Wanli
commited on
Commit
·
2965475
1
Parent(s):
6a397da
Re-quantify some models from per_channel mode to per_tensor mode (#90)
Browse files* re-quantize some models from per_channel mode to per_tensor mode
* remove the description about per_channel
README.md
CHANGED
|
@@ -9,8 +9,6 @@ This model is converted from Tensorflow-JS to ONNX using following tools:
|
|
| 9 |
- tf_saved_model to ONNX: https://github.com/onnx/tensorflow-onnx
|
| 10 |
- simplified by [onnx-simplifier](https://github.com/daquexian/onnx-simplifier)
|
| 11 |
|
| 12 |
-
Also note that the model is quantized in per-channel mode with [Intel's neural compressor](https://github.com/intel/neural-compressor), which gives better accuracy but may lose some speed.
|
| 13 |
-
|
| 14 |
## Demo
|
| 15 |
|
| 16 |
Run the following commands to try the demo:
|
|
|
|
| 9 |
- tf_saved_model to ONNX: https://github.com/onnx/tensorflow-onnx
|
| 10 |
- simplified by [onnx-simplifier](https://github.com/daquexian/onnx-simplifier)
|
| 11 |
|
|
|
|
|
|
|
| 12 |
## Demo
|
| 13 |
|
| 14 |
Run the following commands to try the demo:
|