tanghy2016 comments

Results 5 comments of


                                            tanghy2016

Cannot re-initialize CUDA in forked subprocess. To use CUDA with multiprocessing, you must use the 'spawn' start method

I have a similar problem, have you solved it?

How could I use MMRazor to train YoloX with KD ?

@jvwj @LiuXuan66 Did you succeed? Can you share your configuration? my attempt failed #184

How to get the student model checkpoint after finishing knowledge distillation?

@maxrumi How to deal with object detection? I tried changing ``` model = build_classifier(cfg.model) ``` to ``` model = build_detector(cfg.model) ``` but when loading the model, the weights don't match...

How was trained the model?

> > > select 49 points from 68 points, you can see the intraface code. 你好，请问这49个点是怎么选择的呢？因为DemoDetector中得到的49个点并不是完全在68个点内，我尝试使用全部68个点输入EstimateHeadPose中得到的三个角度看上去不太对。去除68个点的外部轮廓17个点，得到的三个角度看上去还是不对。

about the det2_half.caffemodel

```shell >>> import caffe >>> deploy = "./det2.prototxt" >>> model = "./det2_half.caffemodel" >>> net = caffe.Net(deploy, model, caffe.TEST) ``` WARNING: Logging before InitGoogleLogging() is written to STDERR W0124 13:46:58.335070 14075...