TensorMetadata

class torch.fx.passes.shape_prop.TensorMetadata(shape, dtype, requires_grad, stride, memory_format, is_quantized, qparams)

A structure containing pertinent information about a tensor within a PyTorch program.

Note

Backwards-compatibility for this API is guaranteed.

count(value, /)

Return number of occurrences of value.

dtype: dtype

Alias for field number 1

index(value, start=0, stop=9223372036854775807, /)

Return first index of value.

Raises ValueError if the value is not present.

is_quantized: bool

Alias for field number 5

memory_format: memory_format | None

Alias for field number 4

qparams: dict[str, Any]

Alias for field number 6

requires_grad: bool

Alias for field number 2

shape: Size

Alias for field number 0

stride: tuple[int, ...]

Alias for field number 3