def test_reduction_deriv(transformer_factory):
C = ng.make_axis(name='C')
W = ng.make_axis(name='W')
H = ng.make_axis(name='H')
delta = .001
C.length = 4
W.length = 10
H.length = 10
axes = ng.make_axes([C, W, H])
u = rng.discrete_uniform(1.0, 2.0, 2 * delta, axes)
# Need to test max/min differently since if two elements are extremums
# and we modify one, the derivative will change.
for npred, bered, red in [(np.sum, ng.sum, 'sum')]:
for reduction_axes in [[C],
[W],
[H],
[C, W],
[W, H]]:
p_u = ng.placeholder(axes)
graph_reduce = bered(p_u, reduction_axes=reduction_axes)
check_derivative(graph_reduce, p_u, delta, u, atol=1e-1, rtol=1e-1)
def test_log_sigmoid_deriv(transformer_factory):
"""TODO."""
axes = ng.make_axes([ng.make_axis(20), ng.make_axis(128)])
p_u = ng.placeholder(axes)
u = rng.uniform(-3.0, 3.0, p_u.axes)
log_val_u = ng.log(ng.sigmoid(p_u))
check_derivative(log_val_u, p_u, 0.001, u, atol=1e-2, rtol=1e-2)
def test_softmax_deriv(transformer_factory):
N = ng.make_axis(name='N', batch=True)
W = ng.make_axis(name='W')
W.length = 3
N.length = 10
axes = ng.make_axes([W, N])
x = rng.uniform(0, 1, axes)
p_x = ng.placeholder(axes)
check_derivative(ng.softmax(p_x), p_x, 0.001, x, atol=1e-2, rtol=1e-2)
def test_print_op_bprop():
"""
Ensure bprop of PrintOp is correct (passes through exactly the delta)
"""
A = ng.make_axis(10, name='A')
x = ng.placeholder(ng.make_axes([A]))
# randomly initialize
x_value = rng.uniform(-1, 1, x.axes)
check_derivative(ng.PrintOp(x), x, 0.001, x_value, atol=1e-3, rtol=1e-3)
def test_reciprocal_derivative(transformer_factory):
"""TODO."""
N = ng.make_axis(name='N')
W = ng.make_axis(name='W')
delta = .001
W.length = 20
N.length = 128
axes = ng.make_axes([W, N])
p_u = ng.placeholder(axes)
u = rng.uniform(.1, 5.0, p_u.axes)
rec_u = ng.reciprocal(p_u)
check_derivative(rec_u, p_u, delta, u, atol=1e-2, rtol=1e-2)
def test_dimshuffle_bprop(transformer_factory):
"""
dimshuffle a 2d array and make sure bprop works
"""
A = ng.make_axis(2)
B = ng.make_axis(3)
x = ng.placeholder(ng.make_axes([A, B]))
# randomly initialize
x_value = rng.uniform(-1, 1, x.axes)
check_derivative(
ng.axes_with_order(x, [B, A]),
x, 0.001, x_value,
atol=1e-3, rtol=1e-3
)
def test_elementwise_binary_ops_matched_args_deriv_rhs(transformer_factory):
"""TODO."""
axes = ng.make_axes([ng.make_axis(20), ng.make_axis(20)])
for np_op, be_op in ELEMENTWISE_BINARY_OPS:
# Matched sizes
p_u = ng.placeholder(axes)
p_v = ng.placeholder(axes)
u = rng.uniform(-1.0, 1.0, p_u.axes)
v = rng.uniform(1.0, 2.0, p_v.axes)
check_derivative(
be_op(p_u, p_v), p_v, 0.001, v,
parameters=[p_u],
parameter_values=[u],
atol=1e-3, rtol=1e-3,
)
def test_cross_entropy_softmax_deriv(transformer_factory):
N = ng.make_axis(name='N', batch=True)
W = ng.make_axis(name='W')
W.length = 3
N.length = 10
axes = ng.make_axes([W, N])
p_x = ng.placeholder(axes)
p_t = ng.placeholder(axes)
x = rng.uniform(0, 1, axes)
t = np_softmax(rng.uniform(0, 1, axes), 0)
check_derivative(
ng.cross_entropy_multi(ng.softmax(p_x), p_t),
p_x, 0.001, x,
parameters=[p_t],
parameter_values=[t],
atol=1e-2, rtol=1e-2
)
