Images not being logged after using auto_lr_find

[SubodhDahal]

🐛 Bug

When I use the auto LR finder, the images that I logged are no longer being displayed in Tensorboard. The logging works normally when setting it to False or when not specifying it.

To Reproduce

Steps to reproduce the behavior:

1. Set auto_lr_find = True for the trainer
2. Log image during test/validation step as self.logger.experiment.add_image(title, image, step)
3. Load the Tensorboard WebUI and see that the image is not being logged

Expected behavior

The image that I logged should be visible in Tensorboard, even when using auto LR finder.

Environment

* CUDA:
        - GPU:
                - GeForce GTX 1050 Ti with Max-Q Design
        - available:         True
        - version:           10.2
* Packages:
        - numpy:             1.18.3
        - pyTorch_debug:     False
        - pyTorch_version:   1.5.0
        - pytorch-lightning: 0.7.6
        - tensorboard:       2.2.1
        - tqdm:              4.46.0
* System:
        - OS:                Linux
        - architecture:
                - 64bit
                - ELF
        - processor:         x86_64
        - python:            3.8.2
        - version:           #41~1586790036~18.04~600aeb5-Ubuntu SMP Mon Apr 13 17:47:15 UTC

• How you installed PyTorch (conda, pip, source): pip

Additional context

Not sure if relevant, but I couldn't figure out how to pass the params to the Trainer as I'm using from_argparse_args(hparams), where hparams contains the keys and values obtained from command-line ArgumentParser. So I set hparams.auto_lr_finder = True before instantiating the Trainer. The LR finder work's, so I assume there's no issue with that.

[github-actions]

Hi! thanks for your contribution!, great first issue!

[danielhomola]

I can second this. Same goes for the graph of the net.. Spent almost a full day trying to figure out what am I doing wrong, but once auto_lr_find=False, everything works fine..

[SkafteNicki]

@SubodhDahal and @danielhomola could one of you try out the fix that is on master now, and report back if it worked. If so please close the issue :)

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.