PLSA.data package

Submodules

PLSA.data.processing module

Module for processing data

The function of this Module is served for processing data.

PLSA.data.processing.cut_groups(data, col, cutoffs)

Cut data into subsets according to cutoffs

Parameters:
  • data (pandas.DataFrame) – Data to split.
  • col (str) – Name of column in data to compare with.
  • cutoffs (list(int)) – List of cutoffs, like as [min-value, 30, 60, max-value].
Returns:

List of sub-data as DataFrame.
Return type:

list(pandas.DataFrame)

Examples

>>> cut_groups(data, "X", [0, 0.4, 0.6, 1.0])
[pandas.DataFrame, pandas.DataFrame, pandas.DataFrame]
PLSA.data.processing.parse_surv(x, label)

Parse raw-data for survival analyze(Deep Surival).

Parameters:
  • x (np.array) – two-dimension array indicating variables.
  • label (dict) –

    Contain ‘e’, ‘t’.

    examples as {‘e’: np.array, ‘t’: np.array}.
Returns:

Sorted (x, e, t) tuple, index of people who is failure or at risk, and type of ties.
Return type:

tuple

Examples

>>> parse_surv(data[x_cols].values, {'e': data['e'].values, 't': data['t'].values})
PLSA.data.processing.prepare_surv(x, label)

Prepare data for survival analyze(Deep Surival).

Parameters:
  • x (numpy.array) – Two-dimension array indicating variables.
  • label (dict) –

    Contain ‘e’, ‘t’.

    examples as {‘e’: np.array, ‘t’: np.array}.
Returns:

Sorted (x, label) tuple of survival data.
Return type:

tuple

Examples

>>> prepare_surv(data[x_cols].values, {'e': data['e'].values, 't': data['t'].values})

Module contents