DeepCore 1.4 Now Supports ONNX
DeepCore supports the Open Neural Network Exchange (ONNX) format, an open standard created by a consortium of major companies including Maxar (www.onnx.ai). Facebook, Microsoft, AWS, Nvidia and others started supporting ONNX to foster portability of neural network models. For the last few years deep learning algorithms were built and trained in frameworks with libraries that made deployment in production systems a challenge. TensorFlow, PyTorch, Caffe, and other machine learning libraries have been widely supported by strong open source communities. However, this required systems like DeepCore to support each framework natively. With the release of DeepCore 1.4, some of this challenge is handled by ONNX support based on ONNX 1.5 and the v10 ONNX operator set.
DeepCore model metadata was extended to support ONNX packages. The DeepCore specific metadata is stored on the ONNX model metadata properties. For more information on the metadata fields, please refer to the model packaging reference documents. Prioritizing implementation relevant to current customer requirements, DeepCore supports ONNX capabilities based on matching capabilities in DeepCore’s integrated machine learning frameworks. The supported operators are described in the DeepCore ONNX documentation.
ONNX support in DeepCore expands the universe of models able to run in DeepCore. We’d love to hear from you on your experiences with the ONNX format.