def train_dA(self,data,corruption_level=0.2,num_epochs=200,lr=0.1,output_folder="",iteration=0): train_data = data #print_var = theano.printing.Print()(self.dA.input) #print_fn = theano.function([self.dA.input],print_var) # pdb.set_trace() train_set = SequenceDataset(train_data,batch_size=20,number_batches=None) sgd_optimizer(self.dA.params,[self.dA.input],self.dA.cost,train_set,lr=lr, num_epochs=num_epochs,save=True,output_folder=output_folder,iteration=iteration)
def train_NADE(self, data, num_epochs=200, lr=0.1, output_folder="", iteration=0): train_data = data train_set = SequenceDataset(train_data,batch_size=20,number_batches=None) sgd_optimizer(self.NADE.params,[self.NADE.v],self.NADE.cost,train_set, lr=lr,num_epochs=num_epochs,save=False, output_folder=output_folder,iteration=iteration)
def train_dA(self, data, corruption_level=0.2, num_epochs=200, lr=0.1, output_folder="", iteration=0): train_data = data train_set = SequenceDataset(train_data,batch_size=20,number_batches=None) sgd_optimizer(self.dA.params,[self.dA.input],self.dA.cost,train_set, lr=lr,num_epochs=num_epochs,save=False, output_folder=output_folder,iteration=iteration)
def train_RBM(self,k=20,lr=0.1): train_data = self.genotypes_history.top_x_percent() train_set = SequenceDataset(train_data,batch_size=20,number_batches=None) inputs,params,cost,monitor,updates,consider_constant = self.RBM.build_RBM(k=k) sgd_optimizer(params,[inputs],cost,train_set,updates_old=updates,monitor=monitor, consider_constant=[consider_constant],lr=0.1,num_epochs=10)
def train_RBM(self,data,num_epochs=200,lr=0.1,output_folder="",): train_data = data train_set = SequenceDataset(train_data,batch_size=20,number_batches=None) sgd_optimizer(self.RBM.params,[self.RBM.input],self.RBM.cost,train_set,consider_constant=self.RBM.consider_constant,updates=self.RBM.updates,save=True)
def train_dA(self,corruption_level=0.2): train_data = self.genotypes_history.top_x_percent() train_set = SequenceDataset(train_data,batch_size=20,number_batches=None) sgd_optimizer(self.dA.params,[self.dA.input],self.cost,train_set,lr=0.1,num_epochs=200)