blob: 4cc3f47a354164d3a2250ffa5bcbef0b230265ae [file] [log] [blame]
import torch
import warnings
from sys import maxsize as maxsize
from typing import Set
import torch.onnx
# This import monkey-patches graph manipulation methods on Graph, used for the
# ONNX symbolics
import torch.onnx.utils
from functools import wraps
# Note [Edit Symbolic Files]
# EDITING THIS FILE AND SYMBOLIC_OPSET<VERSION> FILES? READ THIS FIRST!
#
# - These files is ONLY for ATen operators (e.g., operators that show up in the
# trace as aten::blah). If you need to special case a primitive operator,
# look at _run_symbolic_function
# - Parameter ordering does NOT necessarily match what is in VariableType.cpp;
# tensors are always first, then non-tensor arguments.
# - Parameter names must *exactly* match the names in VariableType.cpp, because
# dispatch is done with keyword arguments.
# - Looking for inplace ops? They're detected by the trailing underscore, and
# transparently dispatched to their non inplace versions in
# 'run_symbolic_function'. See Note [Export inplace]
#
# ----------------------------------------------------------------------------------
# A note on Tensor types
# ----------------------------------------------------------------------------------
#
# In general, we should avoid depending on the type of Tensor Values contained
# within the trace graph. However, this is sometimes unavoidable (due to ONNX
# spec requirements, etc). The TensorType object has accessors for these properties
# that return the property if it is statically known and return nullopt otherwise.
#
# In general, we should prefer to rely on the least specific information possible.
# For example, not relying on tensor properties at all is better than relying
# on the number of dimensions which is better than relying on
# concrete shapes. Doing so will make the export symbolics
# more robust to different graphs.
# ---------------------------------------------------------------------------------
# Helper functions
# ---------------------------------------------------------------------------------
# Save some builtins as locals, because we'll shadow them below
_sum = sum
def _parse_arg(value, desc):
if desc == 'none':
return value
if desc == 'v' or not _is_value(value):
return value
if value.node().mustBeNone():
return None
if value.node().kind() == 'onnx::Constant':
tval = value.node()['value']
if desc == 'i':
return int(tval)
elif desc == 'f':
return float(tval)
elif desc == 'b':
return bool(tval)
elif desc == 's':
return str(tval)
elif desc == 't':
return tval
elif desc == 'is':
return [int(v) for v in tval]
elif desc == 'fs':
return [float(v) for v in tval]
else:
raise RuntimeError("ONNX symbolic doesn't know to interpret Constant node")
elif value.node().kind() == 'prim::ListConstruct':
if desc == 'is':
for v in value.node().inputs():
if v.node().kind() != 'onnx::Constant':
raise RuntimeError("Failed to export an ONNX attribute '" + v.node().kind() +
"', since it's not constant, please try to make "
"things (e.g., kernel size) static if possible")
return [int(v.node()['value']) for v in value.node().inputs()]
else:
raise RuntimeError("ONNX symbolic doesn't know to interpret ListConstruct node")
raise RuntimeError("Unexpected node type: {}".format(value.node().kind()))
def _maybe_get_const(value, desc):
if _is_value(value) and value.node().kind() == 'onnx::Constant':
return _parse_arg(value, desc)
return value
def _maybe_get_scalar(value):
value_t = _maybe_get_const(value, 't')
if isinstance(value_t, torch.Tensor) and value_t.shape == ():
return value_t
return value
def _get_const(value, desc, arg_name):
if _is_value(value) and value.node().kind() not in ('onnx::Constant', 'prim::Constant'):
raise RuntimeError("ONNX symbolic expected a constant value of the {} argument, got `{}`".format(arg_name, value))
return _parse_arg(value, desc)
def _unpack_list(list_value):
list_node = list_value.node()
assert list_node.kind() == "prim::ListConstruct"
return list(list_node.inputs())
# Check if list_value is output from prim::ListConstruct
# This is usually called before _unpack_list to ensure the list can be unpacked.
def _is_packed_list(list_value):
return _is_value(list_value) and list_value.node().kind() == "prim::ListConstruct"
def parse_args(*arg_descriptors):
def decorator(fn):
fn._arg_descriptors = arg_descriptors
def wrapper(g, *args, **kwargs):
# some args may be optional, so the length may be smaller
assert len(arg_descriptors) >= len(args)
args = [_parse_arg(arg, arg_desc) for arg, arg_desc in zip(args, arg_descriptors)] # type: ignore
# only support _outputs in kwargs
assert len(kwargs) <= 1
if len(kwargs) == 1:
assert '_outputs' in kwargs
return fn(g, *args, **kwargs)
# In Python 2 functools.wraps chokes on partially applied functions, so we need this as a workaround
try:
wrapper = wraps(fn)(wrapper)
except Exception:
pass
return wrapper
return decorator
def _scalar(x):
"""Convert a scalar tensor into a Python value."""
assert x.numel() == 1
return x.item()
def _if_scalar_type_as(g, self, tensor):
"""
Convert self into the same type of tensor, as necessary.
We only support implicit casting for scalars, so we never
actually need to insert an ONNX cast operator here; just
fix up the scalar.
"""
if isinstance(self, torch._C.Value):
return self
scalar_type = tensor.type().scalarType()
if scalar_type:
ty = scalar_type.lower()
return getattr(self, ty)()
return self
def _is_none(x):
return x.node().mustBeNone()
def _is_value(x):
return isinstance(x, torch._C.Value)
def _is_tensor(x):
return x.type().isSubtypeOf(torch._C.TensorType.get())
def _is_tensor_list(x):
return isinstance(x.type(), torch._C.ListType) and isinstance(x.type().getElementType(), torch._C.TensorType)
def _get_tensor_rank(x):
if not _is_tensor(x) or x.type() is None:
return None
return x.type().dim()
def _get_tensor_sizes(x, allow_nonstatic=True):
if not _is_tensor(x) or x.type() is None:
return None
if allow_nonstatic:
# Each individual symbol is returned as None.
# e.g. [1, 'a', 'b'] -> [1, None, None]
return x.type().varyingSizes()
# returns None, if exists any symbol in sizes.
# e.g. [1, 'a', 'b'] -> None
return x.type().sizes()
def _get_tensor_dim_size(x, dim):
try:
sizes = _get_tensor_sizes(x)
return sizes[dim]
except Exception:
pass
return None
def _unimplemented(op, msg):
warnings.warn("ONNX export failed on " + op + " because " + msg + " not supported")
def _onnx_unsupported(op_name):
raise RuntimeError('Unsupported: ONNX export of operator {}. '
'Please open a bug to request ONNX export support for the missing operator.'.format(op_name))
def _onnx_opset_unsupported(op_name, current_opset, supported_opset):
raise RuntimeError('Unsupported: ONNX export of {} in '
'opset {}. Please try opset version {}.'.format(op_name, current_opset, supported_opset))
def _onnx_opset_unsupported_detailed(op_name, current_opset, supported_opset, reason):
raise RuntimeError('Unsupported: ONNX export of {} in '
'opset {}. {}. Please try opset version {}.'.format(op_name, current_opset, reason, supported_opset))
def _block_list_in_opset(name):
def symbolic_fn(*args, **kwargs):
raise RuntimeError("ONNX export failed on {}, which is not implemented for opset {}. "
"Try exporting with other opset versions."
.format(name, _export_onnx_opset_version))
return symbolic_fn
def _try_get_scalar_type(*args):
for arg in args:
try:
return arg.type().scalarType()
except RuntimeError:
pass
return None
def _select_helper(g, self, dim, index, apply_reshape=True):
index_const = _maybe_get_scalar(index)
index_dim = _get_tensor_rank(index)
if not _is_value(index_const):
# Index is a constant scalar. Make it a size 1 constant tensor.
index = g.op("Constant", value_t=torch.LongTensor([index_const]))
elif index_dim is not None and apply_reshape:
if index_dim == 0:
# Index is a scalar. Reshape it to a size 1 tensor.
index = g.op("Reshape", index, g.op("Constant", value_t=torch.LongTensor([1])))
index_scalar_type = index.type().scalarType()
if index_scalar_type is None or index_scalar_type not in ['Long', 'Int']:
index = g.op("Cast", index, to_i=cast_pytorch_to_onnx["Long"])
return g.op("Gather", self, index, axis_i=dim)
def _slice_helper(g, input, axes, starts, ends, steps=None, dynamic_slice=False):
if _export_onnx_opset_version <= 9:
from torch.onnx.symbolic_opset9 import _slice as _slice9
return _slice9(g, input, axes, starts, ends)
else:
from torch.onnx.symbolic_opset10 import _slice as _slice10
return _slice10(g, input, axes, starts, ends, steps, dynamic_slice)
def _hardtanh_helper(g, input, min_val, max_val):
if _export_onnx_opset_version <= 10:
from torch.onnx.symbolic_opset9 import hardtanh
return hardtanh(g, input, min_val, max_val)
else:
from torch.onnx.symbolic_opset11 import hardtanh # type: ignore[no-redef]
return hardtanh(g, input, min_val, max_val)
def _is_fp(value):
if value:
if isinstance(value, torch.Tensor):
type = value.dtype
return (type == 'torch.float32') or (type == 'torch.float64') or (type == 'torch.float16')
else:
type = value.type().scalarType()
if type is None:
warnings.warn("Type cannot be inferred, which might cause exported graph to produce incorrect results.")
return (type == 'Float') or (type == 'Double') or (type == 'Half')
return False
def _sort_helper(g, input, dim, decending=True, out=None):
if out is not None:
_unimplemented("Sort", "Out parameter is not supported")
shape_ = g.op("Shape", input)
dim_size_ = g.op("Gather", shape_, g.op("Constant", value_t=torch.tensor([dim], dtype=torch.int64)))
if _export_onnx_opset_version <= 10:
if not decending:
_unimplemented("Sort", "Ascending is not supported")
return g.op("TopK", input, dim_size_, axis_i=dim, outputs=2)
else:
return g.op("TopK", input, dim_size_, axis_i=dim, largest_i=decending, outputs=2)
def _topk_helper(g, input, k, dim, largest=True, sorted=False, out=None):
if out is not None:
_unimplemented("TopK", "Out parameter is not supported")
if not _is_value(k):
k = g.op("Constant", value_t=torch.tensor([k], dtype=torch.int64))
else:
k = g.op("Reshape", k, g.op("Constant", value_t=torch.tensor([1])))
if _export_onnx_opset_version <= 10:
if not largest:
_unimplemented("TopK", "Ascending is not supported")
return g.op("TopK", input, k, axis_i=dim, outputs=2)
else:
return g.op("TopK", input, k, axis_i=dim, largest_i=largest, sorted_i=sorted, outputs=2)
def _interpolate_warning(interpolate_mode):
onnx_op = "onnx:Resize" if _export_onnx_opset_version >= 10 else "onnx:Upsample"
warnings.warn("You are trying to export the model with " + onnx_op + " for ONNX opset version "
"" + str(_export_onnx_opset_version) + ". "
"This operator might cause results to not match the expected results by PyTorch.\n"
"ONNX's Upsample/Resize operator did not match Pytorch's Interpolation until opset 11. "
"Attributes to determine how to transform the input were added in onnx:Resize in opset 11 "
"to support Pytorch's behavior (like coordinate_transformation_mode and nearest_mode).\n"
"We recommend using opset 11 and above for models using this operator. ")
def _unsqueeze_helper(g, input, dim):
from torch.onnx.symbolic_opset9 import unsqueeze
return unsqueeze(g, input, dim)
def _interpolate_size_to_scales(g, input, output_size, dim):
output_size = _maybe_get_const(output_size, 'is')
if _is_value(output_size):
offset = 2
offsets = g.op("Constant", value_t=torch.ones(offset, dtype=torch.float32))
dividend = g.op("Cast", output_size, to_i=cast_pytorch_to_onnx["Float"])
divisor = _slice_helper(g, g.op("Shape", input), axes=[0], ends=[maxsize], starts=[offset])
divisor = g.op("Cast", divisor, to_i=cast_pytorch_to_onnx["Float"])
scale_dims = g.op("Div", dividend, divisor)
scales = g.op("Concat", offsets, scale_dims, axis_i=0)
else:
scales_constant = [1. if i < 2 else
float(output_size[-(dim - i)]) / float(input.type().sizes()[-(dim - i)])
for i in range(0, dim)]
scales = g.op("Constant", value_t=torch.tensor(scales_constant, dtype=torch.float32))
return scales
def _interpolate_get_scales_if_available(g, scales):
available_scales = _maybe_get_const(scales[0], 'fs') != -1 and not _is_none(scales[0])
if not available_scales:
return None
offsets = g.op("Constant", value_t=torch.ones(2, dtype=torch.float32))
scales_list = g.op("Constant", value_t=torch.tensor(_maybe_get_const(scales[0], 'fs')))
scales = g.op("Concat", offsets, scales_list, axis_i=0)
return scales
def _get_interpolate_attributes(g, mode, args):
if mode == 'nearest':
align_corners = None
scales = args[0:]
else:
align_corners = args[0]
scales = args[1:]
scales = _interpolate_get_scales_if_available(g, scales)
return scales, align_corners
def _interpolate_get_scales(g, scale_factor, dim):
offsets = g.op("Constant", value_t=torch.ones(2, dtype=torch.float32))
scale_factor_rank = _get_tensor_rank(scale_factor)
if isinstance(scale_factor.type(), torch._C.ListType) or (scale_factor_rank is not None and scale_factor_rank > 0):
return g.op("Concat", offsets, scale_factor, axis_i=0)
else:
scale_factor = _unsqueeze_helper(g, scale_factor, 0)
scale_factor = g.op("Cast", scale_factor, to_i=cast_pytorch_to_onnx["Float"])
scales = [scale_factor for i in range(dim - 2)]
scale_factor = g.op("Concat", offsets, *scales, axis_i=0)
return scale_factor
def _interpolate_get_scales_and_mode(g, input, size, scale_factor, mode , align_corners):
mode = _maybe_get_const(mode, 's')
if 'linear' in mode:
mode = 'linear'
if 'cubic' in mode:
mode = 'cubic'
_interpolate_warning(mode)
align_corners = _maybe_get_const(align_corners, 'b')
if isinstance(align_corners, bool) and align_corners:
return _unimplemented("interpolate", "align_corners == True")
if not input.type().dim():
return _unimplemented("interpolate", "missing input shape")
dim = input.type().dim()
if not _is_none(scale_factor):
scale_factor = _interpolate_get_scales(g, scale_factor, dim)
elif not _is_none(size):
if not _is_packed_list(size):
is_scalar = ((_maybe_get_const(size, 't').dim() == 0))
if is_scalar:
size = _unsqueeze_helper(g, size, 0)
size = [size for i in range(dim - 2)]
size = g.op("Concat", *size, axis_i=0)
scale_factor = _interpolate_size_to_scales(g, input, size, dim)
else:
return _unimplemented("interpolate", "Both size and scales are None in __interpolate")
return scale_factor, mode
def _unbind_helper(g, self, dim, _outputs):
if _export_onnx_opset_version <= 9:
from torch.onnx.symbolic_opset9 import unbind
else:
from torch.onnx.symbolic_opset11 import unbind # type: ignore[no-redef]
return unbind(g, self, dim, _outputs)
def _scatter_helper(g, self, dim, index, src):
if _export_onnx_opset_version <= 10:
from torch.onnx.symbolic_opset9 import scatter
else:
# for mypy, scatter was imported two lines above
from torch.onnx.symbolic_opset11 import scatter # type: ignore
return scatter(g, self, dim, index, src)
def _arange_cast_helper(g, end, start=None, step=None, dtype=None):
def _is_all_integral(scalars):
for scalar in scalars:
try:
if scalar.type().scalarType() != 'Long':
return False
except Exception:
pass
return True
# This logic is based on torch.arange docs. If 'dtype' is provided,
# infer input types from dtype. If not, then check if any of start, stop,
# or step are floating point, and infer the type from get_default.
# Otherwise, the dtype is inferred to be torch.int64.
if dtype is None or (_is_value(dtype) and _is_none(dtype)):
if _is_all_integral([start, end, step]):
type = scalar_type_to_pytorch_type.index(torch.int64)
else:
type = scalar_type_to_pytorch_type.index(torch.get_default_dtype())
else:
type = dtype
start = g.op("Cast", start, to_i=scalar_type_to_onnx[type]) if start else None
end = g.op("Cast", end, to_i=scalar_type_to_onnx[type]) if end else None
step = g.op("Cast", step, to_i=scalar_type_to_onnx[type]) if step else None
return type, end, start, step
def _size_helper(g, self, dim):
full_shape = g.op("Shape", self)
from torch.onnx.symbolic_opset9 import select
return select(g, full_shape, g.op("Constant", value_t=torch.tensor([0])), dim)
def _index_fill_reshape_helper(g, self, dim, index):
# 1. reshape index => [1, ..., 1, dim, 1, ..., 1]
# 2. expand index => [..., dim, ...], same shape as self except for dim.
# 3. expand value as well.
# 4. apply onnx::scatter.
from torch.onnx.symbolic_opset9 import expand
if _export_onnx_opset_version <= 10:
from torch.onnx.symbolic_opset9 import scatter
else:
# for mypy, scatter was imported two lines above
from torch.onnx.symbolic_opset11 import scatter # type: ignore
if self.type().dim() is None:
return _unimplemented("index_fill", "input rank not accesible")
self_dim = self.type().dim()
dim_value = _parse_arg(dim, 'i')
unsqueezed_index = g.op("Unsqueeze", index, axes_i=[i for i in range(self_dim) if i != dim_value])
expanded_index_shape = scatter(g, g.op("Shape", self), 0,
g.op("Unsqueeze", dim, axes_i=[0]), g.op("Shape", index))
expanded_index = expand(g, unsqueezed_index, expanded_index_shape, None)
return expanded_index_shape, expanded_index
def _avgpool_helper(tuple_fn, padding, kernel_size, stride, divisor_override, name):
if divisor_override and divisor_override.node().kind() != 'prim::Constant':
return _unimplemented(name, "divisor_override")
if not stride:
stride = kernel_size
padding = tuple(tuple_fn(padding))
return padding
def assert_training_mode(op_mode, op_name):
global _training_mode
op_mode = True if op_mode == 1 else False
if op_mode != _training_mode:
op_mode = "training " if op_mode else "inference"
training_mode = "training " if _training_mode else "inference"
# setting the model mode could result in op_mode != _training_mode
# if the model is a FuncModule. In this case we warn the user of
# the state and export depending on training_mode
warnings.warn("ONNX export mode is set to " + training_mode +
" mode, but operator " + op_name + " is set to " +
op_mode + " mode. The model will be exported in " +
training_mode + ", as specified by the export mode.")
def _flatten_helper(g, input, start_dim, end_dim, dim):
input_size = g.op("Shape", input)
slice1 = _slice_helper(g, input_size, axes=[0], starts=[0], ends=[start_dim])
slices = [slice1, g.op("Constant", value_t=torch.tensor([-1], dtype=torch.long))]
if end_dim < dim - 1:
slice3 = _slice_helper(g, input_size, axes=[0], starts=[end_dim + 1], ends=[dim])
slices = [slice1, g.op("Constant", value_t=torch.tensor([-1], dtype=torch.long)), slice3]
final_shape = g.op("Concat", *slices, axis_i=0)
from torch.onnx.symbolic_opset9 import _reshape_from_tensor
return _reshape_from_tensor(g, input, final_shape)
def _is_split_static(split_size_or_sizes, _outputs):
if _outputs is None:
return False
if _is_value(split_size_or_sizes) and split_size_or_sizes.node().kind() != 'onnx::Constant':
return False
return True
# ---------------------------------------------------------------------
# ONNX operator version
# ---------------------------------------------------------------------
# READ ME BEFORE EDITING _default_onnx_opset_version:
#
# The variable below controls which ONNX operator set version we are
# targeting. THIS VARIABLE HAS SEMANTIC EFFECT! Say a breaking
# change occurred in version 8. As long as this variable < 8, you can
# export models targeting the old behavior. However, if you bump
# this variable to 8 or later, the breaking change will take into effect:
# you MUST adjust any symbolic affected by breaking changes. The ONNX
# spec publishes a *comprehensive* list of BC-breaking changes for every
# operator revision at:
#
# https://github.com/onnx/onnx/blob/master/docs/Changelog.md
#
# Please be sure to go through and check all of our implementations here before
# increasing this number. This includes symbolic definitions NOT in this
# file, so grep for "OpName" (with quotes)
#
# Besides, opset_version can be specified in the invocation of export()
# and export_to_pretty_string(), and _export_onnx_opset_version will be set
# and the symbolic functions should check it to determine the behavior
# of the exporter.
_default_onnx_opset_version = 9
_onnx_master_opset = 10
_onnx_stable_opsets = [7, 8, 9, 10, 11, 12]
_export_onnx_opset_version = _default_onnx_opset_version
def _set_opset_version(opset_version):
global _export_onnx_opset_version
if opset_version == _default_onnx_opset_version:
_export_onnx_opset_version = opset_version
return
if opset_version in _onnx_stable_opsets + [_onnx_master_opset]:
_export_onnx_opset_version = opset_version
return
raise ValueError("Unsupported ONNX opset version: " + str(opset_version))
_operator_export_type = None
def _set_operator_export_type(operator_export_type):
global _operator_export_type
_operator_export_type = operator_export_type
_training_mode = None
def _set_training_mode(training_mode):
global _training_mode
_training_mode = training_mode
_onnx_shape_inference = False
def _set_onnx_shape_inference(onnx_shape_inference):
global _onnx_shape_inference
_onnx_shape_inference = onnx_shape_inference
# Metaprogram symbolics for each ATen native specialized cast operator.
# For e.g. we specify a function named `_cast_uint8_t` that instantiates an
# ONNX cast node with `to` attribute 'UINT8'
#
# TODO: remove these once we support Type's in the JIT IR and we can once again
# use the unified toType operator
cast_pytorch_to_onnx = {
'Byte': torch.onnx.TensorProtoDataType.UINT8,
'Char': torch.onnx.TensorProtoDataType.INT8,
'Double': torch.onnx.TensorProtoDataType.DOUBLE,
'Float': torch.onnx.TensorProtoDataType.FLOAT,
'Half': torch.onnx.TensorProtoDataType.FLOAT16,
'Int': torch.onnx.TensorProtoDataType.INT32,
'Long': torch.onnx.TensorProtoDataType.INT64,
'Short': torch.onnx.TensorProtoDataType.INT16,
'Bool': torch.onnx.TensorProtoDataType.BOOL,
'ComplexFloat': torch.onnx.TensorProtoDataType.COMPLEX64,
'ComplexDouble': torch.onnx.TensorProtoDataType.COMPLEX128,
'Undefined': torch.onnx.TensorProtoDataType.UNDEFINED,
}
scalar_name_to_pytorch = {
'uint8_t': 'Byte',
'int8_t': 'Char',
'double': 'Double',
'float': 'Float',
'half': 'Half',
'int': 'Int',
'int64_t': 'Long',
'int16_t': 'Short',
'bool': 'Bool',
'complex64': 'ComplexFloat',
'complex128': 'ComplexDouble'
}
# This indicates each scalar type's corresponding
# torch type. Related source:
# https://github.com/pytorch/pytorch/blob/da7468853ae322252270bbb58032668bd21b7457/c10/core/ScalarType.h
scalar_type_to_pytorch_type = [
torch.uint8, # 0
torch.int8, # 1
torch.short, # 2
torch.int, # 3
torch.int64, # 4
torch.half, # 5
torch.float, # 6
torch.double, # 7
torch.complex32, # 8
torch.complex64, # 9
torch.complex128, # 10
torch.bool, # 11
]
def _cast_func_template(to_i, g, input, non_blocking):
return g.op("Cast", input, to_i=to_i)
scalar_type_to_onnx = [
cast_pytorch_to_onnx["Byte"],
cast_pytorch_to_onnx["Char"],
cast_pytorch_to_onnx["Short"],
cast_pytorch_to_onnx["Int"],
cast_pytorch_to_onnx["Long"],
cast_pytorch_to_onnx["Half"],
cast_pytorch_to_onnx["Float"],
cast_pytorch_to_onnx["Double"],
cast_pytorch_to_onnx["Undefined"],
cast_pytorch_to_onnx["ComplexFloat"],
cast_pytorch_to_onnx["ComplexDouble"],
cast_pytorch_to_onnx["Bool"],
]
# Global set to store the list of quantized operators in the network.
# This is currently only used in the conversion of quantized ops from PT -> C2 via ONNX.
_quantized_ops: Set[int] = set()