deep_bottleneck.eval_tools package

Submodules

deep_bottleneck.eval_tools.artifact module

class deep_bottleneck.eval_tools.artifact.Artifact(name, file)[source]

Bases: object

Displays or saves an artifact.

content
extension = ''
save()[source]
class deep_bottleneck.eval_tools.artifact.CSVArtifact(name, file)[source]

Bases: deep_bottleneck.eval_tools.artifact.Artifact

Displays and saves a CSV artifact

extension = 'csv'
show()[source]
class deep_bottleneck.eval_tools.artifact.MP4Artifact(name, file)[source]

Bases: deep_bottleneck.eval_tools.artifact.Artifact

Displays or saves a MP4 artifact

extension = 'mp4'
show()[source]
class deep_bottleneck.eval_tools.artifact.PNGArtifact(name, file)[source]

Bases: deep_bottleneck.eval_tools.artifact.Artifact

Displays or saves a PNG artifact.

extension = 'png'
img
show(figsize=(10, 10))[source]

deep_bottleneck.eval_tools.experiment module

class deep_bottleneck.eval_tools.experiment.Experiment(id_, database, grid_filesystem, config, artifact_links, metric_links)[source]

Bases: object

artifact_name_to_cls = {'activations': <class 'deep_bottleneck.eval_tools.artifact.PNGArtifact'>, 'infoplane': <class 'deep_bottleneck.eval_tools.artifact.PNGArtifact'>, 'infoplane_movie': <class 'deep_bottleneck.eval_tools.artifact.MP4Artifact'>, 'information_measures': <class 'deep_bottleneck.eval_tools.artifact.CSVArtifact'>, 'single_neuron_activations': <class 'deep_bottleneck.eval_tools.artifact.PNGArtifact'>, 'snr': <class 'deep_bottleneck.eval_tools.artifact.PNGArtifact'>}
artifacts

The artifacts belonging to the experiment.

Returns:A mapping from artifact names to artifact objects, that belong to the experiment.
classmethod from_db_object(database, grid_filesystem, experiment_data: dict)[source]
metrics

The metrics belonging to the experiment.

Returns:A mapping from metric names to pandas Series objects, that belong to the experiment.

deep_bottleneck.eval_tools.experiment_loader module

class deep_bottleneck.eval_tools.experiment_loader.ExperimentLoader(mongo_uri='mongodb://<MONGO_INITDB_ROOT_USERNAME>:<MONGO_INITDB_ROOT_PASSWORD>@<server_ip_address>:27017/?authMechanism=SCRAM-SHA-1', db_name='<MONGO_DATABASE>')[source]

Bases: object

Loads artifacts related to experiments.

find_by_config_key[source]

Find experiments based on regex search against an configuration value.

A partial match between configuration value and regex is enough to find the experiment.

Parameters:
  • key – Configuration key to search on.
  • value – Regex that is matched against the experiment’s configuration.
Returns:

The matched experiments.

find_by_id[source]

Find experiment based on its id.

Parameters:experiment_id – The id of the experiment.
Returns:The experiment corresponing to the id.
find_by_ids(experiment_ids: Iterable[int]) → List[deep_bottleneck.eval_tools.experiment.Experiment][source]

Find experiments based on a collection of ids.

Parameters:experiment_ids – Iterable of experiment ids.
Returns:The experiments corresponding to the ids.
find_by_name[source]

Find experiments based on regex search against its name.

A partial match between experiment name and regex is enough to find the experiment.

Parameters:name – Regex that is matched against the experiment name.
Returns:The matched experiments.

deep_bottleneck.eval_tools.utils module

deep_bottleneck.eval_tools.utils.find_differing_config_keys(experiments: Iterable[deep_bottleneck.eval_tools.experiment.Experiment])[source]

Find the config keys that were assigned to different values in a cohort of experiments..

deep_bottleneck.eval_tools.utils.format_config(config, *config_keys)[source]

Module contents