Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

MAINT Decrease code diff and fix Cython-lint #4

Open
wants to merge 1 commit into
base: scarliles/defuse-partitioner
Choose a base branch
from
Open

MAINT Decrease code diff and fix Cython-lint #4

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
13 changes: 10 additions & 3 deletions sklearn/tree/_partitioner.pxd
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10,7 +10,7 @@ cdef float32_t EXTRACT_NNZ_SWITCH = 0.1
# functions. The alternative would have been to use inheritance-based polymorphism
# but it would have resulted in a ~10% overall tree fitting performance
# degradation caused by the overhead frequent virtual method lookups.
#ctypedef fused Partitioner:
# ctypedef fused Partitioner:
# DensePartitioner
# SparsePartitioner

Expand Down Expand Up @@ -67,8 +67,15 @@ cdef class Partitioner:
float32_t* min_feature_value_out,
float32_t* max_feature_value_out,
) noexcept nogil
inline void next_p(self, intp_t* p_prev, intp_t* p) noexcept nogil
inline intp_t partition_samples(self, float64_t current_threshold) noexcept nogil
inline void next_p(
self,
intp_t* p_prev,
intp_t* p
) noexcept nogil
inline intp_t partition_samples(
self,
float64_t current_threshold
) noexcept nogil
inline void partition_samples_final(
self,
intp_t best_pos,
Expand Down
28 changes: 14 additions & 14 deletions sklearn/tree/_partitioner.pyx
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -6,14 +6,14 @@ from scipy.sparse import issparse

import numpy as np

from ._sort cimport sort, sparse_swap, swap, FEATURE_THRESHOLD
from ._splitter cimport sort, sparse_swap, FEATURE_THRESHOLD


cdef class Partitioner:
cdef:
inline void init_node_split(self, intp_t start, intp_t end) noexcept nogil:
self._init_node_split(self, start, end)

inline void sort_samples_and_feature_values(
self,
intp_t current_feature
Expand All @@ -33,7 +33,7 @@ cdef class Partitioner:

inline intp_t partition_samples(self, float64_t current_threshold) noexcept nogil:
return self._partition_samples(self, current_threshold)

inline void partition_samples_final(
self,
intp_t best_pos,
Expand Down Expand Up @@ -536,22 +536,22 @@ cdef inline void sparse_extract_nnz(SparsePartitioner self, intp_t feature) noex
if ((1 - self.is_samples_sorted) * n_samples * log(n_samples) +
n_samples * log(n_indices) < EXTRACT_NNZ_SWITCH * n_indices):
extract_nnz_binary_search(X_indices, X_data,
indptr_start, indptr_end,
samples, self.start, self.end,
index_to_samples,
feature_values,
&self.end_negative, &self.start_positive,
sorted_samples, &self.is_samples_sorted)
indptr_start, indptr_end,
samples, self.start, self.end,
index_to_samples,
feature_values,
&self.end_negative, &self.start_positive,
sorted_samples, &self.is_samples_sorted)

# Using an index to samples technique to extract non zero values
# index_to_samples is a mapping from X_indices to samples
else:
extract_nnz_index_to_samples(X_indices, X_data,
indptr_start, indptr_end,
samples, self.start, self.end,
index_to_samples,
feature_values,
&self.end_negative, &self.start_positive)
indptr_start, indptr_end,
samples, self.start, self.end,
index_to_samples,
feature_values,
&self.end_negative, &self.start_positive)


cdef int compare_SIZE_t(const void* a, const void* b) noexcept nogil:
Expand Down
13 changes: 0 additions & 13 deletions sklearn/tree/_sort.pxd
View file
Open in desktop
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I broke out the sort functions into their own module because

  • they're used both by Splitter and by Partitioner
  • Splitter depends on Partitioner
  • as a practice, circular dependencies are best avoided, so that Partitioner should then not depend on Splitter
  • more generally, dependency graphs are best modeled as DAGs

Copy link
Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Hmm fair point. Let me try to circle this up with scikit learn devs to try to make the code more maintainable.

This file was deleted.

123 changes: 0 additions & 123 deletions sklearn/tree/_sort.pyx
View file
Open in desktop

This file was deleted.

17 changes: 17 additions & 0 deletions sklearn/tree/_splitter.pxd
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -10,6 +10,10 @@ from ._tree cimport ParentInfo
from ..utils._typedefs cimport float32_t, float64_t, intp_t, int8_t, int32_t, uint32_t


# Mitigate precision differences between 32 bit and 64 bit
cdef float32_t FEATURE_THRESHOLD = 1e-7


cdef struct SplitRecord:
# Data to track sample split
intp_t feature # Which feature to split on.
Expand Down Expand Up @@ -132,3 +136,16 @@ cdef void shift_missing_values_to_left_if_required(
intp_t[::1] samples,
intp_t end,
) noexcept nogil


# Sort n-element arrays pointed to by feature_values and samples, simultaneously,
# by the values in feature_values. Algorithm: Introsort (Musser, SP&E, 1997).
cdef void sort(float32_t* feature_values, intp_t* samples, intp_t n) noexcept nogil

cdef void swap(float32_t* feature_values, intp_t* samples, intp_t i, intp_t j) noexcept nogil
cdef void sparse_swap(
intp_t[::1] index_to_samples,
intp_t[::1] samples,
intp_t pos_1,
intp_t pos_2
) noexcept nogil
Loading