Sony Custom Layers (SCL) is an open-source project implementing detection post process NN layers not supported by the TensorFlow Keras API or Torch's torch.nn for the easy integration of those layers into pretrained models.
This section provides an installation and a quick starting guide.
To install the latest stable release of SCL, run the following command:
pip install sony-custom-layers
By default, no framework dependencies are installed. To install SCL including the latest tested dependencies (up to patch version) for TensorFlow:
pip install sony-custom-layers[tf]
To install SCL including the latest tested dependencies (up to patch version) for PyTorch/ONNX/OnnxRuntime:
pip install sony-custom-layers[torch]
| Tested FW versions | Tested Python version | Serialization |
|---|---|---|
| 2.10 | 3.8-3.10 | .h5 |
| 2.11 | 3.8-3.10 | .h5 |
| 2.12 | 3.8-3.11 | .h5 .keras |
| 2.13 | 3.8-3.11 | .keras |
| 2.14 | 3.9-3.11 | .keras |
| 2.15 | 3.9-3.11 | .keras |
| Tested FW versions | Tested Python version | Serialization |
|---|---|---|
| torch 2.0-2.4 torchvision 0.15-0.19 onnxruntime 1.15-1.19 onnxruntime_extensions 0.8-0.12 onnx 1.14-1.16 |
3.8-3.11 | .onnx (via torch.onnx.export) |
For sony-custom-layers API see https://sony.github.io/custom_layers
For TensorFlow layers see KerasAPI
To load a model with custom layers in TensorFlow, see custom_layers_scope
For PyTorch layers see PyTorchAPI
No special handling is required for torch.onnx.export and onnx.load.
For OnnxRuntime support see load_custom_ops