AOFlagger

Automatic flagging in time/freq windows using Andre Offringa’s advanced aoflagger.

inputs

type

type:

string

default:

aoflag

doc:

Case-insensitive step type; must be ‘aoflagger’ (or ‘aoflag’) .

autocorr

type:

boolean

default:

True

doc:

Flag autocorrelations ?

count.path

type:

string

default:

""

doc:

The directory where to create the flag percentages table. If empty, the path of the input MS is used .

count.save

type:

boolean

default:

False

doc:

If true, the flag percentages per frequency are saved to a table with extension .flagfreq and percentages per station to a table with extension .flagstat. The basename of the table is the MS name (without extension) followed by the stepname and extension.

keepstatistics

type:

boolean

default:

True

doc:

Write the quality statistics ?

memorymax

type:

float

doc:

Maximum amount of memory (in GB) to use. <=0 means no maximum. As stated above, this affects flagging accuracy .

default:

0

memoryperc

type:

int

default:

0

doc:

If >0, percentage of the machines memory to use. If memorymax nor memoryperc is given, all memory will be used (minus 2 GB (at most 50%) for other purposes). Accepts only integer values (LOFAR v2.16). Limiting the available memory too much affects flagging accuracy; in general try to use at least 10 GB of memory.

overlapmax

type:

int

doc:

Maximum overlap value (0 is no maximum) .

default:

0

overlapperc

type:

float

default:

0

doc:

If >0, percentage of time window to be added to the left and right side for overlap purposes (to minimize boundary effects). If overlapmax is not given, it defaults to 1%.

pedantic

type:

boolean

default:

False

doc:

Be more pedantic when flagging ?

pulsar

type:

boolean

default:

False

doc:

Use flagging strategy optimized for pulsar observations ?

step_name

type:

string

default:

aoflag

doc:

unique name for the step .

strategy

type:

File?

doc:

The name of the strategy file to use. If no name is given, the default strategy is used which is fine for HBA. For LBA data the strategy LBAdefault should be used.

timewindow

type:

int

default:

0

doc:

Number of time slots to be flagged jointly. The larger the time window, the better the flagging performs. 0 means that it will be deduced from the memory to use. Note that the time window can be extended with an overlap on the left and right side to minimize possible boundary effects .