Please refer to link installing mmcv on NPU Devices.
Here we use 8 NPUs on your computer to train the model with the following command:
bash tools/dist_train.sh configs/ssd/ssd300_coco.py 8Also, you can use only one NPU to train the model with the following command:
python tools/train.py configs/ssd/ssd300_coco.py| Model | box AP | mask AP | Config | Download |
|---|---|---|---|---|
| ssd300 | 25.6 | --- | config | log |
| ssd512 | 29.4 | --- | config | log |
| *ssdlite-mbv2 | 20.2 | --- | config | log |
| retinanet-r50 | 36.6 | --- | config | log |
| *fcos-r50 | 36.1 | --- | config | log |
| solov2-r50 | --- | 34.7 | config | log |
Notes:
- If not specially marked, the results are same between results on the NPU and results on the GPU with FP32.
- (*) The results on the NPU of these models are aligned with the results of the mixed-precision training on the GPU, but are lower than the results of the FP32. This situation is mainly related to the phase of the model itself in mixed-precision training, users please adjust the hyperparameters to achieve the best result by self.
All above models are provided by Huawei Ascend group.