Changelog

The MetaTF documentation is generated from the akida_examples repository.
It relies on the Sphinx Python documentation generator and GitHub Pages documentation for rendering.
Please refer to the repository release area for the full changelog.
For reference, this documentation was generated using the following packages

absl-py==2.1.0, accelerate==1.14.0, aiohappyeyeballs==2.7.1, aiohttp==3.14.3, aiosignal==1.4.0, akida==2.19.3, akida_models==1.14.2, alabaster==1.0.0, anyio==4.14.2, array_record==0.8.3, asttokens==3.0.2, astunparse==1.6.3, attrs==26.1.0, audioread==3.1.0, autodocsumm==0.2.15, babel==2.18.0, backports.zstd==1.6.0, blinker==1.4, brotli==1.2.0, certifi==2025.1.31, cffi==2.1.1, charset-normalizer==3.4.1, cnn2snn==2.19.3, coloredlogs==15.0.1, contourpy==1.3.3, cryptography==3.4.8, cycler==0.12.1, datasets==5.0.1, dbus-python==1.2.18, decorator==5.3.1, dill==0.4.1, distro==1.7.0, distro-info==1.1+ubuntu0.2, dm-tree==0.1.10, docstring_parser==0.18.0, docutils==0.22.4, einops==0.8.2, etils==1.14.0, executing==2.2.1, filelock==3.29.0, flake8==4.0.1, flatbuffers==25.2.10, fonttools==4.63.0, frozenlist==1.8.0, fsspec==2026.4.0, gast==0.6.0, google-pasta==0.2.0, googleapis-common-protos==1.75.0, grpcio==1.71.0, h11==0.16.0, h5py==3.13.0, hf-xet==1.6.0, httpcore==1.0.9, httplib2==0.20.2, httpx==0.28.1, huggingface_hub==0.36.2, humanfriendly==10.0, idna==3.10, imagecorruptions-imaug==1.1.5, ImageIO==2.37.4, imagesize==2.0.0, imaug==0.4.2, immutabledict==4.3.1, importlib-metadata==4.6.4, importlib_resources==7.1.0, inflate64==1.0.4, ipython==9.16.1, ipython_pygments_lexers==1.1.1, jams==0.3.5, jedi==0.20.0, jeepney==0.7.1, Jinja2==3.1.6, joblib==1.5.3, jsonschema==4.26.0, jsonschema-specifications==2025.9.1, keras==3.9.0, keyring==23.5.0, kiwisolver==1.5.0, launchpadlib==1.10.16, lazr.restfulclient==0.14.4, lazr.uri==1.0.6, lazy-loader==0.5, libclang==18.1.1, librosa==0.11.0, llvmlite==0.48.0, Markdown==3.7, markdown-it-py==3.0.0, MarkupSafe==3.0.2, matplotlib==3.11.1, matplotlib-inline==0.2.2, mccabe==0.6.1, mdurl==0.1.2, mediapy==1.2.7, metatf==2.19.3, mir_eval==0.8.2, ml_dtypes==0.5.4, more-itertools==8.10.0, mpmath==1.3.0, msgpack==1.2.1, mtcnn==0.1.1, multidict==6.7.1, multiprocess==0.70.19, multivolumefile==0.2.3, namex==0.0.8, narwhals==2.24.0, networkx==3.6.1, numba==0.66.0, numpy==2.1.3, nvidia-cublas-cu11==11.11.3.6, nvidia-cublas-cu12==12.5.3.2, nvidia-cuda-cupti-cu11==11.8.87, nvidia-cuda-cupti-cu12==12.5.82, nvidia-cuda-nvcc-cu12==12.5.82, nvidia-cuda-nvrtc-cu11==11.8.89, nvidia-cuda-nvrtc-cu12==12.5.82, nvidia-cuda-runtime-cu11==11.8.89, nvidia-cuda-runtime-cu12==12.5.82, nvidia-cudnn-cu11==9.1.0.70, nvidia-cudnn-cu12==9.3.0.75, nvidia-cufft-cu11==10.9.0.58, nvidia-cufft-cu12==11.2.3.61, nvidia-curand-cu11==10.3.0.86, nvidia-curand-cu12==10.3.6.82, nvidia-cusolver-cu11==11.4.1.48, nvidia-cusolver-cu12==11.6.3.83, nvidia-cusparse-cu11==11.7.5.86, nvidia-cusparse-cu12==12.5.1.3, nvidia-nccl-cu11==2.21.5, nvidia-nccl-cu12==2.23.4, nvidia-nvjitlink-cu12==12.5.82, nvidia-nvtx-cu11==11.8.86, oauthlib==3.2.0, onnx==1.22.0, onnx-ir==0.1.12, onnxruntime==1.23.2, onnxruntime_extensions==0.13.0, onnxscript==0.4.0, opencv-python==5.0.0.93, opencv-python-headless==5.0.0.93, opt_einsum==3.4.0, optimum==1.23.3, optree==0.14.1, packaging==24.2, pandas==3.0.5, parso==0.8.7, pexpect==4.9.0, pillow==12.2.0, pip==26.2.1, platformdirs==4.11.1, pooch==1.9.0, promise==2.3, prompt_toolkit==3.0.53, propcache==0.5.2, protobuf==5.29.3, psutil==7.2.2, ptyprocess==0.7.0, pure_eval==0.2.3, py7zr==1.1.3, pyarrow==25.0.1, pybcj==1.0.8, pycodestyle==2.8.0, pycparser==3.0, pycryptodomex==3.23.0, pydub==0.25.1, pyflakes==2.4.0, Pygments==2.19.1, PyGObject==3.42.1, PyJWT==2.3.0, pyparsing==3.3.2, pyppmd==1.3.1, python-apt==2.4.0+ubuntu4, python-dateutil==2.9.0.post0, PyYAML==6.0.3, quantizeml==1.2.5, referencing==0.37.0, regex==2026.7.19, requests==2.32.3, rich==13.9.4, roman-numerals==4.1.0, rpds-py==2026.6.3, safetensors==0.8.0, scikit-image==0.26.0, scikit-learn==1.9.0, scipy==1.17.1, SecretStorage==3.3.1, setuptools==76.0.0, setuptools-scm==10.2.1, shapely==2.1.2, simple-parsing==0.1.9, six==1.17.0, snowballstemmer==3.1.1, sortedcontainers==2.4.0, soundata==1.0.1, soundfile==0.14.0, soxr==1.1.0, Sphinx==9.0.4, sphinx-gallery==0.21.0, sphinx-reredirects==1.1.0, sphinx_design==0.7.0, sphinx_rtd_theme==3.1.0, sphinxcontrib-applehelp==2.0.0, sphinxcontrib-devhelp==2.0.0, sphinxcontrib-htmlhelp==2.1.0, sphinxcontrib-jquery==4.1, sphinxcontrib-jsmath==1.0.1, sphinxcontrib-qthelp==2.0.0, sphinxcontrib-serializinghtml==2.0.0, sphinxcontrib-video==0.4.2, stack-data==0.6.3, sympy==1.14.0, tenns_modules==0.3.0, tensorboard==2.19.0, tensorboard-data-server==0.7.2, tensorboardX==2.6.5, tensorflow==2.19.0, tensorflow-datasets==4.9.10, tensorflow-io-gcs-filesystem==0.37.1, tensorflow-metadata==1.17.3, termcolor==2.5.0, texttable==1.7.0, tf_keras==2.19.0, threadpoolctl==3.6.0, tifffile==2026.3.3, timm==1.0.28, tokenizers==0.20.3, toml==0.10.2, torch==2.7.0+cu118, torchinfo==1.8.0, torchvision==0.22.0+cu118, tqdm==4.70.0, traitlets==5.16.1, transformers==4.46.3, trimesh==5.0.0, triton==3.3.0, typing_extensions==4.16.0, unattended-upgrades==0.1, urllib3==2.3.0, uv==0.12.3, vcs-versioning==2.2.2, wadllib==1.3.6, wcwidth==0.8.2, Werkzeug==3.1.3, wheel==0.45.1, wrapt==1.17.2, xxhash==3.8.1, yarl==1.24.5, zipp==4.1.0

MetaTF 2.19.3

Released in August 2026. Package set: akida/cnn2snn 2.19.3, quantizeml 1.2.5, akida-models 1.14.2. Aligned with FPGA versions 1764 (2-nodes), 1765 (6-nodes), 1766 (6-nodes bittware) and 905 (Pico).

New features:

  • Improved user experience in documentation. Now coming with an extra example in Getting started <./getting_started.rst>, a new See power number <./power_number.rst> page, power number in the model zoo <./model_zoo.rst> and a version switcher.

  • A new MetaTF package is provided on PyPI to ease installation and dependecy pulling.

Bug fixes:

  • Resolved googleapis-common-protos dependency conflict

Previous versions

Previous documentation versions are also reachable from the version switcher at the bottom of the navigation sidebar. They are kept online for reference but are no longer updated.
MetaTF documentation previous versions

MetaTF Beta

Beta releases are not intended for production use and may be unstable. They are provided to allow users to test new features and provide feedback.