"erf",
"sign",
"isnan",
"gather",
"arange",
"masked_fill",
"index_fill",
"index_copy",
"repeat_interleave",
"isnan",
"any",
"all",
]
for block_listed_op in block_listed_operators:
vars()[block_listed_op] = symbolic_helper._block_list_in_opset(block_listed_op)
vars()[block_listed_op].__module__ = "torch.onnx.symbolic_opset8"
def _interpolate(name, dim, interpolate_mode):
def symbolic_fn(g, input, output_size, *args):
scales, align_corners = symbolic_helper._get_interpolate_attributes(
g, interpolate_mode, args
)
symbolic_helper._interpolate_warning(interpolate_mode)
align_corners = symbolic_helper._maybe_get_scalar(align_corners)
if align_corners:
return symbolic_helper._unimplemented(name, "align_corners == True")
output_size = symbolic_helper._maybe_get_const(output_size, "is")
if symbolic_helper._is_value(output_size):
return symbolic_helper._unimplemented(
# MeanVarianceNormalization
#
# Updated operators:
# BatchNormalization: removed spatial attribute.
# Greater, Less, Constant, MatMul, PRelu, Gemm, Flatten: more data types{integers} supported.
# Cast: more data types{string} supported.
# Upsample: moved scales from attribute to input.
# Scan
block_listed_operators = [
"nonzero", "where", "scatter", "scatter_add", "erf", "sign", "isnan",
"gather", "arange", "masked_fill", "index_fill", "index_copy"
]
for block_listed_op in block_listed_operators:
vars()[block_listed_op] = _block_list_in_opset(block_listed_op)
def _interpolate(name, dim, interpolate_mode):
def symbolic_fn(g, input, output_size, *args):
scales, align_corners = sym_help._get_interpolate_attributes(
g, interpolate_mode, args)
sym_help._interpolate_warning(interpolate_mode)
align_corners = sym_help._maybe_get_scalar(align_corners)
if align_corners:
return _unimplemented(name, "align_corners == True")
output_size = sym_help._maybe_get_const(output_size, 'is')
if sym_help._is_value(output_size):
return _unimplemented(name,
"torch._C.Value (output_size) indexing")
if scales is None:
