Files
ascend-tools/README.md
T
2020-10-26 16:28:11 +01:00

41 lines
1.6 KiB
Markdown
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
EN|[CH](README.md)
# tools
#### introduce
Ascend tools
**Please go to the corresponding folder to get the tool according to your needs, or click the link below to select the tool you need to use.**
#### explain
1. [msame](https://gitee.com/ascend/tools/tree/master/msame)
**Model reasoning tool**:Input. Om model and model required input bin file, output model output data file.
2. [img2bin](https://gitee.com/ascend/tools/tree/master/img2bin)
**Bin file generation tool** : Generates input data required for model reasoning, saved in.bin format.
3. [makesd](https://gitee.com/ascend/tools/tree/master/makesd)
**makesd tool**makesd tools packageProvide card making function under ubuntu.
4. [faster_install](https://gitee.com/ascend/tools/tree/master/faster_install)
**faster_install**environment fast install script.
5. [configure_usb_ethernet](https://gitee.com/ascend/tools/tree/master/configure_usb_ethernet)
**configure_usb_ethernet**configuring the IP address of the USB NIC 。
6. [pt2pb](https://gitee.com/ascend/tools/tree/master/pt2pb)
**pytorch model transform to tensorflow pb model tool**input pytorch weights parameters modeltransform to onnx filethen transform to pb model
7. [dnmetis](https://gitee.com/ascend/tools/tree/master/dnmetis)
**Test tool for NPU inference precision and performance**Using Python to encapsulate the C++ interface of ACL, inputting om model and original dataset images and tags, we can execute model inference and give out precision and performance of the om model