fiftyone.server.utils

module documentation

(source)

FiftyOne Server utils.

Function	`cache_dataset`	Caches the given dataset.
Function	`change_label_tags`	Applies the changes to tags to all labels in the specified label field(s) of the collection, if necessary.
Function	`change_sample_tags`	Applies the changes to tags to all samples of the collection, if necessary.
Function	`from_dict`	Wrapping function for `dacite.from_dict` that ensures a common configuration is used.
Function	`load_and_cache_dataset`	Loads the dataset with the given name and caches it.
Function	`meets_type`	Determines whether the field meets type or types, or the field is a `fiftyone.core.fields.ListField` that meets the type or types
Function	`_parse_changes`	Undocumented
Variable	`_cache`	Undocumented
Variable	`_dacite_config`	Undocumented

def cache_dataset(dataset): (source) ¶

Caches the given dataset.

This method ensures that subsequent calls to fiftyone.core.dataset.load_dataset in async calls will return this dataset singleton.

See load_and_cache_dataset for additional details.

Parameters
dataset	a `fiftyone.core.dataset.Dataset`

def change_label_tags(sample_collection, changes, label_fields=None): (source) ¶

Applies the changes to tags to all labels in the specified label field(s) of the collection, if necessary.

Parameters
sample_collection	a `fiftyone.core.collections.SampleCollection`
changes	a dict of tags as keys and bools as values. A `True` value adds the tag to all labels, if necessary. A `False` value removes the tag from all labels, if necessary
label_fields:`None`	an optional name or iterable of names of `fiftyone.core.labels.Label` fields. By default, all label fields are used

def change_sample_tags(sample_collection, changes): (source) ¶

Applies the changes to tags to all samples of the collection, if necessary.

Parameters
sample_collection	a `fiftyone.core.collections.SampleCollection`
changes	a dict of tags as keys and bools as values. A `True` value adds the tag to all samples, if necessary. A `False` value removes the tag from all samples, if necessary

def from_dict(data_class: type[T], data: Data) -> T: (source) ¶

Wrapping function for dacite.from_dict that ensures a common configuration is used.

Parameters
data_class:`type[T]`	a dataclass
data:`Data`	the data with which to instantiate the dataclass instance
Returns
`T`	a dataclass instance

def load_and_cache_dataset(name): (source) ¶

Loads the dataset with the given name and caches it.

This method is a wrapper around fiftyone.core.dataset.load_dataset that stores a reference to every dataset it loads in a TTL cache to ensure that references to recently used datasets exist in memory so that dataset objects aren't garbage collected between async calls.

It is desirable to avoid dataset objects being garbage collected because datasets are singletons and may have objects (eg brain results) that are expensive to load cached on them.

Parameters
name	the dataset name
Returns
a `fiftyone.core.dataset.Dataset`

def meets_type(field: fof.Field, type_or_types): (source) ¶

Determines whether the field meets type or types, or the field is a fiftyone.core.fields.ListField that meets the type or types

Parameters
field:`fof.Field`	a class:`fiftyone.core.fields.Field`
type_or_types	Undocumented
type	a field type or `tuple` of field types

def _parse_changes(changes): (source) ¶

Undocumented

_cache = (source) ¶

Undocumented

_dacite_config = (source) ¶

Undocumented