mirror of
https://github.com/ROCm/jax.git
synced 2025-04-24 21:56:05 +00:00
2d25773c21
fixes #2107, draws from #2356 and #2357, thanks @yingted ! Co-authored-by: Ted Ying <yingted@gmail.com>
198 lines
7.0 KiB
Python
198 lines
7.0 KiB
Python
# Copyright 2019 Google LLC
|
|
#
|
|
# Licensed under the Apache License, Version 2.0 (the "License");
|
|
# you may not use this file except in compliance with the License.
|
|
# You may obtain a copy of the License at
|
|
#
|
|
# https://www.apache.org/licenses/LICENSE-2.0
|
|
#
|
|
# Unless required by applicable law or agreed to in writing, software
|
|
# distributed under the License is distributed on an "AS IS" BASIS,
|
|
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
|
|
# See the License for the specific language governing permissions and
|
|
# limitations under the License.
|
|
|
|
"""Tests for nn module."""
|
|
|
|
import collections
|
|
import itertools
|
|
|
|
from absl.testing import absltest
|
|
from absl.testing import parameterized
|
|
|
|
import numpy as onp
|
|
|
|
from jax import test_util as jtu
|
|
from jax.test_util import check_grads
|
|
from jax import nn
|
|
from jax import random
|
|
import jax
|
|
import jax.numpy as np
|
|
|
|
from jax.config import config
|
|
config.parse_flags_with_absl()
|
|
|
|
class NNFunctionsTest(jtu.JaxTestCase):
|
|
|
|
@jtu.skip_on_flag("jax_skip_slow_tests", True)
|
|
def testSoftplusGrad(self):
|
|
check_grads(nn.softplus, (1e-8,), order=4,
|
|
rtol=1e-2 if jtu.device_under_test() == "tpu" else None)
|
|
|
|
def testSoftplusGradZero(self):
|
|
check_grads(nn.softplus, (0.,), order=1,
|
|
rtol=1e-2 if jtu.device_under_test() == "tpu" else None)
|
|
|
|
def testSoftplusGradInf(self):
|
|
self.assertAllClose(
|
|
1., jax.grad(nn.softplus)(float('inf')), check_dtypes=True)
|
|
|
|
def testSoftplusGradNegInf(self):
|
|
check_grads(nn.softplus, (-float('inf'),), order=1,
|
|
rtol=1e-2 if jtu.device_under_test() == "tpu" else None)
|
|
|
|
def testSoftplusGradNan(self):
|
|
check_grads(nn.softplus, (float('nan'),), order=1,
|
|
rtol=1e-2 if jtu.device_under_test() == "tpu" else None)
|
|
|
|
@parameterized.parameters([
|
|
int, np.int32, float, np.float64, np.float32, np.float64,])
|
|
def testSoftplusZero(self, dtype):
|
|
self.assertEqual(np.log(dtype(2)), nn.softplus(dtype(0)))
|
|
|
|
def testReluGrad(self):
|
|
rtol = 1e-2 if jtu.device_under_test() == "tpu" else None
|
|
check_grads(nn.relu, (1.,), order=3, rtol=rtol)
|
|
check_grads(nn.relu, (-1.,), order=3, rtol=rtol)
|
|
jaxpr = jax.make_jaxpr(jax.grad(nn.relu))(0.)
|
|
self.assertEqual(len(jaxpr.jaxpr.eqns), 2)
|
|
|
|
def testSoftplusValue(self):
|
|
val = nn.softplus(89.)
|
|
self.assertAllClose(val, 89., check_dtypes=False)
|
|
|
|
@jtu.skip_on_flag("jax_skip_slow_tests", True)
|
|
def testEluGrad(self):
|
|
check_grads(nn.elu, (1e4,), order=4, eps=1.)
|
|
|
|
def testEluValue(self):
|
|
val = nn.elu(1e4)
|
|
self.assertAllClose(val, 1e4, check_dtypes=False)
|
|
|
|
@parameterized.parameters(*itertools.product(
|
|
(np.float32, np.bfloat16, np.float16),
|
|
(nn.gelu, nn.relu, nn.softplus, nn.sigmoid)))
|
|
def testDtypeMatchesInput(self, dtype, fn):
|
|
if dtype is np.float16 and jtu.device_under_test() == "tpu":
|
|
self.skipTest("float16 not supported on TPU")
|
|
x = np.zeros((), dtype=dtype)
|
|
out = fn(x)
|
|
self.assertEqual(out.dtype, dtype)
|
|
|
|
@jtu.skip_on_devices("gpu", "tpu")
|
|
def testEluMemory(self):
|
|
# see https://github.com/google/jax/pull/1640
|
|
jax.make_jaxpr(nn.elu)(np.ones((10 ** 12,))) # don't oom
|
|
|
|
@jtu.skip_on_devices("gpu", "tpu")
|
|
def testHardTanhMemory(self):
|
|
# see https://github.com/google/jax/pull/1640
|
|
jax.make_jaxpr(nn.hard_tanh)(np.ones((10 ** 12,))) # don't oom
|
|
|
|
def testOneHot(self):
|
|
actual = nn.one_hot(np.array([0, 1, 2]), 3)
|
|
expected = np.array([[1., 0., 0.],
|
|
[0., 1., 0.],
|
|
[0., 0., 1.]])
|
|
self.assertAllClose(actual, expected, check_dtypes=True)
|
|
|
|
actual = nn.one_hot(np.array([1, 2, 0]), 3)
|
|
expected = np.array([[0., 1., 0.],
|
|
[0., 0., 1.],
|
|
[1., 0., 0.]])
|
|
self.assertAllClose(actual, expected, check_dtypes=True)
|
|
|
|
def testOneHotOutOfBound(self):
|
|
actual = nn.one_hot(np.array([-1, 3]), 3)
|
|
expected = np.array([[0., 0., 0.],
|
|
[0., 0., 0.]])
|
|
self.assertAllClose(actual, expected, check_dtypes=True)
|
|
|
|
def testOneHotNonArrayInput(self):
|
|
actual = nn.one_hot([0, 1, 2], 3)
|
|
expected = np.array([[1., 0., 0.],
|
|
[0., 1., 0.],
|
|
[0., 0., 1.]])
|
|
self.assertAllClose(actual, expected, check_dtypes=True)
|
|
|
|
def testOneHotCustomDtype(self):
|
|
actual = nn.one_hot(np.array([0, 1, 2]), 3, dtype=np.bool_)
|
|
expected = np.array([[True, False, False],
|
|
[False, True, False],
|
|
[False, False, True]])
|
|
self.assertAllClose(actual, expected, check_dtypes=True)
|
|
|
|
InitializerRecord = collections.namedtuple(
|
|
"InitializerRecord",
|
|
["name", "initializer", "shapes"])
|
|
|
|
ALL_SHAPES = [(2,), (2, 2), (2, 3), (3, 2), (2, 3, 4), (4, 3, 2), (2, 3, 4, 5)]
|
|
|
|
def initializer_record(name, initializer, min_dims=2, max_dims=4):
|
|
shapes = [shape for shape in ALL_SHAPES
|
|
if min_dims <= len(shape) <= max_dims]
|
|
return InitializerRecord(name, initializer, shapes)
|
|
|
|
INITIALIZER_RECS = [
|
|
initializer_record("uniform", nn.initializers.uniform, 1),
|
|
initializer_record("normal", nn.initializers.normal, 1),
|
|
initializer_record("he_normal", nn.initializers.he_normal),
|
|
initializer_record("he_uniform", nn.initializers.he_uniform),
|
|
initializer_record("glorot_normal", nn.initializers.glorot_normal),
|
|
initializer_record("glorot_uniform", nn.initializers.glorot_uniform),
|
|
initializer_record("lecun_normal", nn.initializers.lecun_normal),
|
|
initializer_record("lecun_uniform", nn.initializers.lecun_uniform),
|
|
initializer_record("orthogonal", nn.initializers.orthogonal, 2, 2),
|
|
initializer_record("delta_orthogonal", nn.initializers.delta_orthogonal, 4, 4)
|
|
]
|
|
|
|
class NNInitializersTest(jtu.JaxTestCase):
|
|
|
|
@parameterized.named_parameters(jtu.cases_from_list(
|
|
{"testcase_name":
|
|
"_{}_{}".format(
|
|
rec.name,
|
|
jtu.format_shape_dtype_string(shape, dtype)),
|
|
"initializer": rec.initializer(),
|
|
"shape": shape, "dtype": dtype}
|
|
for rec in INITIALIZER_RECS
|
|
for shape in rec.shapes
|
|
for dtype in [onp.float32, onp.float64]))
|
|
def testInitializer(self, initializer, shape, dtype):
|
|
rng = random.PRNGKey(0)
|
|
val = initializer(rng, shape, dtype)
|
|
self.assertEqual(shape, np.shape(val))
|
|
self.assertEqual(jax.dtypes.canonicalize_dtype(dtype), np.dtype(val))
|
|
|
|
@parameterized.named_parameters(jtu.cases_from_list(
|
|
{"testcase_name":
|
|
"_{}_{}".format(
|
|
rec.name,
|
|
jtu.format_shape_dtype_string(shape, dtype)),
|
|
"initializer_provider": rec.initializer,
|
|
"shape": shape, "dtype": dtype}
|
|
for rec in INITIALIZER_RECS
|
|
for shape in rec.shapes
|
|
for dtype in [onp.float32, onp.float64]))
|
|
def testInitializerProvider(self, initializer_provider, shape, dtype):
|
|
rng = random.PRNGKey(0)
|
|
initializer = initializer_provider(dtype=dtype)
|
|
val = initializer(rng, shape)
|
|
|
|
self.assertEqual(shape, np.shape(val))
|
|
self.assertEqual(jax.dtypes.canonicalize_dtype(dtype), np.dtype(val))
|
|
|
|
|
|
if __name__ == "__main__":
|
|
absltest.main()
|