Meta data

Utility classes for meta data storage and handling.

Usage examples:

# load meta data
metadata_container = MetaDataContainer(filename='event_list.txt').load()

# Print content
print(metadata_container)

# Log content
metadata_container.log()

# Filter based on event label
print(metadata_container.filter(event_label='drawer'))

# Filter based on scene label
print(metadata_container.filter(scene_label='office'))

# Filter time segment
print(metadata_container.filter_time_segment(onset=10.0, offset=30.0))

# Add time offset
metadata_container.add_time_offset(100)

# Process events
print(metadata_container.process_events(minimum_event_length=0.5, minimum_event_gap=0.5))

# Combine content
metadata_container2 = MetaDataContainer(filename='event_list.txt').load()
metadata_container += metadata_container2

# Unique file list
metadata_container.file_list

MetaDataItem

Dict based class for storing meta data item (i.e. one row in meta data file).

MetaDataItem(\*args, \*\*kwargs)	Constructor
MetaDataItem.id	Unique item identifier
MetaDataItem.file	Filename
MetaDataItem.scene_label	Scene label
MetaDataItem.event_label	Event label
MetaDataItem.event_onset	Event onset
MetaDataItem.event_offset	Event offset
MetaDataItem.identifier	Identifier
MetaDataItem.source_label	Source label

MetaDataContainer

List of MetaDataItems for storing meta data file in one container.

Reads meta data from CSV-text files. Preferred delimiter is tab, however, other delimiters are supported automatically (they are sniffed automatically).

Supported input formats:

• [file(string)]
• [event_onset (float)][tab][event_offset (float)]
• [file(string)][scene_label(string)]
• [file(string)][scene_label(string)][identifier(string)]
• [event_onset (float)][tab][event_offset (float)][tab][event_label (string)]
• [file(string)][event_onset (float)][tab][event_offset (float)][tab][event_label (string)]
• [file(string)][event_onset (float)][tab][event_offset (float)][tab][event_label (string)][tab][identifier(string)]
• [file(string)[tab][scene_label][tab][event_onset (float)][tab][event_offset (float)][tab][event_label (string)]
• [file(string)[tab][scene_label][tab][event_onset (float)][tab][event_offset (float)][tab][event_label (string)][tab][source_label(string)]

MetaDataContainer(\*args, \*\*kwargs)
MetaDataContainer.log([level])	Log container content
MetaDataContainer.show([show_stats])	Print container content
MetaDataContainer.get_string([show_stats])	Get content in string format
MetaDataContainer.update(data)	Replace content with given list
MetaDataContainer.filter([filename, ...])	Filter content
MetaDataContainer.filter_time_segment([...])	Filter time segment
MetaDataContainer.process_events([...])	Process event content
MetaDataContainer.remove_field(field_name)	Remove field from meta items
MetaDataContainer.slice_field(field_name)	Slice field values into list
MetaDataContainer.filter_time_segment([...])	Filter time segment
MetaDataContainer.add_time_offset(offset)	Add time offset to event onset and offset timestamps
MetaDataContainer.file_list	List of unique files in the container
MetaDataContainer.event_count	Get number of events
MetaDataContainer.scene_label_count	Get number of unique scene labels
MetaDataContainer.event_label_count	Get number of unique event labels
MetaDataContainer.unique_scene_labels	Get unique scene labels
MetaDataContainer.unique_event_labels	Get unique event labels
MetaDataContainer.max_offset	Find the offset (end-time) of last event
MetaDataContainer.load([filename])	Load event list from delimited text file (csv-formated)
MetaDataContainer.save([filename, delimiter])	Save content to csv file
MetaDataContainer.event_stat_counts()	Event count statistics
MetaDataContainer.event_roll([label_list, ...])	Event roll

EventRoll

Class to convert MetaDataContainer to binary matrix indicating event activity withing time segment defined by time_resolution.

EventRoll(metadata_container[, label_list, ...])	Event roll
EventRoll.roll	Event roll
EventRoll.pad(length)	Pad event roll’s length to given length
EventRoll.plot()	Plot Event roll

ProbabilityItem

Dict based class for storing meta data item along with probability.

ProbabilityItem(\*args, \*\*kwargs)	Constructor
ProbabilityItem.id	Unique item identifier
ProbabilityItem.file	Filename
ProbabilityItem.label	Label
ProbabilityItem.timestamp	Returns:
ProbabilityItem.probability	Returns:
ProbabilityItem.get_list()	Return item values in a list with specified order.

ProbabilityContainer

List of ProbabilityItem for storing meta data along with probabilities in one container.

ProbabilityContainer(\*args, \*\*kwargs)
ProbabilityContainer.log([level])	Log container content
ProbabilityContainer.show(\*\*kwargs)	Print container content
ProbabilityContainer.update(data)	Replace content with given list
ProbabilityContainer.file_list	List of unique files in the container
ProbabilityContainer.unique_labels	Get unique labels
ProbabilityContainer.filter([filename, ...])	Filter content
ProbabilityContainer.get_string()	Get content in string format
ProbabilityContainer.load([filename])	Load probability list from delimited text file (csv-formated)
ProbabilityContainer.save([filename, delimiter])	Save content to csv file