​
AWS Lambda

​
wrappers

​
editor_agent

def editor_agent(
    *,
    label_row_metadata_include_args: LabelRowMetadataIncludeArgs | None = None,
    label_row_initialise_labels_args: LabelRowInitialiseLabelsArgs
    | None = None
) -> Callable[[AgentFunction], Callable[[Dict[str, Any], Any], Dict[str,
                                                                    Any]]]
Wrapper to make resources available for AWS Lambda editor agents.

​
Dependencies

​
DAssetPath

Get a local file path to data asset temporarily stored till end of agent execution.

​
dep_asset

def dep_asset(storage_item: StorageItem) -> Generator[Path, None, None]
Returns a local file path to the data asset, temporarily stored for the duration of the agent’s execution. This dependency fetches the underlying data asset using a signed URL. The asset is temporarily stored on disk for the duration of the task and is automatically removed once the task completes. Example: 
from encord_agents.gcp import editor_agent
from encord_agents.gcp.dependencies import dep_asset
...
runner = Runner(project_hash="<project_hash_a>")

@editor_agent()
def my_agent(
    asset: Annotated[Path, Depends(dep_asset)]
) -> None:
    asset.stat()  # read file stats
    ...
Returns: The path to the asset. Raises:
  • ValueError - if the underlying assets are not videos, images, or audio.
  • EncordException - if data type not supported by SDK yet.

​
DEncordClient

Get an authenticated user client.

​
dep_client

def dep_client() -> EncordUserClient
Dependency to provide an authenticated user client. Example: 
from encord.user_client import EncordUserClient
from encord_agents.gcp import editor_agent
from encord_agents.gcp.dependencies import dep_client
...
@editor_agent()
def (
    client: Annotated[EncordUserClient, Depends(dep_client)]
):
    # Client will authenticated and ready to use.
    client.get_dataset("")

​
DObjectCrops

Get all object crops that the agent was triggered on. The instance crop contains the object instance, the frame content (pixel values), and the frame.

​
dep_object_crops

def dep_object_crops(
    filter_ontology_objects: list[Object | str] | None = None
) -> Callable[[FrameData, LabelRowV2, NDArray[np.uint8]], list[InstanceCrop]]
Returns a list of object instances and frame crops associated with each object. One example use-case is to run each crop against a model. Example: 
@editor_agent
def my_agent(crops: Annotated[list[InstanceCrop], Depends[dep_object_crops(filter_ontology_objects=["eBw/75bg"])]]):
    for crop in crops:
        crop.content  # <- this is raw numpy rgb values
        crop.frame    # <- this is the frame number in video
        crop.instance # <- this is the object instance from the label row
        crop.b64_encoding()  # <- a base64 encoding of the image content
    ...
Arguments:
  • filter_ontology_objects - Specify a list of ontology objects to include. If provided, only instances of these object types are included. Strings are matched against feature_node_hashes.
  • Returns - The dependency to be injected into the cloud function.

​
DObjectsInstances

Get all object instances that the agent was triggered on. No pixels, just the annotation.

​
dep_objects

Description not available in Core reference MDX.

​
DSingleFrame

Get the single frame that the agent was triggered on.

​
dep_single_frame

def dep_single_frame(storage_item: StorageItem,
                     frame_data: FrameData) -> NDArray[np.uint8]
Dependency to inject the first frame of the underlying asset. The downloaded asset will be named lr.data_hash.{suffix}. When the function has finished running, the downloaded file is removed from the file system. Example: 
from encord_agents import FrameData
from encord_agents.gcp import editor_agent
from encord_agents.gcp.dependencies import dep_single_frame
...

@editor_agent()
def my_agent(
    frame: Annotated[NDArray[np.uint8], Depends(dep_single_frame)]
):
    assert frame.ndim == 3, "Will work"
Arguments:
  • storage_item - The Storage item. Automatically injected (see example above).
Returns: Numpy array of shape [h, w, 3] RGB colors.

​
DStorageItem

Get the storage item associated with the underlying agent task to, for example, read/write client metadata or read data properties.

​
dependencies.shares

​
DataLookup Objects

class DataLookup()
!!! warning “Deprecated” DataLookup is deprecated and will be removed in version 0.2.10. Migration Guide:
  • For accessing storage items, use dep_storage_item instead: 
    # Old way (deprecated)
from encord_agents.core.dependencies.shares import DataLookup
lookup: Annotated[DataLookup, Depends(dep_data_lookup)]
storage_item = lookup.get_storage_item(data_hash)

# New way (recommended)
from encord_agents.tasks.dependencies import dep_storage_item
# or from encord_agents.aws.dependencies import dep_storage_item
# or from encord_agents.gcp.dependencies import dep_storage_item
# or from encord_agents.fastapi.dependencies import dep_storage_item
storage_item: Annotated[StorageItem, Depends(dep_storage_item)]

​
dep_storage_item

def dep_storage_item(storage_item: StorageItem) -> StorageItem
Get the storage item associated with the underlying agent task. The StorageItem is useful for multiple things like
  • Updating client metadata
  • Reading file properties like storage location, fps, duration, DICOM tags, etc.
Example 
from typing_extensions import Annotated
from encord.storage import StorageItem
from encord_agents.gcp import editor_agent, Depends
from encord_agents.gcp.dependencies import dep_storage_item

@editor_agent()
def my_agent(storage_item: Annotated[StorageItem, Depends(dep_storage_item)]):
    print("uuid", storage_item.uuid)
    print("client_metadata", storage_item.client_metadata)
    ...

​
DVideoIterator

Get a video frame iterator for doing things over many frames.

​
dep_video_iterator

def dep_video_iterator(
        storage_item: StorageItem) -> Generator[Iterator[Frame], None, None]
Dependency to inject a video frame iterator for performing operations over many frames. Example: 
from encord_agents import FrameData
from encord_agents.gcp import editor_agent
from encord_agents.gcp.dependencies import dep_video_iterator
...

@editor_agent()
def my_agent(
    video_frames: Annotated[Iterator[Frame], Depends(dep_video_iterator)]
):
    for frame in video_frames:
        print(frame.frame, frame.content.shape)
Arguments:
  • storage_item - Automatically injected storage item dependency.
Raises:
  • NotImplementedError - Fails for data types other than video.
Yields: An iterator.

​
GCP

​
wrappers

​
editor_agent

def editor_agent(
    *,
    label_row_metadata_include_args: LabelRowMetadataIncludeArgs | None = None,
    label_row_initialise_labels_args: LabelRowInitialiseLabelsArgs
    | None = None,
    custom_cors_regex: str | None = None
) -> Callable[[AgentFunction], Callable[[Request], Response]]
Wrapper to make resources available for gcp editor agents. The editor agents are intended to be used with dependency injections. Arguments:
  • label_row_metadata_include_args - arguments to overwrite default arguments on project.list_label_rows_v2().
  • label_row_initialise_labels_args - Arguments to overwrite default arguments on label_row.initialise_labels(...)
  • custom_cors_regex - A regex to use for the CORS settings. If not provided, the default regex will be used. Only required if the agent is not deployed on Encord’s platform.
Returns: A wrapped function suitable for gcp functions.

​
Dependencies

​
DAssetPath

Get a local file path to data asset temporarily stored till end of agent execution.

​
dep_asset

def dep_asset(storage_item: StorageItem) -> Generator[Path, None, None]
Returns a local file path to the data asset, temporarily stored for the duration of the agent’s execution. This dependency fetches the underlying data asset using a signed URL. The asset is temporarily stored on disk for the duration of the task and is automatically removed once the task completes. Example: 
from encord_agents.gcp import editor_agent
from encord_agents.gcp.dependencies import dep_asset
...
runner = Runner(project_hash="<project_hash_a>")

@editor_agent()
def my_agent(
    asset: Annotated[Path, Depends(dep_asset)]
) -> None:
    asset.stat()  # read file stats
    ...
Returns: The path to the asset. Raises:
  • ValueError - if the underlying assets are not videos, images, or audio.
  • EncordException - if data type not supported by SDK yet.

​
DEncordClient

Get an authenticated user client.

​
dep_client

def dep_client() -> EncordUserClient
Dependency to provide an authenticated user client. Example: 
from encord.user_client import EncordUserClient
from encord_agents.gcp import editor_agent
from encord_agents.gcp.dependencies import dep_client
...
@editor_agent()
def (
    client: Annotated[EncordUserClient, Depends(dep_client)]
):
    # Client will authenticated and ready to use.
    client.get_dataset("")

​
DObjectCrops

Get all object crops that the agent was triggered on. The instance crop contains the object instance, the frame content (pixel values), and the frame.

​
dep_object_crops

def dep_object_crops(
    filter_ontology_objects: list[Object | str] | None = None
) -> Callable[[FrameData, LabelRowV2, NDArray[np.uint8]], list[InstanceCrop]]
Returns a list of object instances and frame crops associated with each object. One example use-case is to run each crop against a model. Example: 
@editor_agent
def my_agent(crops: Annotated[list[InstanceCrop], Depends[dep_object_crops(filter_ontology_objects=["eBw/75bg"])]]):
    for crop in crops:
        crop.content  # <- this is raw numpy rgb values
        crop.frame    # <- this is the frame number in video
        crop.instance # <- this is the object instance from the label row
        crop.b64_encoding()  # <- a base64 encoding of the image content
    ...
Arguments:
  • filter_ontology_objects - Specify a list of ontology objects to include. If provided, only instances of these object types are included. Strings are matched against feature_node_hashes.
  • Returns - The dependency to be injected into the cloud function.

​
DObjectsInstances

Get all object instances that the agent was triggered on. No pixels, just the annotation.

​
dep_objects

Description not available in Core reference MDX.

​
DSingleFrame

Get the single frame that the agent was triggered on.

​
dep_single_frame

def dep_single_frame(storage_item: StorageItem,
                     frame_data: FrameData) -> NDArray[np.uint8]
Dependency to inject the first frame of the underlying asset. The downloaded asset will be named lr.data_hash.{suffix}. When the function has finished running, the downloaded file is removed from the file system. Example: 
from encord_agents import FrameData
from encord_agents.gcp import editor_agent
from encord_agents.gcp.dependencies import dep_single_frame
...

@editor_agent()
def my_agent(
    frame: Annotated[NDArray[np.uint8], Depends(dep_single_frame)]
):
    assert frame.ndim == 3, "Will work"
Arguments:
  • storage_item - The Storage item. Automatically injected (see example above).
Returns: Numpy array of shape [h, w, 3] RGB colors.

​
DStorageItem

Get the storage item associated with the underlying agent task to, for example, read/write client metadata or read data properties.

​
dependencies.shares

​
DataLookup Objects

class DataLookup()
!!! warning “Deprecated” DataLookup is deprecated and will be removed in version 0.2.10. Migration Guide:
  • For accessing storage items, use dep_storage_item instead: 
    # Old way (deprecated)
from encord_agents.core.dependencies.shares import DataLookup
lookup: Annotated[DataLookup, Depends(dep_data_lookup)]
storage_item = lookup.get_storage_item(data_hash)

# New way (recommended)
from encord_agents.tasks.dependencies import dep_storage_item
# or from encord_agents.aws.dependencies import dep_storage_item
# or from encord_agents.gcp.dependencies import dep_storage_item
# or from encord_agents.fastapi.dependencies import dep_storage_item
storage_item: Annotated[StorageItem, Depends(dep_storage_item)]

​
dep_storage_item

def dep_storage_item(storage_item: StorageItem) -> StorageItem
Get the storage item associated with the underlying agent task. The StorageItem is useful for multiple things like
  • Updating client metadata
  • Reading file properties like storage location, fps, duration, DICOM tags, etc.
Example 
from typing_extensions import Annotated
from encord.storage import StorageItem
from encord_agents.gcp import editor_agent, Depends
from encord_agents.gcp.dependencies import dep_storage_item

@editor_agent()
def my_agent(storage_item: Annotated[StorageItem, Depends(dep_storage_item)]):
    print("uuid", storage_item.uuid)
    print("client_metadata", storage_item.client_metadata)
    ...

​
DVideoIterator

Get a video frame iterator for doing things over many frames.

​
dep_video_iterator

def dep_video_iterator(
        storage_item: StorageItem) -> Generator[Iterator[Frame], None, None]
Dependency to inject a video frame iterator for performing operations over many frames. Example: 
from encord_agents import FrameData
from encord_agents.gcp import editor_agent
from encord_agents.gcp.dependencies import dep_video_iterator
...

@editor_agent()
def my_agent(
    video_frames: Annotated[Iterator[Frame], Depends(dep_video_iterator)]
):
    for frame in video_frames:
        print(frame.frame, frame.content.shape)
Arguments:
  • storage_item - Automatically injected storage item dependency.
Raises:
  • NotImplementedError - Fails for data types other than video.
Yields: An iterator.

​
FastAPI

​
cors

Convenience method to easily extend FastAPI servers with the appropriate CORS Middleware to allow interactions from the Encord platform.

​
EncordCORSMiddleware Objects

class EncordCORSMiddleware(CORSMiddleware)
Like a regular fastapi.middleware.cors.CORSMiddleware but matches against the Encord origin by default and handles X-Encord-Editor-Agent test header Example: The CORS middleware will allow POST requests from the Encord domain. 
from fastapi import FastAPI
from encord_agents.fastapi.cors import EncordCORSMiddleware

app = FastAPI()
app.add_middleware(EncordCORSMiddleware)

​
EncordTestHeaderMiddleware Objects

class EncordTestHeaderMiddleware(BaseHTTPMiddleware)

​
dispatch

async def dispatch(request: Request,
                   call_next: RequestResponseEndpoint) -> Response
Middleware to handle the X-Encord-Editor-Agent test header. Arguments: request (Request): call_next (RequestResponseEndpoint): Returns: Response

​
get_encord_app

def get_encord_app(*, custom_cors_regex: str | None = None) -> FastAPI
Get a FastAPI app with the Encord middleware. Arguments:
  • custom_cors_regex str | None, optional - A regex to use for the CORS middleware. Only necessary if you are not using the default Encord domain.
Returns:
  • FastAPI - A FastAPI app with the Encord middleware.

​
utils

​
verify_auth

def verify_auth() -> None
FastAPI lifecycle start hook to fail early if access key is missing. Example: This will make the server fail early if auth is not set up. 
from fastapi import FastAPI

app = FastAPI(
    on_startup=[verify_auth]

​
dependencies

Dependencies for injection in FastAPI servers. This module contains dependencies that you can inject within your api routes. Dependencies that depend on others don’t need to be used together. They’ll work just fine alone. Note that you can also use the function parameter: 
from typing_extensions import Annotated
from fastapi import Form
from encord_agents import FrameData
...
@app.post("/my-agent-route")
def my_agent(
    frame_data: FrameData,
):
    ...
FrameData is automatically injected via the api request body.

​
dep_client

def dep_client() -> EncordUserClient
Dependency to provide an authenticated user client. Example: 
from encord.user_client import EncordUserClient
from encord_agents.fastapi.dependencies import dep_client
...
@app.post("/my-route")
def my_route(
    client: Annotated[EncordUserClient, Depends(dep_client)]
):
    # Client will authenticated and ready to use.

​
dep_label_row_with_args

def dep_label_row_with_args(
    label_row_metadata_include_args: LabelRowMetadataIncludeArgs | None = None,
    label_row_initialise_labels_args: LabelRowInitialiseLabelsArgs
    | None = None
) -> Callable[[FrameData], LabelRowV2]
Dependency to provide an initialized label row. Example: 
from encord_agents.core.data_model import LabelRowMetadataIncludeArgs, LabelRowInitialiseLabelsArgs
from encord_agents.fastapi.dependencies import dep_label_row_with_args
...

include_args = LabelRowMetadataIncludeArgs(
    include_client_metadata=True,
    include_workflow_graph_node=True,
)
init_args = LabelRowInitialiseLabelsArgs(
    include_signed_url=True,
)

@app.post("/my-route")
def my_route(
    lr: Annotated[LabelRowV2, Depends(dep_label_row_with_args(include_args, init_args))]
):
    assert lr.is_labelling_initialised  # will work
    assert lr.client_metadata           # will be available if set already
Arguments:
  • label_row_metadata_include_args - What arguments to include on the metadata front
  • label_row_initialise_labels_args - How and whether to initialize the label rows
Returns: The initialized label row.

​
dep_label_row

def dep_label_row(frame_data: FrameData) -> LabelRowV2
Dependency to provide an initialized label row. Example: 
from encord_agents.fastapi.dependencies import dep_label_row
...

@app.post("/my-route")
def my_route(
    lr: Annotated[LabelRowV2, Depends(dep_label_row)]
):
    assert lr.is_labelling_initialised  # will work
Arguments:
  • frame_data - the frame data from the route. This parameter is automatically injected if it’s a part of your route (see example above)
Returns: The initialized label row.

​
dep_storage_item

def dep_storage_item(
    label_row: Annotated[LabelRowV2, Depends(dep_label_row)],
    user_client: Annotated[EncordUserClient,
                           Depends(dep_client)]) -> StorageItem
Get the storage item associated with the underlying agent task. The StorageItem is useful for multiple things like
  • Updating client metadata
  • Reading file properties like storage location, fps, duration, DICOM tags, etc.
Example 
from encord.storage import StorageItem
from encord_agents.fastapi.dependencies import dep_storage_item

@app.post("/my-agent")
def my_agent(
    storage_item: Annotated[StorageItem, Depends(dep_storage_item)]
):
    # Client will authenticated and ready to use.
    print(storage_item.dicom_study_uid)
    print(storage_item.client_metadata)

​
dep_single_frame

def dep_single_frame(storage_item: Annotated[StorageItem,
                                             Depends(dep_storage_item)],
                     frame_data: FrameData) -> NDArray[np.uint8]
Dependency to inject the underlying asset of the frame data. The downloaded asset will be named lr.data_hash.{suffix}. When the function has finished, the downloaded file will be removed from the file system. Example: 
from encord_agents.fastapi.dependencies import dep_single_frame
...

@app.post("/my-route")
def my_route(
    frame: Annotated[NDArray[np.uint8], Depends(dep_single_frame)]
):
    assert arr.ndim == 3, "Will work"
Arguments:
  • storage_item - The label row. Automatically injected (see example above).
  • frame_data - the frame data from the route. This parameter is automatically injected if it’s a part of your route (see example above).
  • Returns - Numpy array of shape [h, w, 3] RGB colors.

​
dep_asset

def dep_asset(
    storage_item: Annotated[
        StorageItem,
        Depends(dep_storage_item),
    ]
) -> Generator[Path, None, None]
Get a local file path to data asset temporarily stored till end of agent execution. This dependency will fetch the underlying data asset based on a signed url. It will temporarily store the data on disk. Once the task is completed, the asset will be removed from disk again. Example: 
from encord_agents.fastapi.dependencies import dep_asset
...
runner = Runner(project_hash="<project_hash_a>")

@app.post("/my-route")
def my_agent(
    asset: Annotated[Path, Depends(dep_asset)],
) -> str | None:
    asset.stat()  # read file stats
    ...
Returns: The path to the asset. Raises:
  • ValueError - if the underlying assets are not videos, images, or audio.
  • EncordException - if data type not supported by SDK yet.

​
dep_video_iterator

def dep_video_iterator(
    storage_item: Annotated[StorageItem,
                            Depends(dep_storage_item)]
) -> Generator[Iterator[Frame], None, None]
Dependency to inject a video frame iterator for doing things over many frames. Example: 
from encord_agents.fastapi.dependencies import dep_video_iterator, Frame
...

@app.post("/my-route")
def my_route(
    video_frames: Annotated[Iterator[Frame], Depends(dep_video_iterator)]
):
    for frame in video_frames:
        print(frame.frame, frame.content.shape)
Arguments:
  • storage_item - Automatically injected storage item dependency.
Raises:
  • NotImplementedError - Will fail for other data types than video.
Yields: An iterator.

​
dep_project

def dep_project(
        frame_data: FrameData,
        client: Annotated[EncordUserClient,
                          Depends(dep_client)]) -> Project
Dependency to provide an instantiated Project. Example: 
from encord.project import Project
from encord_agents.fastapi.dependencies import dep_project
...
@app.post("/my-route")
def my_route(
    project: Annotated[Project, Depends(dep_project)]
):
    # Project will authenticated and ready to use.
    print(project.title)
Arguments: frame_data: client:

​
dep_data_lookup

def dep_data_lookup(
        lookup: Annotated[DataLookup, Depends(_lookup_adapter)]) -> DataLookup
Get a lookup to easily retrieve data rows and storage items associated with the given task. !!! warning “Deprecated” dep_data_lookup is deprecated and will be removed in version 0.2.10. Use dep_storage_item instead for accessing storage items. Migration Guide: 
## Old way (deprecated)
from encord_agents.fastapi.dependencies import dep_data_lookup, DataLookup

@app.post("/my-agent")
def my_agent(
    frame_data: FrameData,
    lookup: Annotated[DataLookup, Depends(dep_data_lookup)]
):
    storage_item = lookup.get_storage_item(frame_data.data_hash)
    print(storage_item.client_metadata)

## New way (recommended)
from encord_agents.fastapi.dependencies import dep_storage_item

@app.post("/my-agent")
def my_agent(
    frame_data: FrameData,
    storage_item: Annotated[StorageItem, Depends(dep_storage_item)]
):
    # storage_item is directly available
    print(storage_item.client_metadata)
Arguments:
  • lookup - The object that you can use to lookup data rows and storage items. Automatically injected.
Returns: The (shared) lookup object.

​
dep_object_crops

def dep_object_crops(
    filter_ontology_objects: list[Object | str] | None = None
) -> Callable[[FrameData, LabelRowV2, NDArray[np.uint8]], list[InstanceCrop]]
Create a dependency that provides crops of object instances. Useful, e.g., to be able to run each crop against a model. Example: 
@app.post("/object_classification")
async def classify_objects(
    crops: Annotated[
        list[InstanceCrop],
        Depends(dep_object_crops(filter_ontology_objects=[generic_ont_obj])),
    ],
):
    for crop in crops:
        crop.content  # <- this is raw numpy rgb values
        crop.frame    # <- this is the frame number in video
        crop.instance # <- this is the object instance from the label row
        crop.b64_encoding()  # <- a base64 encoding of the image content
    ...
Arguments:
  • filter_ontology_objects - Optional list of ontology objects to filter by. If provided, only instances of these object types will be included. Strings are matched against feature_node_hashes.
Returns: A FastAPI dependency function that yields a list of InstanceCrop.