Result Extraction APIs#
-
da_status da_handle_get_result_int(const da_handle handle, da_result query, da_int *dim, da_int *result)#
-
da_status da_handle_get_result_s(const da_handle handle, da_result query, da_int *dim, float *result)#
-
da_status da_handle_get_result_d(const da_handle handle, da_result query, da_int *dim, double *result)#
Get results stored in a da_handle.
Some solvers will store relevant data in the handle. These functions provide a means to extract it. To check the available data stored by a given API check its associated documentation.
- Parameters
handle – a valid handle used to call any of the solvers.
query – the data of interest, see da_result. If the result is not available or not found in the
handlethen the function returns da_status_unknown_query.dim – the size of the array
result. Ifdimis too small, then on exit it will be overwritten with the correct size and the function will return da_status_invalid_array_dimension.result – location of the array in which to store the data.
- Returns
da_status. The function returns:
da_status_success - the operation completed successfully.
da_status_unknown_query - the
queryis either not available or not found in thehandle. This can happen if you try an extract a result before performing the operation (for example, extracting the coefficient of a linear regression before actually performing a successful fit), or if the handle is of the wrong type (for example, a handle initialized for linear models cannot contain results about a principal component analysis).da_status_wrong_type - the floating point precision used to initialize the
handledoes not match the precision ofresult.da_status_handle_not_initialized - the
handlehas not been initialized or is corrupted.da_status_invalid_array_dimension - the size
dimof theresultarray is too small. After the calldimcontains the correct size.da_status_invalid_pointer - the pointer to
handleis invalid.
da_result enums#
-
typedef enum da_result_ da_result#
Alias for the da_result_ enum.
-
enum da_result_#
Enumeration defining which result to extract from a handle.
Values:
-
enumerator da_rinfo#
General information array, containing a variety of metrics. See each solver’s documentation for further information, since each solver stores different information in this array.
-
enumerator da_linmod_coef#
Optimal fitted coefficients produced by the last call to a linear regression solver.
-
enumerator da_pca_scores#
Matrix of scores computed by the PCA API.
-
enumerator da_pca_variance#
The variance explained by each component computed by the PCA API.
-
enumerator da_pca_principal_components#
Matrix of principal components computed by the PCA API.
-
enumerator da_pca_total_variance#
The total variance of the data matrix supplied to the PCA API.
-
enumerator da_pca_column_means#
Column means of the data matrix supplied to the PCA API.
-
enumerator da_pca_column_sdevs#
Column standard deviations of the data matrix supplied to the PCA API.
-
enumerator da_pca_u#
The matrix \(U\) from the singular value decomposition \(A = U\Sigma V^T\), computed as part of a PCA computation.
-
enumerator da_pca_sigma#
The nonzero diagonal entries of \(\Sigma\) from the singular value decomposition \(A = U\Sigma V^T\), computed as part of a PCA computation.
-
enumerator da_pca_vt#
The matrix \(V^T\) from the singular value decomposition \(A = U\Sigma V^T\), computed as part of a PCA computation.
-
enumerator da_kmeans_cluster_centres#
Matrix of cluster centres computed in k-means clustering.
-
enumerator da_kmeans_labels#
Labels of samples in the data matrix used to compute k-means clustering.
-
enumerator da_knn_model_params#
Model parameters for the trained and fitted k-nearest neighbors.
-
enumerator da_rinfo#