def test_einsum():
for datatype in backends:
T.set_backend(datatype)
x2 = ad.Variable(name="x2", shape=[3, 2])
x3 = ad.Variable(name="x3", shape=[2, 3])
matmul = ad.einsum('ik,kj->ij', x2, x3)
y = ad.sum(matmul)
grad_x2, grad_x3 = ad.gradients(y, [x2, x3])
executor = ad.Executor([y, grad_x2, grad_x3])
x2_val = T.tensor([[1, 2], [3, 4], [5, 6]]) # 3x2
x3_val = T.tensor([[7, 8, 9], [10, 11, 12]]) # 2x3
y_val, grad_x2_val, grad_x3_val = executor.run(feed_dict={
x2: x2_val,
x3: x3_val
})
expected_grad_sum = T.ones_like(T.dot(x2_val, x3_val))
expected_yval = T.sum(T.dot(x2_val, x3_val))
expected_grad_x2_val = T.dot(expected_grad_sum, T.transpose(x3_val))
expected_grad_x3_val = T.dot(T.transpose(x2_val), expected_grad_sum)
assert isinstance(y, ad.Node)
assert T.array_equal(y_val, expected_yval)
assert T.array_equal(grad_x2_val, expected_grad_x2_val)
assert T.array_equal(grad_x3_val, expected_grad_x3_val)
def test_add_mul_mix_3(backendopt):
for datatype in backendopt:
T.set_backend(datatype)
x2 = ad.Variable(name="x2", shape=[3])
x3 = ad.Variable(name="x3", shape=[3])
z = x2 * x2 + x2 + x3 + 3
y = ad.sum(z * z + x3)
grad_x2, grad_x3 = ad.gradients(y, [x2, x3])
executor = ad.Executor([y, grad_x2, grad_x3])
x2_val = 2 * T.ones(3)
x3_val = 3 * T.ones(3)
y_val, grad_x2_val, grad_x3_val = executor.run(feed_dict={
x2: x2_val,
x3: x3_val
})
z_val = x2_val * x2_val + x2_val + x3_val + 3
expected_yval = z_val * z_val + x3_val
expected_grad_x2_val = 2 * \
(x2_val * x2_val + x2_val + x3_val + 3) * (2 * x2_val + 1)
expected_grad_x3_val = 2 * (x2_val * x2_val + x2_val + x3_val + 3) + 1
assert isinstance(y, ad.Node)
assert T.array_equal(y_val, T.sum(expected_yval))
assert T.array_equal(grad_x2_val, expected_grad_x2_val)
assert T.array_equal(grad_x3_val, expected_grad_x3_val)
def test_inner_product_hvp():
for datatype in backends:
T.set_backend(datatype)
x = ad.Variable(name="x", shape=[3, 1])
v = ad.Variable(name="v", shape=[3, 1])
y = ad.sum(ad.einsum("ab,bc->ac", ad.transpose(x), x))
grad_x, = ad.gradients(y, [x])
Hv, = ad.hvp(output_node=y, node_list=[x], vector_list=[v])
executor = ad.Executor([y, grad_x, Hv])
x_val = T.tensor([[1.], [2.], [3]]) # 3x1
v_val = T.tensor([[1.], [2.], [3]]) # 3x1
y_val, grad_x_val, Hv_val = executor.run(feed_dict={
x: x_val,
v: v_val
})
expected_yval = T.sum(T.dot(T.transpose(x_val), x_val))
expected_grad_x_val = 2 * x_val
expected_hv_val = 2 * v_val
assert isinstance(y, ad.Node)
assert T.array_equal(y_val, expected_yval)
assert T.array_equal(grad_x_val, expected_grad_x_val)
assert T.array_equal(Hv_val, expected_hv_val)
def test_hvp2(backendopt):
for datatype in backendopt:
T.set_backend(datatype)
x = ad.Variable(name="x", shape=[3, 1])
H = ad.Variable(name="H", shape=[3, 3])
v = ad.Variable(name="v", shape=[3, 1])
y = ad.sum(
ad.einsum("ab,bc->ac", ad.einsum("ab,bc->ac", ad.transpose(x), H),
x))
grad_x, = ad.gradients(y, [x])
Hv, = ad.hvp(output_node=y, node_list=[x], vector_list=[v])
executor = ad.Executor([y, grad_x, Hv])
x_val = T.tensor([[1.], [2.], [3]]) # 3x1
v_val = T.tensor([[1.], [2.], [3]]) # 3x1
H_val = T.tensor([[2., 0., 0.], [0., 2., 0.], [0., 0., 2.]]) # 3x3
y_val, grad_x_val, Hv_val = executor.run(feed_dict={
x: x_val,
H: H_val,
v: v_val
})
Hx = T.dot(H_val, x_val)
expected_yval = T.sum(T.dot(T.transpose(x_val), Hx))
expected_grad_x_val = 2 * Hx
expected_hv_val = T.tensor([[4.], [8.], [12.]])
assert isinstance(y, ad.Node)
assert T.array_equal(y_val, expected_yval)
assert T.array_equal(grad_x_val, expected_grad_x_val)
assert T.array_equal(Hv_val, expected_hv_val)
def test_add_mul_mix_2(backendopt):
for datatype in backendopt:
T.set_backend(datatype)
x1 = ad.Variable(name="x1", shape=[3])
x2 = ad.Variable(name="x2", shape=[3])
x3 = ad.Variable(name="x3", shape=[3])
x4 = ad.Variable(name="x4", shape=[3])
y = ad.sum(x1 + x2 * x3 * x4)
grad_x1, grad_x2, grad_x3, grad_x4 = ad.gradients(y, [x1, x2, x3, x4])
executor = ad.Executor([y, grad_x1, grad_x2, grad_x3, grad_x4])
x1_val = 1 * T.ones(3)
x2_val = 2 * T.ones(3)
x3_val = 3 * T.ones(3)
x4_val = 4 * T.ones(3)
y_val, grad_x1_val, grad_x2_val, grad_x3_val, grad_x4_val = executor.run(
feed_dict={
x1: x1_val,
x2: x2_val,
x3: x3_val,
x4: x4_val
})
assert isinstance(y, ad.Node)
assert T.array_equal(y_val, T.sum(x1_val + x2_val * x3_val * x4_val))
assert T.array_equal(grad_x1_val, T.ones_like(x1_val))
assert T.array_equal(grad_x2_val, x3_val * x4_val)
assert T.array_equal(grad_x3_val, x2_val * x4_val)
assert T.array_equal(grad_x4_val, x2_val * x3_val)
def test_summation_einsum_2(backendopt):
for datatype in backendopt:
T.set_backend(datatype)
x = ad.Variable(name="x", shape=[2, 2])
y = ad.Variable(name="y", shape=[2, 2])
out = ad.sum(ad.einsum('ij,ab->ab', x, y))
grad_x, = ad.gradients(out, [x])
executor = ad.Executor([out, grad_x])
x_val = T.tensor([[1., 2.], [3., 4.]])
y_val = T.tensor([[5., 6.], [7., 8.]])
out_val, grad_x_val = executor.run(feed_dict={x: x_val, y: y_val})
expected_out_val = T.sum(T.einsum('ij,ab->ab', x_val, y_val))
expected_grad_x_val = T.sum(y_val) * T.ones_like(x_val)
assert T.array_equal(out_val, expected_out_val)
assert T.array_equal(grad_x_val, expected_grad_x_val)
def test_negative(backendopt):
for datatype in backendopt:
T.set_backend(datatype)
x2 = ad.Variable(name="x2", shape=[3])
y = ad.sum(-x2)
grad_x2, = ad.gradients(y, [x2])
executor = ad.Executor([y, grad_x2])
x2_val = 2 * T.ones(3)
y_val, grad_x2_val = executor.run(feed_dict={x2: x2_val})
assert isinstance(y, ad.Node)
assert T.array_equal(y_val, T.sum(-x2_val))
assert T.array_equal(grad_x2_val, -T.ones_like(x2_val))
def test_jacobian_summation_einsum(backendopt):
for datatype in backendopt:
T.set_backend(datatype)
x = ad.Variable(name="x", shape=[2, 2])
x_sum = ad.einsum('ij->', x)
grad_x, = ad.jacobians(x_sum, [x])
executor = ad.Executor([x_sum, grad_x])
x_val = T.tensor([[1., 2.], [3., 4.]])
x_sum_val, grad_x_val = executor.run(feed_dict={x: x_val})
expected_x_sum_val = T.sum(x_val)
expected_grad_x_val = T.ones_like(x_val)
assert T.array_equal(x_sum_val, expected_x_sum_val)
assert T.array_equal(grad_x_val, expected_grad_x_val)
def test_transpose_einsum(backendopt):
for datatype in backendopt:
T.set_backend(datatype)
x = ad.Variable(name="x", shape=[3, 2])
y = ad.sum(ad.einsum("ij->ji", x))
grad_x, = ad.gradients(y, [x])
executor = ad.Executor([y, grad_x])
x_val = T.tensor([[1, 2], [3, 4], [5, 6]]) # 3x2
y_val, grad_x_val = executor.run(feed_dict={x: x_val})
expected_yval = T.sum(T.transpose(x_val))
expected_grad_x_val = T.ones_like(x_val)
assert isinstance(y, ad.Node)
assert T.array_equal(y_val, expected_yval)
assert T.array_equal(grad_x_val, expected_grad_x_val)
def test_einsum_3op(backendopt):
for datatype in backendopt:
T.set_backend(datatype)
x2 = ad.Variable(name="x2", shape=[3, 2])
x3 = ad.Variable(name="x3", shape=[2, 3])
x4 = ad.Variable(name="x4", shape=[3, 2])
matmul = ad.einsum('ik,kj,jl->il', x2, x3, x4)
y = ad.sum(matmul)
grad_x2, grad_x3, grad_x4 = ad.gradients(y, [x2, x3, x4])
executor = ad.Executor([y, grad_x2, grad_x3, grad_x4])
x2_val = T.tensor([[1, 2], [3, 4], [5, 6]]) # 3x2
x3_val = T.tensor([[7, 8, 9], [10, 11, 12]]) # 2x3
x4_val = T.tensor([[1, 2], [3, 4], [5, 6]]) # 3x2
y_val, grad_x2_val, grad_x3_val, grad_x4_val = executor.run(feed_dict={
x2: x2_val,
x3: x3_val,
x4: x4_val
})
expected_grad_sum = T.ones_like(T.dot(T.dot(x2_val, x3_val), x4_val))
expected_yval = T.sum(T.dot(T.dot(x2_val, x3_val), x4_val))
expected_grad_x2_val = T.einsum("il, kj, jl->ik", expected_grad_sum,
x3_val, x4_val)
expected_grad_x3_val = T.einsum("ik, il, jl->kj", x2_val,
expected_grad_sum, x4_val)
expected_grad_x4_val = T.einsum("ik, kj, il->jl", x2_val, x3_val,
expected_grad_sum)
assert isinstance(y, ad.Node)
assert T.array_equal(y_val, expected_yval)
assert T.array_equal(grad_x2_val, expected_grad_x2_val)
assert T.array_equal(grad_x3_val, expected_grad_x3_val)
assert T.array_equal(grad_x4_val, expected_grad_x4_val)
def test_mul_two_vars(backendopt):
for datatype in backendopt:
T.set_backend(datatype)
x2 = ad.Variable(name="x2", shape=[3])
x3 = ad.Variable(name="x3", shape=[3])
y = ad.sum(x2 * x3)
grad_x2, grad_x3 = ad.gradients(y, [x2, x3])
executor = ad.Executor([y, grad_x2, grad_x3])
x2_val = 2 * T.ones(3)
x3_val = 3 * T.ones(3)
y_val, grad_x2_val, grad_x3_val = executor.run(feed_dict={
x2: x2_val,
x3: x3_val
})
assert isinstance(y, ad.Node)
assert T.array_equal(y_val, T.sum(x2_val * x3_val))
assert T.array_equal(grad_x2_val, x3_val)
assert T.array_equal(grad_x3_val, x2_val)
def group_vecnorm(xs):
s = sum([T.sum(x * x) for x in xs])
return s**0.5
def group_dot(xs, ys):
assert len(xs) == len(ys)
return sum([T.sum(x * y) for (x, y) in zip(xs, ys)])
