rocm_jax/jax/test_util.py

# Copyright 2018 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.

from __future__ import absolute_import
from __future__ import division
from __future__ import print_function

import functools
import re
import itertools as it
import os
from unittest import SkipTest

from absl.testing import absltest
from absl.testing import parameterized

import numpy as onp
import numpy.random as npr

from six.moves import xrange

from . import api
from . import dtypes
from .config import flags
from .util import partial
from .tree_util import tree_multimap, tree_all, tree_map, tree_reduce
from .lib import xla_bridge
from .interpreters import xla


FLAGS = flags.FLAGS
flags.DEFINE_enum(
    'jax_test_dut', '',
    enum_values=['', 'cpu', 'gpu', 'tpu'],
    help=
    'Describes the device under test in case special consideration is required.'
)

flags.DEFINE_integer(
  'num_generated_cases',
  int(os.getenv('JAX_NUM_GENERATED_CASES', 10)),
  help='Number of generated cases to test')

EPS = 1e-4
ATOL = 1e-4
RTOL = 1e-4

_dtype = lambda x: getattr(x, 'dtype', None) or onp.asarray(x).dtype


def is_sequence(x):
  try:
    iter(x)
  except TypeError:
    return False
  else:
    return True


def assert_numpy_eq(x, y):
  testing_tpu = FLAGS.jax_test_dut and FLAGS.jax_test_dut.startswith("tpu")
  testing_x32 = not FLAGS.jax_enable_x64
  if testing_tpu or testing_x32:
    onp.testing.assert_allclose(x, y, 1e-3, 1e-3)
  else:
    onp.testing.assert_allclose(x, y)


def assert_numpy_close(a, b, atol=ATOL, rtol=RTOL):
  testing_tpu = FLAGS.jax_test_dut and FLAGS.jax_test_dut.startswith("tpu")
  testing_x32 = not FLAGS.jax_enable_x64
  if testing_tpu or testing_x32:
    atol = max(atol, 1e-1)
    rtol = max(rtol, 1e-1)
  assert a.shape == b.shape
  onp.testing.assert_allclose(a, b, atol=atol * a.size, rtol=rtol * b.size)


def check_eq(xs, ys):
  tree_all(tree_multimap(assert_numpy_eq, xs, ys))


def check_close(xs, ys, atol=ATOL, rtol=RTOL):
  assert_close = partial(assert_numpy_close, atol=atol, rtol=rtol)
  tree_all(tree_multimap(assert_close, xs, ys))


def inner_prod(xs, ys):
  contract = lambda x, y: onp.real(onp.vdot(x, y))
  return tree_reduce(onp.add, tree_multimap(contract, xs, ys))


add = partial(tree_multimap, onp.add)
sub = partial(tree_multimap, onp.subtract)
conj = partial(tree_map, onp.conj)


def scalar_mul(xs, a):
  return tree_map(lambda x: onp.multiply(x, a, dtype=_dtype(x)), xs)


def rand_like(rng, x):
  shape = onp.shape(x)
  dtype = _dtype(x)
  randn = lambda: onp.asarray(rng.randn(*shape), dtype=dtype)
  if dtypes.issubdtype(dtype, onp.complexfloating):
    return randn() + dtype.type(1.0j) * randn()
  else:
    return randn()


def numerical_jvp(f, primals, tangents, eps=EPS):
  delta = scalar_mul(tangents, eps)
  f_pos = f(*add(primals, delta))
  f_neg = f(*sub(primals, delta))
  return scalar_mul(sub(f_pos, f_neg), 0.5 / eps)


def check_jvp(f, f_jvp, args, atol=ATOL, rtol=RTOL, eps=EPS):
  rng = onp.random.RandomState(0)
  tangent = tree_map(partial(rand_like, rng), args)
  v_out, t_out = f_jvp(args, tangent)
  v_out_expected = f(*args)
  t_out_expected = numerical_jvp(f, args, tangent, eps=eps)
  # In principle we should expect exact equality of v_out and v_out_expected,
  # but due to nondeterminism especially on GPU (e.g., due to convolution
  # autotuning) we only require "close".
  check_close(v_out, v_out_expected, atol=atol, rtol=rtol)
  check_close(t_out, t_out_expected, atol=atol, rtol=rtol)


def check_vjp(f, f_vjp, args, atol=ATOL, rtol=RTOL, eps=EPS):
  _rand_like = partial(rand_like, onp.random.RandomState(0))
  v_out, vjpfun = f_vjp(*args)
  v_out_expected = f(*args)
  check_close(v_out, v_out_expected, atol=atol, rtol=rtol)
  tangent = tree_map(_rand_like, args)
  tangent_out = numerical_jvp(f, args, tangent, eps=eps)
  cotangent = tree_map(_rand_like, v_out)
  cotangent_out = conj(vjpfun(conj(cotangent)))
  ip = inner_prod(tangent, cotangent_out)
  ip_expected = inner_prod(tangent_out, cotangent)
  check_close(ip, ip_expected, atol=atol, rtol=rtol)


def check_grads(f, args, order,
                modes=["fwd", "rev"], atol=None, rtol=None, eps=None):
  args = tuple(args)
  default_tol = 1e-6 if FLAGS.jax_enable_x64 else 1e-2
  atol = atol or default_tol
  rtol = rtol or default_tol
  eps = eps or EPS

  _check_jvp = partial(check_jvp, atol=atol, rtol=rtol, eps=eps)
  _check_vjp = partial(check_vjp, atol=atol, rtol=rtol, eps=eps)

  def _check_grads(f, args, order):
    if "fwd" in modes:
      _check_jvp(f, partial(api.jvp, f), args)
      if order > 1:
        _check_grads(partial(api.jvp, f), (args, args), order - 1)

    if "rev" in modes:
      _check_vjp(f, partial(api.vjp, f), args)
      if order > 1:
        def f_vjp(*args):
          out_primal_py, vjp_py = api.vjp(f, *args)
          return vjp_py(out_primal_py)
        _check_grads(f_vjp, args, order - 1)

  _check_grads(f, args, order)

def device_under_test():
  return FLAGS.jax_test_dut or xla_bridge.get_backend().platform

def supported_dtypes():
  if device_under_test() == "tpu":
    return {onp.bool_, onp.int32, onp.int64, onp.uint32, onp.uint64,
            onp.float32, onp.complex64}
  else:
    return {onp.bool_, onp.int8, onp.int16, onp.int32, onp.int64,
            onp.uint8, onp.uint16, onp.uint32, onp.uint64,
            onp.float16, onp.float32, onp.float64, onp.complex64,
            onp.complex128}

def skip_on_devices(*disabled_devices):
  """A decorator for test methods to skip the test on certain devices."""
  def skip(test_method):
    @functools.wraps(test_method)
    def test_method_wrapper(self, *args, **kwargs):
      device = device_under_test()
      if device in disabled_devices:
        test_name = getattr(test_method, '__name__', '[unknown test]')
        raise SkipTest('{} not supported on {}.'
                       .format(test_name, device.upper()))
      return test_method(self, *args, **kwargs)
    return test_method_wrapper
  return skip


def skip_on_flag(flag_name, skip_value):
  """A decorator for test methods to skip the test when flags are set."""
  def skip(test_method):        # pylint: disable=missing-docstring
    @functools.wraps(test_method)
    def test_method_wrapper(self, *args, **kwargs):
      flag_value = getattr(FLAGS, flag_name)
      if flag_value == skip_value:
        test_name = getattr(test_method, '__name__', '[unknown test]')
        raise SkipTest('{} not supported when FLAGS.{} is {}'
                       .format(test_name, flag_name, flag_value))
      return test_method(self, *args, **kwargs)
    return test_method_wrapper
  return skip


def format_test_name_suffix(opname, shapes, dtypes):
  arg_descriptions = (format_shape_dtype_string(shape, dtype)
                      for shape, dtype in zip(shapes, dtypes))
  return '{}_{}'.format(opname.capitalize(), '_'.join(arg_descriptions))


# We use special symbols, represented as singleton objects, to distinguish
# between NumPy scalars, Python scalars, and 0-D arrays.
class ScalarShape(object):
  def __len__(self): return 0
class _NumpyScalar(ScalarShape): pass
class _PythonScalar(ScalarShape): pass
NUMPY_SCALAR_SHAPE = _NumpyScalar()
PYTHON_SCALAR_SHAPE = _PythonScalar()


def _dims_of_shape(shape):
  """Converts `shape` to a tuple of dimensions."""
  if type(shape) in (list, tuple):
    return shape
  elif isinstance(shape, ScalarShape):
    return ()
  else:
    raise TypeError(type(shape))


def _cast_to_shape(value, shape, dtype):
  """Casts `value` to the correct Python type for `shape` and `dtype`."""
  if shape is NUMPY_SCALAR_SHAPE:
    # explicitly cast to NumPy scalar in case `value` is a Python scalar.
    return onp.dtype(dtype).type(value)
  elif shape is PYTHON_SCALAR_SHAPE:
    # explicitly cast to Python scalar via https://stackoverflow.com/a/11389998
    return onp.asarray(value).item()
  elif type(shape) in (list, tuple):
    assert onp.shape(value) == tuple(shape)
    return value
  else:
    raise TypeError(type(shape))


def dtype_str(dtype):
  return onp.dtype(dtype).name


def format_shape_dtype_string(shape, dtype):
  if shape is NUMPY_SCALAR_SHAPE:
    return dtype_str(dtype)
  elif shape is PYTHON_SCALAR_SHAPE:
    return 'py' + dtype_str(dtype)
  elif type(shape) in (list, tuple):
    shapestr = ','.join(str(dim) for dim in shape)
    return '{}[{}]'.format(dtype_str(dtype), shapestr)
  elif type(shape) is int:
    return '{}[{},]'.format(dtype_str(dtype), shape)
  else:
    raise TypeError(type(shape))


def _rand_dtype(rand, shape, dtype, scale=1., post=lambda x: x):
  """Produce random values given shape, dtype, scale, and post-processor.

  Args:
    rand: a function for producing random values of a given shape, e.g. a
      bound version of either onp.RandomState.randn or onp.RandomState.rand.
    shape: a shape value as a tuple of positive integers.
    dtype: a numpy dtype.
    scale: optional, a multiplicative scale for the random values (default 1).
    post: optional, a callable for post-processing the random values (default
      identity).

  Returns:
    An ndarray of the given shape and dtype using random values based on a call
    to rand but scaled, converted to the appropriate dtype, and post-processed.
  """
  r = lambda: onp.asarray(scale * rand(*_dims_of_shape(shape)), dtype)
  if dtypes.issubdtype(dtype, onp.complexfloating):
    vals = r() + 1.0j * r()
  else:
    vals = r()
  return _cast_to_shape(onp.asarray(post(vals), dtype), shape, dtype)


def rand_default():
  randn = npr.RandomState(0).randn
  return partial(_rand_dtype, randn, scale=3)


def rand_nonzero():
  post = lambda x: onp.where(x == 0, 1, x)
  randn = npr.RandomState(0).randn
  return partial(_rand_dtype, randn, scale=3, post=post)


def rand_positive():
  post = lambda x: x + 1
  rand = npr.RandomState(0).rand
  return partial(_rand_dtype, rand, scale=2, post=post)


def rand_small():
  randn = npr.RandomState(0).randn
  return partial(_rand_dtype, randn, scale=1e-3)


def rand_not_small():
  post = lambda x: x + onp.where(x > 0, 10., -10.)
  randn = npr.RandomState(0).randn
  return partial(_rand_dtype, randn, scale=3., post=post)


def rand_small_positive():
  rand = npr.RandomState(0).rand
  return partial(_rand_dtype, rand, scale=2e-5)

def rand_uniform(low=0.0, high=1.0):
  assert low < high
  rand = npr.RandomState(0).rand
  post = lambda x: x * (high - low) + low
  return partial(_rand_dtype, rand, post=post)


def rand_some_equal():
  randn = npr.RandomState(0).randn
  rng = npr.RandomState(0)

  def post(x):
    x_ravel = x.ravel()
    if len(x_ravel) == 0:
      return x
    flips = rng.rand(*onp.shape(x)) < 0.5
    return onp.where(flips, x_ravel[0], x)

  return partial(_rand_dtype, randn, scale=100., post=post)


def rand_some_inf():
  """Return a random sampler that produces infinities in floating types."""
  rng = npr.RandomState(1)
  base_rand = rand_default()

  """
  TODO: Complex numbers are not correctly tested
  If blocks should be switched in order, and relevant tests should be fixed
  """
  def rand(shape, dtype):
    """The random sampler function."""
    if not dtypes.issubdtype(dtype, onp.floating):
      # only float types have inf
      return base_rand(shape, dtype)

    if dtypes.issubdtype(dtype, onp.complexfloating):
      base_dtype = onp.real(onp.array(0, dtype=dtype)).dtype
      return rand(shape, base_dtype) + 1j * rand(shape, base_dtype)

    dims = _dims_of_shape(shape)
    posinf_flips = rng.rand(*dims) < 0.1
    neginf_flips = rng.rand(*dims) < 0.1

    vals = base_rand(shape, dtype)
    vals = onp.where(posinf_flips, onp.inf, vals)
    vals = onp.where(neginf_flips, -onp.inf, vals)

    return _cast_to_shape(onp.asarray(vals, dtype=dtype), shape, dtype)

  return rand

def rand_some_nan():
  """Return a random sampler that produces nans in floating types."""
  rng = npr.RandomState(1)
  base_rand = rand_default()

  def rand(shape, dtype):
    """The random sampler function."""
    if dtypes.issubdtype(dtype, onp.complexfloating):
      base_dtype = onp.real(onp.array(0, dtype=dtype)).dtype
      return rand(shape, base_dtype) + 1j * rand(shape, base_dtype)

    if not dtypes.issubdtype(dtype, onp.floating):
      # only float types have inf
      return base_rand(shape, dtype)

    dims = _dims_of_shape(shape)
    nan_flips = rng.rand(*dims) < 0.1

    vals = base_rand(shape, dtype)
    vals = onp.where(nan_flips, onp.nan, vals)

    return _cast_to_shape(onp.asarray(vals, dtype=dtype), shape, dtype)

  return rand

def rand_some_inf_and_nan():
  """Return a random sampler that produces infinities in floating types."""
  rng = npr.RandomState(1)
  base_rand = rand_default()

  """
  TODO: Complex numbers are not correctly tested
  If blocks should be switched in order, and relevant tests should be fixed
  """
  def rand(shape, dtype):
    """The random sampler function."""
    if not dtypes.issubdtype(dtype, onp.floating):
      # only float types have inf
      return base_rand(shape, dtype)

    if dtypes.issubdtype(dtype, onp.complexfloating):
      base_dtype = onp.real(onp.array(0, dtype=dtype)).dtype
      return rand(shape, base_dtype) + 1j * rand(shape, base_dtype)

    dims = _dims_of_shape(shape)
    posinf_flips = rng.rand(*dims) < 0.1
    neginf_flips = rng.rand(*dims) < 0.1
    nan_flips = rng.rand(*dims) < 0.1

    vals = base_rand(shape, dtype)
    vals = onp.where(posinf_flips, onp.inf, vals)
    vals = onp.where(neginf_flips, -onp.inf, vals)
    vals = onp.where(nan_flips, onp.nan, vals)

    return _cast_to_shape(onp.asarray(vals, dtype=dtype), shape, dtype)

  return rand

# TODO(mattjj): doesn't handle complex types
def rand_some_zero():
  """Return a random sampler that produces some zeros."""
  rng = npr.RandomState(1)
  base_rand = rand_default()

  def rand(shape, dtype):
    """The random sampler function."""
    dims = _dims_of_shape(shape)
    zeros = rng.rand(*dims) < 0.5

    vals = base_rand(shape, dtype)
    vals = onp.where(zeros, 0, vals)

    return _cast_to_shape(onp.asarray(vals, dtype=dtype), shape, dtype)

  return rand


def rand_int(low, high=None):
  randint = npr.RandomState(0).randint
  def fn(shape, dtype):
    return randint(low, high=high, size=shape, dtype=dtype)
  return fn

def rand_bool():
  rng = npr.RandomState(0)
  def generator(shape, dtype):
    return _cast_to_shape(rng.rand(*_dims_of_shape(shape)) < 0.5, shape, dtype)
  return generator

def check_raises(thunk, err_type, msg):
  try:
    thunk()
    assert False
  except err_type as e:
    assert str(e).startswith(msg), "\n{}\n\n{}\n".format(e, msg)

_CACHED_INDICES = {}

def cases_from_list(xs):
  xs = list(xs)
  n = len(xs)
  k = min(n, FLAGS.num_generated_cases)
  # Random sampling for every parameterized test is expensive. Do it once and
  # cache the result.
  indices = _CACHED_INDICES.get(n)
  if indices is None:
    rng = npr.RandomState(42)
    _CACHED_INDICES[n] = indices = rng.permutation(n)
  return [xs[i] for i in indices[:k]]

def cases_from_gens(*gens):
  sizes = [1, 3, 10]
  cases_per_size = int(FLAGS.num_generated_cases / len(sizes)) + 1
  for size in sizes:
    for i in xrange(cases_per_size):
      yield ('_{}_{}'.format(size, i),) + tuple(gen(size) for gen in gens)


class JaxTestCase(parameterized.TestCase):
  """Base class for JAX tests including numerical checks and boilerplate."""

  def assertArraysAllClose(self, x, y, check_dtypes, atol=None, rtol=None):
    """Assert that x and y are close (up to numerical tolerances)."""
    self.assertEqual(x.shape, y.shape)
    dtype = lambda x: str(onp.asarray(x).dtype)
    tol = 1e-2 if str(onp.dtype(onp.float32)) in {dtype(x), dtype(y)} else 1e-5
    atol = atol or tol
    rtol = rtol or tol

    if FLAGS.jax_test_dut == 'tpu':
      atol = max(atol, 0.5)
      rtol = max(rtol, 1e-1)

    onp.testing.assert_allclose(x, y, atol=atol, rtol=rtol)

    if check_dtypes:
      self.assertDtypesMatch(x, y)

  def assertDtypesMatch(self, x, y):
    # special rule for complex128, which XLA doesn't support
    def c128_to_c64(dtype):
      if dtype == onp.complex128:
        return onp.complex64
      else:
        return dtype

    if FLAGS.jax_enable_x64:
      x_dtype = c128_to_c64(onp.asarray(x).dtype)
      y_dtype = c128_to_c64(onp.asarray(y).dtype)
      self.assertEqual(x_dtype, y_dtype)

  def assertAllClose(self, x, y, check_dtypes, atol=None, rtol=None):
    """Assert that x and y, either arrays or nested tuples/lists, are close."""
    if isinstance(x, dict):
      self.assertIsInstance(y, dict)
      self.assertEqual(set(x.keys()), set(y.keys()))
      for k in x.keys():
        self.assertAllClose(x[k], y[k], check_dtypes, atol=atol, rtol=rtol)
    elif is_sequence(x) and not hasattr(x, '__array__'):
      self.assertTrue(is_sequence(y) and not hasattr(y, '__array__'))
      self.assertEqual(len(x), len(y))
      for x_elt, y_elt in zip(x, y):
        self.assertAllClose(x_elt, y_elt, check_dtypes, atol=atol, rtol=rtol)
    elif hasattr(x, '__array__') or onp.isscalar(x):
      self.assertTrue(hasattr(y, '__array__') or onp.isscalar(y))
      x = onp.asarray(x)
      y = onp.asarray(y)
      self.assertArraysAllClose(x, y, check_dtypes, atol=atol, rtol=rtol)
    elif x == y:
        return
    else:
      raise TypeError((type(x), type(y)))

  def _CompileAndCheck(self, fun, args_maker, check_dtypes,
                       rtol=None, atol=None):
    """Helper method for running JAX compilation and allclose assertions."""
    args = args_maker()

    def wrapped_fun(*args):
      self.assertTrue(python_should_be_executing)
      return fun(*args)

    python_should_be_executing = True
    python_ans = fun(*args)

    cache_misses = xla.xla_primitive_callable.cache_info().misses
    python_ans = fun(*args)
    self.assertEqual(
        cache_misses, xla.xla_primitive_callable.cache_info().misses,
        "Compilation detected during second call of {} in op-by-op "
        "mode.".format(fun))

    cfun = api.jit(wrapped_fun)
    python_should_be_executing = True
    monitored_ans = cfun(*args)

    python_should_be_executing = False
    compiled_ans = cfun(*args)

    self.assertAllClose(python_ans, monitored_ans, check_dtypes, rtol, atol)
    self.assertAllClose(python_ans, compiled_ans, check_dtypes, rtol, atol)

    args = args_maker()

    python_should_be_executing = True
    python_ans = fun(*args)

    python_should_be_executing = False
    compiled_ans = cfun(*args)

    self.assertAllClose(python_ans, compiled_ans, check_dtypes, rtol, atol)

  def _CheckAgainstNumpy(self, numpy_reference_op, lax_op, args_maker,
                         check_dtypes=False, tol=1e-5):
    args = args_maker()
    numpy_ans = numpy_reference_op(*args)
    lax_ans = lax_op(*args)
    self.assertAllClose(lax_ans, numpy_ans, check_dtypes=check_dtypes,
                        atol=tol, rtol=tol)