def train_mvn(iv_file, train_list, preproc_file, name, save_tlist,
append_tlist, output_path, **kwargs):
if preproc_file is not None:
preproc = TransformList.load(preproc_file)
else:
preproc = None
vr_args = VR.filter_args(**kwargs)
vr = VR(iv_file, train_list, preproc, **vr_args)
x = vr.read()
t1 = time.time()
model = MVN(name=name)
model.fit(x)
logging.info('Elapsed time: %.2f s.' % (time.time() - t1))
if save_tlist:
if append_tlist and preproc is not None:
preproc.append(model)
model = preproc
else:
model = TransformList(model)
model.save(output_path)
def plot_vector_hist(iv_file, v_list, preproc_file, output_path, num_bins,
normed, **kwargs):
if preproc_file is not None:
preproc = TransformList.load(preproc_file)
else:
preproc = None
vr_args = VR.filter_args(**kwargs)
vr = VR(iv_file, v_list, preproc, **vr_args)
x = vr.read()
t1 = time.time()
if not os.path.exists(output_path):
os.makedirs(ouput_path)
for i in xrange(x.shape[1]):
fig_file = '%s/D%04d.pdf' % (output_path, i)
plt.hist(x[:, i], num_bins, normed=normed)
plt.xlabel('Dim %d' % i)
plt.grid(True)
plt.show()
plt.savefig(fig_file)
plt.clf()
logging.info('Elapsed time: %.2f s.' % (time.time() - t1))
def train_gauss(iv_file, train_list, preproc_file,
save_tlist, append_tlist, input_path, output_path, **kwargs):
if preproc_file is not None:
preproc = TransformList.load(preproc_file)
else:
preproc = None
vr_args = VR.filter_args(**kwargs)
vr = VR(iv_file, train_list, preproc, **vr_args)
x = vr.read()
t1 = time.time()
model_args = Gaussianizer.filter_args(**kwargs)
model = load_model(input_path, **model_args)
model.fit(x)
if save_tlist:
if append_tlist and preproc is not None:
preproc.append(model)
model = preproc
else:
model = TransformList(model)
model.save(output_path)
def eval_plda(iv_file, ndx_file, enroll_file, test_subseg2orig_file,
preproc_file, coh_iv_file, coh_list, coh_nbest,
coh_nbest_discard, model_file, score_file, plda_type, **kwargs):
logging.info('loading data')
if preproc_file is not None:
preproc = TransformList.load(preproc_file)
else:
preproc = None
tdr = TDR(iv_file, ndx_file, enroll_file, None, test_subseg2orig_file,
preproc)
x_e, x_t, enroll, ndx, orig_seg = tdr.read()
logging.info('loading plda model: %s' % (model_file))
model = F.load_plda(plda_type, model_file)
t1 = time.time()
logging.info('computing llr')
scores = model.llr_1vs1(x_e, x_t)
dt = time.time() - t1
num_trials = len(enroll) * x_t.shape[0]
logging.info(
'scoring elapsed time: %.2f s. elapsed time per trial: %.2f ms.' %
(dt, dt / num_trials * 1000))
logging.info('loading cohort data')
vr = VR(coh_iv_file, coh_list, preproc)
x_coh = vr.read()
t2 = time.time()
logging.info('score cohort vs test')
scores_coh_test = model.llr_1vs1(x_coh, x_t)
logging.info('score enroll vs cohort')
scores_enr_coh = model.llr_1vs1(x_e, x_coh)
dt = time.time() - t2
logging.info('cohort-scoring elapsed time: %.2f s.' % (dt))
t2 = time.time()
logging.info('apply s-norm')
snorm = SNorm(nbest=coh_nbest, nbest_discard=coh_nbest_discard)
scores = snorm.predict(scores, scores_coh_test, scores_enr_coh)
dt = time.time() - t2
logging.info('s-norm elapsed time: %.2f s.' % (dt))
dt = time.time() - t1
logging.info(
('total-scoring elapsed time: %.2f s. '
'elapsed time per trial: %.2f ms.') % (dt, dt / num_trials * 1000))
logging.info('combine cluster scores')
scores = combine_diar_scores(ndx, orig_seg, scores)
logging.info('saving scores to %s' % (score_file))
s = TrialScores(enroll, ndx.seg_set, scores)
s = s.align_with_ndx(ndx)
s.save_txt(score_file)
def train_cw(iv_file, train_list, preproc_file, with_lnorm,
save_tlist, append_tlist, input_path, output_path, **kwargs):
if preproc_file is not None:
preproc = TransformList.load(preproc_file)
else:
preproc = None
vr_args = VR.filter_args(**kwargs)
vr = VR(iv_file, train_list, preproc, **vr_args)
x = vr.read()
t1 = time.time()
model_args = CentWhiten.filter_args(**kwargs)
model = load_model(input_path, with_lnorm, **model_args)
model.fit(x)
logging.info('Elapsed time: %.2f s.' % (time.time()-t1))
x = model.predict(x)
gauss=Normal(x_dim=x.shape[1])
gauss.fit(x=x)
logging.debug(gauss.mu[:4])
logging.debug(gauss.Sigma[:4,:4])
if save_tlist:
if append_tlist and preproc is not None:
preproc.append(model)
model = preproc
else:
model = TransformList(model)
model.save(output_path)
model.save(output_path)
if __name__ == "__main__":
parser=argparse.ArgumentParser(
formatter_class=argparse.ArgumentDefaultsHelpFormatter,
fromfile_prefix_chars='@',
description='Train Centering+Whitening')
parser.add_argument('--iv-file', dest='iv_file', required=True)
parser.add_argument('--train-list', dest='train_list', required=True)
parser.add_argument('--preproc-file', dest='preproc_file', default=None)
VR.add_argparse_args(parser)
CentWhiten.add_argparse_args(parser)
parser.add_argument('--input-path', dest='input_path', default=None)
parser.add_argument('--output-path', dest='output_path', required=True)
parser.add_argument('--no-lnorm', dest='with_lnorm',
default=True, action='store_false')
parser.add_argument('--no-save-tlist', dest='save_tlist',
default=True, action='store_false')
parser.add_argument('--no-append-tlist', dest='append_tlist',
default=True, action='store_false')
parser.add_argument('-v', '--verbose', dest='verbose', default=1, choices=[0, 1, 2, 3], type=int)
args=parser.parse_args()
config_logger(args.verbose)