针对pulse-transit的工具

dist/client/sklearn/utils/_cython_blas.pxd

@@ -0,0 +1,41 @@
+from cython cimport floating
+
+
+cpdef enum BLAS_Order:
+    RowMajor  # C contiguous
+    ColMajor  # Fortran contiguous
+
+
+cpdef enum BLAS_Trans:
+    NoTrans = 110  # correspond to 'n'
+    Trans = 116    # correspond to 't'
+
+
+# BLAS Level 1 ################################################################
+cdef floating _dot(int, floating*, int, floating*, int) nogil
+
+cdef floating _asum(int, floating*, int) nogil
+
+cdef void _axpy(int, floating, floating*, int, floating*, int) nogil
+
+cdef floating _nrm2(int, floating*, int) nogil
+
+cdef void _copy(int, floating*, int, floating*, int) nogil
+
+cdef void _scal(int, floating, floating*, int) nogil
+
+cdef void _rotg(floating*, floating*, floating*, floating*) nogil
+
+cdef void _rot(int, floating*, int, floating*, int, floating, floating) nogil
+
+# BLAS Level 2 ################################################################
+cdef void _gemv(BLAS_Order, BLAS_Trans, int, int, floating, floating*, int,
+                floating*, int, floating, floating*, int) nogil
+
+cdef void _ger(BLAS_Order, int, int, floating, floating*, int, floating*, int,
+               floating*, int) nogil
+
+# BLASLevel 3 ################################################################
+cdef void _gemm(BLAS_Order, BLAS_Trans, BLAS_Trans, int, int, int, floating,
+                floating*, int, floating*, int, floating, floating*,
+                int) nogil

dist/client/sklearn/utils/_fast_dict.pxd

@@ -0,0 +1,22 @@
+# Author: Gael Varoquaux
+# License: BSD
+"""
+Uses C++ map containers for fast dict-like behavior with keys being
+integers, and values float.
+"""
+
+from libcpp.map cimport map as cpp_map
+
+# Import the C-level symbols of numpy
+cimport numpy as np
+
+ctypedef np.float64_t DTYPE_t
+
+ctypedef np.intp_t ITYPE_t
+
+###############################################################################
+# An object to be used in Python
+
+cdef class IntFloatDict:
+    cdef cpp_map[ITYPE_t, DTYPE_t] my_map
+    cdef _to_arrays(self, ITYPE_t [:] keys, DTYPE_t [:] values)

dist/client/sklearn/utils/_heap.pxd

@@ -0,0 +1,14 @@
+# Heap routines, used in various Cython implementations.
+
+from cython cimport floating
+
+from ._typedefs cimport ITYPE_t
+
+
+cdef int heap_push(
+    floating* values,
+    ITYPE_t* indices,
+    ITYPE_t size,
+    floating val,
+    ITYPE_t val_idx,
+) nogil

dist/client/sklearn/utils/_openmp_helpers.pxd

@@ -0,0 +1,6 @@
+# Helpers to access OpenMP threads information
+#
+# Those interfaces act as indirections which allows the non-support of OpenMP
+# for implementations which have been written for it.
+
+cdef int _openmp_thread_num() nogil

dist/client/sklearn/utils/_random.pxd

@@ -0,0 +1,44 @@
+# Authors: Arnaud Joly
+#
+# License: BSD 3 clause
+
+
+import numpy as np
+cimport numpy as np
+ctypedef np.npy_uint32 UINT32_t
+
+cdef inline UINT32_t DEFAULT_SEED = 1
+
+cdef enum:
+    # Max value for our rand_r replacement (near the bottom).
+    # We don't use RAND_MAX because it's different across platforms and
+    # particularly tiny on Windows/MSVC.
+    RAND_R_MAX = 0x7FFFFFFF
+
+cpdef sample_without_replacement(np.int_t n_population,
+                                 np.int_t n_samples,
+                                 method=*,
+                                 random_state=*)
+
+# rand_r replacement using a 32bit XorShift generator
+# See http://www.jstatsoft.org/v08/i14/paper for details
+cdef inline UINT32_t our_rand_r(UINT32_t* seed) nogil:
+    """Generate a pseudo-random np.uint32 from a np.uint32 seed"""
+    # seed shouldn't ever be 0.
+    if (seed[0] == 0): seed[0] = DEFAULT_SEED
+
+    seed[0] ^= <UINT32_t>(seed[0] << 13)
+    seed[0] ^= <UINT32_t>(seed[0] >> 17)
+    seed[0] ^= <UINT32_t>(seed[0] << 5)
+
+    # Note: we must be careful with the final line cast to np.uint32 so that
+    # the function behaves consistently across platforms.
+    #
+    # The following cast might yield different results on different platforms:
+    # wrong_cast = <UINT32_t> RAND_R_MAX + 1
+    #
+    # We can use:
+    # good_cast = <UINT32_t>(RAND_R_MAX + 1)
+    # or:
+    # cdef np.uint32_t another_good_cast = <UINT32_t>RAND_R_MAX + 1
+    return seed[0] % <UINT32_t>(RAND_R_MAX + 1)

dist/client/sklearn/utils/_seq_dataset.pxd

@@ -0,0 +1,116 @@
+
+#------------------------------------------------------------------------------
+
+"""
+Dataset abstractions for sequential data access.
+WARNING: Do not edit .pxd file directly, it is generated from .pxd.tp
+"""
+
+cimport numpy as np
+
+# SequentialDataset and its two concrete subclasses are (optionally randomized)
+# iterators over the rows of a matrix X and corresponding target values y.
+
+
+cdef class SequentialDataset64:
+    cdef int current_index
+    cdef np.ndarray index
+    cdef int *index_data_ptr
+    cdef Py_ssize_t n_samples
+    cdef np.uint32_t seed
+
+    cdef void shuffle(self, np.uint32_t seed) nogil
+    cdef int _get_next_index(self) nogil
+    cdef int _get_random_index(self) nogil
+
+    cdef void _sample(self, double **x_data_ptr, int **x_ind_ptr,
+                      int *nnz, double *y, double *sample_weight,
+                      int current_index) nogil
+    cdef void next(self, double **x_data_ptr, int **x_ind_ptr,
+                   int *nnz, double *y, double *sample_weight) nogil
+    cdef int random(self, double **x_data_ptr, int **x_ind_ptr,
+                    int *nnz, double *y, double *sample_weight) nogil
+
+
+cdef class ArrayDataset64(SequentialDataset64):
+    cdef np.ndarray X
+    cdef np.ndarray Y
+    cdef np.ndarray sample_weights
+    cdef Py_ssize_t n_features
+    cdef np.npy_intp X_stride
+    cdef double *X_data_ptr
+    cdef double *Y_data_ptr
+    cdef np.ndarray feature_indices
+    cdef int *feature_indices_ptr
+    cdef double *sample_weight_data
+
+
+cdef class CSRDataset64(SequentialDataset64):
+    cdef np.ndarray X_data
+    cdef np.ndarray X_indptr
+    cdef np.ndarray X_indices
+    cdef np.ndarray Y
+    cdef np.ndarray sample_weights
+    cdef double *X_data_ptr
+    cdef int *X_indptr_ptr
+    cdef int *X_indices_ptr
+    cdef double *Y_data_ptr
+    cdef double *sample_weight_data
+
+#------------------------------------------------------------------------------
+
+"""
+Dataset abstractions for sequential data access.
+WARNING: Do not edit .pxd file directly, it is generated from .pxd.tp
+"""
+
+cimport numpy as np
+
+# SequentialDataset and its two concrete subclasses are (optionally randomized)
+# iterators over the rows of a matrix X and corresponding target values y.
+
+
+cdef class SequentialDataset32:
+    cdef int current_index
+    cdef np.ndarray index
+    cdef int *index_data_ptr
+    cdef Py_ssize_t n_samples
+    cdef np.uint32_t seed
+
+    cdef void shuffle(self, np.uint32_t seed) nogil
+    cdef int _get_next_index(self) nogil
+    cdef int _get_random_index(self) nogil
+
+    cdef void _sample(self, float **x_data_ptr, int **x_ind_ptr,
+                      int *nnz, float *y, float *sample_weight,
+                      int current_index) nogil
+    cdef void next(self, float **x_data_ptr, int **x_ind_ptr,
+                   int *nnz, float *y, float *sample_weight) nogil
+    cdef int random(self, float **x_data_ptr, int **x_ind_ptr,
+                    int *nnz, float *y, float *sample_weight) nogil
+
+
+cdef class ArrayDataset32(SequentialDataset32):
+    cdef np.ndarray X
+    cdef np.ndarray Y
+    cdef np.ndarray sample_weights
+    cdef Py_ssize_t n_features
+    cdef np.npy_intp X_stride
+    cdef float *X_data_ptr
+    cdef float *Y_data_ptr
+    cdef np.ndarray feature_indices
+    cdef int *feature_indices_ptr
+    cdef float *sample_weight_data
+
+
+cdef class CSRDataset32(SequentialDataset32):
+    cdef np.ndarray X_data
+    cdef np.ndarray X_indptr
+    cdef np.ndarray X_indices
+    cdef np.ndarray Y
+    cdef np.ndarray sample_weights
+    cdef float *X_data_ptr
+    cdef int *X_indptr_ptr
+    cdef int *X_indices_ptr
+    cdef float *Y_data_ptr
+    cdef float *sample_weight_data

dist/client/sklearn/utils/_sorting.pxd

@@ -0,0 +1,9 @@
+from ._typedefs cimport DTYPE_t, ITYPE_t
+
+from cython cimport floating
+
+cdef int simultaneous_sort(
+    floating *dist,
+    ITYPE_t *idx,
+    ITYPE_t size,
+) nogil

dist/client/sklearn/utils/_typedefs.pxd

@@ -0,0 +1,17 @@
+#!python
+cimport numpy as np
+
+# Floating point/data type
+ctypedef np.float64_t DTYPE_t  # WARNING: should match DTYPE in typedefs.pyx
+
+cdef enum:
+    DTYPECODE = np.NPY_FLOAT64
+    ITYPECODE = np.NPY_INTP
+    INT32TYPECODE = np.NPY_INT32
+    INT64TYPECODE = np.NPY_INT64
+
+# Index/integer type.
+#  WARNING: ITYPE_t must be a signed integer type or you will have a bad time!
+ctypedef np.intp_t ITYPE_t  # WARNING: should match ITYPE in typedefs.pyx
+ctypedef np.int32_t INT32TYPE_t  # WARNING: should match INT32TYPE in typedefs.pyx
+ctypedef np.int64_t INT64TYPE_t  # WARNING: should match INT32TYPE in typedefs.pyx

dist/client/sklearn/utils/_vector_sentinel.pxd

@@ -0,0 +1,12 @@
+cimport numpy as np
+
+from libcpp.vector cimport vector
+from ..utils._typedefs cimport ITYPE_t, DTYPE_t, INT32TYPE_t, INT64TYPE_t
+
+ctypedef fused vector_typed:
+    vector[DTYPE_t]
+    vector[ITYPE_t]
+    vector[INT32TYPE_t]
+    vector[INT64TYPE_t]
+
+cdef np.ndarray vector_to_nd_array(vector_typed * vect_ptr)

dist/client/sklearn/utils/_weight_vector.pxd

@@ -0,0 +1,45 @@
+
+# WARNING: Do not edit this .pyx file directly, it is generated from its .pyx.tp
+cimport numpy as np
+
+cdef class WeightVector64(object):
+    cdef readonly double[::1] w
+    cdef readonly double[::1] aw
+    cdef double *w_data_ptr
+    cdef double *aw_data_ptr
+    cdef double wscale
+    cdef double average_a
+    cdef double average_b
+    cdef int n_features
+    cdef double sq_norm
+
+    cdef void add(self, double *x_data_ptr, int *x_ind_ptr,
+                  int xnnz, double c) nogil
+    cdef void add_average(self, double *x_data_ptr, int *x_ind_ptr,
+                          int xnnz, double c, double num_iter) nogil
+    cdef double dot(self, double *x_data_ptr, int *x_ind_ptr,
+                    int xnnz) nogil
+    cdef void scale(self, double c) nogil
+    cdef void reset_wscale(self) nogil
+    cdef double norm(self) nogil
+
+cdef class WeightVector32(object):
+    cdef readonly float[::1] w
+    cdef readonly float[::1] aw
+    cdef float *w_data_ptr
+    cdef float *aw_data_ptr
+    cdef float wscale
+    cdef float average_a
+    cdef float average_b
+    cdef int n_features
+    cdef float sq_norm
+
+    cdef void add(self, float *x_data_ptr, int *x_ind_ptr,
+                  int xnnz, float c) nogil
+    cdef void add_average(self, float *x_data_ptr, int *x_ind_ptr,
+                          int xnnz, float c, float num_iter) nogil
+    cdef float dot(self, float *x_data_ptr, int *x_ind_ptr,
+                    int xnnz) nogil
+    cdef void scale(self, float c) nogil
+    cdef void reset_wscale(self) nogil
+    cdef float norm(self) nogil

dist/client/sklearn/utils/murmurhash.pxd

@@ -0,0 +1,21 @@
+"""Export fast murmurhash C/C++ routines + cython wrappers"""
+
+cimport numpy as np
+
+# The C API is disabled for now, since it requires -I flags to get
+# compilation to work even when these functions are not used.
+#cdef extern from "MurmurHash3.h":
+#    void MurmurHash3_x86_32(void* key, int len, unsigned int seed,
+#                            void* out)
+#
+#    void MurmurHash3_x86_128(void* key, int len, unsigned int seed,
+#                             void* out)
+#
+#    void MurmurHash3_x64_128(void* key, int len, unsigned int seed,
+#                             void* out)
+
+
+cpdef np.uint32_t murmurhash3_int_u32(int key, unsigned int seed)
+cpdef np.int32_t murmurhash3_int_s32(int key, unsigned int seed)
+cpdef np.uint32_t murmurhash3_bytes_u32(bytes key, unsigned int seed)
+cpdef np.int32_t murmurhash3_bytes_s32(bytes key, unsigned int seed)