import { Attribute } from '../../../attribute';
import { UnaryOp } from '../../../ops/unary-op';
import { Tensor } from '../../../tensor';
import { CpuInferenceHandler } from '../inference-handler';
declare type UnaryOpCoreFunction<T> = (input: Tensor.NumberType, output: Tensor.NumberType, attributes?: T) => void;
export declare class CpuUnaryOp<T = unknown> extends UnaryOp {
    private func;
    private attributesInitializer?;
    private attributes?;
    constructor(typeConstraint: ReadonlyArray<Tensor.DataType>, func: UnaryOpCoreFunction<T>, attributesInitializer?: ((attributes: Attribute) => T) | undefined, resultType?: Tensor.DataType);
    initialize(attributes: Attribute): void;
    run(inferenceHandler: CpuInferenceHandler, inputs: Tensor[]): Tensor[];
}
export declare function unaryOp<T>(x: Tensor, func: UnaryOpCoreFunction<T>, attributes: T, resultType?: Tensor.DataType): Tensor;
export declare function abs(input: Tensor.NumberType, output: Tensor.NumberType): void;
export declare function acos(input: Tensor.NumberType, output: Tensor.NumberType): void;
export declare function acosh(input: Tensor.NumberType, output: Tensor.NumberType): void;
export declare function asin(input: Tensor.NumberType, output: Tensor.NumberType): void;
export declare function asinh(input: Tensor.NumberType, output: Tensor.NumberType): void;
export declare function atan(input: Tensor.NumberType, output: Tensor.NumberType): void;
export declare function atanh(input: Tensor.NumberType, output: Tensor.NumberType): void;
export declare function ceil(input: Tensor.NumberType, output: Tensor.NumberType): void;
export declare function clipInitializer(attributes: Attribute): {
    min: number;
    max: number;
};
export declare function clip(input: Tensor.NumberType, output: Tensor.NumberType, attributes: {
    min: number;
    max: number;
}): void;
export declare function cos(input: Tensor.NumberType, output: Tensor.NumberType): void;
export declare function cosh(input: Tensor.NumberType, output: Tensor.NumberType): void;
export declare function eluInitializer(attributes: Attribute): number;
export declare function elu(input: Tensor.NumberType, output: Tensor.NumberType, attributes: number): void;
export declare function exp(input: Tensor.NumberType, output: Tensor.NumberType): void;
export declare function floor(input: Tensor.NumberType, output: Tensor.NumberType): void;
export declare function isNan(input: Tensor.NumberType, output: Tensor.BooleanType): void;
export declare function leakyReluInitializer(attributes: Attribute): number;
export declare function leakyRelu(input: Tensor.NumberType, output: Tensor.NumberType, attributes: number): void;
export declare function log(input: Tensor.NumberType, output: Tensor.NumberType): void;
export declare function neg(input: Tensor.NumberType, output: Tensor.NumberType): void;
export declare function not(input: Tensor.BooleanType, output: Tensor.BooleanType): void;
export declare function reciprocal(input: Tensor.NumberType, output: Tensor.NumberType): void;
export declare function relu(input: Tensor.NumberType, output: Tensor.NumberType): void;
export declare function sigmoid(input: Tensor.NumberType, output: Tensor.NumberType): void;
export declare function sign(input: Tensor.NumberType, output: Tensor.NumberType): void;
export declare function sin(input: Tensor.NumberType, output: Tensor.NumberType): void;
export declare function sinh(input: Tensor.NumberType, output: Tensor.NumberType): void;
export declare function sqrt(input: Tensor.NumberType, output: Tensor.NumberType): void;
export declare function tan(input: Tensor.NumberType, output: Tensor.NumberType): void;
export declare function tanh(input: Tensor.NumberType, output: Tensor.NumberType): void;
export {};