APeiZou comments

Results 36 comments of


                                            APeiZou

Why do I need to load 2D data when training 3D data？

@mks0601 Can I only train in Human36 datasets ? How to init_weight of the model when training? Do I need to load the pretrained model?

Why do I need to load 2D data when training 3D data？

The values of vis are all zero when I load Human36 dataset，and the loss is zero。 ![image](https://user-images.githubusercontent.com/49524647/123779423-bda90780-d904-11eb-86ce-94fee64d8c3f.png)

Why do I need to load 2D data when training 3D data？

![image](https://user-images.githubusercontent.com/49524647/123780374-b6cec480-d905-11eb-89c2-e5c2559d0127.png) The vis Attributes of target is zero and the loss are always zero。

Why do I need to load 2D data when training 3D data？

@mks0601 I have not bbox_root_human36m_output.json file ，can I train the model in Human36 datasets ? ![image](https://user-images.githubusercontent.com/49524647/123784711-4b3b2600-d90a-11eb-9f3b-985ec7a37e53.png)

coco数据集输入数据是原图去训练还是输入标注部分人体ROI区域进入网络训练？

@leoxiaobin

你好，你们有blaze-pose的tflite版本的模型吗？

Hello, do you have a model of the tflite version of blaze-pose?@vietanhdev

您好，怎么在COCO的预训练模型上进行迁移学习啊

optimizer: lr: 0.14 momentum: 0.9 name: SGD weight_decay: 0.0001 total_epochs: 3 val_intervals: 10 warmup: name: linear ratio: 0.1 steps: 20 [root][02-25 11:47:42]INFO:loaded /data/project_all/project_mmdet/nanodet/nanodet_m.pth, epoch 70 [root][02-25 11:47:42]INFO:Skip loading parameter head.gfl_cls.0.weight,...