riptable.Utils.conversion_utils

Functions

append_dataset_dict(ds_dict, key_field_name)

Converts a dictionary of Datasets to a single Dataset appending the dictionary

dataset_as_matrix(ds[, save_metadata, column_data])

dset_dict_to_list(ds_dict, key_field_name[, ...])

Converts a dict of Datasets to a list, appending the keyname as a new field key_field_name.

numpy2d_to_dict(arr, columns)

Converts arr 2D ndarray and column names to arr dict (is ordered) of ndarray's

numpy_array_to_dataset(inarray[, columns])

numpy_array_to_dict(inarray[, columns])

riptable.Utils.conversion_utils.append_dataset_dict(ds_dict, key_field_name)

Converts a dictionary of Datasets to a single Dataset appending the dictionary keys as key_field_name to distinguish them. NB: This modifies the original Datasets!

TODO: add support for harmonizing fields, e.g.,

fill missing values and deal with differing types

Parameters:

  • ds_dict – dictionary of Datasets. Keys MUST be ascii strings (or bytes)!

  • key_field_name – New column to add to each to which Dataset will be assigned the constant value of the key.

Returns:

New dataset.

riptable.Utils.conversion_utils.dataset_as_matrix(ds, save_metadata=True, column_data={})
riptable.Utils.conversion_utils.dset_dict_to_list(ds_dict, key_field_name, allow_overwrite=False)

Converts a dict of Datasets to a list, appending the keyname as a new field key_field_name. NB: This modifies the Datasets!

TODO: allow option of inplace or copy

Parameters:

  • ds_dict – dictionary of Datasets. Keys MUST be ascii strings (or bytes)!

  • key_field_name – New column to add to each to which Dataset will be assigned the constant value of the key.

  • allow_overwrite – Unless set to True the key_field_name may not exist in any of the input Datasets.

Returns:

list of original Datasets _modified_.

riptable.Utils.conversion_utils.numpy2d_to_dict(arr, columns)

Converts arr 2D ndarray and column names to arr dict (is ordered) of ndarray’s suitable for the Dataset constructor:

Parameters:

  • arr – numpy NxM ndarray

  • columns – list of M column names

Returns:

dictionary suitable for rt.Dataset constructor

Example: import numpy as np import riptable as rt from rt.Utils.conversion_utils import numpy2d_to_dict

arr = np.array([[1, 2, 3], [4, 5, 6]]) columns = [‘c1’, ‘c2’, ‘c3’] dset = rt.Dataset(numpy2d_to_dict(arr, columns)) print(dset)

riptable.Utils.conversion_utils.numpy_array_to_dataset(inarray, columns=None)
riptable.Utils.conversion_utils.numpy_array_to_dict(inarray, columns=None)