Reducer algorithm that returns the position of the maximum value. The identity is meaningless and should be covered using mask = true
.
More...
#include <Reducer.h>
Public Member Functions | |
const std::string | name () const override |
Name of the reducer algorithm: "argmax" . | |
util::dtype | preferred_dtype () const override |
Data type to prefer, as a NumPy dtype, if the array has UnknownType. | |
util::dtype | return_dtype (util::dtype given_dtype) const override |
Return type for a given_dtype as a NumPy dtype. | |
virtual bool | returns_positions () const override |
True if this reducer returns index positions; false otherwise. | |
const std::shared_ptr< void > | apply_bool (const bool *data, const Index64 &parents, int64_t outlength) const override |
Apply the reducer algorithm to an array of boolean values. | |
const std::shared_ptr< void > | apply_int8 (const int8_t *data, const Index64 &parents, int64_t outlength) const override |
Apply the reducer algorithm to an array of signed 8-bit integer values. | |
const std::shared_ptr< void > | apply_uint8 (const uint8_t *data, const Index64 &parents, int64_t outlength) const override |
Apply the reducer algorithm to an array of unsigned 8-bit integer values. | |
const std::shared_ptr< void > | apply_int16 (const int16_t *data, const Index64 &parents, int64_t outlength) const override |
Apply the reducer algorithm to an array of signed 16-bit integer values. | |
const std::shared_ptr< void > | apply_uint16 (const uint16_t *data, const Index64 &parents, int64_t outlength) const override |
Apply the reducer algorithm to an array of unsigned 16-bit integer values. | |
const std::shared_ptr< void > | apply_int32 (const int32_t *data, const Index64 &parents, int64_t outlength) const override |
Apply the reducer algorithm to an array of signed 32-bit integer values. | |
const std::shared_ptr< void > | apply_uint32 (const uint32_t *data, const Index64 &parents, int64_t outlength) const override |
Apply the reducer algorithm to an array of unsigned 32-bit integer values. | |
const std::shared_ptr< void > | apply_int64 (const int64_t *data, const Index64 &parents, int64_t outlength) const override |
Apply the reducer algorithm to an array of signed 64-bit integer values. | |
const std::shared_ptr< void > | apply_uint64 (const uint64_t *data, const Index64 &parents, int64_t outlength) const override |
Apply the reducer algorithm to an array of unsigned 64-bit integer values. | |
const std::shared_ptr< void > | apply_float32 (const float *data, const Index64 &parents, int64_t outlength) const override |
Apply the reducer algorithm to an array of 32-bit floating-point values. | |
const std::shared_ptr< void > | apply_float64 (const double *data, const Index64 &parents, int64_t outlength) const override |
Apply the reducer algorithm to an array of 64-bit floating-point values. | |
const std::shared_ptr< void > | apply_complex64 (const std::complex< float > *data, const Index64 &parents, int64_t outlength) const override |
Apply the reducer algorithm to an array of complex 32-bit floating-point values. | |
const std::shared_ptr< void > | apply_complex128 (const std::complex< double > *data, const Index64 &parents, int64_t outlength) const override |
Apply the reducer algorithm to an array of complex 64-bit floating-point values. | |
const std::shared_ptr< void > | apply_datetime (const int64_t *data, const Index64 &parents, int64_t outlength) const override |
Apply the reducer algorithm to an array of datetime values. | |
const std::shared_ptr< void > | apply_timedelta (const int64_t *data, const Index64 &parents, int64_t outlength) const override |
Apply the reducer algorithm to an array of timedelta values. | |
Reducer algorithm that returns the position of the maximum value. The identity is meaningless and should be covered using mask = true
.
Reducers have no parameters or state. They are classes for convenience, to separate Content::reduce_next
, determining which values to combine, from the choice of reducer algorithm.
For the complex numbers array it returns the position of the maximum absolute value (also known as norm, modulus, or magnitude) of a complex number.
|
overridevirtual |
Apply the reducer algorithm to an array of boolean values.
data | The array to reduce. |
parents | An integer array indicating which group each element belongs to. |
outlength | The length of the output array (equal to the number of groups). |
Implements Reducer.
|
overridevirtual |
Apply the reducer algorithm to an array of complex 64-bit floating-point values.
data | The array to reduce. |
parents | An integer array indicating which group each element belongs to. |
outlength | The length of the output array (equal to the number of groups). |
Implements Reducer.
|
overridevirtual |
Apply the reducer algorithm to an array of complex 32-bit floating-point values.
data | The array to reduce. |
parents | An integer array indicating which group each element belongs to. |
outlength | The length of the output array (equal to the number of groups). |
Implements Reducer.
|
overridevirtual |
Apply the reducer algorithm to an array of datetime values.
data | The array to reduce. |
parents | An integer array indicating which group each element belongs to. |
outlength | The length of the output array (equal to the number of groups). |
Implements Reducer.
|
overridevirtual |
Apply the reducer algorithm to an array of 32-bit floating-point values.
data | The array to reduce. |
parents | An integer array indicating which group each element belongs to. |
outlength | The length of the output array (equal to the number of groups). |
Implements Reducer.
|
overridevirtual |
Apply the reducer algorithm to an array of 64-bit floating-point values.
data | The array to reduce. |
parents | An integer array indicating which group each element belongs to. |
outlength | The length of the output array (equal to the number of groups). |
Implements Reducer.
|
overridevirtual |
Apply the reducer algorithm to an array of signed 16-bit integer values.
data | The array to reduce. |
parents | An integer array indicating which group each element belongs to. |
outlength | The length of the output array (equal to the number of groups). |
Implements Reducer.
|
overridevirtual |
Apply the reducer algorithm to an array of signed 32-bit integer values.
data | The array to reduce. |
parents | An integer array indicating which group each element belongs to. |
outlength | The length of the output array (equal to the number of groups). |
Implements Reducer.
|
overridevirtual |
Apply the reducer algorithm to an array of signed 64-bit integer values.
data | The array to reduce. |
parents | An integer array indicating which group each element belongs to. |
outlength | The length of the output array (equal to the number of groups). |
Implements Reducer.
|
overridevirtual |
Apply the reducer algorithm to an array of signed 8-bit integer values.
data | The array to reduce. |
parents | An integer array indicating which group each element belongs to. |
outlength | The length of the output array (equal to the number of groups). |
Implements Reducer.
|
overridevirtual |
Apply the reducer algorithm to an array of timedelta values.
data | The array to reduce. |
parents | An integer array indicating which group each element belongs to. |
outlength | The length of the output array (equal to the number of groups). |
Implements Reducer.
|
overridevirtual |
Apply the reducer algorithm to an array of unsigned 16-bit integer values.
data | The array to reduce. |
parents | An integer array indicating which group each element belongs to. |
outlength | The length of the output array (equal to the number of groups). |
Implements Reducer.
|
overridevirtual |
Apply the reducer algorithm to an array of unsigned 32-bit integer values.
data | The array to reduce. |
parents | An integer array indicating which group each element belongs to. |
outlength | The length of the output array (equal to the number of groups). |
Implements Reducer.
|
overridevirtual |
Apply the reducer algorithm to an array of unsigned 64-bit integer values.
data | The array to reduce. |
parents | An integer array indicating which group each element belongs to. |
outlength | The length of the output array (equal to the number of groups). |
Implements Reducer.
|
overridevirtual |
Apply the reducer algorithm to an array of unsigned 8-bit integer values.
data | The array to reduce. |
parents | An integer array indicating which group each element belongs to. |
outlength | The length of the output array (equal to the number of groups). |
Implements Reducer.
|
overridevirtual |
Name of the reducer algorithm: "argmax"
.
Implements Reducer.
|
overridevirtual |
Data type to prefer, as a NumPy dtype, if the array has UnknownType.
The preferred type for ReducerArgmax is int64
.
Implements Reducer.
|
overridevirtual |
Return type for a given_dtype
as a NumPy dtype.
The return type for ReducerArgmax is int64
.
Reimplemented from Reducer.
True if this reducer returns index positions; false otherwise.
This is always true.
Reimplemented from Reducer.