np.random.seed(args.seed) torch.manual_seed(args.seed) torch.multiprocessing.set_sharing_strategy('file_system') if args.cuda: cudnn.benchmark = True # Define visulaize SummaryWriter instance kwargs = {'num_workers': args.nj, 'pin_memory': False} if args.cuda else {} l2_dist = nn.CosineSimilarity( dim=1, eps=1e-6) if args.cos_sim else PairwiseDistance(2) if args.input_length == 'var': transform = transforms.Compose([ # concateinputfromMFB(num_frames=c.NUM_FRAMES_SPECT, remove_vad=False), varLengthFeat(remove_vad=args.remove_vad), to2tensor() ]) transform_T = transforms.Compose([ # concateinputfromMFB(num_frames=c.NUM_FRAMES_SPECT, input_per_file=args.test_input_per_file, remove_vad=False), varLengthFeat(remove_vad=args.remove_vad), to2tensor() ]) elif args.input_length == 'fix': transform = transforms.Compose( [concateinputfromMFB(remove_vad=args.remove_vad), to2tensor()]) transform_T = transforms.Compose([ concateinputfromMFB(input_per_file=args.test_input_per_file, remove_vad=args.remove_vad), to2tensor()
# num_pro = 1. # for datum in voxceleb: # # Data/voxceleb1/ # # /data/voxceleb/voxceleb1_wav/ # GenerateSpect(wav_path='/data/voxceleb/voxceleb1_wav/' + datum['filename']+'.wav', # write_path=args.dataroot +'/spectrogram/voxceleb1_wav/' + datum['filename']+'.npy') # print('\rprocessed {:2f}% {}/{}.'.format(num_pro/len(voxceleb), num_pro, len(voxceleb)), end='\r') # num_pro += 1 # print('\nComputing Spectrograms success!') # exit(1) if args.acoustic_feature == 'fbank': transform = transforms.Compose([ # concateinputfromMFB(), # truncatedinputfromMFB(), varLengthFeat(), totensor() ]) transform_T = transforms.Compose([ # truncatedinputfromMFB(input_per_file=args.test_input_per_file), concateinputfromMFB(input_per_file=args.test_input_per_file), totensor() ]) file_loader = read_MFB elif args.acoustic_feature == 'spectrogram': # Start from spectrogram transform = transforms.Compose([truncatedinputfromMFB(), totensor()]) transform_T = transforms.Compose([ truncatedinputfromMFB(input_per_file=args.test_input_per_file), totensor()
np.random.seed(args.seed) torch.manual_seed(args.seed) if args.cuda: cudnn.benchmark = True # create logger # Define visulaize SummaryWriter instance kwargs = {'num_workers': 12, 'pin_memory': True} if args.cuda else {} l2_dist = nn.CosineSimilarity( dim=1, eps=1e-6) if args.cos_sim else PairwiseDistance(2) if args.acoustic_feature == 'fbank': transform = transforms.Compose([varLengthFeat(), totensor()]) transform_T = transforms.Compose([varLengthFeat(), totensor()]) file_loader = read_mat else: transform = transforms.Compose([ truncatedinput(), toMFB(), totensor(), # tonormal() ]) file_loader = read_audio # pdb.set_trace() train_dir = KaldiExtractDataset(dir=args.train_dir, loader=file_loader, transform=transform)
opt_kwargs = { 'lr': args.lr, 'lr_decay': args.lr_decay, 'weight_decay': args.weight_decay, 'dampening': args.dampening, 'momentum': args.momentum } l2_dist = nn.CosineSimilarity( dim=1, eps=1e-6) if args.cos_sim else PairwiseDistance(2) if args.acoustic_feature == 'fbank': transform = transforms.Compose([ # concateinputfromMFB(num_frames=c.NUM_FRAMES_SPECT, remove_vad=False), varLengthFeat(remove_vad=True), to2tensor() ]) transform_T = transforms.Compose([ # concateinputfromMFB(num_frames=c.NUM_FRAMES_SPECT, input_per_file=args.test_input_per_file, remove_vad=False), varLengthFeat(remove_vad=True), to2tensor(), # tonormal() ]) transform_V = transforms.Compose( [varLengthFeat(remove_vad=args.remove_vad), to2tensor()]) else: transform = transforms.Compose([ truncatedinput(),
opt_kwargs = { 'lr': args.lr, 'lr_decay': args.lr_decay, 'weight_decay': args.weight_decay, 'dampening': args.dampening, 'momentum': args.momentum } l2_dist = nn.CosineSimilarity( dim=1, eps=1e-6) if args.cos_sim else PairwiseDistance(2) if args.acoustic_feature == 'fbank': transform = transforms.Compose([ # concateinputfromMFB(num_frames=c.NUM_FRAMES_SPECT, remove_vad=False), varLengthFeat(remove_vad=True), to2tensor() ]) transform_T = transforms.Compose([ # concateinputfromMFB(num_frames=c.NUM_FRAMES_SPECT, input_per_file=args.test_input_per_file, remove_vad=False), varLengthFeat(remove_vad=True), to2tensor(), # tonormal() ]) else: transform = transforms.Compose([ truncatedinput(), toMFB(), totensor(), # tonormal()