def __init__(self):
super().__init__()
MUS = [-0.9, -0.7, -0.5, -0.3, -0.1, 0.1, 0.3, 0.5, 0.7, 0.9, 1.0]
SIGMAS = [0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.001]
self.bert_ranker = VanillaBertRanker()
self.simmat = modeling_util.SimmatModule()
self.kernels = modeling_util.KNRMRbfKernelBank(MUS, SIGMAS)
self.combine = torch.nn.Linear(self.kernels.count() * self.CHANNELS + self.BERT_SIZE, 1)
def __init__(self):
super().__init__()
## KNRM
MUS = [-0.9, -0.7, -0.5, -0.3, -0.1, 0.1, 0.3, 0.5, 0.7, 0.9, 1.0]
SIGMAS = [0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.001]
self.kernels = modeling_util.KNRMRbfKernelBank(MUS, SIGMAS)
#self.batchnorm = torch.nn.BatchNorm1d(self.kernels.count()*self.CHANNELS)
self.combine = torch.nn.Linear(self.kernels.count() * self.CHANNELS, 1)
#self.dropout_two = torch.nn.Dropout(0.1).to('cuda:0')
self.simmat = modeling_util.SimmatModule()
def __init__(self, config_path):
super().__init__(config_path)
# MUS,SIGMAS = modeling_util.init_kernels(21)
MUS = [-0.9, -0.7, -0.5, -0.3, -0.1, 0.1, 0.3, 0.5, 0.7, 0.9, 1.0]
SIGMAS = [0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.001]
self.simmat = modeling_util.SimmatModule()
self.kernels = modeling_util.KNRMRbfKernelBank(MUS, SIGMAS)
self.combine = torch.nn.Linear(self.kernels.count(), 1) #
def __init__(self,QLEN):
super().__init__(QLEN)
MUS,SIGMAS = modeling_util.init_kernels(21)
#MUS = [-0.9, -0.7, -0.5, -0.3, -0.1, 0.1, 0.3, 0.5, 0.7, 0.9, 1.0]
#SIGMAS = [0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.1, 0.001]
# self.bert_ranker = VanillaBertRanker()
self.simmat = modeling_util.SimmatModule()
self.kernels = modeling_util.KNRMRbfKernelBank(MUS, SIGMAS)
self.combine = torch.nn.Linear(self.kernels.count() * self.CHANNELS, 1)