imhr.Webgazer.processing¶

Classes

class imhr.Webgazer.processing.Processing(config, isLibrary=False, isDebug=False)[source]¶

Bases: object

Hub for running processing and analyzing raw data.

Methods

Methods

append_classify(self, df, cg_df)	Appending classification to Dataframe.
classify(self, config, df[, ctype, …])	I-DT algorithm takes into account the distribution or spatial proximity of eye position points in the eye-movement trace.
dwell(self, df[, cores, isMultiprocessing])	Calculate dwell time for sad and neutral images.
filter_data(self, df, filter_type, config)	Butterworth: Design an Nth-order digital or analog Butterworth filter and return the filter coefficients.
getData(self[, path])	preparing data for use in analysis.
getEstimatedMonitor(self, diagonal, window)	calculate estimate monitor size (w,h;cm) using estimated diagonal monitor (hypotenuse; cm).
onset_diff(self, df0[, merge, cores])	Calculate differences in onset presentation (stimulus, dotloc) using bokeh, seaborn, and pandas.
preprocess(self, df, window)	Initial data cleaning.
process(self, window, filters, gxy_df, trial)	Plotting and preparing data for classification.
roi(self[, filters, flt, df, manual, …])	Check if fixation is within bounds.
run(self, path[, task_type, single_subject, …])	Processing of data.
subject_metadata(self, fpath, spath)	Collect all subjects metadata.
variables(self, df)	Output list of variables for easy html viewing.

append_classify(self, df, cg_df)	Appending classification to Dataframe.
classify(self, config, df[, ctype, …])	I-DT algorithm takes into account the distribution or spatial proximity of eye position points in the eye-movement trace.
dwell(self, df[, cores, isMultiprocessing])	Calculate dwell time for sad and neutral images.
filter_data(self, df, filter_type, config)	Butterworth: Design an Nth-order digital or analog Butterworth filter and return the filter coefficients.
getData(self[, path])	preparing data for use in analysis.
getEstimatedMonitor(self, diagonal, window)	calculate estimate monitor size (w,h;cm) using estimated diagonal monitor (hypotenuse; cm).
onset_diff(self, df0[, merge, cores])	Calculate differences in onset presentation (stimulus, dotloc) using bokeh, seaborn, and pandas.
preprocess(self, df, window)	Initial data cleaning.
process(self, window, filters, gxy_df, trial)	Plotting and preparing data for classification.
roi(self[, filters, flt, df, manual, …])	Check if fixation is within bounds.
run(self, path[, task_type, single_subject, …])	Processing of data.
subject_metadata(self, fpath, spath)	Collect all subjects metadata.
variables(self, df)	Output list of variables for easy html viewing.

getEstimatedMonitor(self, diagonal, window)[source]¶

calculate estimate monitor size (w,h;cm) using estimated diagonal monitor (hypotenuse; cm).

Attributes:	df_raw : pandas.DataFrame Pandas dataframe of subjects.

preprocess(self, df, window)[source]¶

Initial data cleaning.

Parameters:	df : pandas.DataFrame Pandas dataframe of raw data. window : tuple horizontal, vertical resolution
Attributes:	m_delta : int Maxinum one-sample change in velocity

Notes

remove_missing:

Remove samples with null values.

remove_bounds:

Remove samples outside of window bounds (1920,1080).

remove_spikes:

remove one-sample spikes if x and y-axis delta is greater than 5.

getData(self, path=None)[source]¶

preparing data for use in analysis.

Parameters:	path : str The directory path of the subject data
Attributes:	path : str Specific directory path used.
Returns:	df : pandas.DataFrame Pandas dataframe of raw data. _path : list list of files used for analysis.

Notes

You can either get data from all subjects within a directory, or from a specific subject (subject_session).

Examples

>>> #if using path:
>>> df_raw = getData(path=self.config['path']['raw'])

>>> #if getting data for single subject:
>>> df_raw = getData(path=self.config['path']['raw'],subject_session=['1099','1', '0'])

filter_data(self, df, filter_type, config)[source]¶

Butterworth: Design an Nth-order digital or analog Butterworth filter and return the filter coefficients.

Parameters:	df : pandas.DataFrame Pandas dataframe of raw data. filter_type : str, optional Type of filter. config : dict Configuration data. i.e. trial number, location.
Attributes:	filter_type : str Filter type: ‘butterworth’

classify(self, config, df, ctype='ivt', filter_type=None, v_th=None, dr_th=None, di_th=None, missing=None, maxdist=None, mindur=None)[source]¶

I-DT algorithm takes into account the distribution or spatial proximity of eye position points in the eye-movement trace.

In the I-VT model, the velocity value is computed for every eye position sample. The velocity value is then compared to the threshold. If the sampled velocity is less than the threshold, the corresponding eye-position sample is marked as part of a fixation, otherwise it is marked as a part of a saccade.

The simple model detects fixations, defined as consecutive samples with an inter-sample distance of less than a set amount of pixels (disregarding missing data)

Parameters:	config : dict Configuration data. i.e. trial number, location. df : pandas.DataFrame Pandas dataframe of classified data. ctype : str Classification type: ‘ivt’ filter_type : [type], optional Filter type: ‘butter’ ctype : int, optional velocity threshold (ivt), dispersion threshold (idt; used by SR-Research and Tobii), or simple v_th : str Velocity threshold in pix/sec (ivt) dr_th : str Fixation duration threshold in pix/msec (idt) di_th : str Dispersion threshold in pixels (idt) missing : str value to be used for missing data (simple) maxdist : str maximal inter sample distance in pixels (simple) mindur : str minimal duration of a fixation in milliseconds; detected fixation cadidates will be disregarded if they are below this duration (simple)
Returns:	df : pandas.DataFrame Pandas dataframe of classified data.
Raises:	ValueError Unknown classification type.

roi(self, filters=None, flt=None, df=None, manual=False, monitorSize=None)[source]¶

Check if fixation is within bounds.

Attributes:	manual : str Whether or not processing.roi() is access manually. monitorSize : list Monitor size. filters : list Filter parameters. Default [[‘SavitzkyGolay’,’sg’]]. df : pandas.DataFrame Pandas dataframe of classified data.
Returns:	df : pandas.DataFrame Pandas dataframe of classified data.

process(self, window, filters, gxy_df, trial, _classify=True, ctype='simple', _param='', log=False, v_th=20, dr_th=200, di_th=20, _missing=0.0, _maxdist=25, _mindur=50)[source]¶

Plotting and preparing data for classification. Combined plot of each filter.

Parameters:	window : list horizontal, vertical resolution filters : list List of filters along with short-hand names. gxy_df : pandas.DataFrame Pandas dataframe of raw data. Unfiltered raw data. trial : str Trial number. _classify : bool parameter to include classification ctype : str classification type. simple, idt, ivt _param : str [description] (the default is ‘’, which [default_description]) log : bool [description] (the default is False, which [default_description]) v_th : str Velocity threshold in px/sec (ivt) dr_th : str Fixation duration threshold in px/msec (idt) di_th : str Dispersion threshold in px (idt) _missing : bool value to be used for missing data (simple) _maxdist : str maximal inter sample distance in pixels (simple) _mindur : str minimal duration of a fixation in milliseconds; detected fixation cadidates will be disregarded if they are below this duration (simple) (default = 100)
Attributes:	_fxy_df : pandas.DataFrame Pandas dataframe of raw data. Filtered data. Subset of _fgxy_df.
Returns:	_fgxy_df : pandas.DataFrame Pandas dataframe of filtered data. c_xy : pandas.DataFrame Pandas dataframe of classified data.

append_classify(self, df, cg_df)[source]¶

Appending classification to Dataframe.

Parameters:	df : list Pandas dataframe of raw data. gxy_df : pandas.DataFrame Pandas dataframe of raw data of classification events.

run(self, path, task_type='eyetracking', single_subject=False, single_trial=False, subject=0, trial=0, isMultiprocessing=True, cores=1)[source]¶

Processing of data. Steps here include: cleaning data, fixation identification, and exporting data.

Parameters:

path : string Path of raw data. task_type : string Running analysis on eyetracking or behavioral data. single_subject : bool Whether to run function with all or single subject. single_trial : bool Whether to run function with all or single trial. subject : int Subject number. Only if single_subject = True. trial : int Trial number. Only if single_trial = True. isMultiprocessing : bool Whether multiprocessing of data will be used. Only if single_subject = False. cores : int Number of cores to use for multiprocessing. Only if single_subject = False & isMultiprocessing=True.

Attributes:

process : bool Process all data for export.

subject_metadata(self, fpath, spath)[source]¶

Collect all subjects metadata.

Parameters:	fpath : str The directory path of all participant data. spath : str The directory path of all participant data.
Returns:	df : ndarray Pandas dataframe of subject metadata.

variables(self, df)[source]¶

Output list of variables for easy html viewing.

Parameters:	df : pandas.DataFrame Pandas dataframe of raw data. This is used as a filter to prevent unused participants from being included in the data. path : str The directory path save and read the hdf5 dataframe.
Returns:	df_definitions : pandas.DataFrame

dwell(self, df, cores=1, isMultiprocessing=False)[source]¶

Calculate dwell time for sad and neutral images.

Parameters:	df : pandas.DataFrame Pandas dataframe of raw data. This is used as a filter to prevent unused participants from being included in the data. cores : int Number of cores to use for multiprocessing.
Returns:	df : pandas.DataFrame Pandas dataframe with dwell time. error : list List of participants that were not included in dataframe.

onset_diff(self, df0, merge=None, cores=1)[source]¶

Calculate differences in onset presentation (stimulus, dotloc) using bokeh, seaborn, and pandas.

Parameters:	df0 : pandas.DataFrame Pandas dataframe of raw data. This is used to merge variables that may be useful for analysis. merge : list or None Variables to merge into returned df. cores : int Number of cores to use for multiprocessing.
Returns:	df1 : pandas.DataFrame Pandas dataframe. error : pandas.DataFrame Dataframe of each participants and the amount trials included in their data. drop : list List of participants that are 3 SD from median.