How to use the dragon.vm.tensorflow.protobuf.types_pb2 function in dragon
To help you get started, we’ve selected a few dragon examples, based on popular ways it is used in public projects.
Secure your code as it's written. Use Snyk Code to scan source code in minutes - no build needed - and fix issues immediately.
seetaresearch / Dragon / Dragon / python / dragon / vm / tensorflow / framework / dtypes.py View on Github 
types_pb2.DT_UINT16: "uint16",
types_pb2.DT_INT16: "int16",
types_pb2.DT_INT8: "int8",
types_pb2.DT_STRING: "string",
types_pb2.DT_COMPLEX64: "complex64",
types_pb2.DT_COMPLEX128: "complex128",
types_pb2.DT_INT64: "int64",
types_pb2.DT_BOOL: "bool",
types_pb2.DT_QINT8: "qint8",
types_pb2.DT_QUINT8: "quint8",
types_pb2.DT_QINT16: "qint16",
types_pb2.DT_QUINT16: "quint16",
types_pb2.DT_QINT32: "qint32",
types_pb2.DT_BFLOAT16: "bfloat16",
types_pb2.DT_RESOURCE: "resource",
types_pb2.DT_HALF_REF: "float16_ref",
types_pb2.DT_FLOAT_REF: "float32_ref",
types_pb2.DT_DOUBLE_REF: "float64_ref",
types_pb2.DT_INT32_REF: "int32_ref",
types_pb2.DT_UINT8_REF: "uint8_ref",
types_pb2.DT_UINT16_REF: "uint16_ref",
types_pb2.DT_INT16_REF: "int16_ref",
types_pb2.DT_INT8_REF: "int8_ref",
types_pb2.DT_STRING_REF: "string_ref",
types_pb2.DT_COMPLEX64_REF: "complex64_ref",
types_pb2.DT_COMPLEX128_REF: "complex128_ref",
types_pb2.DT_INT64_REF: "int64_ref",
types_pb2.DT_BOOL_REF: "bool_ref",
types_pb2.DT_QINT8_REF: "qint8_ref",
types_pb2.DT_QUINT8_REF: "quint8_ref",
types_pb2.DT_QINT16_REF: "qint16_ref",
types_pb2.DT_QUINT16_REF: "quint16_ref",np.uint64: (0, 2 ** 64 - 1),
np.int32: (-2 ** 31, 2 ** 31 - 1),
np.uint32: (0, 2 ** 32 - 1),
np.float32: (-1, 1),
np.float64: (-1, 1)}
# Define standard wrappers for the types_pb2.DataType enum.
resource = DType(types_pb2.DT_RESOURCE)
float16 = DType(types_pb2.DT_HALF)
half = float16
float32 = DType(types_pb2.DT_FLOAT)
float64 = DType(types_pb2.DT_DOUBLE)
double = float64
int32 = DType(types_pb2.DT_INT32)
uint8 = DType(types_pb2.DT_UINT8)
uint16 = DType(types_pb2.DT_UINT16)
int16 = DType(types_pb2.DT_INT16)
int8 = DType(types_pb2.DT_INT8)
string = DType(types_pb2.DT_STRING)
complex64 = DType(types_pb2.DT_COMPLEX64)
complex128 = DType(types_pb2.DT_COMPLEX128)
int64 = DType(types_pb2.DT_INT64)
bool = DType(types_pb2.DT_BOOL)
qint8 = DType(types_pb2.DT_QINT8)
quint8 = DType(types_pb2.DT_QUINT8)
qint16 = DType(types_pb2.DT_QINT16)
quint16 = DType(types_pb2.DT_QUINT16)
qint32 = DType(types_pb2.DT_QINT32)
bfloat16 = DType(types_pb2.DT_BFLOAT16)types_pb2.DT_QINT16: "qint16",
types_pb2.DT_QUINT16: "quint16",
types_pb2.DT_QINT32: "qint32",
types_pb2.DT_BFLOAT16: "bfloat16",
types_pb2.DT_RESOURCE: "resource",
types_pb2.DT_HALF_REF: "float16_ref",
types_pb2.DT_FLOAT_REF: "float32_ref",
types_pb2.DT_DOUBLE_REF: "float64_ref",
types_pb2.DT_INT32_REF: "int32_ref",
types_pb2.DT_UINT8_REF: "uint8_ref",
types_pb2.DT_UINT16_REF: "uint16_ref",
types_pb2.DT_INT16_REF: "int16_ref",
types_pb2.DT_INT8_REF: "int8_ref",
types_pb2.DT_STRING_REF: "string_ref",
types_pb2.DT_COMPLEX64_REF: "complex64_ref",
types_pb2.DT_COMPLEX128_REF: "complex128_ref",
types_pb2.DT_INT64_REF: "int64_ref",
types_pb2.DT_BOOL_REF: "bool_ref",
types_pb2.DT_QINT8_REF: "qint8_ref",
types_pb2.DT_QUINT8_REF: "quint8_ref",
types_pb2.DT_QINT16_REF: "qint16_ref",
types_pb2.DT_QUINT16_REF: "quint16_ref",
types_pb2.DT_QINT32_REF: "qint32_ref",
types_pb2.DT_BFLOAT16_REF: "bfloat16_ref",
types_pb2.DT_RESOURCE_REF: "resource_ref",
}
# Standard mappings between types_pb2.DataType values and numpy.dtypes.
# Numpy representation for quantized dtypes.
_np_qint8 = np.dtype([("qint8", np.int8, 1)])
_np_quint8 = np.dtype([("quint8", np.uint8, 1)])types_pb2.DT_RESOURCE: "resource",
types_pb2.DT_HALF_REF: "float16_ref",
types_pb2.DT_FLOAT_REF: "float32_ref",
types_pb2.DT_DOUBLE_REF: "float64_ref",
types_pb2.DT_INT32_REF: "int32_ref",
types_pb2.DT_UINT8_REF: "uint8_ref",
types_pb2.DT_UINT16_REF: "uint16_ref",
types_pb2.DT_INT16_REF: "int16_ref",
types_pb2.DT_INT8_REF: "int8_ref",
types_pb2.DT_STRING_REF: "string_ref",
types_pb2.DT_COMPLEX64_REF: "complex64_ref",
types_pb2.DT_COMPLEX128_REF: "complex128_ref",
types_pb2.DT_INT64_REF: "int64_ref",
types_pb2.DT_BOOL_REF: "bool_ref",
types_pb2.DT_QINT8_REF: "qint8_ref",
types_pb2.DT_QUINT8_REF: "quint8_ref",
types_pb2.DT_QINT16_REF: "qint16_ref",
types_pb2.DT_QUINT16_REF: "quint16_ref",
types_pb2.DT_QINT32_REF: "qint32_ref",
types_pb2.DT_BFLOAT16_REF: "bfloat16_ref",
types_pb2.DT_RESOURCE_REF: "resource_ref",
}
# Standard mappings between types_pb2.DataType values and numpy.dtypes.
# Numpy representation for quantized dtypes.
_np_qint8 = np.dtype([("qint8", np.int8, 1)])
_np_quint8 = np.dtype([("quint8", np.uint8, 1)])
_np_qint16 = np.dtype([("qint16", np.int16, 1)])
_np_quint16 = np.dtype([("quint16", np.uint16, 1)])
_np_qint32 = np.dtype([("qint32", np.int32, 1)])types_pb2.DT_UINT8: np.uint8,
types_pb2.DT_UINT16: np.uint16,
types_pb2.DT_INT16: np.int16,
types_pb2.DT_INT8: np.int8,
# NOTE(touts): For strings we use np.object as it supports variable length
# strings.
types_pb2.DT_STRING: np.object,
types_pb2.DT_COMPLEX64: np.complex64,
types_pb2.DT_COMPLEX128: np.complex128,
types_pb2.DT_INT64: np.int64,
types_pb2.DT_BOOL: np.bool,
types_pb2.DT_QINT8: _np_qint8,
types_pb2.DT_QUINT8: _np_quint8,
types_pb2.DT_QINT16: _np_qint16,
types_pb2.DT_QUINT16: _np_quint16,
types_pb2.DT_QINT32: _np_qint32,
types_pb2.DT_BFLOAT16: np.uint16,
}
def as_dtype(type_value):
if isinstance(type_value, DType):
return type_value
if isinstance(type_value, np.dtype):
if type_value.type == np.string_ or type_value.type == np.unicode_:
return string
for key, val in _NP_TO_TF:
try:
if key == type_value:
return valtypes_pb2.DT_INT16: "int16",
types_pb2.DT_INT8: "int8",
types_pb2.DT_STRING: "string",
types_pb2.DT_COMPLEX64: "complex64",
types_pb2.DT_COMPLEX128: "complex128",
types_pb2.DT_INT64: "int64",
types_pb2.DT_BOOL: "bool",
types_pb2.DT_QINT8: "qint8",
types_pb2.DT_QUINT8: "quint8",
types_pb2.DT_QINT16: "qint16",
types_pb2.DT_QUINT16: "quint16",
types_pb2.DT_QINT32: "qint32",
types_pb2.DT_BFLOAT16: "bfloat16",
types_pb2.DT_RESOURCE: "resource",
types_pb2.DT_HALF_REF: "float16_ref",
types_pb2.DT_FLOAT_REF: "float32_ref",
types_pb2.DT_DOUBLE_REF: "float64_ref",
types_pb2.DT_INT32_REF: "int32_ref",
types_pb2.DT_UINT8_REF: "uint8_ref",
types_pb2.DT_UINT16_REF: "uint16_ref",
types_pb2.DT_INT16_REF: "int16_ref",
types_pb2.DT_INT8_REF: "int8_ref",
types_pb2.DT_STRING_REF: "string_ref",
types_pb2.DT_COMPLEX64_REF: "complex64_ref",
types_pb2.DT_COMPLEX128_REF: "complex128_ref",
types_pb2.DT_INT64_REF: "int64_ref",
types_pb2.DT_BOOL_REF: "bool_ref",
types_pb2.DT_QINT8_REF: "qint8_ref",
types_pb2.DT_QUINT8_REF: "quint8_ref",
types_pb2.DT_QINT16_REF: "qint16_ref",
types_pb2.DT_QUINT16_REF: "quint16_ref",
types_pb2.DT_QINT32_REF: "qint32_ref",types_pb2.DT_UINT16: np.uint16,
types_pb2.DT_INT16: np.int16,
types_pb2.DT_INT8: np.int8,
# NOTE(touts): For strings we use np.object as it supports variable length
# strings.
types_pb2.DT_STRING: np.object,
types_pb2.DT_COMPLEX64: np.complex64,
types_pb2.DT_COMPLEX128: np.complex128,
types_pb2.DT_INT64: np.int64,
types_pb2.DT_BOOL: np.bool,
types_pb2.DT_QINT8: _np_qint8,
types_pb2.DT_QUINT8: _np_quint8,
types_pb2.DT_QINT16: _np_qint16,
types_pb2.DT_QUINT16: _np_quint16,
types_pb2.DT_QINT32: _np_qint32,
types_pb2.DT_BFLOAT16: np.uint16,
}
def as_dtype(type_value):
if isinstance(type_value, DType):
return type_value
if isinstance(type_value, np.dtype):
if type_value.type == np.string_ or type_value.type == np.unicode_:
return string
for key, val in _NP_TO_TF:
try:
if key == type_value:
return val
except TypeError as e:float16 = DType(types_pb2.DT_HALF)
half = float16
float32 = DType(types_pb2.DT_FLOAT)
float64 = DType(types_pb2.DT_DOUBLE)
double = float64
int32 = DType(types_pb2.DT_INT32)
uint8 = DType(types_pb2.DT_UINT8)
uint16 = DType(types_pb2.DT_UINT16)
int16 = DType(types_pb2.DT_INT16)
int8 = DType(types_pb2.DT_INT8)
string = DType(types_pb2.DT_STRING)
complex64 = DType(types_pb2.DT_COMPLEX64)
complex128 = DType(types_pb2.DT_COMPLEX128)
int64 = DType(types_pb2.DT_INT64)
bool = DType(types_pb2.DT_BOOL)
qint8 = DType(types_pb2.DT_QINT8)
quint8 = DType(types_pb2.DT_QUINT8)
qint16 = DType(types_pb2.DT_QINT16)
quint16 = DType(types_pb2.DT_QUINT16)
qint32 = DType(types_pb2.DT_QINT32)
bfloat16 = DType(types_pb2.DT_BFLOAT16)
# Standard mappings between types_pb2.DataType values and string names.
_TYPE_TO_STRING = {
types_pb2.DT_HALF: "float16",
types_pb2.DT_FLOAT: "float32",
types_pb2.DT_DOUBLE: "float64",
types_pb2.DT_INT32: "int32",
types_pb2.DT_UINT8: "uint8",
types_pb2.DT_UINT16: "uint16",
types_pb2.DT_INT16: "int16",types_pb2.DT_FLOAT_REF: "float32_ref",
types_pb2.DT_DOUBLE_REF: "float64_ref",
types_pb2.DT_INT32_REF: "int32_ref",
types_pb2.DT_UINT8_REF: "uint8_ref",
types_pb2.DT_UINT16_REF: "uint16_ref",
types_pb2.DT_INT16_REF: "int16_ref",
types_pb2.DT_INT8_REF: "int8_ref",
types_pb2.DT_STRING_REF: "string_ref",
types_pb2.DT_COMPLEX64_REF: "complex64_ref",
types_pb2.DT_COMPLEX128_REF: "complex128_ref",
types_pb2.DT_INT64_REF: "int64_ref",
types_pb2.DT_BOOL_REF: "bool_ref",
types_pb2.DT_QINT8_REF: "qint8_ref",
types_pb2.DT_QUINT8_REF: "quint8_ref",
types_pb2.DT_QINT16_REF: "qint16_ref",
types_pb2.DT_QUINT16_REF: "quint16_ref",
types_pb2.DT_QINT32_REF: "qint32_ref",
types_pb2.DT_BFLOAT16_REF: "bfloat16_ref",
types_pb2.DT_RESOURCE_REF: "resource_ref",
}
# Standard mappings between types_pb2.DataType values and numpy.dtypes.
# Numpy representation for quantized dtypes.
_np_qint8 = np.dtype([("qint8", np.int8, 1)])
_np_quint8 = np.dtype([("quint8", np.uint8, 1)])
_np_qint16 = np.dtype([("qint16", np.int16, 1)])
_np_quint16 = np.dtype([("quint16", np.uint16, 1)])
_np_qint32 = np.dtype([("qint32", np.int32, 1)])
_NP_TO_TF = frozenset([
(np.float16, float16),types_pb2.DT_COMPLEX128: "complex128",
types_pb2.DT_INT64: "int64",
types_pb2.DT_BOOL: "bool",
types_pb2.DT_QINT8: "qint8",
types_pb2.DT_QUINT8: "quint8",
types_pb2.DT_QINT16: "qint16",
types_pb2.DT_QUINT16: "quint16",
types_pb2.DT_QINT32: "qint32",
types_pb2.DT_BFLOAT16: "bfloat16",
types_pb2.DT_RESOURCE: "resource",
types_pb2.DT_HALF_REF: "float16_ref",
types_pb2.DT_FLOAT_REF: "float32_ref",
types_pb2.DT_DOUBLE_REF: "float64_ref",
types_pb2.DT_INT32_REF: "int32_ref",
types_pb2.DT_UINT8_REF: "uint8_ref",
types_pb2.DT_UINT16_REF: "uint16_ref",
types_pb2.DT_INT16_REF: "int16_ref",
types_pb2.DT_INT8_REF: "int8_ref",
types_pb2.DT_STRING_REF: "string_ref",
types_pb2.DT_COMPLEX64_REF: "complex64_ref",
types_pb2.DT_COMPLEX128_REF: "complex128_ref",
types_pb2.DT_INT64_REF: "int64_ref",
types_pb2.DT_BOOL_REF: "bool_ref",
types_pb2.DT_QINT8_REF: "qint8_ref",
types_pb2.DT_QUINT8_REF: "quint8_ref",
types_pb2.DT_QINT16_REF: "qint16_ref",
types_pb2.DT_QUINT16_REF: "quint16_ref",
types_pb2.DT_QINT32_REF: "qint32_ref",
types_pb2.DT_BFLOAT16_REF: "bfloat16_ref",
types_pb2.DT_RESOURCE_REF: "resource_ref",
}dragon
A powerful toolkit targeting Apple development, research, and packaging.
Package Health Score
50 / 100Popular dragon functions
- dragon.core.logging.info
- dragon.core.tensor.Tensor
- dragon.core.tensor.Tensor.CreateOperator
- dragon.core.tensor.Tensor.Ref
- dragon.core.workspace
- dragon.core.workspace.FeedTensor
- dragon.ops
- dragon.transpiler.lark.transformers.haxe_transformer.HaxeTransformer
- dragon.vm.caffe.layers.Convolution
- dragon.vm.caffe.layers.ReLU
- dragon.vm.tensorflow.framework.ops
- dragon.vm.tensorflow.protobuf.types_pb2
- dragon.vm.torch.nn.Module
- dragon.vm.torch.nn.modules.utils._pair
- dragon.vm.torch.ops.factory.get_module
- dragon.vm.torch.ops.modules.base.BaseModule
- dragon.vm.torch.ops.primitive.MakeContext
- dragon.vm.torch.ops.primitive.MakeDevice
- dragon.vm.torch.ops.primitive.WrapScalar
- dragon.vm.torch.tensor.Tensor