Censor

Censor is a batch testing and repairing library for scientific data.

Features

• Checks for correctness of data
• Repair data for errors
• Provide reports on data health

Installation

clone censor from GitHub repository:

git clone https://github.com/tomography/censor

then:

cd censor

python setup.py install

Usage

checking data:

Create check:parameters dictionary:

checks_dir =

{censor.common.constants.IS_NPARRAY:(),

censor.common.constants.HAS_NO_NEGATIVE:(),

censor.common.constants.HAS_NO_NAN:(),

censor.common.constants.IS_INT:(),

censor.common.constants.IS_SIZE:(360, 1024, 1024) }

Run the verification:

result = censor.checks.check(arr, checks_dir[data_tag, logger, axis])

Fixing data:

Create fixer:parameters dictionary:

fixers_dir =

{censor.common.constants.REPLACE_NEGATIVE:(0),

censor.common.constants.REPLACE_NAN: (0),

censor.common.constants.TO_TYPE:(np.dtype(np.cfloat))}

Run the repair:

arr = censor.repairs.replace(arr, fixers_dir, [data_tag, logger])