio = 0.8 ; jitterType = "uniRatio" } :
{ type = "Scale" ; width = 32 ; height = 32 ; channels = 3 ; interpolations = "linear" } :
{ type = "Transpose" }
)}
labels = { labelDim = 10 }
}
})
}
}
# Eval action
Eval = {
action = "eval"
minibatchSize = 16
evalNodeNames = errs
reader = {
verbosity = 0 ; randomize = true
deserializers = ({
type = "ImageDeserializer" ; module = "ImageReader"
file = "$dataDir$/cifar-10-batches-py/test_map.txt"
input = {
features = { transforms = (
{ type = "Scale" ; width = 32 ; height = 32 ; channels = 3 ; interpolations = "linear" } :
{ type = "Transpose" }
)}
labels = { labelDim = 10 }
}
})
}
}