Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

SSL4EO-S12: add new dataset/datamodule #1151

Merged
merged 31 commits into from Apr 15, 2023
Merged

SSL4EO-S12: add new dataset/datamodule #1151

Changes from 1 commit
Commits
Show all changes
31 commits
Select commit Hold shift + click to select a range
e7afd1f
SSL4EO-S12: add new dataset
adamjstewart Feb 28, 2023
44f2953
Style fixes
adamjstewart Feb 28, 2023
45fc445
100% coverage
adamjstewart Feb 28, 2023
729d377
fix mypy
adamjstewart Feb 28, 2023
a2f39d5
black fixes
adamjstewart Feb 28, 2023
0f0e02d
mypy fix
adamjstewart Mar 1, 2023
1ad0fb6
Convert from db to power
adamjstewart Mar 6, 2023
d2a8049
Don't cast to numpy
adamjstewart Mar 7, 2023
3df7ec1
Remove comments referring to SeCo
adamjstewart Mar 7, 2023
a424a77
SSL4EO: add extraction time
adamjstewart Mar 9, 2023
ec0379a
Add RandomSeasonContrast
adamjstewart Mar 10, 2023
6172fff
Fix axes indexing
adamjstewart Mar 10, 2023
4b0774c
Add datamodule
adamjstewart Mar 17, 2023
7b32aa3
fix tests
adamjstewart Mar 18, 2023
bd5962a
mypy fixes
adamjstewart Mar 18, 2023
0265ccf
fix missing import
adamjstewart Mar 18, 2023
6529150
Fix tests
adamjstewart Mar 18, 2023
8268f6e
isort fix
adamjstewart Mar 18, 2023
05dab7d
Typo fix
adamjstewart Mar 19, 2023
1fd7a2b
s2c: add B10
adamjstewart Mar 27, 2023
80a5c3a
Update test channels
adamjstewart Mar 27, 2023
fe28fe9
S2 plotting was broken
calebrob6 Mar 29, 2023
651347b
Fix plotting
calebrob6 Mar 29, 2023
5cae88d
Merge branch 'main' into datasets/ssl4eo
adamjstewart Mar 29, 2023
b4e4410
Black fix
adamjstewart Mar 29, 2023
c4a5c75
Merge branch 'main' into datasets/ssl4eo
adamjstewart Mar 30, 2023
5dbcab6
Rename conf files
adamjstewart Mar 30, 2023
8af0a53
Remove file introduced by bad merge
adamjstewart Mar 31, 2023
87f068e
Fix pixel size of bands
adamjstewart Apr 9, 2023
d553b57
black fix
adamjstewart Apr 9, 2023
90714a0
Better S1 plotting
adamjstewart Apr 9, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Jump to
Jump to file
Failed to load files.
Diff view
Diff view
3 changes: 1 addition & 2 deletions torchgeo/datasets/ssl4eo.py
View file
Open in desktop
Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@calebrob6 this directly contradicts the change made in #1027. Should we revert that PR too? Given that:

  1. All of our datasets use a different scale factor, ignoring which satellite their images come from
  2. All images are normalized before they are returned from the dataset or datamodule, changing the appropriate scale factor

I'm starting to think we should use one of the default enhancements from QGIS:

  1. Min / max
  2. 2% / 98%
  3. ± 2 std dev

There could even be an option to select which of these you want to use. We can decide and implement this later, but I want to make sure we're not contradicting ourselves and won't revert #1027 multiple times until we come up with a uniform policy. This also relates to #496.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

dividing by 10k obviously didn't work though

Copy link
Collaborator Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Then why did we merge #1027?

Expand Up @@ -238,9 +238,8 @@ def plot(
(co_polarization, cross_polarization, ratio), dim=-1
)
else:
# See Sentinel2.plot
image = image[[3, 2, 1]].permute(1, 2, 0)
image = torch.clamp(image / 10000, min=0, max=1)
image = torch.clamp(image / 3000, min=0, max=1)

axes[i].imshow(image)
axes[i].axis("off")
Expand Down