fiftyone.utils.eta.ETAModel

class documentation

class ETAModel(fom.Model, fom.EmbeddingsMixin, fom.LogitsMixin): (source)

Constructor: ETAModel(config, _model)

Wrapper for running an eta.core.learning.Model model.

Parameters
config	an `ETAModelConfig`

Class Method	`from_eta_model`	Builds an `ETAModel` for running the provided `eta.core.learning.Model` instance.
Method	`__enter__`	Undocumented
Method	`__exit__`	Undocumented
Method	`__init__`	Undocumented
Method	`embed`	Generates an embedding for the given data.
Method	`embed_all`	Generates embeddings for the given iterable of data.
Method	`get_embeddings`	Returns the embeddings generated by the last forward pass of the model.
Method	`predict`	Performs prediction on the given data.
Method	`predict_all`	Performs prediction on the given iterable of data.
Method	`preprocess.setter`	Undocumented
Instance Variable	`config`	Undocumented
Property	`has_embeddings`	Whether this instance can generate embeddings.
Property	`has_logits`	Whether this instance can generate logits for its predictions.
Property	`media_type`	The media type processed by the model.
Property	`preprocess`	Whether to apply `transforms` during inference (True) or to assume that they have already been applied (False).
Property	`ragged_batches`	True/False whether `transforms` may return tensors of different sizes. If True, then passing ragged lists of data to `predict_all` is not allowed.
Property	`transforms`	The preprocessing function that will/must be applied to each input before prediction, or `None` if no preprocessing is performed.
Method	`_ensure_embeddings`	Undocumented
Method	`_parse_predictions`	Undocumented
Instance Variable	`_model`	Undocumented

Inherited from Model:

Property can_embed_prompts Whether this instance can generate prompt embeddings.

Inherited from LogitsMixin (via Model, EmbeddingsMixin):

Method	`store_logits.setter`	Undocumented
Property	`store_logits`	Whether the model should store logits in its predictions.
Instance Variable	`_store_logits`	Undocumented

@classmethod
def from_eta_model(cls, model): (source) ¶

Builds an ETAModel for running the provided eta.core.learning.Model instance.

Parameters
model	an `eta.core.learning.Model` instance
Returns
an `ETAModel`

def __enter__(self): (source) ¶

overrides fiftyone.core.models.Model.__enter__

Undocumented

def __exit__(self, *args): (source) ¶

overrides fiftyone.core.models.Model.__exit__

Undocumented

def __init__(self, config, _model=None): (source) ¶

overrides fiftyone.core.models.LogitsMixin.__init__

Undocumented

def embed(self, arg): (source) ¶

overrides fiftyone.core.models.EmbeddingsMixin.embed

Generates an embedding for the given data.

Subclasses can override this method to increase efficiency, but, by default, this method simply calls predict and then returns get_embeddings.

Parameters
arg	the data. See `predict` for details
Returns
a numpy array containing the embedding

def embed_all(self, args): (source) ¶

overrides fiftyone.core.models.EmbeddingsMixin.embed_all

Generates embeddings for the given iterable of data.

Subclasses can override this method to increase efficiency, but, by default, this method simply iterates over the data and applies embed to each.

Parameters
args	an iterable of data. See `predict_all` for details
Returns
a numpy array containing the embeddings stacked along axis 0

def get_embeddings(self): (source) ¶

overrides fiftyone.core.models.EmbeddingsMixin.get_embeddings

Returns the embeddings generated by the last forward pass of the model.

By convention, this method should always return an array whose first axis represents batch size (which will always be 1 when predict was last used).

Returns
a numpy array containing the embedding(s)

def predict(self, arg): (source) ¶

overrides fiftyone.core.models.Model.predict

Performs prediction on the given data.

Image models should support, at minimum, processing arg values that are uint8 numpy arrays (HWC).

Video models should support, at minimum, processing arg values that are eta.core.video.VideoReader instances.

Parameters
arg	the data
Returns
a `fiftyone.core.labels.Label` instance or dict of `fiftyone.core.labels.Label` instances containing the predictions

def predict_all(self, args): (source) ¶

overrides fiftyone.core.models.Model.predict_all

Performs prediction on the given iterable of data.

Image models should support, at minimum, processing args values that are either lists of uint8 numpy arrays (HWC) or numpy array tensors (NHWC).

Video models should support, at minimum, processing args values that are lists of eta.core.video.VideoReader instances.

Subclasses can override this method to increase efficiency, but, by default, this method simply iterates over the data and applies predict to each.

Parameters
args	an iterable of data
Returns
a list of `fiftyone.core.labels.Label` instances or a list of dicts of `fiftyone.core.labels.Label` instances containing the predictions

@preprocess.setter
def preprocess(self, value): (source) ¶

overrides fiftyone.core.models.Model.preprocess.setter

Undocumented

config = (source) ¶

Undocumented

@property
has_embeddings = (source) ¶

overrides fiftyone.core.models.Model.has_embeddings

Whether this instance can generate embeddings.

This method returns False by default. Methods that can generate embeddings will override this via implementing the EmbeddingsMixin interface.

@property
has_logits = (source) ¶

overrides fiftyone.core.models.Model.has_logits

Whether this instance can generate logits for its predictions.

This method returns False by default. Methods that can generate logits will override this via implementing the LogitsMixin interface.

@property
media_type = (source) ¶

overrides fiftyone.core.models.Model.media_type

The media type processed by the model.

Supported values are "image" and "video".

@property
preprocess = (source) ¶

overrides fiftyone.core.models.Model.preprocess

Whether to apply transforms during inference (True) or to assume that they have already been applied (False).

@property
ragged_batches = (source) ¶

overrides fiftyone.core.models.Model.ragged_batches

True/False whether transforms may return tensors of different sizes. If True, then passing ragged lists of data to predict_all is not allowed.

@property
transforms = (source) ¶

overrides fiftyone.core.models.Model.transforms

The preprocessing function that will/must be applied to each input before prediction, or None if no preprocessing is performed.

def _ensure_embeddings(self): (source) ¶

Undocumented

def _parse_predictions(self, eta_labels_or_batch): (source) ¶

Undocumented

_model = (source) ¶

Undocumented