Conversation
|
Warning This pull request is not mergeable via GitHub because a downstack PR is open. Once all requirements are satisfied, merge this PR as a stack on Graphite. This stack of pull requests is managed by Graphite. Learn more about stacking. |
aidang3019
force-pushed
the
aidan/bounds-check
branch
from
023180b to
49ef806
Compare
|
|
||
| return data | ||
|
|
||
| def set_norm_stats(self, norm_stats: dict) -> None: |
There was a problem hiding this comment.
Why is this needed? I thought when we instantiate ZarrDataset we pass in the norm stats?
| ) | ||
| return (next_idx, origin, attempts) | ||
|
|
||
| def _get_image_fallback_idx( |
There was a problem hiding this comment.
Why does get_image_fallback_idx have diff logic from get_fallback_idx
There was a problem hiding this comment.
i thought we didnt want to random resample frames, but just do closest clean frame
| @@ -798,6 +967,7 @@ def get_item_keys( | |||
| idx: int, | |||
There was a problem hiding this comment.
[Re: line +965]
This function isn't being used so all these changes can be deleted, right?
See this comment inline on Graphite.
| ), | ||
| ) | ||
|
|
||
| # Propagate norm stats to all zarrdatasets out of bounds check in getitem |
There was a problem hiding this comment.
Is there a reason we can't just directly set dataset.norm_stats = data_schematic.norm_stats. I wish we didn't have to add norm stats as an attribute after creating the dataset, but I'm not sure if there's a clean pattern to avoid this
3 participants
No description provided.