Closed lichen14 closed 4 months ago
这里使用了all_data[fname]的['example_clip_features']属性,但是我实际按照您给的代码顺序,跑完 https://github.com/Hzzone/PseCo/blob/e68019ae6ec3c5401f62f2553721915d4325c335/fsc147/1_generate_data.ipynb 的In [38] '' torch.save(all_data, f'{project_root}/data/fsc147/sam/all_data_vitl.pth') '' 时,保存的all data 并无example_clip_features属性。
example_clip_features属性真正出现在该第一步文件的In [66] 中,但此时已经没有更新all data的保存。
因此,我有两个问题:
一些小的建议或者疑问,仅供您参考 :)
你好:
https://github.com/Hzzone/PseCo/blob/e68019ae6ec3c5401f62f2553721915d4325c335/fscd_lvis/4_1_train_roi_head.py#L79
Hi, it seems that the wandb.init() has no attribute 'print_freq'?
Is there any wrong?