def __init__(self,ShapeData,GradType,dimState=30):
super(model_GradUpdate3, self).__init__()
with torch.no_grad():
self.GradType = GradType
self.shape = ShapeData
self.DimState = dimState
self.layer_dim = 1
self.compute_Grad = Compute_Grad(ShapeData,GradType)
if len(self.shape) == 2: ## 1D Data
self.convLayer = torch.Linear(dimState,self.shape[0]*self.shape[1])
self.lstm = torch.nn.LSTM(self.shape[0]*self.shape[1],self.DimState,self.layer_dim)
else:
self.convLayer = torch.Linear(dimState,self.shape[0]*self.shape[1]*self.shape[2])
self.lstm = torch.nn.LSTM(self.shape[0]*self.shape[1]*self.shape[2],self.DimState,self.layer_dim)
def __init__(self, d_model, vocab):
super(Generator, self).__init__()
self.fc = torch.Linear(d_model, vocab)
def __init__(self, num_user, num_movies):
super().__init__()
self.user_embed = nn.embedding(num_user, 32)
self.movie_embed = nn.embedding(num_movies, 32)
self.out = nn.Linear(64, 1)
self.step_scheduler_after = 'epoch'
def __init__(self, input_size, hidden_size, output_size):
super(pytorchNetwork, self).__init__()
self.layer1 = nn.Linear(input_size, hidden_size)
self.layer2 = nn.Linear(hidden_size, output_size)
def __init__(self):
super().__init__()
self.layer1 = torch.Linear(28 * 28, 10)