baohan
/
scikit-learn
mirror of https://github.com/scikit-learn/scikit-learn.git
0
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity
scikit-learn/sklearn/utils/seq_dataset.pxd

54 lines
1.7 KiB
Cython
Raw Normal View History

moved WeightVector und SequentialDataset into seperate modules.
2012-03-08 03:09:05 +08:00
"""Dataset abstractions for sequential data access. """
move WeightVector and Dataset abstraction to new module
2012-02-16 04:15:45 +08:00
cimport numpy as np
ENH use threads instead of multiprocessing in SGD
2014-03-13 00:03:59 +08:00
# SequentialDataset and its two concrete subclasses are (optionally randomized)
# iterators over the rows of a matrix X and corresponding target values y.
move WeightVector and Dataset abstraction to new module
2012-02-16 04:15:45 +08:00
moved WeightVector und SequentialDataset into seperate modules.
2012-03-08 03:09:05 +08:00
cdef class SequentialDataset:
ENH use threads instead of multiprocessing in SGD
2014-03-13 00:03:59 +08:00
cdef int current_index
cdef np.ndarray index
cdef int *index_data_ptr
move WeightVector and Dataset abstraction to new module
2012-02-16 04:15:45 +08:00
cdef Py_ssize_t n_samples
ENH add sag solver in LogisticRegression and Ridge
2015-05-19 16:24:22 +08:00
cdef np.uint32_t seed
move WeightVector and Dataset abstraction to new module
2012-02-16 04:15:45 +08:00
ENH SGD Cython improvements * nogil declarations where possible (but no with nogil blocks, since the performance benefit of those is not clear) * use BLAS where appropriate * replaced DOUBLE by double, INTEGER and int32 by int/np.intc, since that's the type used by scipy.sparse * explicit checks that dataset dims don't exceed INT_MAX * replaced an np.all by an explicit loop (npymath library for C99 isfinite)
2013-09-07 22:38:53 +08:00
cdef void next(self, double **x_data_ptr, int **x_ind_ptr,
int *nnz, double *y, double *sample_weight) nogil
ENH use threads instead of multiprocessing in SGD
2014-03-13 00:03:59 +08:00
cdef void shuffle(self, np.uint32_t seed) nogil
Adding Implementation of SAG
2014-10-30 19:00:11 +08:00
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 int random(self, double **x_data_ptr, int **x_ind_ptr,
int *nnz, double *y, double *sample_weight) nogil
move WeightVector and Dataset abstraction to new module
2012-02-16 04:15:45 +08:00
moved WeightVector und SequentialDataset into seperate modules.
2012-03-08 03:09:05 +08:00
cdef class ArrayDataset(SequentialDataset):
ENH add sag solver in LogisticRegression and Ridge
2015-05-19 16:24:22 +08:00
cdef np.ndarray X
cdef np.ndarray Y
cdef np.ndarray sample_weights
move WeightVector and Dataset abstraction to new module
2012-02-16 04:15:45 +08:00
cdef Py_ssize_t n_features
cdef int stride
ENH SGD Cython improvements * nogil declarations where possible (but no with nogil blocks, since the performance benefit of those is not clear) * use BLAS where appropriate * replaced DOUBLE by double, INTEGER and int32 by int/np.intc, since that's the type used by scipy.sparse * explicit checks that dataset dims don't exceed INT_MAX * replaced an np.all by an explicit loop (npymath library for C99 isfinite)
2013-09-07 22:38:53 +08:00
cdef double *X_data_ptr
cdef double *Y_data_ptr
move WeightVector and Dataset abstraction to new module
2012-02-16 04:15:45 +08:00
cdef np.ndarray feature_indices
ENH SGD Cython improvements * nogil declarations where possible (but no with nogil blocks, since the performance benefit of those is not clear) * use BLAS where appropriate * replaced DOUBLE by double, INTEGER and int32 by int/np.intc, since that's the type used by scipy.sparse * explicit checks that dataset dims don't exceed INT_MAX * replaced an np.all by an explicit loop (npymath library for C99 isfinite)
2013-09-07 22:38:53 +08:00
cdef int *feature_indices_ptr
cdef double *sample_weight_data
move WeightVector and Dataset abstraction to new module
2012-02-16 04:15:45 +08:00
moved WeightVector und SequentialDataset into seperate modules.
2012-03-08 03:09:05 +08:00
cdef class CSRDataset(SequentialDataset):
ENH add sag solver in LogisticRegression and Ridge
2015-05-19 16:24:22 +08:00
cdef np.ndarray X_data
cdef np.ndarray X_indptr
cdef np.ndarray X_indices
cdef np.ndarray Y
cdef np.ndarray sample_weights
move WeightVector and Dataset abstraction to new module
2012-02-16 04:15:45 +08:00
cdef int stride
ENH SGD Cython improvements * nogil declarations where possible (but no with nogil blocks, since the performance benefit of those is not clear) * use BLAS where appropriate * replaced DOUBLE by double, INTEGER and int32 by int/np.intc, since that's the type used by scipy.sparse * explicit checks that dataset dims don't exceed INT_MAX * replaced an np.all by an explicit loop (npymath library for C99 isfinite)
2013-09-07 22:38:53 +08:00
cdef double *X_data_ptr
cdef int *X_indptr_ptr
cdef int *X_indices_ptr
cdef double *Y_data_ptr
move WeightVector and Dataset abstraction to new module
2012-02-16 04:15:45 +08:00
cdef np.ndarray feature_indices
ENH SGD Cython improvements * nogil declarations where possible (but no with nogil blocks, since the performance benefit of those is not clear) * use BLAS where appropriate * replaced DOUBLE by double, INTEGER and int32 by int/np.intc, since that's the type used by scipy.sparse * explicit checks that dataset dims don't exceed INT_MAX * replaced an np.all by an explicit loop (npymath library for C99 isfinite)
2013-09-07 22:38:53 +08:00
cdef int *feature_indices_ptr
cdef double *sample_weight_data