def test_n_iter_without_progress():
# Use a dummy negative n_iter_without_progress and check output on stdout
random_state = check_random_state(0)
X = random_state.randn(100, 10)
for method in ["barnes_hut", "exact"]:
tsne = TSNE(
n_iter_without_progress=-1,
verbose=2,
learning_rate=1e8,
random_state=0,
method=method,
n_iter=351,
init="random",
)
tsne._N_ITER_CHECK = 1
tsne._EXPLORATION_N_ITER = 0
old_stdout = sys.stdout
sys.stdout = StringIO()
try:
tsne.fit_transform(X)
finally:
out = sys.stdout.getvalue()
sys.stdout.close()
sys.stdout = old_stdout
# The output needs to contain the value of n_iter_without_progress
assert (
"did not make any progress during the " "last -1 episodes. Finished." in out
)
def boostrap_do_embedding(X, shared_queue=None):
"""
Boostrap to start doing embedding:
Initialize the tsne object, setup params
"""
print("[TSNEX] Thread to do embedding is starting ... ")
shared_data['queue'] = shared_queue
shared_data['fixed_ids'] = []
shared_data['fixed_pos'] = []
shared_data['errors'] = []
shared_data['grad_norms'] = []
shared_data['z_info'] = np.zeros(X.shape[0])
sklearn.manifold.t_sne._gradient_descent = my_gradient_descent
tsne = TSNE(
n_components=2,
random_state=0,
init='random',
method='exact', # use this method to hook into kl_divergence
perplexity=30.0,
early_exaggeration=12.0,
learning_rate=100.0,
n_iter_without_progress=500,
verbose=2)
tsne._EXPLORATION_N_ITER = 150
X_projected = tsne.fit_transform(X)
return X_projected
def test_bh_match_exact():
# check that the ``barnes_hut`` method match the exact one when
# ``angle = 0`` and ``perplexity > n_samples / 3``
random_state = check_random_state(0)
n_features = 10
X = random_state.randn(30, n_features).astype(np.float32)
X_embeddeds = {}
n_iter = {}
for method in ["exact", "barnes_hut"]:
tsne = TSNE(
n_components=2,
method=method,
learning_rate=1.0,
init="random",
random_state=0,
n_iter=251,
perplexity=30.0,
angle=0,
)
# Kill the early_exaggeration
tsne._EXPLORATION_N_ITER = 0
X_embeddeds[method] = tsne.fit_transform(X)
n_iter[method] = tsne.n_iter_
assert n_iter["exact"] == n_iter["barnes_hut"]
assert_allclose(X_embeddeds["exact"], X_embeddeds["barnes_hut"], rtol=1e-4)
def main(args):
data_path = './data/%s.pkl' % args.dataset
with open(data_path, 'rb') as f:
X, labels = pickle.load(f)
tsne = TSNE(n_iter=args.early_iters, verbose=True)
tsne._EXPLORATION_N_ITER = args.early_iters
Y_seq = extract_sequence(tsne, X)
with open('results/res.pkl', 'wb') as f:
pickle.dump(Y_seq, f)
if not os.path.exists('figures'):
os.mkdir('figures')
lo = Y_seq.min(axis=0).min(axis=0).max()
hi = Y_seq.max(axis=0).max(axis=0).min()
limits = ([lo, hi], [lo, hi])
fig_name = '%s-%d-%d-tsne' % (args.dataset, args.num_iters,
args.early_iters)
fig_path = './figures/%s.gif' % fig_name
savegif(Y_seq, labels, fig_name, fig_path, limits=limits)
