Gong Xinyao

this is my dataset file：animal_lambpose_dataset.py my dataset has 18 keypoint # Copyright (c) OpenMMLab. All rights reserved. import os.path as osp import tempfile import warnings from collections import OrderedDict, defaultdict...

This is the network file：litehrnet_18_coco_256x192.py _base_ = [ '../../../../_base_/default_runtime.py', '../../../../_base_/datasets/lamb.py' ] evaluation = dict(interval=1, metric='mAP', save_best='AP') optimizer = dict( type='Adam', lr=5e-4, ) optimizer_config = dict(grad_clip=None) # learning policy lr_config =...

Okay, I'll turn it up a little and see the results, thank you~

When I ran 210 epochs, the AP was still 0

I found that from the beginning to the end of the training, my loss always kept a low value without ups and downs, and the data_ time is basically 0,...

Okay, I'll check that, thank you~

很抱歉，我也暂时还没有找到是哪儿的问题

> 好的好的，谢谢你呀，我去检查一下我的json文件