onnx_diagnostic.reference.torch_ops.shape_ops

class onnx_diagnostic.reference.torch_ops.shape_ops.ConstantOfShape_9(node: NodeProto, version: int | None = None, device: device | None = None, verbose: int = 0)

ConstantOfShape

classmethod device_dependent() → bool

Returns True if the kernel needs a device to be efficiently initialized.

run(shape: OpRunTensor) → OpRunTensor

Kernel implementation.

class onnx_diagnostic.reference.torch_ops.shape_ops.Expand_8(node: NodeProto, version: int | None = None, verbose: int = 0, custom_kernels: Dict[Tuple[str, str], type] | None = None)

Expand

run(data: OpRunTensor, shape: OpRunTensor) → OpRunTensor

Kernel implementation.

class onnx_diagnostic.reference.torch_ops.shape_ops.Reshape_14(node: NodeProto, version: int | None = None, verbose: int = 0)

Reshape

run(data: OpRunTensor, shape: OpRunTensor) → OpRunTensor

Kernel implementation.

class onnx_diagnostic.reference.torch_ops.shape_ops.Shape_15(node: NodeProto, version: int | None = None, verbose: int = 0)

run(data: OpRunTensor) → OpRunTensor

Kernel implementation.

class onnx_diagnostic.reference.torch_ops.shape_ops.Split_18(node: NodeProto, version: int | None = None, verbose: int = 0)

run(data: OpRunTensor, split: OpRunTensor | None = None) → Tuple[OpRunTensor, ...]

Kernel implementation.

class onnx_diagnostic.reference.torch_ops.shape_ops.Squeeze_13(node: NodeProto, version: int | None = None, verbose: int = 0, custom_kernels: Dict[Tuple[str, str], type] | None = None)

Squeeze

run(data: OpRunTensor, axes: OpRunTensor | None = None) → OpRunTensor

Kernel implementation.

class onnx_diagnostic.reference.torch_ops.shape_ops.Unsqueeze_13(node: NodeProto, version: int | None = None, verbose: int = 0, custom_kernels: Dict[Tuple[str, str], type] | None = None)

Unsqueeze

run(data: OpRunTensor, axes: OpRunTensor) → OpRunTensor

Kernel implementation.