Crate edgefirst_client 

EdgeFirst Studio Client Library

The EdgeFirst Studio Client Library provides a Rust client for interacting with EdgeFirst Studio, a comprehensive platform for computer vision and machine learning workflows. This library enables developers to programmatically manage datasets, annotations, training sessions, and other Studio resources.

Features

• Authentication: Secure token-based authentication with automatic renewal
• Dataset Management: Upload, download, and manage datasets with various file types
• Annotation Management: Create, update, and retrieve annotations for computer vision tasks
• Training & Validation: Manage machine learning training and validation sessions
• Project Organization: Organize work into projects with hierarchical structure
• Polars Integration: Optional integration with Polars DataFrames for data analysis

Quick Start

use edgefirst_client::{Client, Error};

#[tokio::main]
async fn main() -> Result<(), Error> {
    // Create a new client
    let client = Client::new()?;

    // Authenticate with username and password
    let client = client.with_login("username", "password").await?;

    // List available projects
    let projects = client.projects(None).await?;
    println!("Found {} projects", projects.len());

    Ok(())
}

Optional Features

• polars: Enables integration with Polars DataFrames for enhanced data manipulation

Modules

coco

COCO Dataset Format Support

format

EdgeFirst Dataset Format utilities.

instrument

Conditional instrumentation support for profiling.

upload_stats

Reliable, Instant-based upload timing accumulators (profiling builds only).

Macros

span

Conditional span creation macro - compiles to no-op when profiling is disabled.

trace_event

Conditional event recording macro - compiles to no-op when profiling is disabled.

Structs

Annotation

AnnotationSet

The AnnotationSet class represents a collection of annotations in a dataset. A dataset can have multiple annotation sets, each containing annotations for different tasks or purposes.

AnnotationSetID

Unique identifier for an annotation set within a dataset.

AppId

Unique identifier for an application in EdgeFirst Studio.

Artifact

Box2d

Box3d

ChangelogEntry

A single entry in the dataset changelog, recording one modification.

ChangelogResponse

Paginated response from the version.changelog endpoint.

Client

Main client for interacting with EdgeFirst Studio Server.

Dataset

A dataset in EdgeFirst Studio containing sensor data and annotations.

DatasetID

Unique identifier for a dataset within a project.

DatasetParams

DatasetSummary

Cached metrics summary for a dataset’s current state.

Experiment

ExperimentID

Unique identifier for an experiment within a project.

FileTokenStorage

File-based token storage for desktop platforms.

GpsData

GPS location data (latitude and longitude).

Group

A dataset group for organizing samples into logical subsets.

ImageId

Unique identifier for an image in EdgeFirst Studio.

ImuData

IMU orientation data (roll, pitch, yaw in degrees).

Job

A job (app run) entry returned by Client::jobs.

Label

A label used to identify annotations in a dataset.

Location

Location and pose information for a sample.

MaskData

A raster mask stored as PNG-encoded bytes.

MemoryTokenStorage

In-memory token storage (no persistence).

NewTrainingSession

Result of client::Client::start_training_session.

NewValidationSession

Result of client::Client::start_validation_session.

Organization

Organization information and metadata.

OrganizationID

Unique identifier for an organization in EdgeFirst Studio.

Polygon

PresignedUrl

A presigned URL for uploading a file to S3.

Progress

Progress information for long-running operations.

Project

The project class represents a project in the EdgeFirst Studio. A project contains datasets, experiments, and other resources related to a specific task or workflow.

ProjectID

Unique identifier for a project within EdgeFirst Studio.

RestoreResult

Result from version.tag.restore.

RestoredCounts

Counts of entities restored.

RestoredFrom

Source tag information in a restore result.

Sample

A sample in a dataset, typically representing a single image with metadata and optional sensor data.

SampleDimensionUpdate

A single sample dimension update entry.

SampleFile

A file associated with a sample (e.g., LiDAR point cloud, radar data).

SampleID

Unique identifier for a sample within a dataset.

SamplesCountResult

SamplesPopulateParams

Parameters for populating (importing) samples into a dataset.

SamplesPopulateResult

Result from the samples.populate2 API call.

SamplesUpdateDimensionsResult

Result from the samples.update_dimensions API call.

SchemaField

A single field descriptor from a trainer or validator parameter schema.

SchemaOption

One selectable option of a select SchemaField.

SequenceId

Unique identifier for a sequence in EdgeFirst Studio.

ServerAnnotation

Annotation data for the server-side annotation.add_bulk API.

Snapshot

SnapshotFromDatasetResult

Result of creating a snapshot from an existing dataset.

SnapshotID

Unique identifier for a snapshot in EdgeFirst Studio.

SnapshotRestoreResult

Stage

StartTrainingRequest

Request payload for client::Client::start_training_session.

StartValidationRequest

Inputs for client::Client::start_validation_session.

Tag

A dataset version tag, as returned by Client::dataset_tags.

Task

TaskDataList

List of data and chart artefacts attached to a task.

TaskID

Unique identifier for a task in EdgeFirst Studio.

TaskInfo

Timing

Pipeline timing measurements for a sample, in nanoseconds.

TrainerSchemaInfo

Catalog entry describing an available trainer type.

TrainingSession

TrainingSessionID

Unique identifier for a training session within an experiment.

UsageSummary

Billing usage summary for the authenticated user’s organization.

ValidationSession

ValidationSessionID

Unique identifier for a validation session within an experiment.

ValidatorSchema

A validator parameter schema, as returned by Client::validator_schemas.

VersionCurrentResponse

Response from version.current with serial, tags, and summary.

VersionTag

A named version tag that captures a complete dataset state snapshot at a specific serial number. Tags are immutable once created and enable reproducible training and validation by referencing an exact dataset state.

Enums

AnnotationType

Annotation types supported for labeling data in EdgeFirst Studio.

Error

Comprehensive error type for EdgeFirst Studio Client operations.

FileType

File types supported in EdgeFirst Studio datasets.

Parameter

Generic parameter value used in API requests and configuration.

RetryScope

Retry scope classification for URL-based retry policies.

SchemaFieldType

The kind of input a SchemaField describes.

StorageError

Error type for token storage operations.

Traits

TokenStorage

Trait for persistent token storage.

Functions

classify_url

Classifies a URL to determine which retry policy to apply.

samples_dataframe

Create a DataFrame from a slice of samples with the 2026.04 schema.

unflatten_polygon_coordinates

Unflatten coordinates with NaN separators back to nested polygon structure.