Evaluation

Edge Impulse APIs: INTRODUCTION Edge Impulse Documentation Copy page Copy page Copy page Copy page Welcome to Edge Impulse!

Public dataset collection for edge ML use cases.

Edge Impulse Developer Plan: INTRODUCTION Edge Impulse Documentation Copy page Copy page Copy page Copy page Welcome to Edge Impulse!

Edge Impulse Enterprise Plan: INTRODUCTION Edge Impulse Documentation Copy page Copy page Copy page Copy page Welcome to Edge Impulse!

SDKs for inference, data ingestion, and Studio interaction.

Linux-specific CLI for Edge Impulse project interaction.

Two-way protocol for controlling devices from Studio.

Data acquisition, annotation formats and file structure specs.

LLM observability platform for tracing, monitoring, and evaluating AI agents.

NVIDIA RAG Blueprint is a reference architecture for building retrieval-augmented generation applications, with deployment options for Docker, Kubernetes, and Python.

Traceloop offers two deployment models: hybrid and full self-hosting for generative AI application workflows.

Command line interface for logging in, running jobs, executing sweeps, and managing artifacts.

Query language for selecting and aggregating data in query panels.

Observability and evaluation platform for LLM apps and agents.

TypeScript SDK for W&B Weave, enabling LLM tracing and evaluation with agents, tools, and sessions.

Evaluate single-turn and multi-turn AI agents in Weave using the agents workflow and EvaluationLogger, scored with an LLM judge.

Python SDK for Weights & Biases Weave, providing classes and functions for LLM development, evaluation, and tracing.

Python and TypeScript libraries for tracing, evaluating, and improving LLM and agent applications.

Documentation index for Weights & Biases, providing links to language-specific docs and OpenAPI specs.