8 Releases 8 Tags

RSS Feed

• v0.1.7 dd305aabeb

  Compare

  RIA Toolkit OSS v0.1.7 - Release Notes Stable

  G gillian released this 2026-05-26 16:02:38 -04:00 | 0 commits to main since this release

  RIA Toolkit OSS v0.1.7 - Release Notes

  This release hardens the ria-agent register flow that connects on-prem RIA Agents to a RIA Hub instance. Newly registered agents now receive a friendly default name (e.g. swift-teal-falcon), and registration errors surface as actionable English instead of raw HTTP status codes. The release also fixes two real-world blockers seen in the field: Cloudflare-fronted hubs rejecting the CLI as a bot, and the CLI hanging forever when a hub is unreachable.

  🌟 New Features

  • Human-readable agent names — When --name is omitted, ria-agent register now auto-generates a memorable adjective-colour-animal name (e.g. swift-teal-falcon) via the new namegen module, instead of registering the agent with an empty string. Operators get a recognizable label for each agent without having to invent one up front.
  • Structured registration error messages — ria-agent register now translates known hub responses into clear, actionable English. The CLI recognizes the documented failure reasons — invalid_key, expired, revoked, already_consumed — as well as rate-limit responses (HTTP 429), and explains what the user should do next, rather than dumping a raw HTTP 4xx body.

  🚀 Enhancements

  • --api-key help text now documents the per-user registration flow. The help string for ria-agent register --api-key describes the personal ria_reg_* registration key flow — minted from Settings → RIA Agents on the hub and shown once at mint time — which is now the preferred way to register agents.
  • ria-agent register success output is more informative. On success, the CLI prints both the hub-assigned agent ID and the chosen name: Registered agent: (), so operators can correlate the local CLI run with the agent record on the hub.

  🛠️ Performance Improvements

  • None in this release.

  🐛 Bug Fixes

  • ria-agent register was being blocked by Cloudflare on hubs behind it. The CLI now sends an explicit User-Agent header — ria-agent/ (+https://riahub.ai/qoherent/ria-toolkit-oss) — instead of the default Python-urllib/, which Cloudflare's Browser Integrity Check rejects with an HTTP 403 (edge error code 1010). The version is read from package metadata so it tracks releases automatically.
  • ria-agent register could hang indefinitely against an unreachable hub. Added an explicit 15-second timeout on the HTTP request to the hub. Previously, urllib's default behavior (no timeout) meant a stuck or unreachable hub would block the CLI forever; the command now fails fast with a clear error.

  ⚠️ Deprecations

  • Shared [wac] API_KEY for agent registration is soft-deprecated in favor of per-user ria_reg_* registration keys minted from Settings → RIA Agents on the hub. The hub still accepts the legacy shared key for backwards compatibility, but the CLI documentation and tooling now point to the per-user flow as the recommended path.

  🗑️ Removals

  • No removals or interface changes in this release.

  🚧 Known Issues

  • None at time of release. Please report any issues via the project's issue tracker.

  💡 Future Plans

  • Continued improvements to the RIA Agent lifecycle (registration, key rotation, health reporting) as part of the broader RIA Hub ↔ on-prem agent integration.

  ℹ️ Additional Information

  SHA256 Checksums:

  7a04816c54b40759f5692671ca8f400974bfdea5f55658fc42ed4d1a6526e44c  ria_toolkit_oss-0.1.7-py3-none-any.whl
  109929b93a99abd3d4a9b20dab5a0e029e5a1c3c7aa8e61c146437f75919505b  ria_toolkit_oss-0.1.7.tar.gz
  

  Changelog: v0.1.6...v0.1.7

  If you have any questions or concerns regarding this release, wish to provide feedback, or would like to suggest an improvement to the project, please feel free to contact the project maintainers.

  Downloads

  • Source Code (ZIP)
  • Source Code (TAR.GZ)
  • ria_toolkit_oss-0.1.7-py3-none-any.whl
    489 KiB
  • ria_toolkit_oss-0.1.7.tar.gz
    42 MiB
• v0.1.6 3295b285f4

  Compare

  RIA Toolkit OSS v0.1.6 - Release Notes Stable

  G gillian released this 2026-05-13 14:48:57 -04:00 | 11 commits to main since this release

  RIA Toolkit OSS v0.1.6 - Release Notes

  This release expands the orchestration layer with full two-machine TX/RX campaign support, introduces multi-pass campaign loops, and improves reliability for long-running capture sessions.

  🌟 New Features

  TX Executor:

  • New orchestration/tx_executor.py coordinates the transmit side of a campaign, streaming a waveform from a remote RIA agent synchronized with the RX executor.

  Two-Machine TX/RX Campaigns:

  • Campaigns can now pair a local (or remote) receiver with a remote transmitter on a second machine.
  • The executor dispatches transmission via the agent API and waits for a handshake before starting capture.

  Campaign Loop Support:

  • The campaign executor now supports multi-pass/repeat loops, cycling through the full capture-label-QA pipeline for a configurable number of iterations.

  Campaign Status Reporting:

  • The agent now streams live campaign progress (current pass, captures completed, QA results) back to RIA Hub during a run.

  🚀 Enhancements

  • Improved getting started guide and installation documentation readability.

  🛠️ Performance Improvements

  • Fixed a large memory issue affecting recording and dataset processing.

  🐛 Bug Fixes

  • Fixed chunk timeout errors during long captures, improving timeout handling and error surfacing in the streaming path.
  • Fixed remote transmitter cleanup, removed stale orchestrator references that caused connection errors after campaign teardown.
  • Hardened annotation pipeline and CLI robustness.

  🗑️ Removals

  • The datatypes package has been renamed to data to align with the rest of the toolkit. Users importing from ria_toolkit_oss.datatypes will need to update their imports to ria_toolkit_oss.data.

  💡 Future Plans

  • Continued clean up of new packages and further integration across the orchestration and remote control layers.
  • Wider range of support for various transmission types and devices.
  • Expanded documentation coverage for the Agent, Orchestration, and Remote Control packages.

  ℹ️ Additional Information

  SHA256 Checksums:

  0642109ca7d24d1e1ec90dcc62c42c99bebc84799ab706063ada7fe3676b8f5a  ria_toolkit_oss-0.1.6-py3-none-any.whl
  bcf5c888ddc1bc5e1d9faed6834cb7526aa6fab4011f29e2b4d956097616e82d  ria_toolkit_oss-0.1.6.tar.gz
  

  Changelog: v0.1.5...v0.1.6

  If you have any questions or concerns regarding this release, wish to provide feedback, or would like to suggest an improvement to the project, please feel free to contact the project maintainers.

  Downloads

  • Source Code (ZIP)
  • Source Code (TAR.GZ)
  • ria_toolkit_oss-0.1.6-py3-none-any.whl
    488 KiB
  • ria_toolkit_oss-0.1.6.tar.gz
    66 MiB
• v0.1.5 50438558d4

  Compare

  RIA Toolkit OSS v0.1.5 - Release Notes Stable

  M madrigal released this 2026-04-21 11:54:30 -04:00 | 52 commits to main since this release

  RIA Toolkit OSS v0.1.5 - Release Notes

  This release adds several new packages, including Agent, Orchestration, Remote Control, and Server. Many of these
  packages have been added to provide support for RIA Hub features. Some of these packages are still in progress or need minor cleanup.

  This release also includes the Addition of the Annotation package, which includes tools for generation and manipulation of various
  types of annotations.

  🌟 New Features

  • Agent Package:

    • Provides functionality for local SDRs to communicate with RIA server.

  • Annotation Package:

    • Added automatic annotation creation tools, including the cusum annotator, energy detector, and threshold qualifier.
    • Added annotation transformation tools, including the parallel signal separator.

  • Orchestration Package:

    • Provides functionality for capture campaigns, monitoring training runs, and running live RF inference demos.

  • Remote Control Package:

    • Added preliminary package for providing remote control of a transmitter.

  • SDR Package:

    • Added option for simultaneous dual transmit on Pluto SDRs

  • Server Package:

    • Contains connection capabilities for agent package.

  • View Package:

    • Added view_annotations(), which allows users to view an annotated signal.

  • CLI:

    • Added annotate to the CLI, with all the automatic annotation generation tools in the Annotations Package.
    • Added serve to the CLI, which allows SDRs to send commands to RIA.

  🚀 Enhancements

  • Docs:

    • Added Getting Started page.

  🛠️ Performance Improvements

  - Significant performance improvements for faster rendering/processing of recording data.
  - Broad fixes and optimizations across dataset builder, H5 helpers, radio dataset, recording I/O, Pluto SDR, campaign orchestration, inference server, and signal generator.
  

  🐛 Bug Fixes

  - Fixed merging and import errors across CLI commands after initial OSS port.
  - Fixed capture CLI error messages.
  - Improved Linting standards and coding passes.
  - Resolved a race condition in the Pluto TX streaming path that could cause underruns or unsafe concurrent transmit state.
  - Fixed a bug where recordings could silently overwrite existing files.
  

  💡 Future Plans

  - Clean up of new packages and further integration.
  - Facilitating remote Reciever and Transmission Code Through RIA Capture Campaign.
  - Wider range of support for various Transmissions and Devices.
  

  ℹ️ Additional Information

  SHA256 Checksums:

  ria_toolkit_oss-0.1.5-py3-none-any.whl: 100964bd9df271a6701b67ecb3b6c7984d1525343012c00040bf3ef0ca371460
  ria_toolkit_oss-0.1.5.tar.gz: fcc98dde85d316506083426442e59d123b18eab7125aa89749ebf4c62743a98d
  

  Changelog: v0.1.4..v0.1.5

  If you have any questions or concerns regarding this release, wish to provide feedback, or would
  like to suggest an improvement to the project, please feel free to contact the project maintainers.

  Downloads

  • Source Code (ZIP)
  • Source Code (TAR.GZ)
  • ria_toolkit_oss-0.1.5-py3-none-any.whl
    489 KiB
  • ria_toolkit_oss-0.1.5.tar.gz
    395 KiB
• v0.1.4 b1e3ebf74f

  Compare

  RIA Toolkit OSS v0.1.4 - Release Notes Stable

  benchinnery released this 2026-02-02 11:02:59 -05:00 | 121 commits to main since this release

  RIA Toolkit OSS v0.1.4 - Release Notes
  Minor Update which added support for Pytorch Viewing and Parsing Widgets/

  🌟 New Features

  • PyTorch Model Visualization Support:

    • Added a new visualization module for PyTorch .py model files using AST-based static analysis.
    • Automatically detects torch.nn.Module subclasses and extracts architectural information without running model code.

  • Multiple Interactive Model Visualizations:

    • Model architecture overview (layer-by-layer visualization)
    • Model complexity analysis (layer type distribution)
    • Model metadata dashboard (layers, estimated parameters, docstring, dependencies)
    • Code structure visualization (method count, arguments, and complexity)

  🚀 Enhancements

  • Updated documentation and package version alignment to reflect wider useage of vizualizers

  🛠️ Performance Improvements

  • None in this release.

  🐛 Bug Fixes

  • None in this release.

  ℹ️ Additional Information

  SHA256 Checksums:

  ria_toolkit_oss-0.1.4-py3-none-any.whl cf35d5fbbc3a69bbd15a7c8df39558defced09fac5a45483eafe5d2e6a3c5b20
  ria_toolkit_oss-0.1.4.tar.gz: c49583cc88b0b3a13d9100b7256ed8501d1f3c40b45541c8a4ac8d8f84c4b1d0
  

  Changelog: v0.1.3..v0.1.4

  If you have any questions or concerns regarding this release, wish to provide feedback, or would
  like to suggest an improvement to the project, please feel free to contact the project maintainers.

  Downloads

  • Source Code (ZIP)
  • Source Code (TAR.GZ)
• v0.1.3 0f246e9c69

  Compare

  RIA Toolkit OSS v0.1.3 Stable

  benchinnery released this 2025-10-27 14:39:40 -04:00 | 189 commits to main since this release

  RIA Toolkit OSS v0.1.3 - Release Notes

  This release adds support for viewing recordings via a new package: ria_toolkit_oss.view. It allows users to call recording.view() or recording.simple_view() to generate and save plots of Recording objects. Support has also been added for Library asset visualization via a new package: ria_toolkit_oss.viz.

  This release also includes updates and improvements to the SDR package, including adding support for RTLSDR and ThinkRF radios.

  🌟 New Features

  • SDR Package:

    • Added support for RTLSDR and ThinkRF radios.
    • Added support for setting the bias tee in relevant radios.
    • Added get() methods to SDR for sample rate, center frequency, and gain.

  • View Package:

    • Added view_sig() and view_simple_sig(), which support recording.view() and recording.simple_view().

  • Viz Package:

    • Added Widget data Viewing & Visualization support for numerous supported asset types such as Recordings, Pytorch State Dictionaries, Radio Datasets, and Onnx Graphs.

  🚀 Enhancements

  • Docs:

    • Added new SDR packages (ThinkRF and RTLSDR).

  🛠️ Performance Improvements

  • Recording Package:

    • The to_npy() and to_sigmf() methods have been updated to include file overwrite protection.

  🐛 Bug Fixes

  • SDR Package:

    • Some of the record() and tx_recording() methods did not properly implement the ability to input either num_samples or rx/tx_time; this has been fixed.

  💡 Future Plans

  - Extending asset visualization for a wider variety of asset types and file formats.
  

  ℹ️ Additional Information

  SHA256 Checksums:

  ria_toolkit_oss-0.1.1-py3-none-any.whl: 606197289f1f38ec2ea3d0b3f992c5c36862e2ab5ac3aa55833b0f558ea76a3e
  ria_toolkit_oss-0.1.1.tar.gz: 606197289f1f38ec2ea3d0b3f992c5c36862e2ab5ac3aa55833b0f558ea76a3e
  ria-toolkit-oss-0.1.1-py_0.conda: 35a048956e9f0dab497f5fe49cde1dce63da3e51a4d31ac113db2329fcb5768e
  

  Changelog: v0.1.2..v0.1.3

  If you have any questions or concerns regarding this release, wish to provide feedback, or would
  like to suggest an improvement to the project, please feel free to contact the project maintainers.

  Downloads

  • Source Code (ZIP)
  • Source Code (TAR.GZ)
  • ria-toolkit-oss-0.1.3-py_0.conda
    99 KiB
  • ria_toolkit_oss-0.1.3-py3-none-any.whl
    120 KiB
  • ria_toolkit_oss-0.1.3.tar.gz
    101 KiB
• v0.1.2 43068c153d

  Compare

  SDR package: USRP, BladeRF, PlutoSDR, HackRF Stable

  michael released this 2025-10-03 14:24:12 -04:00 | 225 commits to main since this release

  RIA Toolkit OSS v0.1.2 - Release Notes

  This release adds support for for reception and transmission from SDR devices via a new package: ria_toolkit_oss.sdr. The following SDR devices are currently supported: USRP, BladeRF, PlutoSDR, and HackRF. This package provides a unified interface for SDR control, so the same code to works seamlessly across all supported devices.

  🌟 New Features

  • SDR Package (ria_toolkit_oss.sdr) for interfacing with SDR devices. Currently supported devices: USRP, BladeRF, PlutoSDR, and HackRF. New devices can be added by subclassing the ria_toolkit_oss.sdr.SDR base class and implementing hardware-specific reception and transmission methods.
  • Added code examples demonstrating SDR reception and transmission with bladeRF.
  • Added SDR guides summarizing supported models, device capabilities, and setup instructions for using SDRs with RIA Toolkit OSS in Radioconda on Linux.

  🚀 Enhancements

  • Reorganized the project documentation: moved the Sphinx-generated API (previously at the top-level) into its own section called "RIA Toolkit OSS". Added new top-level sections for SDR guides and usage examples.
  • Minor typo and bug fixes

  💡 Future Plans

  • Add support for the following SDR devices: RTL-SDR, HackRF, and thinkRF.
  • Add GNU Radio integration.
  • Add signal synthesis utilities, including GNU Radio integration.
  • Add our framework for building, controlling, and operating SDR testbeds.

  🙌 New Contributors

  • Gillian in e88148c312
  • Madrigal in 0f5f36b103

  ℹ️ Additional Information

  SHA256 Checksums:

  ria_toolkit_oss-0.1.1-py3-none-any.whl: f0c97a30b9b3f3f6d145f57eb22d8325c88dd32ddd5e2dd140ed0fc41e15c39e
  ria_toolkit_oss-0.1.1.tar.gz: ddbc7e7dc228b32142ff67c2288e9477e685fd87effdcd9a2512e9413602d3b7
  ria-toolkit-oss-0.1.1-py_0.conda: a90cf2f1ed4195d15e59fe360e97af4fd6f1877462d5cdfe752df9cfeaaaeefe
  

  Changelog: v0.1.1..v0.1.2

  If you have any questions or concerns regarding this release, wish to provide feedback, or would
  like to suggest an improvement to the project, please feel free to contact the project maintainers.

  Downloads

  • Source Code (ZIP)
  • Source Code (TAR.GZ)
  • ria-toolkit-oss-0.1.2-py_0.conda
    73 KiB
  • ria_toolkit_oss-0.1.2-py3-none-any.whl
    89 KiB
  • ria_toolkit_oss-0.1.2.tar.gz
    76 KiB
• v0.1.1 49857071d2

  Compare

  Conda support Stable

  michael released this 2025-09-10 15:41:13 -04:00 | 253 commits to main since this release

  RIA Toolkit OSS v0.1.1 - Release Notes

  This release adds Conda support for RIA Toolkit OSS. Conda is now the recommended way to install the toolkit.

  🌟 New Features

  • Added support for Conda. Conda packages are available on the RIA Hub Conda Package Registry: ria-toolkit-oss.

  🚀 Enhancements

  • Updated and clarified installation instructions, now including steps for installing with Conda.

  🐛 Bug Fixes

  • Minor docstring formatting fixes throughout the datasets package.

  🚧 Known Issues

  It’s still early days for RIA Toolkit OSS, and we appreciate your patience and support while we work through a few kinks. Please report any bugs to the project issues board: Issues.

  💡 Future Plans

  Please refer to the Initial Release for more details regarding future plans for this project.

  ℹ️ Additional Information

  SHA256 Checksums

  ria_toolkit_oss-0.1.1-py3-none-any.whl: 7b12f5efb4b33c46ae6c8414251f4aed03ced905a91733d9e0eb6d0187e663a1
  ria_toolkit_oss-0.1.1.tar.gz: b1cd38a6f2c7cea5d79ed63c1b4cf88b92c78936a08c36724696d78127afab2c
  ria-toolkit-oss-0.1.1-py_0.conda: 26c61454910a3895cc976d4d966dbc2ed6d71b47b13d3cf31424a19cd2ff02ab
  

  Changelog: v0.1.0..v0.1.1

  If you have any questions or concerns regarding this release, wish to provide feedback, or would
  like to suggest an improvement to the project, please feel free to contact the project maintainers.

  Downloads

  • Source Code (ZIP)
  • Source Code (TAR.GZ)
  • ria-toolkit-oss-0.1.1-py_0.conda
    58 KiB
  • ria_toolkit_oss-0.1.1-py3-none-any.whl
    65 KiB
  • ria_toolkit_oss-0.1.1.tar.gz
    58 KiB
• v0.1.0 a18d39afcd

  Compare

  Initial release Stable

  michael released this 2025-09-05 11:30:43 -04:00 | 262 commits to main since this release

  RIA Toolkit OSS - Release Notes

  Introducing RIA Toolkit OSS: An open-source version of the RIA Toolkit, providing the fundamental components to help engineers and researchers get started building, testing, and deploying radio intelligence applications.

  🌟 New Features

  On the admin side of things, we have completed the following:

  • Set up Poetry for packaging, build, and dependency management.
  • Configured pytest and tox for automated testing.
  • Set up Sphinx documentation with our our custom theme and hosted it on Read the Docs: RIA Toolkit OSS Documentation.
  • Established code style and formatting standards.
  • Added CI workflows for building the project, building the docs, testing with tox.
  • Drafted README.md file along with instructions for installation, getting started, and contributing.

  On the development side of things, we have initialized the following packages:

  • Datatypes Package:

    • Added Recording, Annotation, and RadioDataset datatypes.

  • Utils Package:

    • Added array conversion utilities for converting back and forth between the complex 1xN format and the real 2xN
      format.

  • IO Package:

    • Added basic read/write for Recording class, including support for NumPy binary files and the SigMF standard.

  • Viz Package:

    • Added functionality to create interactive Plotly visualizations for recordings, including spectrograms, time series, frequency spectra, and constellation plots.

  • Transforms Package:

    • Functionals for numerous IQ augmentation including time reversal, spectral inversion, and channel swap.
    • Functionals for numerous IQ impairments including time shift, frequency shit, and phase shift.

  🚧 Known Issues

  It’s still early days for RIA Toolkit OSS, and we appreciate your patience and support while we work out a few kinks. Please report any bugs to the project issues board: Issues.

  💡 Future Plans

  • A unified interface for interacting with software-defined radios, including USRP, BladeRF, PlutoSDR, RTL-SDR, HackRF, and thinkRF.
  • Basic model training and testing utilities.
  • Upgrading our framework for Radio Datasets to offer greater flexibility and integration with industry-standard tools, more updates coming soon!

  🙌 New Contributors

  This initial release is a collaborative effort from the Qoherent team, including various components extracted from the RIA Toolkit.

  ℹ️ Additional Information

  SHA256 Checksums

  ria_toolkit_oss-0.1.0-py3-none-any.whl: 09dd72e6a0aff172e0b1aa3d25b620905fe780e76d3f763962ec992b65490e88
  ria_toolkit_oss-0.1.0.tar.gz: 4d54544e05944ef89ab2e50058a8b2f2c0f4a1a34198b9e104a0d7103088d0ef
  

  Changelog: be7fef3..v0.1.0

  If you have any questions or concerns regarding this release, wish to provide feedback, or would
  like to suggest an improvement to the project, please feel free to contact the project maintainers.

  Downloads

  • Source Code (ZIP)
  • Source Code (TAR.GZ)
  • ria_toolkit_oss-0.1.0-py3-none-any.whl
    65 KiB
  • ria_toolkit_oss-0.1.0.tar.gz
    58 KiB