Python theano.compile() Examples

def __init__(self, name, shared=tensor._shared,
                 sub=tensor.Subtensor,
                 inc_sub=tensor.IncSubtensor,
                 adv_sub1=tensor.AdvancedSubtensor1,
                 adv_incsub1=tensor.AdvancedIncSubtensor1,
                 mode=None,
                 dtype=theano.config.floatX,
                 type=tensor.TensorType,
                 ignore_topo=DeepCopyOp):
        self.shared = shared
        self.sub = sub
        self.inc_sub = inc_sub
        self.adv_sub1 = adv_sub1
        self.adv_incsub1 = adv_incsub1
        if mode is None:
            mode = theano.compile.mode.get_default_mode()
        self.mode = mode
        self.dtype = dtype
        self.type = type
        self.ignore_topo = ignore_topo
        self.fast_compile = theano.config.mode == 'FAST_COMPILE'
        self.ops = (sub, inc_sub, adv_sub1, adv_incsub1)
        return super(T_subtensor, self).__init__(name) 

def __init__(self, name,
                 shared=tensor._shared,
                 sub=tensor.AdvancedSubtensor,
                 inc_sub=tensor.AdvancedIncSubtensor,
                 mode=None,
                 dtype=theano.config.floatX,
                 ignore_topo=DeepCopyOp):
        self.shared = shared
        self.sub = sub
        self.inc_sub = inc_sub
        if mode is None:
            mode = theano.compile.mode.get_default_mode()
        self.mode = mode
        self.dtype = dtype
        self.ignore_topo = ignore_topo
        super(TestAdvancedSubtensor, self).__init__(name) 

def __init__(self, name, shared=tensor._shared,
                 sub=tensor.Subtensor,
                 inc_sub=tensor.IncSubtensor,
                 adv_sub1=tensor.AdvancedSubtensor1,
                 adv_incsub1=tensor.AdvancedIncSubtensor1,
                 mode=None,
                 dtype=theano.config.floatX,
                 type=tensor.TensorType,
                 ignore_topo=DeepCopyOp):
        self.shared = shared
        self.sub = sub
        self.inc_sub = inc_sub
        self.adv_sub1 = adv_sub1
        self.adv_incsub1 = adv_incsub1
        if mode is None:
            mode = theano.compile.mode.get_default_mode()
        self.mode = mode
        self.dtype = dtype
        self.type = type
        self.ignore_topo = ignore_topo
        self.fast_compile = theano.config.mode == 'FAST_COMPILE'
        self.ops = (sub, inc_sub, adv_sub1, adv_incsub1)
        return super(T_subtensor, self).__init__(name) 

def handle_shared_float32(tf):
    """
    Set the default shared type for float32 tensor to CudaNdarrayType.

    This function is intended to be called from use(gpu_index), not directly.

    """
    if tf:
        theano.compile.shared_constructor(float32_shared_constructor)
    else:
        theano.compile.shared_constructor(float32_shared_constructor, True)
        assert (float32_shared_constructor not in
                theano.compile.shared.constructors)

# We can't test the driver during import here as this cause circular
# import dependency. So we also test it in the file theano/__init__.py 

def dnn_version():
    """Return the current cuDNN version we compile with.

    This returns a tuple with the header version and the library
    version we link with. For older cudnn version without version
    information, we return -1.

    """
    if not dnn_available():
        raise Exception(
            "We can't determine the cudnn version as it is not available",
            dnn_available.msg)

    if dnn_version.v is None:
        f = theano.function([], DnnVersion()(),
                            theano.Mode(optimizer=None),
                            profile=False)
        dnn_version.v = f()
    return dnn_version.v 

def dnn_version():
    """Return the current cuDNN version we compile with.

    This returns a tuple with the header version and the library
    version we link with. For older cudnn version without version
    information, we return -1.

    """
    if not dnn_available():
        raise Exception(
            "We can't determine the cudnn version as it is not available",
            dnn_available.msg)

    if dnn_version.v is None:
        f = theano.function([], DnnVersion()(),
                            theano.Mode(optimizer=None),
                            profile=False)
        dnn_version.v = f()
    return dnn_version.v 

def handle_shared_float32(tf):
    """
    Set the default shared type for float32 tensor to CudaNdarrayType.

    This function is intended to be called from use(gpu_index), not directly.

    """
    if tf:
        theano.compile.shared_constructor(float32_shared_constructor)
    else:
        theano.compile.shared_constructor(float32_shared_constructor, True)
        assert (float32_shared_constructor not in
                theano.compile.shared.constructors)

# We can't test the driver during import here as this cause circular
# import dependency. So we also test it in the file theano/__init__.py 

def __init__(self, name,
                 shared=tensor._shared,
                 sub=tensor.AdvancedSubtensor,
                 inc_sub=tensor.AdvancedIncSubtensor,
                 mode=None,
                 dtype=theano.config.floatX,
                 ignore_topo=DeepCopyOp):
        self.shared = shared
        self.sub = sub
        self.inc_sub = inc_sub
        if mode is None:
            mode = theano.compile.mode.get_default_mode()
        self.mode = mode
        self.dtype = dtype
        self.ignore_topo = ignore_topo
        super(TestAdvancedSubtensor, self).__init__(name) 

def set_cuda_disabled():
    """
    Function used to disable cuda.

    A warning is displayed, so that the user is aware that cuda-based code is
    not going to work.
    Note that there is no point calling this function from outside of
    `cuda.__init__`, since it has no effect once the module is loaded.

    """
    global cuda_available, cuda_warning_is_displayed
    cuda_available = False

# cuda_ndarray compile and import 

def test_reallocation():
    x = tensor.scalar('x')
    y = tensor.scalar('y')
    z = tensor.tanh(3 * x + y) + tensor.cosh(x + 5 * y)
    # The functinality is currently implement for non lazy and non c VM only.
    for l in [vm.VM_Linker(allow_gc=False, lazy=False, use_cloop=False),
              vm.VM_Linker(allow_gc=True, lazy=False, use_cloop=False)]:
        m = theano.compile.get_mode(theano.Mode(linker=l))
        m = m.excluding('fusion', 'inplace')

        f = theano.function([x, y], z, name="test_reduce_memory",
                            mode=m)
        output = f(1, 2)
        assert output
        storage_map = f.fn.storage_map

        def check_storage(storage_map):
            from theano.tensor.var import TensorConstant
            for i in storage_map:
                if not isinstance(i, TensorConstant):
                    keys_copy = list(storage_map.keys())[:]
                    keys_copy.remove(i)
                    for o in keys_copy:
                        if (storage_map[i][0] and
                                storage_map[i][0] is storage_map[o][0]):
                            return [True, storage_map[o][0]]
            return [False, None]

        assert check_storage(storage_map)[0]
        assert len(set(id(v) for v in
                       itervalues(storage_map))) < len(storage_map) 

def test_duplicate_inputs(self):
        x = theano.tensor.lscalar('x')
        self.assertRaises(theano.compile.UnusedInputError,
                theano.function, [x, x, x], x) 

def test_check_isfinite(self):
        x = theano.tensor.vector()
        f = theano.function([x], (x + 2) * 5, mode='DEBUG_MODE')
        g = theano.function([x], theano.tensor.log(x), mode='DEBUG_MODE')

        # this should work
        f(numpy.log([3, 4, 5]).astype(config.floatX))

        # if TensorType.filter_checks_isfinite were true, these would raise
        # ValueError
        # if not, DebugMode will check internally, and raise InvalidValueError
        # passing an invalid value as an input should trigger ValueError
        self.assertRaises(debugmode.InvalidValueError, f,
                numpy.log([3, -4, 5]).astype(config.floatX))
        self.assertRaises(debugmode.InvalidValueError, f,
                (numpy.asarray([0, 1.0, 0]) / 0).astype(config.floatX))
        self.assertRaises(debugmode.InvalidValueError, f,
                (numpy.asarray([1.0, 1.0, 1.0]) / 0).astype(config.floatX))

        # generating an invalid value internally should trigger
        # InvalidValueError
        self.assertRaises(debugmode.InvalidValueError, g,
                numpy.asarray([3, -4, 5], dtype=config.floatX))

        # this should disable the exception
        theano.tensor.TensorType.filter_checks_isfinite = False
        theano.compile.mode.predefined_modes[
                'DEBUG_MODE'].check_isfinite = False
        # insert several Inf
        f(numpy.asarray(numpy.asarray([1.0, 1.0, 1.0]) / 0,
            dtype=config.floatX)) 

def tearDown(self):
        theano.tensor.TensorType.filter_checks_isfinite = self.old_ts
        theano.compile.mode.predefined_modes[
                'DEBUG_MODE'].check_isfinite = self.old_dm 

def setUp(self):
        self.old_ts = theano.tensor.TensorType.filter_checks_isfinite
        self.old_dm = theano.compile.mode.predefined_modes[
                'DEBUG_MODE'].check_isfinite 

def test_empty_givens_updates():
    """
    Regression test for bug fixed in 8625e03.
    """
    # Empty givens / updates dictionaries were not properly detected before,
    # triggering useless crashes at compile time.
    x = T.scalar()
    y = x * 2
    function([theano.In(x)], y, givens={})
    function([theano.In(x)], y, updates={}) 

def test_constant_output(self):
        # Test that if the output is a constant, we respect the theano memory interface
        f = theano.function([], theano.tensor.constant([4]))
        # print f.maker.fgraph.toposort()
        out = f()
        assert (out == 4).all()
        out[0] = 3
        out2 = f()
        # If the following 2 asserts fail it mean Theano broke it's memory contract.
        assert out2 is not out
        assert (out2 == 4).all()

        # Test that if the output is a constant and borrow, we respect the theano memory interface
        f = theano.function([], Out(theano.tensor.constant([4]), borrow=True))
        # print f.maker.fgraph.toposort()
        out = f()
        assert (out == 4).all()
        out[0] = 3
        out2 = f()

        if isinstance(theano.compile.mode.get_default_mode(),
                      theano.compile.DebugMode):
            # In DebugMode, we don't implement optimization based on borrow on the output.
            assert (out2 == 4).all()
        else:
            assert out2 is out
            assert (out2 == 3).all() 

def compile_gpu_func(nan_is_error, inf_is_error, big_is_error):
    """ compile utility function used by contains_nan and contains_inf
    """
    global f_gpumin, f_gpumax, f_gpuabsmax
    if not cuda.cuda_available:
        return
    guard_input = cuda.fvector('nan_guard')
    cuda_compile_failed = False
    if (nan_is_error or inf_is_error) and f_gpumin is None:
        try:
            f_gpumin = theano.function(
                [guard_input], T.min(guard_input),
                mode='FAST_RUN'
            )
        except RuntimeError:
            # This can happen if cuda is available, but the
            # device is in exclusive mode and used by another
            # process.
            cuda_compile_failed = True
    if inf_is_error and not cuda_compile_failed and f_gpumax is None:
        try:
            f_gpumax = theano.function(
                [guard_input], T.max(guard_input),
                mode='FAST_RUN'
            )
        except RuntimeError:
            # This can happen if cuda is available, but the
            # device is in exclusive mode and used by another
            # process.
            cuda_compile_failed = True
    if big_is_error and not cuda_compile_failed and f_gpuabsmax is None:
        try:
            f_gpuabsmax = theano.function(
                [guard_input], T.max(T.abs_(guard_input)),
                mode='FAST_RUN'
                )
        except RuntimeError:
            # This can happen if cuda is available, but the
            # device is in exclusive mode and used by another
            # process.
            cuda_compile_failed = True 

def test_filter_float():
    theano.compile.shared_constructor(gpuarray_shared_constructor)
    try:
        s = theano.shared(numpy.array(0.0, dtype='float32'),
                          target=test_ctx_name)
        theano.function([], updates=[(s, 0.0)])
    finally:
        del theano.compile.sharedvalue.shared.constructors[-1] 

def test_1d_set_adv_selection(self):
        a = set_subtensor(self.v[self.adv1q], self.v[self.adv1q])

        assert a.type == self.v.type

        # TODO: compile a function and verify that the subtensor is removed
        #      completely, because the whole expression is redundant.

        f = theano.function([self.v, self.adv1q], a, allow_input_downcast=True)
        aval = f([.4, .9, .1], [1, 2])
        assert numpy.allclose(aval, [.4, 0.9, 0.1]) 

def test_filter_float():
    theano.compile.shared_constructor(gpuarray_shared_constructor)
    try:
        s = theano.shared(numpy.array(0.0, dtype='float32'),
                          target=test_ctx_name)
        theano.function([], updates=[(s, 0.0)])
    finally:
        del theano.compile.sharedvalue.shared.constructors[-1] 

def set_cuda_disabled():
    """
    Function used to disable cuda.

    A warning is displayed, so that the user is aware that cuda-based code is
    not going to work.
    Note that there is no point calling this function from outside of
    `cuda.__init__`, since it has no effect once the module is loaded.

    """
    global cuda_available, cuda_warning_is_displayed
    cuda_available = False

# cuda_ndarray compile and import 

def test_reallocation():
    x = tensor.scalar('x')
    y = tensor.scalar('y')
    z = tensor.tanh(3 * x + y) + tensor.cosh(x + 5 * y)
    # The functinality is currently implement for non lazy and non c VM only.
    for l in [vm.VM_Linker(allow_gc=False, lazy=False, use_cloop=False),
              vm.VM_Linker(allow_gc=True, lazy=False, use_cloop=False)]:
        m = theano.compile.get_mode(theano.Mode(linker=l))
        m = m.excluding('fusion', 'inplace')

        f = theano.function([x, y], z, name="test_reduce_memory",
                            mode=m)
        output = f(1, 2)
        assert output
        storage_map = f.fn.storage_map

        def check_storage(storage_map):
            from theano.tensor.var import TensorConstant
            for i in storage_map:
                if not isinstance(i, TensorConstant):
                    keys_copy = list(storage_map.keys())[:]
                    keys_copy.remove(i)
                    for o in keys_copy:
                        if (storage_map[i][0] and
                                storage_map[i][0] is storage_map[o][0]):
                            return [True, storage_map[o][0]]
            return [False, None]

        assert check_storage(storage_map)[0]
        assert len(set(id(v) for v in
                       itervalues(storage_map))) < len(storage_map) 

def test_duplicate_inputs(self):
        x = theano.tensor.lscalar('x')
        self.assertRaises(theano.compile.UnusedInputError,
                          theano.function, [x, x, x], x) 

def test_check_isfinite(self):
        x = theano.tensor.vector()
        f = theano.function([x], (x + 2) * 5, mode='DEBUG_MODE')
        g = theano.function([x], theano.tensor.log(x), mode='DEBUG_MODE')

        # this should work
        f(numpy.log([3, 4, 5]).astype(config.floatX))

        # if TensorType.filter_checks_isfinite were true, these would raise
        # ValueError
        # if not, DebugMode will check internally, and raise InvalidValueError
        # passing an invalid value as an input should trigger ValueError
        self.assertRaises(debugmode.InvalidValueError, f,
                          numpy.log([3, -4, 5]).astype(config.floatX))
        self.assertRaises(debugmode.InvalidValueError, f,
                          (numpy.asarray([0, 1.0, 0]) / 0).astype(config.floatX))
        self.assertRaises(debugmode.InvalidValueError, f,
                          (numpy.asarray([1.0, 1.0, 1.0]) / 0).astype(config.floatX))

        # generating an invalid value internally should trigger
        # InvalidValueError
        self.assertRaises(debugmode.InvalidValueError, g,
                          numpy.asarray([3, -4, 5], dtype=config.floatX))

        # this should disable the exception
        theano.tensor.TensorType.filter_checks_isfinite = False
        theano.compile.mode.predefined_modes[
            'DEBUG_MODE'].check_isfinite = False
        # insert several Inf
        f(numpy.asarray(numpy.asarray([1.0, 1.0, 1.0]) / 0,
                        dtype=config.floatX)) 

def tearDown(self):
        theano.tensor.TensorType.filter_checks_isfinite = self.old_ts
        theano.compile.mode.predefined_modes[
            'DEBUG_MODE'].check_isfinite = self.old_dm 

def setUp(self):
        self.old_ts = theano.tensor.TensorType.filter_checks_isfinite
        self.old_dm = theano.compile.mode.predefined_modes[
            'DEBUG_MODE'].check_isfinite 

def test_empty_givens_updates():
    """
    Regression test for bug fixed in 8625e03.
    """
    # Empty givens / updates dictionaries were not properly detected before,
    # triggering useless crashes at compile time.
    x = T.scalar()
    y = x * 2
    function([theano.In(x)], y, givens={})
    function([theano.In(x)], y, updates={}) 

def test_constant_output(self):
        # Test that if the output is a constant, we respect the theano memory interface
        f = theano.function([], theano.tensor.constant([4]))
        # print f.maker.fgraph.toposort()
        out = f()
        assert (out == 4).all()
        out[0] = 3
        out2 = f()
        # If the following 2 asserts fail it mean Theano broke it's memory contract.
        assert out2 is not out
        assert (out2 == 4).all()

        # Test that if the output is a constant and borrow, we respect the theano memory interface
        f = theano.function([], Out(theano.tensor.constant([4]), borrow=True))
        # print f.maker.fgraph.toposort()
        out = f()
        assert (out == 4).all()
        out[0] = 3
        out2 = f()

        if isinstance(theano.compile.mode.get_default_mode(),
                      theano.compile.DebugMode):
            # In DebugMode, we don't implement optimization based on borrow on the output.
            assert (out2 == 4).all()
        else:
            assert out2 is out
            assert (out2 == 3).all() 

def compile_gpu_func(nan_is_error, inf_is_error, big_is_error):
    """ compile utility function used by contains_nan and contains_inf
    """
    global f_gpumin, f_gpumax, f_gpuabsmax
    if not cuda.cuda_available:
        return
    guard_input = cuda.fvector('nan_guard')
    cuda_compile_failed = False
    if (nan_is_error or inf_is_error) and f_gpumin is None:
        try:
            f_gpumin = theano.function(
                [guard_input], T.min(guard_input),
                mode='FAST_RUN'
            )
        except RuntimeError:
            # This can happen if cuda is available, but the
            # device is in exclusive mode and used by another
            # process.
            cuda_compile_failed = True
    if inf_is_error and not cuda_compile_failed and f_gpumax is None:
        try:
            f_gpumax = theano.function(
                [guard_input], T.max(guard_input),
                mode='FAST_RUN'
            )
        except RuntimeError:
            # This can happen if cuda is available, but the
            # device is in exclusive mode and used by another
            # process.
            cuda_compile_failed = True
    if big_is_error and not cuda_compile_failed and f_gpuabsmax is None:
        try:
            f_gpuabsmax = theano.function(
                [guard_input], T.max(T.abs_(guard_input)),
                mode='FAST_RUN'
                )
        except RuntimeError:
            # This can happen if cuda is available, but the
            # device is in exclusive mode and used by another
            # process.
            cuda_compile_failed = True 

def test_1d_set_adv_selection(self):
        a = set_subtensor(self.v[self.adv1q], self.v[self.adv1q])

        assert a.type == self.v.type

        # TODO: compile a function and verify that the subtensor is removed
        #      completely, because the whole expression is redundant.

        f = theano.function([self.v, self.adv1q], a, allow_input_downcast=True)
        aval = f([.4, .9, .1], [1, 2])
        assert numpy.allclose(aval, [.4, 0.9, 0.1])