wasedd

Results 3 issues of wasedd

I run my code using voc format dataset training setting as follows: "AUGMENT": False, "BATCH_SIZE": 2, "MULTI_SCALE_TRAIN": True, "IOU_THRESHOLD_LOSS": 0.5, "YOLO_EPOCHS": 50, "Mobilenet_YOLO_EPOCHS": 120, "NUMBER_WORKERS": 4, "MOMENTUM": 0.9, "WEIGHT_DECAY": 0.0005,...

下面是运行eval_voc.py之后的结果, ![屏幕截图 2021-05-24 172308](https://user-images.githubusercontent.com/54884901/120130656-485cf080-c1f9-11eb-91f5-ad23e550031b.png) 生成output文件,但内容为空。没有测试结果的图片。请问是什么问题,该怎么解决? ps:yolov4模型,无注意力模块 参数设置如下图: ![屏幕截图 2021-05-31 101924](https://user-images.githubusercontent.com/54884901/120130941-ea7cd880-c1f9-11eb-8937-ebca84a8b934.png) 还有,得到的预测结果pred_result如下,依然存在较多候选框,似乎nms没有起到筛选作用。这如何解决? ![屏幕截图 2021-05-31 102403](https://user-images.githubusercontent.com/54884901/120131284-b5bd5100-c1fa-11eb-8e19-5c81785f8f51.png)

I run my code using voc format dataset training setting as follows: "AUGMENT": False, "BATCH_SIZE": 2, "MULTI_SCALE_TRAIN": True, "IOU_THRESHOLD_LOSS": 0.5, "YOLO_EPOCHS": 50, "Mobilenet_YOLO_EPOCHS": 120, "NUMBER_WORKERS": 4, "MOMENTUM": 0.9, "WEIGHT_DECAY": 0.0005,...