avocado.Dataset

class avocado.Dataset(name, metadata, observations=None, objects=None, chunk=None, num_chunks=None, object_class=<class 'avocado.astronomical_object.AstronomicalObject'>)

A dataset of many astronomical objects.

Parameters:

• name (str) – Name of the dataset. This will be used to determine the filenames of various outputs such as computed features and predictions.
• metadata (pandas.DataFrame) – DataFrame where each row is the metadata for an object in the dataset. See AstronomicalObject for details.
• observations (pandas.DataFrame) – Observations of all of the objects’ light curves. See AstronomicalObject for details.
• objects (list) – A list of AstronomicalObject instances. Either this or observations can be specified but not both.
• chunk (int (optional)) – If the dataset was loaded in chunks, this indicates the chunk number.
• num_chunks (int (optional)) – If the dataset was loaded in chunks, this is the total number of chunks used.

__init__(name, metadata, observations=None, objects=None, chunk=None, num_chunks=None, object_class=<class 'avocado.astronomical_object.AstronomicalObject'>)

Create a new Dataset from a set of metadata and observations

Methods

__init__(name, metadata[, observations, …])	Create a new Dataset from a set of metadata and observations
extract_raw_features(featurizer[, keep_models])	Extract raw features from the dataset.
from_objects(name, objects, **kwargs)	Load a dataset from a list of AstronomicalObject instances.
get_bands()	Return a list of all of the bands in the dataset.
get_models_path([tag])	Return the path to where the models for this dataset should lie on disk
get_object([index, object_class, object_id])	Parse keywords to pull a specific object out of the dataset
get_predictions_path([classifier])	Return the path to where the predicitons for this dataset for a given classifier should lie on disk.
get_raw_features_path([tag])	Return the path to where the raw features for this dataset should lie on disk
label_folds([num_folds, random_state])	Separate the dataset into groups for k-folding
load(name[, metadata_only, chunk, …])	Load a dataset that has been saved in HDF5 format in the data directory.
load_predictions([classifier])	Load the predictions for a classifier from disk.
load_raw_features([tag])	Load the raw features from disk.
plot_interactive()	Make an interactive plot of the light curves in the dataset.
plot_light_curve(*args, **kwargs)	Plot the light curve for an object in the dataset.
predict(classifier)	Generate predictions using a classifier.
read_object(object_id[, object_class])	Read an object with a given object_id.
select_features(featurizer)	Select features from the dataset for classification.
write([overwrite])	Write the dataset out to disk.
write_models([tag])	Write the models of the light curves to disk.
write_predictions([classifier])	Write predictions for this classifier to disk.
write_raw_features([tag])	Write the raw features out to disk.

Attributes

path	Return the path to where this dataset should lie on disk