Skip to content

App op (atleast_1d and atleast_2d and atleast_3d) | feat (torchlib) #767

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 6 commits into from
Jun 6, 2023
Merged

App op (atleast_1d and atleast_2d and atleast_3d) | feat (torchlib) #767

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
4114272
Add op(atleast_1d and atleast_2d and atleast_3d) | feat(torchlib)
titaiwangms Jun 5, 2023
d21c79e
add constant node for requirement
titaiwangms Jun 5, 2023
873957f
USe torch logic
titaiwangms Jun 6, 2023
7efaf27
Use SequemceMap
titaiwangms Jun 6, 2023
a916464
Merge branch 'main' into titaiwang/add_hstack_vstack
titaiwangms Jun 6, 2023
2be4465
refactor core and add songle tensor support
titaiwangms Jun 6, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
78 changes: 71 additions & 7 deletions onnxscript/function_libs/torch_lib/ops/core.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -14,7 +14,7 @@
import math
from typing import Any, Optional, Sequence, Tuple, Union

from onnxscript import BOOL, DOUBLE, FLOAT, INT8, INT16, INT32, INT64
from onnxscript import BOOL, DOUBLE, FLOAT, INT8, INT16, INT32, INT64, graph
from onnxscript.function_libs.torch_lib.registration import torch_op
from onnxscript.function_libs.torch_lib.tensor_typing import (
IntType,
Expand Down Expand Up @@ -666,22 +666,86 @@ def aten_atanh(self: TFloat) -> TFloat:
return op.Atanh(self)


def aten_atleast_1d(self: TensorType) -> TensorType:
@torch_op("aten::atleast_1d")
def aten_atleast_1d(self: Sequence[TTensor]) -> TTensor:
"""atleast_1d(Tensor self) -> Tensor"""

raise NotImplementedError()
@graph()
def reshape_to_1d(tensor):
shape = op.Shape(tensor)
rank = op.Size(shape)
if rank == 0:
tensor = op.Reshape(tensor, op.Constant(value_ints=[1]))
return tensor

return op.SequenceMap(self, body=reshape_to_1d)


@torch_op("aten::atleast_1d")
def aten_atleast_1d_single_tensor(self: TTensor) -> TTensor:
"""atleast_1d(Tensor self) -> Tensor"""

shape = op.Shape(self)
rank = op.Size(shape)
if rank == 0:
self = op.Reshape(self, op.Constant(value_ints=[1]))
return self


def aten_atleast_2d(self: TensorType) -> TensorType:
@torch_op("aten::atleast_2d")
def aten_atleast_2d(self: Sequence[TTensor]) -> TTensor:
"""atleast_2d(Tensor self) -> Tensor"""

raise NotImplementedError()
@graph()
def reshape_to_2d(tensor):
shape = op.Shape(tensor)
rank = op.Size(shape)
if rank <= 1:
tensor = op.Reshape(tensor, op.Constant(value_ints=[1, -1]))
return tensor

return op.SequenceMap(self, body=reshape_to_2d)


@torch_op("aten::atleast_2d")
def aten_atleast_2d_single_tensor(self: TTensor) -> TTensor:
"""atleast_2d(Tensor self) -> Tensor"""

shape = op.Shape(self)
rank = op.Size(shape)
if rank <= 1:
self = op.Reshape(self, op.Constant(value_ints=[1, -1]))
return self


def aten_atleast_3d(self: TensorType) -> TensorType:
@torch_op("aten::atleast_3d")
def aten_atleast_3d(self: Sequence[TTensor]) -> TTensor:
"""atleast_3d(Tensor self) -> Tensor"""

raise NotImplementedError()
@graph()
def reshape_to_3d(tensor):
shape = op.Shape(tensor)
rank = op.Size(shape)
if rank <= 1:
tensor = op.Reshape(tensor, op.Constant(value_ints=[1, -1, 1]))
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

You may want to double-check the downstream usage to make sure whether you want to reshape a tensor of size N to 1x1xN or Nx1x1 ... the above reshapes it to 1xNx1 .... which may be okay, but just wondering.

elif rank == 2:
tensor = op.Unsqueeze(tensor, op.Constant(value_ints=[-1]))
return tensor

return op.SequenceMap(self, body=reshape_to_3d)


@torch_op("aten::atleast_3d")
def aten_atleast_3d_single_tensor(self: TTensor) -> TTensor:
"""atleast_3d(Tensor self) -> Tensor"""

shape = op.Shape(self)
rank = op.Size(shape)
if rank <= 1:
self = op.Reshape(self, op.Constant(value_ints=[1, -1, 1]))
elif rank == 2:
self = op.Unsqueeze(self, op.Constant(value_ints=[-1]))
return self


def aten_avg_pool1d(
Expand Down
50 changes: 50 additions & 0 deletions onnxscript/tests/function_libs/torch_lib/ops_test_data.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -344,6 +344,12 @@ def _where_input_wrangler(
"atan": core_ops.aten_atan,
"atan2": core_ops.aten_atan2,
"atanh": core_ops.aten_atanh,
"atleast_1d": core_ops.aten_atleast_1d,
"atleast_1d_single_tensor": core_ops.aten_atleast_1d_single_tensor,
"atleast_2d": core_ops.aten_atleast_2d,
"atleast_2d_single_tensor": core_ops.aten_atleast_2d_single_tensor,
"atleast_3d": core_ops.aten_atleast_3d,
"atleast_3d_single_tensor": core_ops.aten_atleast_3d_single_tensor,
"baddbmm": core_ops.aten_baddbmm,
"bmm": core_ops.aten_bmm,
"broadcast_to": core_ops.aten_broadcast_to,
Expand Down Expand Up @@ -808,6 +814,21 @@ def _where_input_wrangler(
matcher=lambda sample: len(sample.args) != 2,
reason="arange_start_step overload takes three arguments (input, start, step)",
),
skip(
"atleast_1d_single_tensor",
matcher=lambda sample: isinstance(sample.input, (list, tuple)),
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

So we don’t take a Sequence as input? Where is this op used? How are the inputs produced?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

It accepts both tensor and list.

Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Oh I see there are two variants

reason="atleast_1d_single_tensor overload takes single tensor as input",
),
skip(
"atleast_2d_single_tensor",
matcher=lambda sample: isinstance(sample.input, (list, tuple)),
reason="atleast_2d_single_tensor overload takes single tensor as input",
),
skip(
"atleast_3d_single_tensor",
matcher=lambda sample: isinstance(sample.input, (list, tuple)),
reason="atleast_3d_single_tensor overload takes single tensor as input",
),
skip(
"cat",
matcher=lambda sample: sample.input[0].equal(torch.tensor([])),
Expand Down Expand Up @@ -1166,6 +1187,11 @@ def _where_input_wrangler(
),
)

ops_test_common.duplicate_opinfo(OPS_DB, "atleast_1d", ("atleast_1d_single_tensor",))
ops_test_common.duplicate_opinfo(OPS_DB, "atleast_2d", ("atleast_2d_single_tensor",))
ops_test_common.duplicate_opinfo(OPS_DB, "atleast_3d", ("atleast_3d_single_tensor",))


ops_test_common.duplicate_opinfo(OPS_DB, "full_like", ("full_like_dtype",))

ops_test_common.duplicate_opinfo(OPS_DB, "index_put", ("index_put_bool",))
Expand Down Expand Up @@ -1480,6 +1506,30 @@ def _where_input_wrangler(
torch.float32,
torch.float16,
),
"atleast_1d": (
torch.float32,
torch.float16,
),
"atleast_1d_single_tensor": (
torch.float32,
torch.float16,
),
"atleast_2d": (
torch.float32,
torch.float16,
),
"atleast_2d_single_tensor": (
torch.float32,
torch.float16,
),
"atleast_3d": (
torch.float32,
torch.float16,
),
"atleast_3d_single_tensor": (
torch.float32,
torch.float16,
),
"baddbmm": (
torch.float32,
torch.float16,
Expand Down