zthcool

2.3.3梯度结果问题请教

1

**bug描述** 在按照教程2.3.3梯度章节中 “ # 再来反向传播一次，注意grad是累加的 out2 = x.sum() out2.backward() print(x.grad) ” 时我的返回结果是 tensor([[1., 1.], [1., 1.]]) 如图  请问是我的输入有问题么？ **版本信息** pytorch: 1.9.0 torchvision: 0.10.0 torchtext: None ...

how `send_processdata()` and `receive_processdata()` work?

4

Hi @bnjmnp : This is my first time contact with pysoem and soem,In the past few days,I have built communication with the driver (Panasonic),and make it run to a position...

Cannot find tb3_control which suitable to your project,

1

In the line "Next, download the “tb3_simulation” package from Canvas and unzip it. then put this “tb3_control” folder under and copy and paste it to /catkin_ws/src." I cannot figure out...

there is no data in '/orb_slam2/odom',when I run load map model,is that right? How can I use the vo in ros?

麻烦大佬给个示例可以么

1

自己刚刚接触python，想用libfacedetection库进行人脸检测，但是对 facedetect_cnn(image: Union[str, Path, Image, np.ndarray, bytes], width: int = 20, height: int = 20, step: int = 1) -> Faces 这个函数具体调用方法不是太明白，我该如何写这个函数中的各个变量 比如我写的 Path=(r'D:\Users\BC\PycharmProjects\F_C_05_PyFaceDet(libfacedetection)') Image=('000019.jpg') 还是请大佬说明一下吧