rocm_jax/examples/examples_test.py

70 lines
2.0 KiB
Python
Raw Permalink Normal View History

# Copyright 2018 The JAX Authors.
2018-12-12 13:07:46 -08:00
#
# 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.
import os
import sys
import zlib
2018-12-12 13:07:46 -08:00
from absl.testing import absltest
from absl.testing import parameterized
import numpy as np
import jax
2019-04-03 13:32:14 +01:00
from jax import random
import jax.numpy as jnp
from jax._src import test_util as jtu
del jtu # Needed for flags
2018-12-12 13:07:46 -08:00
sys.path.append(os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
2018-12-12 17:21:00 -08:00
from examples import kernel_lsq
2018-12-12 13:07:46 -08:00
sys.path.pop()
jax.config.parse_flags_with_absl()
2018-12-12 13:07:46 -08:00
def _CheckShapeAgreement(test_case, init_fun, apply_fun, input_shape):
jax_rng = random.key(0)
2019-04-03 13:32:14 +01:00
result_shape, params = init_fun(jax_rng, input_shape)
result = apply_fun(params, test_case.rng.normal(size=input_shape).astype("float32"))
2018-12-12 13:07:46 -08:00
test_case.assertEqual(result.shape, result_shape)
class ExamplesTest(parameterized.TestCase):
def setUp(self):
self.rng = np.random.default_rng(zlib.adler32(self.__class__.__name__.encode()))
2018-12-12 13:07:46 -08:00
2018-12-12 17:21:00 -08:00
def testKernelRegressionGram(self):
n, d = 100, 20
xs = self.rng.normal(size=(n, d))
kernel = lambda x, y: jnp.dot(x, y)
np.testing.assert_allclose(kernel_lsq.gram(kernel, xs), jnp.dot(xs, xs.T), atol=1E-5)
2018-12-12 17:21:00 -08:00
@jax.default_matmul_precision("float32")
2018-12-12 17:21:00 -08:00
def testKernelRegressionTrainAndPredict(self):
n, d = 100, 20
truth = self.rng.normal(size=d)
xs = self.rng.normal(size=(n, d))
ys = jnp.dot(xs, truth)
kernel = lambda x, y: jnp.dot(x, y)
2018-12-12 17:21:00 -08:00
predict = kernel_lsq.train(kernel, xs, ys)
np.testing.assert_allclose(predict(xs), ys, atol=1e-3, rtol=1e-3)
2018-12-12 17:21:00 -08:00
2018-12-12 13:07:46 -08:00
if __name__ == "__main__":
absltest.main()