module testing.experimental_c¶
Short summary¶
module mlprodict.testing.experimental_c
C++ experimental implementations.
Documentation¶
C++ experimental implementations.
- mlprodict.testing.experimental_c.BroadcastMatrixAddLeftInplaceDouble(arg0: numpy.ndarray[numpy.float64], arg1: numpy.ndarray[numpy.float64]) None¶
Inplace addition, does X += Y. The function only allows broadcast in one way.
- mlprodict.testing.experimental_c.BroadcastMatrixAddLeftInplaceFloat(arg0: numpy.ndarray[numpy.float32], arg1: numpy.ndarray[numpy.float32]) None¶
Inplace addition, does X += Y. The function only allows broadcast in one way.
- mlprodict.testing.experimental_c.BroadcastMatrixAddLeftInplaceInt64(arg0: numpy.ndarray[numpy.int64], arg1: numpy.ndarray[numpy.int64]) None¶
Inplace addition, does X += Y. The function only allows broadcast in one way.
- mlprodict.testing.experimental_c.code_optimisation() str¶
Returns a string giving some insights about optimisations.
- mlprodict.testing.experimental_c.custom_einsum_double(equation: str, x: numpy.ndarray[numpy.float64], y: numpy.ndarray[numpy.float64], nthread: int = 0) numpy.ndarray[numpy.float64]¶
Custom C++ implementation of operator einsum with double. The function only works with contiguous arrays. It does not any explicit transposes. It does not support diagonal operator (repetition of the same letter). See python’s version
custom_einsum.
- mlprodict.testing.experimental_c.custom_einsum_float(equation: str, x: numpy.ndarray[numpy.float32], y: numpy.ndarray[numpy.float32], nthread: int = 0) numpy.ndarray[numpy.float32]¶
Custom C++ implementation of operator einsum with float. The function only works with contiguous arrays. It does not any explicit transposes. It does not support diagonal operator (repetition of the same letter). See python’s version
custom_einsum.
- mlprodict.testing.experimental_c.custom_einsum_int32(equation: str, x: numpy.ndarray[numpy.int32], y: numpy.ndarray[numpy.int32], nthread: int = 0) numpy.ndarray[numpy.int32]¶
Custom C++ implementation of operator einsum with int32. The function only works with contiguous arrays. It does not any explicit transposes. It does not support diagonal operator (repetition of the same letter). See python’s version
custom_einsum.
- mlprodict.testing.experimental_c.custom_einsum_int64(equation: str, x: numpy.ndarray[numpy.int64], y: numpy.ndarray[numpy.int64], nthread: int = 0) numpy.ndarray[numpy.int64]¶
Custom C++ implementation of operator einsum with int64. The function only works with contiguous arrays. It does not any explicit transposes. It does not support diagonal operator (repetition of the same letter). See python’s version
custom_einsum.
- mlprodict.testing.experimental_c.custom_reducesum_rk_double(x: numpy.ndarray[numpy.float64], nthread: int = 0) numpy.ndarray[numpy.float64]¶
Custom C++ implementation of operator ReduceSum with double when the reduced matrix has two dimensions and the reduced axis is the first one. x is the reduced matrix. nthread specifies the number of threads used to distribute. Negative means OMP default values.
- mlprodict.testing.experimental_c.custom_reducesum_rk_float(x: numpy.ndarray[numpy.float32], nthread: int = 0) numpy.ndarray[numpy.float32]¶
Custom C++ implementation of operator ReduceSum with float when the reduced matrix has two dimensions and the reduced axis is the first one. x is the reduced matrix. nthread specifies the number of threads used to distribute. Negative means OMP default values.