loader Module
Utilities for loading data sets.
@author: drusk
-
pml.data.loader.load(path, has_ids=True, has_header=True, has_labels=True, delimiter=', ')[source]
Loads a data set from a delimited text file.
- Args:
- path:
- the path to the file containing the data set.
- has_ids: boolean
- set to False if the first column in the loaded dataset should not be
interpreted as a feature instead of sample identifiers. Defaults to
True, i.e. first column are interpreted as sample identifiers.
- has_header: boolean
- set to False if the data being loaded does not have column headers on
the first line. Defaults to true.
- has_labels: boolean
- set to False if the data being loaded does not have classification
labels for each sample. Defaults to True. The labels should be the
last column in the dataset being loaded.
- delimiter: string
- the symbol used to separate columns in the file. Default value is
‘,’. Hint: delimiter for tab-delimited files is ‘ ‘.
- Returns:
- A DataSet object.
-
pml.data.loader.shell_load(path, has_ids=True, has_header=True, has_labels=True, delimiter=', ')[source]
Loads a data set from a delimited text file. Will search through sample
data sets.
- Args:
- path:
- the path to the file containing the data set.
- has_ids: boolean
- set to False if the first column in the loaded dataset should not be
interpreted as a feature instead of sample identifiers. Defaults to
True, i.e. first column are interpreted as sample identifiers.
- has_header: boolean
- set to False if the data being loaded does not have column headers on
the first line. Defaults to true.
- has_labels: boolean
- set to False if the data being loaded does not have classification
labels for each sample. Defaults to True. The labels should be the
last column in the dataset being loaded.
- delimiter: string
- the symbol used to separate columns in the file. Default value is
‘,’. Hint: delimiter for tab-delimited files is ‘ ‘.
- Returns:
- A DataSet object.