Fixing a small issue in trainer logging

[guydav]

Before submitting

• Was this discussed/approved via a Github issue? (no need for typos and docs improvements)
• Did you read the contributor guideline, Pull Request section?
• Did you make sure to update the docs? No changes required
• Did you write any new necessary tests? I don't think this requires any new tests, does it?
• If you made a notable change (that affects users), did you update the CHANGELOG? Should this be on the changelog?

What does this PR do?

Fixes half of #1520. This is the simpler half, where the epoch is logged into a dict that isn't passed to the loggers.

PR review

Anyone in the community is free to review the PR once the tests have passed.
If we didn't discuss your PR in Github issues there's a high chance it will not be merged.

Did you have fun?

As much as can be in a one-line fix.

[guydav]

Unsurprisingly, this broke the logging tests, probably since the epoch field now appears in the dictionary that arrives in the logger. Can I get approval to change the tests to account for this new field?

[Borda]

good catch, LGTM 🤖

[guydav]

I added the change to the tests. Let's see that everything passes now.

Two questions:

1. Do you merge directly to master? Or is there a develop or some other branch that this should be merged into?
2. Is this (the epoch field arriving properly to the loggers) a change that should be documented anywhere?

[williamFalcon]

1. yes
2. i don't think so right now? maybe if people bring it up but it seems like intuitive default behavior.

[guydav]

Okay cool. The only test that's failing is the PyTorch-v1_5 one that timed out. Does that make sense? As far as I'm concerned it's ready to merge, but if there's a way to return this test, that might be even better.

[Borda]

@guydav rebased