ML::Explainer Namespace Reference

Classes
class	TreePathInfo

Typedefs
using	TreePathHandle = std::variant< std::shared_ptr< TreePathInfo< float > >, std::shared_ptr< TreePathInfo< double > >>
using	FloatPointer = std::variant< float *, double * >

Functions
void	kernel_dataset (const raft::handle_t &handle, float *X, int nrows_X, int ncols, float *background, int nrows_background, float *dataset, float *observation, int *nsamples, int len_nsamples, int maxsample, uint64_t seed=0ULL)
void	kernel_dataset (const raft::handle_t &handle, float *X, int nrows_X, int ncols, double *background, int nrows_background, double *dataset, double *observation, int *nsamples, int len_nsamples, int maxsample, uint64_t seed=0ULL)
void	permutation_shap_dataset (const raft::handle_t &handle, float *dataset, const float *background, int nrows_bg, int ncols, const float *row, int *idx, bool row_major)
void	permutation_shap_dataset (const raft::handle_t &handle, double *dataset, const double *background, int nrows_bg, int ncols, const double *row, int *idx, bool row_major)
void	shap_main_effect_dataset (const raft::handle_t &handle, float *dataset, const float *background, int nrows_bg, int ncols, const float *row, int *idx, bool row_major)
void	shap_main_effect_dataset (const raft::handle_t &handle, double *dataset, const double *background, int nrows_bg, int ncols, const double *row, int *idx, bool row_major)
void	update_perm_shap_values (const raft::handle_t &handle, float *shap_values, const float *y_hat, const int ncols, const int *idx)
void	update_perm_shap_values (const raft::handle_t &handle, double *shap_values, const double *y_hat, const int ncols, const int *idx)
TreePathHandle	extract_path_info (TreeliteModelHandle model)
void	gpu_treeshap (TreePathHandle path_info, const FloatPointer data, std::size_t n_rows, std::size_t n_cols, FloatPointer out_preds, std::size_t out_preds_size)
void	gpu_treeshap_interventional (TreePathHandle path_info, const FloatPointer data, std::size_t n_rows, std::size_t n_cols, const FloatPointer background_data, std::size_t background_n_rows, std::size_t background_n_cols, FloatPointer out_preds, std::size_t out_preds_size)
void	gpu_treeshap_interactions (TreePathHandle path_info, const FloatPointer data, std::size_t n_rows, std::size_t n_cols, FloatPointer out_preds, std::size_t out_preds_size)
void	gpu_treeshap_taylor_interactions (TreePathHandle path_info, const FloatPointer data, std::size_t n_rows, std::size_t n_cols, FloatPointer out_preds, std::size_t out_preds_size)

Typedef Documentation

FloatPointer

using ML::Explainer::FloatPointer = typedef std::variant<float*, double*>

TreePathHandle

using ML::Explainer::TreePathHandle = typedef std::variant<std::shared_ptr<TreePathInfo<float> >, std::shared_ptr<TreePathInfo<double> >>

Function Documentation

extract_path_info()

TreePathHandle ML::Explainer::extract_path_info

(

TreeliteModelHandle

model

)

gpu_treeshap()

void ML::Explainer::gpu_treeshap	(	TreePathHandle	path_info,
		const FloatPointer	data,
		std::size_t	n_rows,
		std::size_t	n_cols,
		FloatPointer	out_preds,
		std::size_t	out_preds_size
	)

gpu_treeshap_interactions()

void ML::Explainer::gpu_treeshap_interactions	(	TreePathHandle	path_info,
		const FloatPointer	data,
		std::size_t	n_rows,
		std::size_t	n_cols,
		FloatPointer	out_preds,
		std::size_t	out_preds_size
	)

gpu_treeshap_interventional()

void ML::Explainer::gpu_treeshap_interventional	(	TreePathHandle	path_info,
		const FloatPointer	data,
		std::size_t	n_rows,
		std::size_t	n_cols,
		const FloatPointer	background_data,
		std::size_t	background_n_rows,
		std::size_t	background_n_cols,
		FloatPointer	out_preds,
		std::size_t	out_preds_size
	)

gpu_treeshap_taylor_interactions()

void ML::Explainer::gpu_treeshap_taylor_interactions	(	TreePathHandle	path_info,
		const FloatPointer	data,
		std::size_t	n_rows,
		std::size_t	n_cols,
		FloatPointer	out_preds,
		std::size_t	out_preds_size
	)

kernel_dataset() [1/2]

void ML::Explainer::kernel_dataset	(	const raft::handle_t &	handle,
		float *	X,
		int	nrows_X,
		int	ncols,
		double *	background,
		int	nrows_background,
		double *	dataset,
		double *	observation,
		int *	nsamples,
		int	len_nsamples,
		int	maxsample,
		uint64_t	seed = 0ULL
	)

kernel_dataset() [2/2]

void ML::Explainer::kernel_dataset	(	const raft::handle_t &	handle,
		float *	X,
		int	nrows_X,
		int	ncols,
		float *	background,
		int	nrows_background,
		float *	dataset,
		float *	observation,
		int *	nsamples,
		int	len_nsamples,
		int	maxsample,
		uint64_t	seed = 0ULL
	)

Generates samples of dataset for kernel shap algorithm.

Parameters

[in]	handle	cuML handle
[in,out]	X	generated data [on device] 1-0 (row major)
[in]	nrows_X	number of rows in X
[in]	ncols	number of columns in X, background and dataset
[in]	background	background data [on device]
[in]	nrows_background	number of rows in background dataset
[out]	dataset	generated data [on device] observation=background (row major)
[in]	observation	row to scatter
[in]	nsamples	vector with number of entries that are randomly sampled
[in]	len_nsamples	number of entries to be sampled
[in]	maxsample	size of the biggest sampled observation
[in]	seed	Seed for the random number generator

Kernel distrubutes exact part of the kernel shap dataset Each block scatters the data of a row of observations into the (number of rows of background) in dataset, based on the row of X. So, given: background = [[0, 1, 2], [3, 4, 5]] observation = [100, 101, 102] X = [[1, 0, 1], [0, 1, 1]]

dataset (output): [[100, 1, 102], [100, 4, 102] [0, 101, 102], [3, 101, 102]] The first thread of each block calculates the sampling of k entries of observation to scatter into dataset. Afterwards each block scatters the data of a row of X into the (number of rows of background) in dataset. So, given: background = [[0, 1, 2, 3], [5, 6, 7, 8]] observation = [100, 101, 102, 103] nsamples = [3, 2]

X (output) [[1, 0, 1, 1], [0, 1, 1, 0]]

dataset (output): [[100, 1, 102, 103], [100, 6, 102, 103] [0, 101, 102, 3], [5, 101, 102, 8]]

permutation_shap_dataset() [1/2]

void ML::Explainer::permutation_shap_dataset	(	const raft::handle_t &	handle,
		double *	dataset,
		const double *	background,
		int	nrows_bg,
		int	ncols,
		const double *	row,
		int *	idx,
		bool	row_major
	)

permutation_shap_dataset() [2/2]

void ML::Explainer::permutation_shap_dataset	(	const raft::handle_t &	handle,
		float *	dataset,
		const float *	background,
		int	nrows_bg,
		int	ncols,
		const float *	row,
		int *	idx,
		bool	row_major
	)

Generates a dataset by tiling the background matrix into out, while adding a forward and backward permutation pass of the observation row on the positions defined by idx. Example:

background = [[0, 1, 2], [3, 4, 5], [6, 7, 8]] idx = [2, 0, 1] row = [100, 101, 102] output: [[ 0, 1, 2] [ 3, 4, 5] [ 6, 7, 8] [ 0, 1, 102] [ 3, 4, 102] [ 6, 7, 102] [100, 1, 102] [100, 4, 102] [100, 7, 102] [100, 101, 102] [100, 101, 102] [100, 101, 102] [100, 101, 2] [100, 101, 5] [100, 101, 8] [ 0, 101, 2] [ 3, 101, 5] [ 6, 101, 8] [ 0, 1, 2] [ 3, 4, 5] [ 6, 7, 8]]

Parameters

[in]	handle	cuML handle
[out]	dataset	generated data in either row major or column major format, depending on the row_major parameter [on device] [dim = (2 * ncols * nrows_bg + nrows_bg) * ncols]
[in]	background	background data [on device] [dim = ncols * nrows_bg]
[in]	nrows_bg	number of rows in background dataset
[in]	ncols	number of columns
[in]	row	row to scatter in a permutated fashion [dim = ncols]
[in]	idx	permutation indexes [dim = ncols]
[in]	row_major	boolean to generate either row or column major data

shap_main_effect_dataset() [1/2]

void ML::Explainer::shap_main_effect_dataset	(	const raft::handle_t &	handle,
		double *	dataset,
		const double *	background,
		int	nrows_bg,
		int	ncols,
		const double *	row,
		int *	idx,
		bool	row_major
	)

shap_main_effect_dataset() [2/2]

void ML::Explainer::shap_main_effect_dataset	(	const raft::handle_t &	handle,
		float *	dataset,
		const float *	background,
		int	nrows_bg,
		int	ncols,
		const float *	row,
		int *	idx,
		bool	row_major
	)

Generates a dataset by tiling the background matrix into out, while adding a forward and backward permutation pass of the observation row on the positions defined by idx. Example:

background = [[0, 1, 2], [3, 4, 5], [6, 7, 8]] idx = [2, 0, 1] row = [100, 101, 102] output: [[ 0, 1, 2] [ 3, 4, 5] [ 6, 7, 8] [ 0, 1, 102] [ 3, 4, 102] [ 6, 7, 102] [100, 1, 2] [100, 4, 5] [100, 7, 8] [ 0, 101, 2] [ 3, 101, 5] [ 6, 101, 8]]

Parameters

[in]	handle	cuML handle
[out]	dataset	generated data [on device] [dim = (2 * ncols * nrows_bg + nrows_bg) * ncols]
[in]	background	background data [on device] [dim = ncols * nrows_bg]
[in]	nrows_bg	number of rows in background dataset
[in]	ncols	number of columns
[in]	row	row to scatter in a permutated fashion [dim = ncols]
[in]	idx	permutation indexes [dim = ncols]
[in]	row_major	boolean to generate either row or column major data

update_perm_shap_values() [1/2]

void ML::Explainer::update_perm_shap_values	(	const raft::handle_t &	handle,
		double *	shap_values,
		const double *	y_hat,
		const int	ncols,
		const int *	idx
	)

update_perm_shap_values() [2/2]

void ML::Explainer::update_perm_shap_values	(	const raft::handle_t &	handle,
		float *	shap_values,
		const float *	y_hat,
		const int	ncols,
		const int *	idx
	)

Function that aggregates averages of the averatge of results of the model called with the permutation dataset, to estimate the SHAP values. It is equivalent to the Python code: for i,ind in enumerate(idx): shap_values[ind] += y_hat[i + 1] - y_hat[i] for i,ind in enumerate(idx): shap_values[ind] += y_hat[i + ncols] - y_hat[i + ncols + 1]

Parameters

[in]	handle	cuML handle
[out]	shap_values	Array where the results are aggregated [dim = ncols]
[in]	y_hat	Results to use for the aggregation [dim = ncols + 1]
[in]	ncols	number of columns
[in]	idx	permutation indexes [dim = ncols]