CompiledTensorEvaluator

class CompiledTensorEvaluator

Native Symbolica evaluator returning tensors. Construct with TensorEvaluator.compile.

evaluate uses the number_type selected at compilation, as in Symbolica’s compiled evaluators. scalar_evaluator exposes the underlying native evaluator.

Attributes

Name Description
input_size Number of entries in each input row.
output_shape Logical shape of each result tensor.
parameters Symbolic inputs in evaluation order.
scalar_evaluator Underlying Symbolica compiled evaluator.
supports_real Whether compilation selected a real number type.

input_size

CompiledTensorEvaluator.input_size: builtins.int

Number of entries in each input row.

output_shape

CompiledTensorEvaluator.output_shape: tuple[int, ...]

Logical shape of each result tensor.

parameters

CompiledTensorEvaluator.parameters: builtins.list[Expression]

Symbolic inputs in evaluation order.

scalar_evaluator

CompiledTensorEvaluator.scalar_evaluator: typing.Any

Underlying Symbolica compiled evaluator.

supports_real

CompiledTensorEvaluator.supports_real: builtins.bool

Whether compilation selected a real number type.

Methods

Name Description
__repr__
evaluate Evaluate real or complex inputs, according to the compiled number type.

__repr__

CompiledTensorEvaluator.__repr__() -> builtins.str

evaluate

CompiledTensorEvaluator.evaluate(inputs: numpy.typing.ArrayLike) -> builtins.list[Tensor]

Evaluate real or complex inputs, according to the compiled number type.