def __init__(self, *, nx, nt, cnnSize=32, cp1=(64, 3, 2), cp2=(128, 5, 2)):
super(Cnn1d, self).__init__()
self.nx = nx
self.nt = nt
cOut, f, p = cp1
self.conv1 = nn.Conv1d(nx, cOut, f)
self.pool1 = nn.MaxPool1d(p)
lTmp = int(cnn.calConvSize(nt, f, 0, 1, 1) / p)
cIn = cOut
cOut, f, p = cp2
self.conv2 = nn.Conv1d(cIn, cOut, f)
self.pool2 = nn.MaxPool1d(p)
lTmp = int(cnn.calConvSize(lTmp, f, 0, 1, 1) / p)
self.flatLength = int(cOut * lTmp)
self.fc1 = nn.Linear(self.flatLength, cnnSize)
self.fc2 = nn.Linear(cnnSize, cnnSize)
self.name = "Cnn1d"
self.is_legacy = True
def __init__(self,
*,
nx,
ny,
nobs,
hiddenSize,
nkernel=(10, 5),
kernelSize=(3, 3),
stride=(2, 1),
dr=0.5,
poolOpt=None,
cnndr=0.0):
# two convolutional layer
super(CNN1dLCmodel, self).__init__()
self.nx = nx
self.ny = ny
self.obs = nobs
self.hiddenSize = hiddenSize
nlayer = len(nkernel)
self.features = nn.Sequential()
ninchan = 1 # need to modify the hardcode: 4 for smap and 1 for FDC
Lout = nobs
for ii in range(nlayer):
ConvLayer = cnn.CNN1dkernel(
ninchannel=ninchan,
nkernel=nkernel[ii],
kernelSize=kernelSize[ii],
stride=stride[ii],
)
self.features.add_module("CnnLayer%d" % (ii + 1), ConvLayer)
if cnndr != 0.0:
self.features.add_module("dropout%d" % (ii + 1),
nn.Dropout(p=cnndr))
ninchan = nkernel[ii]
Lout = cnn.calConvSize(lin=Lout,
kernel=kernelSize[ii],
stride=stride[ii])
self.features.add_module("Relu%d" % (ii + 1), nn.ReLU())
if poolOpt is not None:
self.features.add_module("Pooling%d" % (ii + 1),
nn.MaxPool1d(poolOpt[ii]))
Lout = cnn.calPoolSize(lin=Lout, kernel=poolOpt[ii])
self.Ncnnout = int(
Lout * nkernel[-1]) # total CNN feature number after convolution
Nf = self.Ncnnout + nx
self.linearIn = torch.nn.Linear(Nf, hiddenSize)
self.lstm = rnn.CudnnLstm(inputSize=hiddenSize,
hiddenSize=hiddenSize,
dr=dr)
self.linearOut = torch.nn.Linear(hiddenSize, ny)
self.gpu = 1
self.name = "CNN1dLCmodel"
self.is_legacy = True
def __init__(self,
*,
nx,
ny,
rho,
nkernel=(10, 5),
kernelSize=(3, 3),
stride=(2, 1),
padding=(1, 1),
dr=0.5,
poolOpt=None):
# two convolutional layer
super(LstmCnn1d, self).__init__()
self.nx = nx
self.ny = ny
self.rho = rho
nlayer = len(nkernel)
self.features = nn.Sequential()
ninchan = nx
Lout = rho
for ii in range(nlayer):
# First layer: no dimension reduction
ConvLayer = cnn.CNN1dkernel(
ninchannel=ninchan,
nkernel=nkernel[ii],
kernelSize=kernelSize[ii],
stride=stride[ii],
padding=padding[ii],
)
self.features.add_module("CnnLayer%d" % (ii + 1), ConvLayer)
ninchan = nkernel[ii]
Lout = cnn.calConvSize(lin=Lout,
kernel=kernelSize[ii],
stride=stride[ii])
if poolOpt is not None:
self.features.add_module("Pooling%d" % (ii + 1),
nn.MaxPool1d(poolOpt[ii]))
Lout = cnn.calPoolSize(lin=Lout, kernel=poolOpt[ii])
self.Ncnnout = int(
Lout * nkernel[-1]) # total CNN feature number after convolution
self.name = "LstmCnn1d"
self.is_legacy = True