def test_dump_concise():
one = 1
two = 2.1
three = 3.01
exact = 1.000000000000001
# loses the last decimal place
almost = 1.0000000000000001
X = [[one, two, three, exact, almost],
[1e9, 2e18, 3e27, 0, 0],
[0, 0, 0, 0, 0],
[0, 0, 0, 0, 0],
[0, 0, 0, 0, 0]]
y = [one, two, three, exact, almost]
f = BytesIO()
dump_svmlight_file(X, y, f)
f.seek(0)
# make sure it's using the most concise format possible
assert_equal(f.readline(),
b("1 0:1 1:2.1 2:3.01 3:1.000000000000001 4:1\n"))
assert_equal(f.readline(), b("2.1 0:1000000000 1:2e+18 2:3e+27\n"))
assert_equal(f.readline(), b("3.01 \n"))
assert_equal(f.readline(), b("1.000000000000001 \n"))
assert_equal(f.readline(), b("1 \n"))
f.seek(0)
# make sure it's correct too :)
X2, y2 = load_svmlight_file(f)
assert_array_almost_equal(X, X2.toarray())
assert_array_equal(y, y2)
def test_load_zero_based_auto():
data1 = b("-1 1:1 2:2 3:3\n")
data2 = b("-1 0:0 1:1\n")
f1 = BytesIO(data1)
X, y = load_svmlight_file(f1, zero_based="auto")
assert_equal(X.shape, (1, 3))
f1 = BytesIO(data1)
f2 = BytesIO(data2)
X1, y1, X2, y2 = load_svmlight_files([f1, f2], zero_based="auto")
assert_equal(X1.shape, (1, 4))
assert_equal(X2.shape, (1, 4))
def test_dump_multilabel():
X = [[1, 0, 3, 0, 5],
[0, 0, 0, 0, 0],
[0, 5, 0, 1, 0]]
y_dense = [[0, 1, 0], [1, 0, 1], [1, 1, 0]]
y_sparse = sp.csr_matrix(y_dense)
for y in [y_dense, y_sparse]:
f = BytesIO()
dump_svmlight_file(X, y, f, multilabel=True)
f.seek(0)
# make sure it dumps multilabel correctly
assert_equal(f.readline(), b("1 0:1 2:3 4:5\n"))
assert_equal(f.readline(), b("0,2 \n"))
assert_equal(f.readline(), b("0,1 1:5 3:1\n"))
def test_dump_comment():
X, y = load_svmlight_file(datafile)
X = X.toarray()
f = BytesIO()
ascii_comment = "This is a comment\nspanning multiple lines."
dump_svmlight_file(X, y, f, comment=ascii_comment, zero_based=False)
f.seek(0)
X2, y2 = load_svmlight_file(f, zero_based=False)
assert_array_almost_equal(X, X2.toarray())
assert_array_equal(y, y2)
# XXX we have to update this to support Python 3.x
utf8_comment = b("It is true that\n\xc2\xbd\xc2\xb2 = \xc2\xbc")
f = BytesIO()
assert_raises(UnicodeDecodeError,
dump_svmlight_file, X, y, f, comment=utf8_comment)
unicode_comment = utf8_comment.decode("utf-8")
f = BytesIO()
dump_svmlight_file(X, y, f, comment=unicode_comment, zero_based=False)
f.seek(0)
X2, y2 = load_svmlight_file(f, zero_based=False)
assert_array_almost_equal(X, X2.toarray())
assert_array_equal(y, y2)
f = BytesIO()
assert_raises(ValueError,
dump_svmlight_file, X, y, f, comment="I've got a \0.")
def test_default_load_files(test_category_dir_1, test_category_dir_2,
load_files_root):
res = load_files(load_files_root)
assert_equal(len(res.filenames), 1)
assert_equal(len(res.target_names), 2)
assert_equal(res.DESCR, None)
assert_equal(res.data, [b("Hello World!\n")])
def test_load_with_long_qid():
# load svmfile with longint qid attribute
data = b("""
1 qid:0 0:1 1:2 2:3
0 qid:72048431380967004 0:1440446648 1:72048431380967004 2:236784985
0 qid:-9223372036854775807 0:1440446648 1:72048431380967004 2:236784985
3 qid:9223372036854775807 0:1440446648 1:72048431380967004 2:236784985""")
X, y, qid = load_svmlight_file(BytesIO(data), query_id=True)
true_X = [[1, 2, 3],
[1440446648, 72048431380967004, 236784985],
[1440446648, 72048431380967004, 236784985],
[1440446648, 72048431380967004, 236784985]]
true_y = [1, 0, 0, 3]
trueQID = [0, 72048431380967004, -9223372036854775807, 9223372036854775807]
assert_array_equal(y, true_y)
assert_array_equal(X.toarray(), true_X)
assert_array_equal(qid, trueQID)
f = BytesIO()
dump_svmlight_file(X, y, f, query_id=qid, zero_based=True)
f.seek(0)
X, y, qid = load_svmlight_file(f, query_id=True, zero_based=True)
assert_array_equal(y, true_y)
assert_array_equal(X.toarray(), true_X)
assert_array_equal(qid, trueQID)
f.seek(0)
X, y = load_svmlight_file(f, query_id=False, zero_based=True)
assert_array_equal(y, true_y)
assert_array_equal(X.toarray(), true_X)
def setup_load_files():
global TEST_CATEGORY_DIR1
global TEST_CATEGORY_DIR2
TEST_CATEGORY_DIR1 = tempfile.mkdtemp(dir=LOAD_FILES_ROOT)
TEST_CATEGORY_DIR2 = tempfile.mkdtemp(dir=LOAD_FILES_ROOT)
sample_file = tempfile.NamedTemporaryFile(dir=TEST_CATEGORY_DIR1, delete=False)
sample_file.write(b("Hello World!\n"))
sample_file.close()
def test_category_dir_1(load_files_root):
test_category_dir1 = tempfile.mkdtemp(dir=load_files_root)
sample_file = tempfile.NamedTemporaryFile(dir=test_category_dir1,
delete=False)
sample_file.write(b("Hello World!\n"))
sample_file.close()
yield str(test_category_dir1)
_remove_dir(test_category_dir1)
def test_default_load_files():
try:
setup_load_files()
res = load_files(LOAD_FILES_ROOT)
assert_equal(len(res.filenames), 1)
assert_equal(len(res.target_names), 2)
assert_equal(res.DESCR, None)
assert_equal(res.data, [b("Hello World!\n")])
finally:
teardown_load_files()
def _dump_svmlight_dense(X, y, f, one_based, comment, query_id):
is_sp = int(hasattr(X, "tocsr"))
if X.dtype.kind == 'i':
value_pattern = u("%d:%d")
else:
value_pattern = u("%d:%.16g")
if y.dtype.kind == 'i':
line_pattern = u("%d")
else:
line_pattern = u("%.16g")
if query_id is not None:
line_pattern += u(" qid:%d")
line_pattern += u(" %s\n")
if comment:
f.write(b("# Generated by dump_svmlight_file from scikit-learn %s\n"
% __version__))
f.write(b("# Column indices are %s-based\n"
% ["zero", "one"][one_based]))
f.write(b("#\n"))
f.writelines(b("# %s\n" % line) for line in comment.splitlines())
for i in range(X.shape[0]):
if is_sp:
#print 'is_sp'
span = slice(X.indptr[i], X.indptr[i + 1])
row = zip(X.indices[span], X.data[span])
else:
#nz = X[i] != 0
#row = zip(np.where(nz)[0], X[i, nz])
row = [(j,X[i][j]) for j in range(len(X[i]))]
#print row
s = " ".join(value_pattern % (j + one_based, x) for j, x in row)
if query_id is not None:
feat = (y[i], query_id[i], s)
else:
feat = (y[i], s)
f.write((line_pattern % feat).encode('ascii'))
def _fetch_lfw_pairs(index_file_path, data_folder_path, slice_=None,
color=False, resize=None):
"""Perform the actual data loading for the LFW pairs dataset
This operation is meant to be cached by a joblib wrapper.
"""
# parse the index file to find the number of pairs to be able to allocate
# the right amount of memory before starting to decode the jpeg files
with open(index_file_path, 'rb') as index_file:
split_lines = [ln.strip().split(b('\t')) for ln in index_file]
pair_specs = [sl for sl in split_lines if len(sl) > 2]
n_pairs = len(pair_specs)
# interating over the metadata lines for each pair to find the filename to
# decode and load in memory
target = np.zeros(n_pairs, dtype=np.int)
file_paths = list()
for i, components in enumerate(pair_specs):
if len(components) == 3:
target[i] = 1
pair = (
(components[0], int(components[1]) - 1),
(components[0], int(components[2]) - 1),
)
elif len(components) == 4:
target[i] = 0
pair = (
(components[0], int(components[1]) - 1),
(components[2], int(components[3]) - 1),
)
else:
raise ValueError("invalid line %d: %r" % (i + 1, components))
for j, (name, idx) in enumerate(pair):
try:
person_folder = join(data_folder_path, name)
except TypeError:
person_folder = join(data_folder_path, str(name, 'UTF-8'))
filenames = list(sorted(listdir(person_folder)))
file_path = join(person_folder, filenames[idx])
file_paths.append(file_path)
pairs = _load_imgs(file_paths, slice_, color, resize)
shape = list(pairs.shape)
n_faces = shape.pop(0)
shape.insert(0, 2)
shape.insert(0, n_faces // 2)
pairs.shape = shape
return pairs, target, np.array(['Different persons', 'Same person'])
def setup_module():
"""Test fixture run once and common to all tests of this module"""
if not pillow_installed:
raise SkipTest("PIL not installed.")
if not os.path.exists(LFW_HOME):
os.makedirs(LFW_HOME)
random_state = random.Random(42)
np_rng = np.random.RandomState(42)
# generate some random jpeg files for each person
counts = {}
for name in FAKE_NAMES:
folder_name = os.path.join(LFW_HOME, 'lfw_funneled', name)
if not os.path.exists(folder_name):
os.makedirs(folder_name)
n_faces = np_rng.randint(1, 5)
counts[name] = n_faces
for i in range(n_faces):
file_path = os.path.join(folder_name, name + '_%04d.jpg' % i)
uniface = np_rng.randint(0, 255, size=(250, 250, 3))
try:
imsave(file_path, uniface)
except ImportError:
raise SkipTest("PIL not installed")
# add some random file pollution to test robustness
with open(os.path.join(LFW_HOME, 'lfw_funneled', '.test.swp'), 'wb') as f:
f.write(six.b('Text file to be ignored by the dataset loader.'))
# generate some pairing metadata files using the same format as LFW
with open(os.path.join(LFW_HOME, 'pairsDevTrain.txt'), 'wb') as f:
f.write(six.b("10\n"))
more_than_two = [
name for name, count in six.iteritems(counts) if count >= 2
]
for i in range(5):
name = random_state.choice(more_than_two)
first, second = random_state.sample(range(counts[name]), 2)
f.write(six.b('%s\t%d\t%d\n' % (name, first, second)))
for i in range(5):
first_name, second_name = random_state.sample(FAKE_NAMES, 2)
first_index = random_state.choice(np.arange(counts[first_name]))
second_index = random_state.choice(np.arange(counts[second_name]))
f.write(
six.b('%s\t%d\t%s\t%d\n' %
(first_name, first_index, second_name, second_index)))
with open(os.path.join(LFW_HOME, 'pairsDevTest.txt'), 'wb') as f:
f.write(six.b("Fake place holder that won't be tested"))
with open(os.path.join(LFW_HOME, 'pairs.txt'), 'wb') as f:
f.write(six.b("Fake place holder that won't be tested"))
def test_load_with_qid():
# load svmfile with qid attribute
data = b("""
3 qid:1 1:0.53 2:0.12
2 qid:1 1:0.13 2:0.1
7 qid:2 1:0.87 2:0.12""")
X, y = load_svmlight_file(BytesIO(data), query_id=False)
assert_array_equal(y, [3, 2, 7])
assert_array_equal(X.todense(), [[.53, .12], [.13, .1], [.87, .12]])
res1 = load_svmlight_files([BytesIO(data)], query_id=True)
res2 = load_svmlight_file(BytesIO(data), query_id=True)
for X, y, qid in (res1, res2):
assert_array_equal(y, [3, 2, 7])
assert_array_equal(qid, [1, 1, 2])
assert_array_equal(X.todense(), [[.53, .12], [.13, .1], [.87, .12]])
def test_load_with_qid():
# load svmfile with qid attribute
data = b("""
3 qid:1 1:0.53 2:0.12
2 qid:1 1:0.13 2:0.1
7 qid:2 1:0.87 2:0.12""")
X, y = load_svmlight_file(BytesIO(data), query_id=False)
assert_array_equal(y, [3, 2, 7])
assert_array_equal(X.todense(), [[.53, .12], [.13, .1], [.87, .12]])
res1 = load_svmlight_files([BytesIO(data)], query_id=True)
res2 = load_svmlight_file(BytesIO(data), query_id=True)
for X, y, qid in (res1, res2):
assert_array_equal(y, [3, 2, 7])
assert_array_equal(qid, [1, 1, 2])
assert_array_equal(X.todense(), [[.53, .12], [.13, .1], [.87, .12]])
def setup_module():
"""Test fixture run once and common to all tests of this module"""
if imsave is None:
raise SkipTest
if not os.path.exists(LFW_HOME):
os.makedirs(LFW_HOME)
random_state = random.Random(42)
np_rng = np.random.RandomState(42)
# generate some random jpeg files for each person
counts = {}
for name in FAKE_NAMES:
folder_name = os.path.join(LFW_HOME, 'lfw_funneled', name)
if not os.path.exists(folder_name):
os.makedirs(folder_name)
n_faces = np_rng.randint(1, 5)
counts[name] = n_faces
for i in range(n_faces):
file_path = os.path.join(folder_name, name + '_%04d.jpg' % i)
uniface = np_rng.randint(0, 255, size=(250, 250, 3))
try:
imsave(file_path, uniface)
except ImportError:
# PIL is not properly installed, skip those tests
raise SkipTest
# add some random file pollution to test robustness
with open(os.path.join(LFW_HOME, 'lfw_funneled', '.test.swp'), 'wb') as f:
f.write(six.b('Text file to be ignored by the dataset loader.'))
# generate some pairing metadata files using the same format as LFW
with open(os.path.join(LFW_HOME, 'pairsDevTrain.txt'), 'wb') as f:
f.write(six.b("10\n"))
more_than_two = [name for name, count in six.iteritems(counts)
if count >= 2]
for i in range(5):
name = random_state.choice(more_than_two)
first, second = random_state.sample(range(counts[name]), 2)
f.write(six.b('%s\t%d\t%d\n' % (name, first, second)))
for i in range(5):
first_name, second_name = random_state.sample(FAKE_NAMES, 2)
first_index = random_state.choice(np.arange(counts[first_name]))
second_index = random_state.choice(np.arange(counts[second_name]))
f.write(six.b('%s\t%d\t%s\t%d\n' % (first_name, first_index,
second_name, second_index)))
with open(os.path.join(LFW_HOME, 'pairsDevTest.txt'), 'wb') as f:
f.write(six.b("Fake place holder that won't be tested"))
with open(os.path.join(LFW_HOME, 'pairs.txt'), 'wb') as f:
f.write(six.b("Fake place holder that won't be tested"))
def test_dump_comment():
X, y = load_svmlight_file(datafile)
X = X.toarray()
f = BytesIO()
ascii_comment = "This is a comment\nspanning multiple lines."
dump_svmlight_file(X, y, f, comment=ascii_comment, zero_based=False)
f.seek(0)
X2, y2 = load_svmlight_file(f, zero_based=False)
assert_array_almost_equal(X, X2.toarray())
assert_array_equal(y, y2)
# XXX we have to update this to support Python 3.x
utf8_comment = b("It is true that\n\xc2\xbd\xc2\xb2 = \xc2\xbc")
f = BytesIO()
assert_raises(UnicodeDecodeError,
dump_svmlight_file,
X,
y,
f,
comment=utf8_comment)
unicode_comment = utf8_comment.decode("utf-8")
f = BytesIO()
dump_svmlight_file(X, y, f, comment=unicode_comment, zero_based=False)
f.seek(0)
X2, y2 = load_svmlight_file(f, zero_based=False)
assert_array_almost_equal(X, X2.toarray())
assert_array_equal(y, y2)
f = BytesIO()
assert_raises(ValueError,
dump_svmlight_file,
X,
y,
f,
comment="I've got a \0.")
def test_default_load_files():
res = load_files(LOAD_FILES_ROOT)
assert_equal(len(res.filenames), 1)
assert_equal(len(res.target_names), 2)
assert_equal(res.DESCR, None)
assert_equal(res.data, [b("Hello World!\n")])
def test_load_zero_based():
f = BytesIO(b("-1 4:1.\n1 0:1\n"))
load_svmlight_file(f, zero_based=False)
def test_default_load_files():
res = load_files(LOAD_FILES_ROOT)
assert_equal(len(res.filenames), 1)
assert_equal(len(res.target_names), 2)
assert_equal(res.DESCR, None)
assert_equal(res.data, [b("Hello World!\n")])
def test_mmhash3_bytes():
assert_equal(murmurhash3_32(b('foo'), 0), -156908512)
assert_equal(murmurhash3_32(b('foo'), 42), -1322301282)
assert_equal(murmurhash3_32(b('foo'), 0, positive=True), 4138058784)
assert_equal(murmurhash3_32(b('foo'), 42, positive=True), 2972666014)
def test_mmhash3_bytes():
assert_equal(murmurhash3_32(b('foo'), 0), -156908512)
assert_equal(murmurhash3_32(b('foo'), 42), -1322301282)
assert_equal(murmurhash3_32(b('foo'), 0, positive=True), 4138058784)
assert_equal(murmurhash3_32(b('foo'), 42, positive=True), 2972666014)
def test_load_zero_based():
f = BytesIO(b("-1 4:1.\n1 0:1\n"))
assert_raises(ValueError, load_svmlight_file, f, zero_based=False)
def test_load_zero_based():
f = BytesIO(b("-1 4:1.\n1 0:1\n"))
load_svmlight_file(f, zero_based=False)
