opt.is_regu_sigma = False
##############################################################################################################
if not opt.is_clip:
opt.CR = 0
######################################## OFDM setting ###########################################
size_after_compress = (opt.size //
(opt.n_downsample**2))**2 * (opt.C_channel // 2)
opt.N = opt.batchSize # Batch size
opt.P = 1 # Number of symbols
opt.M = 64 # Number of subcarriers per symbol
opt.K = 16 # Length of CP
opt.L = 8 # Number of paths
opt.decay = 4
opt.S = size_after_compress // opt.M # Number of packets
opt.is_cfo = False
opt.is_trick = True
opt.is_cfo_random = False
opt.max_ang = 1.7
opt.ang = 1.7
if opt.CE not in ['LS', 'MMSE', 'TRUE']:
raise Exception("Channel estimation method not implemented")
if opt.EQ not in ['ZF', 'MMSE']:
raise Exception("Equalization method not implemented")
dataset_size = len(dataset)
print('#training images = %d' % dataset_size)
elif opt.dataset_mode == 'CelebA':
opt.dataroot = './data/celeba/CelebA_test'
opt.load_size = 80
opt.crop_size = 64
opt.size = 64
dataset = create_dataset(opt) # create a dataset given opt.dataset_mode and other options
dataset_size = len(dataset)
print('#training images = %d' % dataset_size)
else:
raise Exception('Not implemented yet')
opt.K = 256
# Display setting
opt.checkpoints_dir = './Checkpoints/'+ opt.dataset_mode + '_VQVAE'
opt.name = opt.gan_mode + '_K' + str(opt.K)
output_path = './Images/' + opt.dataset_mode + '_VQVAE/' + opt.name
# Choose the neural network model
opt.model = 'VQVAE'
opt.num_test = 10000
opt.how_many_channel = 5
opt.N = opt.how_many_channel