Created by: pele228

如何用自己的数据集蒸馏剪裁模型？ 按照文档步骤https://github.com/PaddlePaddle/PaddleDetection/blob/release/0.2/slim/extensions/distill_pruned_model/README.md

aistudio@jupyter-135123-311635:~/work/PaddleDetection/slim/extensions/distill_pruned_model$ python distill_pruned_model.py -c ../../../configs/yolov3_mobilenet_v1_voc_mydata.yml -t ../../../configs/yolov3_r34_voc_mydata.yml --teacher_pretrained=https://paddlemodels.bj.bcebos.com/object_detection/yolov3_r34_voc.tar --pruned_params "yolo_block.0.0.0.conv.weights,yolo_block.0.0.1.conv.weights,yolo_block.0.1.0.conv.weights" --pruned_ratios="0.2,0.3,0.4" -o use_fine_grained_loss=true pretrain_weights=https://paddlemodels.bj.bcebos.com/object_detection/yolov3_mobilenet_v1_voc.tar 2020-03-19 15:36:27,347-INFO: Config dataset_dir dataset/mydata is not exits, dataset config is not valid 2020-03-19 15:36:27,347-INFO: Dataset /home/aistudio/work/PaddleDetection/slim/extensions/distill_pruned_model/dataset/mydata is not valid for reason above, try searching /home/aistudio/.cache/paddle/dataset or downloading dataset... Traceback (most recent call last): File "distill_pruned_model.py", line 365, in main() File "distill_pruned_model.py", line 149, in main devices_num, cfg) File "/home/aistudio/work/PaddleDetection/ppdet/data/reader.py", line 406, in create_reader reader = Reader(**cfg)() File "/home/aistudio/work/PaddleDetection/ppdet/data/reader.py", line 202, in init self._roidbs = self._dataset.get_roidb() File "/home/aistudio/work/PaddleDetection/ppdet/data/source/dataset.py", line 65, in get_roidb self.image_dir) File "/home/aistudio/work/PaddleDetection/ppdet/utils/download.py", line 145, in get_dataset_path format(path, osp.split(path)[-1])) ValueError: Dataset dataset/mydata is not valid and cannot parse dataset type 'mydata' for automaticly downloading, which only supports 'voc' , 'coco', 'wider_face' and 'fruit' currently