Multiple commits in the experiments table

[sroy3]

Part of #1966

Quite straight forward. Props to @mattseddon for thinking ahead, because this just works.

[sroy3]

Screen.Recording.2022-09-14.at.2.33.55.PM.mov

Doing some tests, haven't found anything "weird" yet.

[mattseddon]

Initially, I was very confused by the chevron placement of the second commit but it does seem like everything is working as expected:

I do think we now need to nest experiments underneath their respective commit in the experiments tree:

The top of the tree is flat and fairly confusing. We should follow up and make that change ASAP.

There is at least one function (collectWorkspaceRunningCheckpoint) which is (probably) checking previously commits when it doesn't need to be.

We should also add a small multi-commit test fixture 🙏

[mattseddon]

Unable to apply an experiment from a previous commit:

Screen.Recording.2022-09-15.at.4.54.44.pm.mov

This doesn't work for either the commit/short sha or the name. It looks like exp apply will only accept names from the current commit. That means we need to raise a bug in DVC and suppress the option from the tree/palette/table for previous commits (for the time being).

 ────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
  Experiment                Created        State    avg_prec   roc_auc   prepare.split   prepare.seed   featurize.max_features   featurize.ngrams   train.seed   train.n_est   train.min_
 ────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
  workspace                 -              -         0.92438   0.94473   0.21            20170428       200                      2                  20170428     50            0.01      
  test-multi                04:29 PM       -           0.925   0.94602   0.2             20170428       200                      2                  20170428     50            0.01      
  ├── a7c62f2 [exp-1e491]   04:34 PM       -         0.92438   0.94473   0.21            20170428       200                      2                  20170428     50            0.01      
  └── 397549b               04:29 PM       Queued          -         -   0.2             20170428       200                      2                  20170428     50            0.01      
  bigrams-experiment        May 20, 2022   -           0.925   0.94602   0.2             20170428       200                      2                  20170428     50            0.01      
  ├── 15d24d7 [exp-df115]   Aug 16, 2022   -         0.94254   0.96158   0.2             20170428       448                      2                  20170428     50            0.01      
  ├── 58f48c7 [exp-869d9]   Aug 16, 2022   -         0.94489   0.96199   0.2             20170428       512                      2                  20170428     50            0.01      
  ├── 806879b [exp-41525]   Aug 16, 2022   -         0.94559   0.96137   0.2             20170428       576                      2                  20170428     50            0.01      
  ├── 8e112e6 [exp-53de3]   Aug 16, 2022   -         0.94679   0.96299   0.2             20170428       640                      2                  20170428     50            0.01      
  ├── d4e2429 [exp-c37c1]   Aug 16, 2022   -         0.84277   0.89615   0.2             20170428       64                       2                  20170428     50            0.01      
  ├── 62b8ff1 [exp-f5669]   Aug 16, 2022   -         0.94144   0.95794   0.2             20170428       320                      2                  20170428     50            0.01      
  └── cce158f [exp-44136]   Jul 19, 2022   -               -         -   0.2             20170428       200                      2                  20170428     50            0.01      
 ────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
~/p/example-get-started test-multi *1 !7 ❯ dvc exp apply 58f48c7
ERROR: '58f48c7' does not appear to be an experiment commit.: Experiment derived from 'a470fae', expected 'dc91a3d'.
~/p/example-get-started test-multi *1 !7 ❯ dvc exp apply exp-869d9
ERROR: 'exp-869d9' does not appear to be an experiment commit.: Experiment derived from 'a470fae', expected 'dc91a3d'.```

[julieg18]

Great work! Agreed with Matt comments. I was also confused about the placement of the commit rows since they're at the same level as the main branch.

[julieg18]

Looks like the row borders are now a bit buggy which I believe is due to this line in our styles:

[julieg18]

If I remember correctly, the CSS lines were added to make the last row on the table not have a border 🤔

[codeclimate]

Code Climate has analyzed commit d8b04fb and detected 0 issues on this pull request.

The test coverage on the diff in this pull request is 91.3% (85% is the threshold).

This pull request will bring the total coverage in the repository to 96.9% (0.0% change).

View more on Code Climate.