fiftyone.zoo.models

package documentation

(source)

The FiftyOne Model Zoo.

Module torch FiftyOne Zoo models provided by torchvision:torchvision.models.

From __init__.py:

Class	`HasZooModel`	Mixin class for Config classes of `fiftyone.core.models.Model` instances whose models are stored in the FiftyOne Model Zoo.
Class	`RemoteModelManager`	Undocumented
Class	`RemoteModelManagerConfig`	Undocumented
Class	`RemoteZooModel`	Undocumented
Class	`RemoteZooModelsManifest`	Class that describes the collection of remotely-sourced models in the FiftyOne Model Zoo.
Class	`ZooModel`	Class describing a model in the FiftyOne Model Zoo.
Class	`ZooModelsManifest`	Class that describes the collection of models in the FiftyOne Model Zoo.
Function	`delete_zoo_model`	Deletes the zoo model from local disk, if necessary.
Function	`delete_zoo_model_source`	Deletes the specified remote source and all downloaded models associated with it.
Function	`download_zoo_model`	Downloads the specified model from the FiftyOne Model Zoo.
Function	`ensure_zoo_model_requirements`	Ensures that the package requirements for the specified zoo model are satisfied.
Function	`find_zoo_model`	Returns the path to the zoo model on disk.
Function	`get_zoo_model`	Returns the `ZooModel` instance for the specified zoo model.
Function	`install_zoo_model_requirements`	Installs any package requirements for the specified zoo model.
Function	`is_zoo_model_downloaded`	Determines whether the zoo model of the given name is downloaded.
Function	`list_downloaded_zoo_models`	Returns information about the zoo models that have been downloaded.
Function	`list_zoo_model_sources`	Returns the list of remote model sources that are registered locally.
Function	`list_zoo_models`	Returns the list of available models in the FiftyOne Model Zoo.
Function	`load_zoo_model`	Loads the specified model from the FiftyOne Model Zoo.
Function	`register_zoo_model_source`	Registers a remote source of models, if necessary.
Constant	`MODELS_MANIEST_FILENAME`	Undocumented
Variable	`logger`	Undocumented
Function	`_download_archive`	Undocumented
Function	`_download_model_metadata`	Undocumented
Function	`_download_remote_model`	Undocumented
Function	`_get_cache_key`	Undocumented
Function	`_get_exact_model`	Undocumented
Function	`_get_latest_model`	Undocumented
Function	`_get_model`	Undocumented
Function	`_get_model_in_dir`	Undocumented
Function	`_import_zoo_module`	Undocumented
Function	`_iter_model_manifests`	Undocumented
Function	`_list_zoo_models`	Undocumented
Function	`_load_remote_model`	Undocumented
Function	`_load_zoo_models_manifest`	Undocumented
Function	`_merge_manifest`	Undocumented
Function	`_merge_remote_manifest`	Undocumented
Function	`_normalize_ref`	Undocumented
Function	`_parse_model_identifier`	Undocumented
Constant	`_BUILTIN_MODELS_MANIFEST_PATT`	Undocumented
Constant	`_MODELS`	Undocumented
Constant	`_THIS_DIR`	Undocumented

def delete_zoo_model(name): (source) ¶

Deletes the zoo model from local disk, if necessary.

Parameters
name	the name of the zoo model, which can have `@<ver>` appended to refer to a specific version of the model

def delete_zoo_model_source(url_or_gh_repo): (source) ¶

Deletes the specified remote source and all downloaded models associated with it.

Parameters

url_or_gh_repo

the remote source to delete, which can be:

a GitHub repo URL like https://github.com/<user>/<repo>
a GitHub ref like https://github.com/<user>/<repo>/tree/<branch> or https://github.com/<user>/<repo>/commit/<commit>
a GitHub ref string like <user>/<repo>[/<ref>]
a publicly accessible URL of an archive (eg zip or tar) file

def download_zoo_model(name_or_url, model_name=None, overwrite=False): (source) ¶

Downloads the specified model from the FiftyOne Model Zoo.

If the model is already downloaded, it is not re-downloaded unless overwrite == True is specified.

Note

To download from a private GitHub repository that you have access to, provide your GitHub personal access token by setting the GITHUB_TOKEN environment variable.

Parameters
name_or_url	the name of the zoo model to download, which can have `@<ver>` appended to refer to a specific version of the model, or the remote source to download it from, which can be: a GitHub repo URL like `https://github.com/<user>/<repo>` a GitHub ref like `https://github.com/<user>/<repo>/tree/<branch>` or `https://github.com/<user>/<repo>/commit/<commit>` a GitHub ref string like `<user>/<repo>[/<ref>]` a publicly accessible URL of an archive (eg zip or tar) file
model_name:`None`	the specific model to download, if `name_or_url` is a remote source
overwrite:`False`	whether to overwrite any existing files
Returns
tuple of	model: the `ZooModel` for the model model_path: the path to the downloaded model on disk

def ensure_zoo_model_requirements(name, error_level=None, log_success=True): (source) ¶

Ensures that the package requirements for the specified zoo model are satisfied.

Parameters
name	the name of the zoo model, which can have `@<ver>` appended to refer to a specific version of the model
error_level:`None`	the error level to use when installing/ensuring requirements, defined as: 0: raise error if a requirement is not satisfied 1: log warning if a requirement is not satisfied 2: ignore unsatisfied requirements By default, `fo.config.requirement_error_level` is used
log_success:`True`	whether to generate a log message when a requirement is satisfied

def find_zoo_model(name): (source) ¶

Returns the path to the zoo model on disk.

The model must be downloaded. Use download_zoo_model to download models.

Parameters
name	the name of the zoo model, which can have `@<ver>` appended to refer to a specific version of the model
Returns
the path to the model on disk
Raises
`ValueError`	if the model does not exist or has not been downloaded

def get_zoo_model(name): (source) ¶

Returns the ZooModel instance for the specified zoo model.

Parameters
name	the name of the zoo model, which can have `@<ver>` appended to refer to a specific version of the model
Returns
a `ZooModel`

def install_zoo_model_requirements(name, error_level=None): (source) ¶

Installs any package requirements for the specified zoo model.

Parameters

name the name of the zoo model, which can have @<ver> appended to refer to a specific version of the model

error_level:None

the error level to use, defined as:

0: raise error if a requirement install fails
1: log warning if a requirement install fails
2: ignore install fails requirements

By default, fo.config.requirement_error_level is used

def is_zoo_model_downloaded(name): (source) ¶

Determines whether the zoo model of the given name is downloaded.

Parameters
name	the name of the zoo model, which can have `@<ver>` appended to refer to a specific version of the model
Returns
True/False

def list_downloaded_zoo_models(): (source) ¶

Returns information about the zoo models that have been downloaded.

Returns
a dict mapping model names to (model path, `ZooModel`) tuples

def list_zoo_model_sources(): (source) ¶

Returns the list of remote model sources that are registered locally.

Returns
the list of remote sources

def list_zoo_models(tags=None, source=None): (source) ¶

Returns the list of available models in the FiftyOne Model Zoo.

Also includes models from any remote sources that you've registered.

Example usage:

import fiftyone as fo
import fiftyone.zoo as foz

#
# List all zoo models
#

names = foz.list_zoo_models()
print(names)

#
# List all zoo models with the specified tag(s)
#

names = foz.list_zoo_models(tags="torch")
print(names)

Parameters
tags:`None`	only include models that have the specified tag or list of tags
source:`None`	only include models available via the given remote source
Returns
a list of model names

def load_zoo_model(name_or_url, model_name=None, download_if_necessary=True, ensure_requirements=True, install_requirements=False, error_level=None, cache=True, **kwargs): (source) ¶

Loads the specified model from the FiftyOne Model Zoo.

By default, the model will be downloaded if necessary, and any documented package requirements will be checked to ensure that they are installed.

Note

To download from a private GitHub repository that you have access to, provide your GitHub personal access token by setting the GITHUB_TOKEN environment variable.

Parameters
name_or_url	the name of the zoo model to load, which can have `@<ver>` appended to refer to a specific version of the model, or the remote source to load it from, which can be: a GitHub repo URL like `https://github.com/<user>/<repo>` a GitHub ref like `https://github.com/<user>/<repo>/tree/<branch>` or `https://github.com/<user>/<repo>/commit/<commit>` a GitHub ref string like `<user>/<repo>[/<ref>]` a publicly accessible URL of an archive (eg zip or tar) file
model_name:`None`	the specific model to load, if `name_or_url` is a remote source
download_if_necessary:`True`	whether to download the model if necessary
ensure_requirements:`True`	whether to ensure any requirements are installed before loading the model
install_requirements:`False`	whether to install any requirements before loading the model
error_level:`None`	the error level to use when installing/ensuring requirements, defined as: 0: raise error if a requirement is not satisfied 1: log warning if a requirement is not satisfied 2: ignore unsatisfied requirements By default, `fo.config.requirement_error_level` is used
cache:`True`	whether to store a weak reference to the model so that running this method again will return the same instance while the model is still in use
**kwargs	keyword arguments to inject into the model's `Config` instance
Returns
a `fiftyone.core.models.Model`

def register_zoo_model_source(url_or_gh_repo, overwrite=False): (source) ¶

Registers a remote source of models, if necessary.

Note

To download from a private GitHub repository that you have access to, provide your GitHub personal access token by setting the GITHUB_TOKEN environment variable.

Parameters

url_or_gh_repo

the remote source to register, which can be:

a GitHub repo URL like https://github.com/<user>/<repo>
a GitHub ref like https://github.com/<user>/<repo>/tree/<branch> or https://github.com/<user>/<repo>/commit/<commit>
a GitHub ref string like <user>/<repo>[/<ref>]
a publicly accessible URL of an archive (eg zip or tar) file

overwrite:False whether to overwrite any existing files

MODELS_MANIEST_FILENAME: str = (source) ¶

Undocumented

Value

'manifest.json'

logger = (source) ¶

Undocumented

def _download_archive(url, outdir): (source) ¶

Undocumented

def _download_model_metadata(url_or_gh_repo, overwrite=False): (source) ¶

Undocumented

def _download_remote_model(model_name, model_path): (source) ¶

Undocumented

def _get_cache_key(name, **kwargs): (source) ¶

Undocumented

def _get_exact_model(name): (source) ¶

Undocumented

def _get_latest_model(base_name): (source) ¶

Undocumented

def _get_model(name_or_url, model_name=None): (source) ¶

Undocumented

def _get_model_in_dir(name_or_url, model_name=None): (source) ¶

Undocumented

def _import_zoo_module(model_dir): (source) ¶

Undocumented

def _iter_model_manifests(root_dir=None): (source) ¶

Undocumented

def _list_zoo_models(tags=None, source=None): (source) ¶

Undocumented

def _load_remote_model(model_name, model_path, **kwargs): (source) ¶

Undocumented

def _load_zoo_models_manifest(): (source) ¶

Undocumented

def _merge_manifest(manifest, manifest_path, sources=None): (source) ¶

Undocumented

def _merge_remote_manifest(manifest, sources, manifest_path): (source) ¶

Undocumented

def _normalize_ref(url_or_gh_repo): (source) ¶

Undocumented

def _parse_model_identifier(url_or_gh_repo, overwrite=False): (source) ¶

Undocumented

_BUILTIN_MODELS_MANIFEST_PATT = (source) ¶

Undocumented

Value

os.path.join(_THIS_DIR, 'manifest-*.json')

_MODELS = (source) ¶

Undocumented

Value

weakref.WeakValueDictionary()

_THIS_DIR = (source) ¶

Undocumented

Value

os.path.dirname(os.path.abspath(__file__))