Metadata-Version: 2.4
Name: metatf
Version: 2.19.3
Summary: MetaTF: BrainChip Akida development tools - installs akida, cnn2snn, akida-models and quantizeml as a single validated set
Author-email: Kevin Tsiknos <ktsiknos@brainchip.com>
License-Expression: Apache-2.0
Project-URL: Documentation, https://doc.brainchipinc.com
Requires-Python: >=3.10
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: akida-models~=1.14.2
Dynamic: license-file

# MetaTF

MetaTF is BrainChip's Akida development environment, delivered as a single
meta-package. Installing `metatf` installs the complete, validated tool set:

| Package | Description |
|---|---|
| [akida](https://pypi.org/project/akida/) | Akida Execution Engine (software simulator and hardware interface) |
| [cnn2snn](https://pypi.org/project/cnn2snn/) | Keras/TensorFlow to Akida model converter |
| [akida-models](https://pypi.org/project/akida-models/) | Model zoo and training pipelines |
| [quantizeml](https://pypi.org/project/quantizeml/) | Base layers and quantization tools |

TensorFlow and tf-keras are pulled in automatically as dependencies of the
set.

`metatf` itself contains no Python code — it only pins the packages above. There is no `metatf`
module to import. Import and use the four packages directly (`akida`, `cnn2snn`, `akida_models`,
`quantizeml`).

Documentation: <https://doc.brainchipinc.com>.
