Releases: zhenhuaw-me/tflite2onnx
Releases · zhenhuaw-me/tflite2onnx
tflite2onnx 0.4.1
Update links. No functional changes.
tflite2onnx 0.4.0
2021-08-31, Project
- More operators, and API
getSupportedOperators()
to list the supported operators. - Relax data type check, most for FP16 and INT8.
- FP16 quantization: fold FP16 tensors to unblock MediaPipe models. See #35 for details.
- API
enableDebugLog()
to dump log for debugging purpose.
Thanks for the contribution of @erizmr @briangrifiin and @IkbeomJeon!
tflite2onnx 0.3.2
- New API
getSupportedOperators()
to know what operators have been supported. No longer need to maintain a list manually. - FP16 quantization: fold FP16 tensors to unblock MediaPipe models. See #35 for details.
tflite2onnx 0.3.1
2020-12-28, Project
- More operators, check the support list.
- Relax data type check, most for FP16 and INT8.
- Interface
enableDebugLog()
to dump log for debugging purpose.
Thanks for the contribution of @erizmr @briangrifiin and @IkbeomJeon!
tflite2onnx 0.3.0
- Now open source with annocement blog.
- Quantization support enabled, and tried quantized MobileNetV1 an MobileNetV2.
- Drop Transpose based layout handling to save the effort of managing quantization.
- More operators added, and tested models:
- MobileNetV1
- MobileNetV2
- DenseNet
- EfficientNet
- MnasNet
- SqueezeNet
- NasNet
tflite2onnx 0.2.0
- Operator support of MobileNetV2.
- Infrastructure improvements.
- Propagation based layout handling.
tflite2onnx 0.1.0
MobileNetV1 passes.