def __init__(self,
vocab_size,
max_num_hidden_layers,
embedding_dim,
n_classes,
output_dim,
dropout,
batch_size,
b=0.99,
n=0.01,
s=0.2,
use_cuda=False):
super().__init__(vocab_size,
max_num_hidden_layers,
embedding_dim,
n_classes,
output_dim,
dropout,
batch_size,
b=b,
n=n,
s=s,
use_cuda=use_cuda)
self.e = Parameter(torch.tensor(e), requires_grad=False)
self.arms_values = Parameter(torch.arange(n_classes),
requires_grad=False)
self.explorations_mab = []
for i in range(n_classes):
self.explorations_mab.append(algs.ThompsomSampling(len(e)))
def __init__(self,
features_size,
max_num_hidden_layers,
qtd_neuron_per_hidden_layer,
n_classes,
b=0.99,
n=0.01,
s=0.2,
e=[0.5, 0.35, 0.2, 0.1, 0.05],
use_cuda=False):
super().__init__(features_size,
max_num_hidden_layers,
qtd_neuron_per_hidden_layer,
n_classes,
b=b,
n=n,
s=s,
use_cuda=use_cuda)
self.e = Parameter(torch.tensor(e), requires_grad=False)
self.arms_values = Parameter(torch.arange(n_classes),
requires_grad=False)
self.explorations_mab = []
for i in range(n_classes):
self.explorations_mab.append(algs.ThompsomSampling(len(e)))
def get_algorithm(self, name, number_of_arms):
"""
This method instantiate the algorithm class.
:param name: Name of algorithm.
:param number_of_arms: Number of arms.
:return: Returns the algorithm instance.
"""
if name == 'ths':
alg = algs.ThompsomSampling(number_of_arms)
elif name == 'tuned':
alg = algs.UCBTuned(number_of_arms)
elif name == 'ucb1':
alg = algs.UCB1(number_of_arms)
return alg