def test_invalid_shape_error():
s = sparse.random((3, 4), density=0.5, format='coo')
with pytest.raises(ValueError):
sparse.as_coo(s, shape=(2, 3))
with pytest.raises(ValueError):
COO(s, shape=(2, 3))
def test_invalid_attrs_error(s):
with pytest.raises(ValueError):
sparse.as_coo(s, shape=(2, 3))
with pytest.raises(ValueError):
COO(s, shape=(2, 3))
with pytest.raises(ValueError):
sparse.as_coo(s, fill_value=0.0)
with pytest.raises(ValueError):
COO(s, fill_value=0.0)
def test_invalid_attrs_error():
s = sparse.random((3, 4), density=0.5, format="coo")
with pytest.raises(ValueError):
sparse.as_coo(s, shape=(2, 3))
with pytest.raises(ValueError):
COO(s, shape=(2, 3))
with pytest.raises(ValueError):
sparse.as_coo(s, fill_value=0.0)
with pytest.raises(ValueError):
COO(s, fill_value=0.0)
def as_pydata_coo(a, shape=None, fill_value=None):
"""Convert a to pydata coo array
Examples:
>>> import sparse
>>> a = np.random.random((3, 4, 5))
>>> a2 = as_pydata_coo(a)
>>> isinstance(a2, sparse.COO)
True
>>> np.testing.assert_array_equal(a, a2.todense())
>>> a2 = as_pydata_coo(sparse.as_coo(a))
>>> isinstance(a2, sparse.COO)
True
>>> np.testing.assert_array_equal(a, a2.todense())
>>> import tensorflow as tf
>>> indices = [[0,0], [1,1]]
>>> values = [1.0, 2.0]
>>> a = tf.SparseTensorValue(indices=indices, values=values,
... dense_shape=(2,2))
>>> a2 = as_pydata_coo(a)
>>> isinstance(a2, sparse.COO)
True
>>> np.testing.assert_array_equal(a2.todense(),
... np.array([[1.0, 0], [0, 2.0]]))
"""
_config.assert_has_package('sparse')
import sparse
if not (scipy.sparse.issparse(a) or isinstance(a, np.ndarray) or
is_pydata_sparse(a)):
a = as_scipy_coo(a)
return sparse.as_coo(a, shape, fill_value)
def test_as_coo(format):
x = format(sparse.random((3, 4), density=0.5, format="coo").todense())
s1 = sparse.as_coo(x)
s2 = COO(x)
assert_eq(x, s1)
assert_eq(x, s2)
def test_empty(self):
x = np.array([])
assert_eq(np.roll(x, 1), sparse.roll(sparse.as_coo(x), 1))
K = A.shape[1] if X is None else X.shape[0]
nC = Y.shape[1]
W = None
if args.weighted:
W = utils.calc_class_weights(Y[..., idx_train, :])
# ************************************************************
# calculate node features
# ************************************************************
vals = []
X = X.astype(np.float32)
EYE = scipy.sparse.eye(K, dtype=np.float32, format='coo')
A = A.astype(np.float32)
# normalized x
rowsum = sparse.as_coo(X).sum(axis=-1, keepdims=True)
#rowsum.data[rowsum.data==0] = 1e-10
rowsum.data = 1.0 / rowsum.data
vals.append((sparse.as_coo(X) * rowsum).to_scipy_sparse())
nodes = scipy.sparse.hstack(vals)
nodes = nodes.toarray()
# ************************************************************
# calculate edge features
# ************************************************************
EYE = scipy.sparse.eye(K, dtype=np.float32, format='coo')
curvs_used = ['both', 'ollivier', 'forman']
for i, curv in enumerate(curvs_used):
print("curv: ", curv)
if curv == 'both':
W = None
if args.weighted:
W = utils.calc_class_weights(Y[..., idx_train, :])
# ************************************************************
# calculate edge features
# ************************************************************
EYE = scipy.sparse.eye(K, dtype=np.float32, format='coo')
edge_feat_list = [ollivier_curv_vals, forman_curv_vals]
vals = []
for mat in edge_feat_list:
vals.append((mat + mat.transpose() + EYE > 0).astype(np.float32))
if args.encode_edge_direction:
vals.append((mat + EYE > 0).astype(np.float32))
vals.append((mat.transpose() + EYE > 0).astype(np.float32))
vals = [sparse.as_coo(x) for x in vals]
vals = sparse.stack(vals, axis=0)
vals = vals.todense()
vals = aml_graph.normalize_adj(vals,
args.edge_norm,
assume_symmetric_input=False)
temp_vals = vals
vals = [vals]
ret = np.concatenate(vals, 1)
edges = np.transpose(ret, [1, 2, 0])
for j in range(args.num_trials):
