124 lines
4.5 KiB
Cython
124 lines
4.5 KiB
Cython
# Author: Peter Prettenhofer, Brian Holt, Gilles Louppe
|
|
# License: BSD Style.
|
|
|
|
# See _tree.pyx for details.
|
|
|
|
cimport numpy as np
|
|
|
|
ctypedef np.float32_t DTYPE_t
|
|
ctypedef np.float64_t DOUBLE_t
|
|
ctypedef np.int8_t BOOL_t
|
|
|
|
|
|
# ==============================================================================
|
|
# Criterion
|
|
# ==============================================================================
|
|
|
|
cdef class Criterion:
|
|
# Methods
|
|
cdef void init(self, DOUBLE_t* y, int y_stride, BOOL_t* sample_mask,
|
|
int n_samples, int n_total_samples)
|
|
|
|
cdef void reset(self)
|
|
|
|
cdef int update(self, int a, int b, DOUBLE_t* y, int y_stride,
|
|
int* X_argsorted_i, BOOL_t* sample_mask)
|
|
|
|
cdef double eval(self)
|
|
|
|
cdef void init_value(self, double* buffer_value)
|
|
|
|
|
|
# ==============================================================================
|
|
# Tree
|
|
# ==============================================================================
|
|
|
|
cdef class Tree:
|
|
# Input/Output layout
|
|
cdef public int n_features
|
|
cdef int* n_classes
|
|
cdef public int n_outputs
|
|
|
|
cdef public int max_n_classes
|
|
cdef public int value_stride
|
|
|
|
# Parameters
|
|
cdef public Criterion criterion
|
|
cdef public double max_depth
|
|
cdef public int min_samples_split
|
|
cdef public int min_samples_leaf
|
|
cdef public double min_density
|
|
cdef public int max_features
|
|
cdef public int find_split_algorithm
|
|
cdef public object random_state
|
|
|
|
# Inner structures
|
|
cdef public int node_count
|
|
cdef public int capacity
|
|
cdef int* children_left
|
|
cdef int* children_right
|
|
cdef int* feature
|
|
cdef double* threshold
|
|
cdef double* value
|
|
cdef double* best_error
|
|
cdef double* init_error
|
|
cdef int* n_samples
|
|
|
|
# Methods
|
|
cdef void resize(self, int capacity=*)
|
|
|
|
cpdef build(self, np.ndarray X, np.ndarray y,
|
|
np.ndarray sample_mask=*, np.ndarray X_argsorted=*)
|
|
|
|
cdef void recursive_partition(self,
|
|
np.ndarray[DTYPE_t, ndim=2, mode="fortran"] X,
|
|
np.ndarray[np.int32_t, ndim=2, mode="fortran"] X_argsorted,
|
|
np.ndarray[DOUBLE_t, ndim=2, mode="c"] y,
|
|
np.ndarray sample_mask,
|
|
int n_node_samples,
|
|
int depth,
|
|
int parent,
|
|
int is_left_child,
|
|
double* buffer_value)
|
|
|
|
cdef int add_split_node(self, int parent, int is_left_child, int feature,
|
|
double threshold, double* value,
|
|
double best_error, double init_error,
|
|
int n_samples)
|
|
|
|
cdef int add_leaf(self, int parent, int is_left_child, double* value,
|
|
double error, int n_samples)
|
|
|
|
cdef void find_split(self, DTYPE_t* X_ptr, int X_stride,
|
|
int* X_argsorted_ptr, int X_argsorted_stride,
|
|
DOUBLE_t* y_ptr, int y_stride, BOOL_t* sample_mask_ptr,
|
|
int n_node_samples, int n_total_samples, int* _best_i,
|
|
double* _best_t, double* _best_error,
|
|
double* _initial_error)
|
|
|
|
cdef void find_best_split(self, DTYPE_t* X_ptr, int X_stride,
|
|
int* X_argsorted_ptr, int X_argsorted_stride,
|
|
DOUBLE_t* y_ptr, int y_stride,
|
|
BOOL_t* sample_mask_ptr, int n_node_samples,
|
|
int n_total_samples, int* _best_i,
|
|
double* _best_t, double* _best_error,
|
|
double* _initial_error)
|
|
|
|
cdef void find_random_split(self, DTYPE_t* X_ptr, int X_stride,
|
|
int* X_argsorted_ptr, int X_argsorted_stride,
|
|
DOUBLE_t* y_ptr, int y_stride,
|
|
BOOL_t* sample_mask_ptr, int n_node_samples,
|
|
int n_total_samples, int* _best_i,
|
|
double* _best_t, double* _best_error,
|
|
double* _initial_error)
|
|
|
|
cpdef predict(self, np.ndarray[DTYPE_t, ndim=2] X)
|
|
|
|
cpdef apply(self, np.ndarray[DTYPE_t, ndim=2] X)
|
|
|
|
cpdef compute_feature_importances(self, method=*)
|
|
|
|
cdef inline double _compute_feature_importances_gini(self, int node)
|
|
|
|
cdef inline double _compute_feature_importances_squared(self, int node)
|