def _check_agreement_with_empirical(self, W_std, b_std, filter_size, is_conv,
is_ntk, is_res, layer_norm, padding, phi, proj_into_2d, same_inputs,
strides, use_pooling, width):
key = random.PRNGKey(1)
x1, x2 = _get_inputs(key, is_conv, same_inputs, INPUT_SHAPE)
init_fn, apply_fn, kernel_fn = _get_net(W_std, b_std, filter_size,
is_conv, use_pooling, is_res,
padding, phi, strides, width,
is_ntk, proj_into_2d, layer_norm)
x1_out_shape, params = init_fn(key, x1.shape)
if x2 is None:
x2_out_shape = x1_out_shape
else:
x2_out_shape, params = init_fn(key, x2.shape)
del(params)
def _get_empirical(n_samples, get):
kernel_fn_empirical = monte_carlo.monte_carlo_kernel_fn(
init_fn, apply_fn, key, n_samples)
return kernel_fn_empirical(x1, x2, get)
if proj_into_2d == 'ATTN_PARAM':
# no analytic kernel available, just test forward/backward pass
_get_empirical(1, 'ntk' if is_ntk else 'nngp')
else:
if is_ntk:
exact, shape1, shape2 = kernel_fn(x1, x2, ('ntk', 'shape1', 'shape2'))
empirical = np.reshape(_get_empirical(N_SAMPLES, 'ntk'), exact.shape)
else:
exact, shape1, shape2 = kernel_fn(x1, x2, ('nngp', 'shape1', 'shape2'))
empirical = _get_empirical(N_SAMPLES, 'nngp')
utils.assert_close_matrices(self, empirical, exact, RTOL)
self.assertEqual(shape1, x1_out_shape)
self.assertEqual(shape2, x2_out_shape)
def test_exact(self, model, width, strides, padding, phi, same_inputs,
filter_size, use_pooling, is_ntk, is_res, proj_into_2d):
is_conv = 'conv' in model
# Check for duplicate / incorrectly-shaped NN configs / wrong backend.
if is_conv:
if xla_bridge.get_backend().platform == 'cpu':
raise jtu.SkipTest('Not running CNN models on CPU to save time.')
if (is_res and is_conv and ((strides is not None and strides != (1, 1)) or
(padding == 'VALID' and filter_size !=
(1, 1)))):
raise jtu.SkipTest('Different paths in a residual models need to return'
' outputs of the same shape.')
elif (filter_size != FILTER_SIZES[0] or padding != PADDINGS[0] or
strides != STRIDES[0] or proj_into_2d != PROJECTIONS[0] or
use_pooling):
raise jtu.SkipTest('FC models do not have these parameters.')
if (proj_into_2d.startswith('ATTN') and strides == (2, 1) and
padding == 'VALID' and xla_bridge.get_backend().platform == 'tpu'):
#TODO(jirihron): speed up the vmap alternative impl or fix the current one
raise jtu.SkipTest('ATTN forward pass on TPU is broken if one of'
' the spatial dimensions is singleton.')
W_std, b_std = 2.**0.5, 0.5**0.5
key = random.PRNGKey(1)
x1, x2 = _get_inputs(key, is_conv, same_inputs, INPUT_SHAPE)
init_fn, apply_fn, kernel_fn = _get_net(W_std, b_std, filter_size,
is_conv, use_pooling, is_res,
padding, phi, strides, width,
is_ntk, proj_into_2d)
def _get_empirical(n_samples, get):
kernel_fn_empirical = monte_carlo.monte_carlo_kernel_fn(
init_fn, apply_fn, key, n_samples)
return kernel_fn_empirical(x1, x2, get)
if proj_into_2d == 'ATTN_PARAM':
# no analytic kernel available, just test forward/backward pass
_get_empirical(1, 'ntk' if is_ntk else 'nngp')
else:
if is_ntk:
exact = kernel_fn(x1, x2, 'ntk')
empirical = np.reshape(_get_empirical(N_SAMPLES, 'ntk'), exact.shape)
else:
exact = kernel_fn(x1, x2, 'nngp')
empirical = _get_empirical(N_SAMPLES, 'nngp')
utils.assert_close_matrices(self, empirical, exact, RTOL)
def test_exact(self, model, width, strides, padding, phi, same_inputs,
filter_size, use_pooling, is_ntk, is_res):
is_conv = 'conv' in model
# Check for duplicate / incorrectly-shaped NN configs / wrong backend.
if is_conv:
if xla_bridge.get_backend().platform == 'cpu':
raise jtu.SkipTest(
'Not running CNN models on CPU to save time.')
if use_pooling and not same_inputs:
raise jtu.SkipTest(
'Pooling layers for different inputs or for same '
'padding not implemented.')
if (is_res and is_conv
and ((strides is not None and strides !=
(1, 1)) or (padding == 'VALID' and filter_size !=
(1, 1)))):
raise jtu.SkipTest(
'Different paths in a residual models need to return'
' outputs of the same shape.')
elif (filter_size != FILTER_SIZES[0] or padding != PADDINGS[0]
or strides != STRIDES[0] or use_pooling):
raise jtu.SkipTest('FC models do not have these parameters.')
W_std, b_std = 2.**0.5, 0.5**0.5
key = random.PRNGKey(1)
x1, x2 = _get_inputs(key, is_conv, same_inputs, INPUT_SHAPE)
init_fun, apply_fun, ker_fun = _get_net(W_std, b_std, filter_size,
is_conv, use_pooling, is_res,
padding, phi, strides, width,
is_ntk)
if is_ntk:
exact = ker_fun(x1, x2).ntk
ker_fun_empirical = monte_carlo.get_ker_fun_monte_carlo(
init_fun, apply_fun, False, True)
empirical = ker_fun_empirical(x1, x2, key, N_SAMPLES).ntk
empirical = np.reshape(empirical, exact.shape)
else:
exact = ker_fun(x1, x2, compute_ntk=False).nngp
ker_fun_empirical = monte_carlo.get_ker_fun_monte_carlo(
init_fun, apply_fun, True, False)
empirical = ker_fun_empirical(x1, x2, key, N_SAMPLES).nngp
utils.assert_close_matrices(self, empirical, exact, RTOL)
def test_sample_vs_analytic_nngp(self, batch_size, device_count,
store_on_device):
utils.stub_out_pmap(batch, device_count)
x1, x2, init_fn, apply_fn, stax_kernel_fn, key = _get_inputs_and_model(
1024, 256,
xla_bridge.get_backend().platform == 'tpu')
sample = monte_carlo.monte_carlo_kernel_fn(init_fn, apply_fn, key, 200,
batch_size, device_count,
store_on_device)
ker_empirical = sample(x1, x2, 'nngp')
ker_analytic = stax_kernel_fn(x1, x2, 'nngp')
utils.assert_close_matrices(self, ker_analytic, ker_empirical, 2e-2)
def test_monte_carlo_vs_analytic_ntk(self, batch_size, device_count,
store_on_device):
utils.stub_out_pmap(batch, device_count)
x1, x2, init_fn, apply_fn, stax_kernel_fn, key = _get_inputs_and_model(
256, 2,
xla_bridge.get_backend().platform == 'tpu')
sample = monte_carlo.monte_carlo_kernel_fn(init_fn, apply_fn, key, 100,
batch_size, device_count,
store_on_device)
ker_empirical = sample(x1, x2, 'ntk')
ker_empirical = (np.sum(ker_empirical, axis=(-1, -2)) /
ker_empirical.shape[-1])
ker_analytic = stax_kernel_fn(x1, x2, 'ntk')
utils.assert_close_matrices(self, ker_analytic, ker_empirical, 2e-2)
def test_sample_vs_analytic_nngp(self, batch_size, device_count,
store_on_device):
utils.stub_out_pmap(batch, device_count)
x1, x2, init_fun, apply_fun, stax_ker_fun, key = _get_inputs_and_model(
512, 512)
sample = monte_carlo.get_ker_fun_monte_carlo(init_fun, apply_fun, True,
False, batch_size,
device_count,
store_on_device)
ker_empirical = sample(x1, x2, key, 200).nngp
ker_analytic = stax_ker_fun(x1,
x2,
compute_ntk=False,
compute_nngp=True)
ker_analytic = ker_analytic.nngp
utils.assert_close_matrices(self, ker_analytic, ker_empirical, 1e-2)
def _check_agreement_with_empirical(self, net, same_inputs, is_conv,
use_dropout, is_ntk, proj_into_2d):
(init_fn, apply_fn, kernel_fn), input_shape = net
num_samples = N_SAMPLES * 5 if use_dropout else N_SAMPLES
key = random.PRNGKey(1)
x1, x2 = _get_inputs(key, is_conv, same_inputs, input_shape)
x1_out_shape, params = init_fn(key, x1.shape)
if same_inputs:
assert (x2 is None)
if x2 is None:
x2_out_shape = x1_out_shape
else:
x2_out_shape, params = init_fn(key, x2.shape)
del (params)
def _get_empirical(n_samples, get):
kernel_fn_empirical = monte_carlo.monte_carlo_kernel_fn(
init_fn, apply_fn, key, n_samples)
if same_inputs:
assert (x2 is None)
return kernel_fn_empirical(x1, x2, get)
if proj_into_2d == 'ATTN_PARAM':
# no analytic kernel available, just test forward/backward pass
_get_empirical(1, 'ntk' if is_ntk else 'nngp')
else:
if is_ntk:
exact, shape1, shape2 = kernel_fn(x1, x2, ('ntk', 'shape1', 'shape2'))
empirical = np.reshape(_get_empirical(num_samples, 'ntk'), exact.shape)
else:
exact, shape1, shape2 = kernel_fn(x1, x2, ('nngp', 'shape1', 'shape2'))
empirical = _get_empirical(num_samples, 'nngp')
utils.assert_close_matrices(self, exact, empirical, RTOL)
self.assertEqual(shape1, x1_out_shape)
self.assertEqual(shape2, x2_out_shape)
def test_monte_carlo_vs_analytic_ntk(self, batch_size, device_count,
store_on_device):
utils.stub_out_pmap(batch, device_count)
x1, x2, init_fun, apply_fun, stax_ker_fun, key = _get_inputs_and_model(
512, 2)
sample = monte_carlo.get_ker_fun_monte_carlo(init_fun, apply_fun,
False, True, batch_size,
device_count,
store_on_device)
ker_empirical = sample(x1, x2, key, 100).ntk
ker_empirical = (np.sum(ker_empirical, axis=(-1, -2)) /
ker_empirical.shape[-1])
ker_analytic = stax_ker_fun(x1,
x2,
compute_ntk=True,
compute_nngp=True)
ker_analytic = ker_analytic.ntk
utils.assert_close_matrices(self, ker_analytic, ker_empirical, 1e-2)
def _check_agreement_with_empirical(self, W_std, b_std, filter_size,
is_conv, is_ntk, is_res, layer_norm,
padding, phi, proj_into_2d,
same_inputs, strides, use_pooling,
width, parameterization, use_dropout):
if is_conv:
# Select a random dimension order.
default_spec = 'NHWC'
if xla_bridge.get_backend().platform == 'tpu':
# Keep batch dimension leading for TPU for batching to work.
specs = ['NHWC', 'NHCW', 'NCHW']
else:
specs = ['NHWC', 'NHCW', 'NCHW', 'CHWN', 'CHNW', 'CNHW']
spec = prandom.choice(specs)
input_shape = tuple(INPUT_SHAPE[default_spec.index(c)]
for c in spec)
if layer_norm:
layer_norm = tuple(spec.index(c) for c in layer_norm)
else:
# Only `NC` dimension order is supported and is enforced by layers.
spec = None
input_shape = INPUT_SHAPE
if layer_norm:
layer_norm = prandom.choice([(1, ), (-1, )])
num_samples = N_SAMPLES * 5 if use_dropout else N_SAMPLES
key = random.PRNGKey(1)
dimension_numbers = (spec, 'HWIO', spec)
x1, x2 = _get_inputs(key, is_conv, same_inputs, input_shape)
init_fn, apply_fn, kernel_fn = _get_net(W_std, b_std, filter_size,
is_conv, use_pooling, is_res,
padding, phi, strides, width,
is_ntk, proj_into_2d,
layer_norm, parameterization,
use_dropout, dimension_numbers)
x1_out_shape, params = init_fn(key, x1.shape)
if same_inputs:
assert (x2 is None)
if x2 is None:
x2_out_shape = x1_out_shape
else:
x2_out_shape, params = init_fn(key, x2.shape)
del (params)
def _get_empirical(n_samples, get):
kernel_fn_empirical = monte_carlo.monte_carlo_kernel_fn(
init_fn, apply_fn, key, n_samples)
if same_inputs:
assert (x2 is None)
return kernel_fn_empirical(x1, x2, get)
if proj_into_2d == 'ATTN_PARAM':
# no analytic kernel available, just test forward/backward pass
_get_empirical(1, 'ntk' if is_ntk else 'nngp')
else:
if is_ntk:
exact, shape1, shape2 = kernel_fn(x1, x2,
('ntk', 'shape1', 'shape2'))
empirical = np.reshape(_get_empirical(num_samples, 'ntk'),
exact.shape)
else:
exact, shape1, shape2 = kernel_fn(x1, x2,
('nngp', 'shape1', 'shape2'))
empirical = _get_empirical(num_samples, 'nngp')
utils.assert_close_matrices(self, empirical, exact, RTOL)
self.assertEqual(shape1, x1_out_shape)
self.assertEqual(shape2, x2_out_shape)
def test_fan_in_conv(self,
same_inputs,
axis,
n_branches,
get,
branch_in,
readout):
if xla_bridge.get_backend().platform == 'cpu':
raise jtu.SkipTest('Not running CNNs on CPU to save time.')
if axis in (None, 0, 1, 2) and branch_in == 'dense_after_branch_in':
raise jtu.SkipTest('`FanInSum` and `FanInConcat(0/1/2)` '
'require `is_gaussian`.')
if axis == 3 and branch_in == 'dense_before_branch_in':
raise jtu.SkipTest('`FanInConcat` on feature axis requires a dense layer '
'after concatenation.')
key = random.PRNGKey(1)
X0_1 = random.normal(key, (2, 5, 6, 3))
X0_2 = None if same_inputs else random.normal(key, (3, 5, 6, 3))
if xla_bridge.get_backend().platform == 'tpu':
width = 2048
n_samples = 1024
tol = 0.02
else:
width = 1024
n_samples = 512
tol = 0.01
conv = stax.Conv(out_chan=width,
filter_shape=(3, 3),
padding='SAME',
W_std=1.25,
b_std=0.1)
input_layers = [conv,
stax.FanOut(n_branches)]
branches = []
for b in range(n_branches):
branch_layers = [FanInTest._get_phi(b)]
for i in range(b):
branch_layers += [
stax.Conv(
out_chan=width,
filter_shape=(i + 1, 4 - i),
padding='SAME',
W_std=1.25 + i,
b_std=0.1 + i),
FanInTest._get_phi(i)]
if branch_in == 'dense_before_branch_in':
branch_layers += [conv]
branches += [stax.serial(*branch_layers)]
output_layers = [
stax.FanInSum() if axis is None else stax.FanInConcat(axis),
stax.Relu(),
stax.GlobalAvgPool() if readout == 'pool' else stax.Flatten()
]
if branch_in == 'dense_after_branch_in':
output_layers.insert(1, conv)
nn = stax.serial(*(input_layers + [stax.parallel(*branches)] +
output_layers))
init_fn, apply_fn, kernel_fn = stax.serial(
nn, stax.Dense(1 if get == 'ntk' else width, 1.25, 0.5))
kernel_fn_mc = monte_carlo.monte_carlo_kernel_fn(
init_fn,
apply_fn,
key,
n_samples,
device_count=0 if axis in (0, -4) else -1)
exact = kernel_fn(X0_1, X0_2, get=get)
empirical = kernel_fn_mc(X0_1, X0_2, get=get)
empirical = empirical.reshape(exact.shape)
utils.assert_close_matrices(self, empirical, exact, tol)
def test_fan_in_fc(self, same_inputs, axis, n_branches, get, branch_in):
if axis in (None, 0) and branch_in == 'dense_after_branch_in':
raise jtu.SkipTest('`FanInSum` and `FanInConcat(0)` '
'require `is_gaussian`.')
if axis == 1 and branch_in == 'dense_before_branch_in':
raise jtu.SkipTest('`FanInConcat` on feature axis requires a dense layer'
'after concatenation.')
key = random.PRNGKey(1)
X0_1 = random.normal(key, (10, 20))
X0_2 = None if same_inputs else random.normal(key, (8, 20))
if xla_bridge.get_backend().platform == 'tpu':
width = 2048
n_samples = 1024
tol = 0.02
else:
width = 1024
n_samples = 256
tol = 0.01
dense = stax.Dense(width, 1.25, 0.1)
input_layers = [dense,
stax.FanOut(n_branches)]
branches = []
for b in range(n_branches):
branch_layers = [FanInTest._get_phi(b)]
for i in range(b):
branch_layers += [
stax.Dense(width, 1. + 2 * i, 0.5 + i),
FanInTest._get_phi(i)]
if branch_in == 'dense_before_branch_in':
branch_layers += [dense]
branches += [stax.serial(*branch_layers)]
output_layers = [
stax.FanInSum() if axis is None else stax.FanInConcat(axis),
stax.Relu()
]
if branch_in == 'dense_after_branch_in':
output_layers.insert(1, dense)
nn = stax.serial(*(input_layers + [stax.parallel(*branches)] +
output_layers))
if get == 'nngp':
init_fn, apply_fn, kernel_fn = nn
elif get == 'ntk':
init_fn, apply_fn, kernel_fn = stax.serial(nn, stax.Dense(1, 1.25, 0.5))
else:
raise ValueError(get)
kernel_fn_mc = monte_carlo.monte_carlo_kernel_fn(
init_fn, apply_fn, key, n_samples, device_count=0)
exact = kernel_fn(X0_1, X0_2, get=get)
empirical = kernel_fn_mc(X0_1, X0_2, get=get)
empirical = empirical.reshape(exact.shape)
utils.assert_close_matrices(self, empirical, exact, tol)
