Browse Source

Simplify arguments for ConstellationNet

master
Mattéo Delabre 4 years ago
parent
commit
d8a140d793
Signed by: matteo GPG Key ID: AE3FBD02DC583ABB
  1. 47
      constellation/ConstellationNet.py
  2. 4
      constellation/GaussianChannel.py
  3. 2
      train.py

47
constellation/ConstellationNet.py

@ -13,9 +13,8 @@ class ConstellationNet(nn.Module):
def __init__( def __init__(
self, self,
order=2, order=2,
encoder_layers_sizes=(),
decoder_layers_sizes=(),
channel_model=GaussianChannel()
encoder_layers=(),
decoder_layers=(),
): ):
""" """
Create an autoencoder. Create an autoencoder.
@ -23,56 +22,54 @@ class ConstellationNet(nn.Module):
:param order: Order of the constellation, i.e. the number of messages :param order: Order of the constellation, i.e. the number of messages
that are to be transmitted, or equivalently the number of symbols whose that are to be transmitted, or equivalently the number of symbols whose
placements in the constellation have to be learned. placements in the constellation have to be learned.
:param encoder_layers_sizes: Shape of the encoders hidden layers. The
:param encoder_layers: Shape of the encoders hidden layers. The
size of this sequence is the number of hidden layers, with each element size of this sequence is the number of hidden layers, with each element
being a number which specifies the number of neurons in its channel. being a number which specifies the number of neurons in its channel.
:param decoder_layers_sizes: Shape of the decoders hidden layers. Uses
:param decoder_layers: Shape of the decoders hidden layers. Uses
the same convention as `encoder_layers_sizes` above. the same convention as `encoder_layers_sizes` above.
:param channel_model: Instance of the channel model to use between the
encoder and decoder network.
""" """
super().__init__() super().__init__()
self.order = order self.order = order
# Build the encoder network taking a one-hot encoded message as input # Build the encoder network taking a one-hot encoded message as input
# and outputting an I/Q vector. The network additionally uses hidden # and outputting an I/Q vector. The network additionally uses hidden
# layers as specified in `encoder_layers_sizes`
# layers as specified in `encoder_layers`
prev_layer_size = order prev_layer_size = order
encoder_layers = []
encoder_layers_list = []
for layer_size in encoder_layers_sizes:
encoder_layers.append(nn.Linear(prev_layer_size, layer_size))
encoder_layers.append(nn.ReLU())
encoder_layers.append(nn.BatchNorm1d(layer_size))
for layer_size in encoder_layers:
encoder_layers_list.append(nn.Linear(prev_layer_size, layer_size))
encoder_layers_list.append(nn.ReLU())
encoder_layers_list.append(nn.BatchNorm1d(layer_size))
prev_layer_size = layer_size prev_layer_size = layer_size
encoder_layers += [
encoder_layers_list += [
nn.Linear(prev_layer_size, 2), nn.Linear(prev_layer_size, 2),
NormalizePower(), NormalizePower(),
] ]
self.encoder = nn.Sequential(*encoder_layers)
self.channel = channel_model
self.encoder = nn.Sequential(*encoder_layers_list)
self.channel = GaussianChannel()
# Build the decoder network taking the noisy I/Q vector received from # Build the decoder network taking the noisy I/Q vector received from
# the channel as input and outputting a probability vector for each # the channel as input and outputting a probability vector for each
# original message. The network additionally uses hidden layers as # original message. The network additionally uses hidden layers as
# specified in `decoder_layers_sizes`
# specified in `decoder_layers`
prev_layer_size = 2 prev_layer_size = 2
decoder_layers = []
decoder_layers_list = []
for layer_size in decoder_layers_sizes:
decoder_layers.append(nn.Linear(prev_layer_size, layer_size))
encoder_layers.append(nn.ReLU())
decoder_layers.append(nn.BatchNorm1d(layer_size))
for layer_size in decoder_layers:
decoder_layers_list.append(nn.Linear(prev_layer_size, layer_size))
encoder_layers_list.append(nn.ReLU())
decoder_layers_list.append(nn.BatchNorm1d(layer_size))
prev_layer_size = layer_size prev_layer_size = layer_size
# Softmax is not used at the end of the network because the # Softmax is not used at the end of the network because the
# CrossEntropyLoss criterion is used for training, which includes # CrossEntropyLoss criterion is used for training, which includes
# LogSoftmax # LogSoftmax
decoder_layers.append(nn.Linear(prev_layer_size, order))
decoder_layers_list.append(nn.Linear(prev_layer_size, order))
self.decoder = nn.Sequential(*decoder_layers)
self.decoder = nn.Sequential(*decoder_layers_list)
def forward(self, x): def forward(self, x):
""" """
@ -88,7 +85,7 @@ class ConstellationNet(nn.Module):
def get_constellation(self): def get_constellation(self):
""" """
Extract symbol constellation out of the trained encoder.
Extract the symbol constellation out of the trained encoder.
:return: Matrix containing `order` rows with the nᵗʰ one being the I/Q :return: Matrix containing `order` rows with the nᵗʰ one being the I/Q
vector that is the result of encoding the nᵗʰ message. vector that is the result of encoding the nᵗʰ message.

4
constellation/GaussianChannel.py

@ -4,6 +4,10 @@ import math
class GaussianChannel(nn.Module): class GaussianChannel(nn.Module):
"""
Simulated communication channel that assumes a Gaussian noise model for
taking in account interference.
"""
def __init__(self): def __init__(self):
super().__init__() super().__init__()

2
train.py

@ -5,7 +5,7 @@ from matplotlib import pyplot
from mpl_toolkits.axisartist.axislines import SubplotZero from mpl_toolkits.axisartist.axislines import SubplotZero
import warnings import warnings
torch.manual_seed(42)
torch.manual_seed(57)
# Number of symbols to learn # Number of symbols to learn
order = 16 order = 16

Loading…
Cancel
Save