Don't see any output from TensorboardLoggerHook

Prerequisite

• I have searched the existing and past issues but cannot get the expected help.
• I have read the FAQ documentation but cannot get the expected help.
• The bug has not been fixed in the latest version.

🐞 Describe the bug

On v3.0.0rc1, I no longer see any output while using TensorboardLoggerHook. I’ve added the following log_config to yolox_s_8xb8-300e_coco.py:

log_config = dict(  # config to register logger hook
    interval=10,  # Interval to print the log
    hooks=[
        dict(type='TextLoggerHook'),
        dict(type='TensorboardLoggerHook', log_dir='/opt/ml/checkpoints', by_epoch=False)
    ]
)

However at the end of training I can’t find any tensorboard logs in the log_dir location.

Environment

Traceback (most recent call last):
  File "mmedit/utils/collect_env.py", line 3, in <module>
    from mmcv.utils import get_git_hash
ImportError: cannot import name 'get_git_hash' from 'mmcv.utils' (/opt/conda/lib/python3.8/site-packages/mmcv/utils/__init__.py)

---

Python 3.8.13

nvcc: NVIDIA ® Cuda compiler driver Copyright © 2005-2021 NVIDIA Corporation Built on Mon_May__3_19:15:13_PDT_2021 Cuda compilation tools, release 11.3, V11.3.109 Build cuda_11.3.r11.3/compiler.29920130_0

mmcv==2.0.0rc1 mmengine==0.1.0

Additional information

1. No
2. Small version of COCO
3. IIRC, it seemed to work before upgrading to v3.0 and using MMEngine

[austinmw]

I think this method was deprecated with mmcv 2.x and mmdet 3.x in favor of something like this:

visualizer = dict(
    type='DetLocalVisualizer',
    vis_backends=[
        dict(type='TensorboardVisBackend'),
    ],
    save_dir='/path/to/save',
    name='visualizer'
)

But I haven’t tried it yet myself

[austinmw]

Using TensorboardVisBackend worked for me to display images in TensorBoard