Add SolarPlantsBrazil dataset for photovoltaic panel detection (semantic segmentation) #2797

FederCO23 · 2025-05-24T00:01:28Z

This PR adds the SolarPlantsBrazil dataset to TorchGeo.

SolarPlantsBrazil is a geospatial dataset curated for binary semantic segmentation, focused on detecting photovoltaic (PV) power stations in satellite imagery. It includes 272 manually annotated 256×256 GeoTIFF image patches from various regions of Brazil. Each image patch comes with:

A 4-band image (RGB + NIR, float32)
A binary mask indicating solar panel regions

Changes introduced:

Added SolarPlantsBrazil dataset class under torchgeo.datasets
Added SolarPlantsBrazilDataModule datamodule class under torchgeo.datamodules
Integrated automatic download from Hugging Face: FederCO23/solar-plants-brazil
Parsed RGB + NIR float32 GeoTIFFs and binary mask labels
Implemented a plot() method using RGB bands for sample visualization
Registered dataset in __init__.py and non_geo_datasets.csv
Added full unit test coverage under tests/datasets/test_solar_plants_brazil.py

Example output from `plot()`:

References:

FederCO23 · 2025-05-24T00:43:53Z

@microsoft-github-policy-service agree

FederCO23 · 2025-05-24T00:46:38Z

@microsoft-github-policy-service agree

…

On Fri, May 23, 2025 at 9:02 PM microsoft-github-policy-service[bot] < ***@***.***> wrote: *microsoft-github-policy-service[bot]* left a comment (microsoft/torchgeo#2797) <#2797 (comment)> @FederCO23 <https://github.com/FederCO23> please read the following Contributor License Agreement(CLA). If you agree with the CLA, please reply with the following information. @microsoft-github-policy-service agree [company="{your company}"] Options: - (default - no company specified) I have sole ownership of intellectual property rights to my Submissions and I am not making Submissions in the course of work for my employer. @microsoft-github-policy-service agree - (when company given) I am making Submissions in the course of work for my employer (or my employer has intellectual property rights in my Submissions by contract or applicable law). I have permission from my employer to make Submissions and enter into this Agreement on behalf of my employer. By signing below, the defined term “You” includes me and my employer. @microsoft-github-policy-service agree company="Microsoft" Contributor License Agreement Contribution License Agreement This Contribution License Agreement (*“Agreement”*) is agreed to by the party signing below (*“You”*), and conveys certain license rights to Microsoft Corporation and its affiliates (“Microsoft”) for Your contributions to Microsoft open source projects. This Agreement is effective as of the latest signature date below. 1. *Definitions*. *“Code”* means the computer software code, whether in human-readable or machine-executable form, that is delivered by You to Microsoft under this Agreement. *“Project”* means any of the projects owned or managed by Microsoft and offered under a license approved by the Open Source Initiative (www.opensource.org). *“Submit”* is the act of uploading, submitting, transmitting, or distributing code or other content to any Project, including but not limited to communication on electronic mailing lists, source code control systems, and issue tracking systems that are managed by, or on behalf of, the Project for the purpose of discussing and improving that Project, but excluding communication that is conspicuously marked or otherwise designated in writing by You as “Not a Submission.” *“Submission”* means the Code and any other copyrightable material Submitted by You, including any associated comments and documentation. 2. *Your Submission*. You must agree to the terms of this Agreement before making a Submission to any Project. This Agreement covers any and all Submissions that You, now or in the future (except as described in Section 4 below), Submit to any Project. 3. *Originality of Work*. You represent that each of Your Submissions is entirely Your original work. Should You wish to Submit materials that are not Your original work, You may Submit them separately to the Project if You (a) retain all copyright and license information that was in the materials as You received them, (b) in the description accompanying Your Submission, include the phrase “Submission containing materials of a third party:” followed by the names of the third party and any licenses or other restrictions of which You are aware, and (c) follow any other instructions in the Project’s written guidelines concerning Submissions. 4. *Your Employer*. References to “employer” in this Agreement include Your employer or anyone else for whom You are acting in making Your Submission, e.g. as a contractor, vendor, or agent. If Your Submission is made in the course of Your work for an employer or Your employer has intellectual property rights in Your Submission by contract or applicable law, You must secure permission from Your employer to make the Submission before signing this Agreement. In that case, the term “You” in this Agreement will refer to You and the employer collectively. If You change employers in the future and desire to Submit additional Submissions for the new employer, then You agree to sign a new Agreement and secure permission from the new employer before Submitting those Submissions. 5. *Licenses*. - *Copyright License*. You grant Microsoft, and those who receive the Submission directly or indirectly from Microsoft, a perpetual, worldwide, non-exclusive, royalty-free, irrevocable license in the Submission to reproduce, prepare derivative works of, publicly display, publicly perform, and distribute the Submission and such derivative works, and to sublicense any or all of the foregoing rights to third parties. - *Patent License*. You grant Microsoft, and those who receive the Submission directly or indirectly from Microsoft, a perpetual, worldwide, non-exclusive, royalty-free, irrevocable license under Your patent claims that are necessarily infringed by the Submission or the combination of the Submission with the Project to which it was Submitted to make, have made, use, offer to sell, sell and import or otherwise dispose of the Submission alone or with the Project. - *Other Rights Reserved*. Each party reserves all rights not expressly granted in this Agreement. No additional licenses or rights whatsoever (including, without limitation, any implied licenses) are granted by implication, exhaustion, estoppel or otherwise. 6. *Representations and Warranties*. You represent that You are legally entitled to grant the above licenses. You represent that each of Your Submissions is entirely Your original work (except as You may have disclosed under Section 3). You represent that You have secured permission from Your employer to make the Submission in cases where Your Submission is made in the course of Your work for Your employer or Your employer has intellectual property rights in Your Submission by contract or applicable law. If You are signing this Agreement on behalf of Your employer, You represent and warrant that You have the necessary authority to bind the listed employer to the obligations contained in this Agreement. You are not expected to provide support for Your Submission, unless You choose to do so. UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING, AND EXCEPT FOR THE WARRANTIES EXPRESSLY STATED IN SECTIONS 3, 4, AND 6, THE SUBMISSION PROVIDED UNDER THIS AGREEMENT IS PROVIDED WITHOUT WARRANTY OF ANY KIND, INCLUDING, BUT NOT LIMITED TO, ANY WARRANTY OF NONINFRINGEMENT, MERCHANTABILITY, OR FITNESS FOR A PARTICULAR PURPOSE. 7. *Notice to Microsoft*. You agree to notify Microsoft in writing of any facts or circumstances of which You later become aware that would make Your representations in this Agreement inaccurate in any respect. 8. *Information about Submissions*. You agree that contributions to Projects and information about contributions may be maintained indefinitely and disclosed publicly, including Your name and other information that You submit with Your Submission. 9. *Governing Law/Jurisdiction*. This Agreement is governed by the laws of the State of Washington, and the parties consent to exclusive jurisdiction and venue in the federal courts sitting in King County, Washington, unless no federal subject matter jurisdiction exists, in which case the parties consent to exclusive jurisdiction and venue in the Superior Court of King County, Washington. The parties waive all defenses of lack of personal jurisdiction and forum non-conveniens. 10. *Entire Agreement/Assignment*. This Agreement is the entire agreement between the parties, and supersedes any and all prior agreements, understandings or communications, written or oral, between the parties relating to the subject matter hereof. This Agreement may be assigned by Microsoft. — Reply to this email directly, view it on GitHub <#2797 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/BIQQR3ODLK34VV6GRDRWWKD276ZJFAVCNFSM6AAAAAB5Z3M4TOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDSMBWGA2TQNRRG4> . You are receiving this because you were mentioned.Message ID: ***@***.***>

FederCO23 · 2025-05-24T00:47:38Z

@microsoft-github-policy-service agree

…

On Fri, May 23, 2025 at 9:02 PM microsoft-github-policy-service[bot] < ***@***.***> wrote: *microsoft-github-policy-service[bot]* left a comment (microsoft/torchgeo#2797) <#2797 (comment)> @FederCO23 <https://github.com/FederCO23> please read the following Contributor License Agreement(CLA). If you agree with the CLA, please reply with the following information. @microsoft-github-policy-service agree [company="{your company}"] Options: - (default - no company specified) I have sole ownership of intellectual property rights to my Submissions and I am not making Submissions in the course of work for my employer. @microsoft-github-policy-service agree - (when company given) I am making Submissions in the course of work for my employer (or my employer has intellectual property rights in my Submissions by contract or applicable law). I have permission from my employer to make Submissions and enter into this Agreement on behalf of my employer. By signing below, the defined term “You” includes me and my employer. @microsoft-github-policy-service agree company="Microsoft" Contributor License Agreement Contribution License Agreement This Contribution License Agreement (*“Agreement”*) is agreed to by the party signing below (*“You”*), and conveys certain license rights to Microsoft Corporation and its affiliates (“Microsoft”) for Your contributions to Microsoft open source projects. This Agreement is effective as of the latest signature date below. 1. *Definitions*. *“Code”* means the computer software code, whether in human-readable or machine-executable form, that is delivered by You to Microsoft under this Agreement. *“Project”* means any of the projects owned or managed by Microsoft and offered under a license approved by the Open Source Initiative (www.opensource.org). *“Submit”* is the act of uploading, submitting, transmitting, or distributing code or other content to any Project, including but not limited to communication on electronic mailing lists, source code control systems, and issue tracking systems that are managed by, or on behalf of, the Project for the purpose of discussing and improving that Project, but excluding communication that is conspicuously marked or otherwise designated in writing by You as “Not a Submission.” *“Submission”* means the Code and any other copyrightable material Submitted by You, including any associated comments and documentation. 2. *Your Submission*. You must agree to the terms of this Agreement before making a Submission to any Project. This Agreement covers any and all Submissions that You, now or in the future (except as described in Section 4 below), Submit to any Project. 3. *Originality of Work*. You represent that each of Your Submissions is entirely Your original work. Should You wish to Submit materials that are not Your original work, You may Submit them separately to the Project if You (a) retain all copyright and license information that was in the materials as You received them, (b) in the description accompanying Your Submission, include the phrase “Submission containing materials of a third party:” followed by the names of the third party and any licenses or other restrictions of which You are aware, and (c) follow any other instructions in the Project’s written guidelines concerning Submissions. 4. *Your Employer*. References to “employer” in this Agreement include Your employer or anyone else for whom You are acting in making Your Submission, e.g. as a contractor, vendor, or agent. If Your Submission is made in the course of Your work for an employer or Your employer has intellectual property rights in Your Submission by contract or applicable law, You must secure permission from Your employer to make the Submission before signing this Agreement. In that case, the term “You” in this Agreement will refer to You and the employer collectively. If You change employers in the future and desire to Submit additional Submissions for the new employer, then You agree to sign a new Agreement and secure permission from the new employer before Submitting those Submissions. 5. *Licenses*. - *Copyright License*. You grant Microsoft, and those who receive the Submission directly or indirectly from Microsoft, a perpetual, worldwide, non-exclusive, royalty-free, irrevocable license in the Submission to reproduce, prepare derivative works of, publicly display, publicly perform, and distribute the Submission and such derivative works, and to sublicense any or all of the foregoing rights to third parties. - *Patent License*. You grant Microsoft, and those who receive the Submission directly or indirectly from Microsoft, a perpetual, worldwide, non-exclusive, royalty-free, irrevocable license under Your patent claims that are necessarily infringed by the Submission or the combination of the Submission with the Project to which it was Submitted to make, have made, use, offer to sell, sell and import or otherwise dispose of the Submission alone or with the Project. - *Other Rights Reserved*. Each party reserves all rights not expressly granted in this Agreement. No additional licenses or rights whatsoever (including, without limitation, any implied licenses) are granted by implication, exhaustion, estoppel or otherwise. 6. *Representations and Warranties*. You represent that You are legally entitled to grant the above licenses. You represent that each of Your Submissions is entirely Your original work (except as You may have disclosed under Section 3). You represent that You have secured permission from Your employer to make the Submission in cases where Your Submission is made in the course of Your work for Your employer or Your employer has intellectual property rights in Your Submission by contract or applicable law. If You are signing this Agreement on behalf of Your employer, You represent and warrant that You have the necessary authority to bind the listed employer to the obligations contained in this Agreement. You are not expected to provide support for Your Submission, unless You choose to do so. UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING, AND EXCEPT FOR THE WARRANTIES EXPRESSLY STATED IN SECTIONS 3, 4, AND 6, THE SUBMISSION PROVIDED UNDER THIS AGREEMENT IS PROVIDED WITHOUT WARRANTY OF ANY KIND, INCLUDING, BUT NOT LIMITED TO, ANY WARRANTY OF NONINFRINGEMENT, MERCHANTABILITY, OR FITNESS FOR A PARTICULAR PURPOSE. 7. *Notice to Microsoft*. You agree to notify Microsoft in writing of any facts or circumstances of which You later become aware that would make Your representations in this Agreement inaccurate in any respect. 8. *Information about Submissions*. You agree that contributions to Projects and information about contributions may be maintained indefinitely and disclosed publicly, including Your name and other information that You submit with Your Submission. 9. *Governing Law/Jurisdiction*. This Agreement is governed by the laws of the State of Washington, and the parties consent to exclusive jurisdiction and venue in the federal courts sitting in King County, Washington, unless no federal subject matter jurisdiction exists, in which case the parties consent to exclusive jurisdiction and venue in the Superior Court of King County, Washington. The parties waive all defenses of lack of personal jurisdiction and forum non-conveniens. 10. *Entire Agreement/Assignment*. This Agreement is the entire agreement between the parties, and supersedes any and all prior agreements, understandings or communications, written or oral, between the parties relating to the subject matter hereof. This Agreement may be assigned by Microsoft. — Reply to this email directly, view it on GitHub <#2797 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/BIQQR3ODLK34VV6GRDRWWKD276ZJFAVCNFSM6AAAAAB5Z3M4TOVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDSMBWGA2TQNRRG4> . You are receiving this because you were mentioned.Message ID: ***@***.***>

torchgeo/datasets/solar_plants_brazil.py

robmarkcole · 2025-05-25T05:30:27Z

@FederCO23 care to add a datamodule too? Also this dataset is only 4 channel, don't suppose you investigated adding the other channels too?

adamjstewart

Not a bad data loader, but it doesn't yet follow the style of other TorchGeo datasets/tests. Take a look at https://torchgeo.readthedocs.io/en/latest/tutorials/contribute_non_geo_dataset.html or any other builtin dataset for inspiration.

tests/datasets/test_solar_plants_brazil.py

torchgeo/datasets/solar_plants_brazil.py

FederCO23 · 2025-05-27T05:43:06Z

Hi @adamjstewart , just circling back on your previous review.

The requested changes have been addressed in the latest commits:

Replaced unittest.mock.patch with pytest.MonkeyPatch
Moved dummy data generation to tests/data/solar_plants_brazil/data.py
Removed unused citation block and huggingface_hub dependency
Switched to relative imports
Updated exception handling to use DatasetNotFoundError
Refactored tests into a single TestSolarPlantsBrazil class following TorchGeo style
Added full unit test coverage for dataset + datamodule (100%)
Lint clean (ruff, mypy)

The dataset currently uses 4 bands (RGB + NIR), which matched the original training setup. For now, I’ve kept it limited to these bands for consistency, but I’m open to extending it in future updates.

I’ve also replied to your earlier comments individually with a bit more context, in case further clarification is helpful.

Please let me know if there’s anything else you’d like me to adjust, and thanks again for your time and feedback!

robmarkcole · 2025-05-28T13:46:35Z

Test failures same as #2707 (comment)

robmarkcole · 2025-05-30T09:04:23Z

@FederCO23 hit the Update branch button as the failing test is fixed on main

FederCO23 · 2025-05-31T18:59:12Z

@FederCO23 hit the Update branch button as the failing test is fixed on main

@robmarkcole, I’ve updated the branch as requested and all checks are passing.
I also addressed all the requested changes from @adamjstewart and left a comment summarizing them above.
Let me know if anything else is needed on my end. Thanks again!

adamjstewart · 2025-05-31T19:51:16Z

Swamped with deadlines, will review soon (Tue or Wed).

docs/api/datasets/non_geo_datasets.csv

docs/api/datasets.rst

tests/datamodules/test_solar_plants_brazil.py

tests/datasets/test_solar_plants_brazil.py

torchgeo/datasets/solar_plants_brazil.py

FederCO23 · 2025-06-08T15:36:19Z

@adamjstewart, thanks again for the detailed feedback!
All 16 requested changes have been addressed. I replied to each one directly above.
Let me know if there’s anything else needed on my end. Happy to contribute!

@robmarkcole, just tagging you here to keep you in the loop

adamjstewart · 2025-06-15T10:54:18Z

Accidentally deleted my notification, but remind me to review this again this week.

adamjstewart

Still quite a lot of differences between this dataset and the other builtin datasets

adamjstewart · 2025-06-19T09:19:41Z

torchgeo/datasets/solar_plants_brazil.py

+
+    """
+
+    url = 'https://huggingface.co/datasets/FederCO23/solar-plants-brazil/resolve/main/solarplantsbrazil.zip'


Suggested change

url = 'https://huggingface.co/datasets/FederCO23/solar-plants-brazil/resolve/main/solarplantsbrazil.zip'

url = 'https://huggingface.co/datasets/FederCO23/solar-plants-brazil/resolve/1dc13a453ef6acabf08a1781c523fd1db3d9bcc5/solarplantsbrazil.zip'

Use a stable commit hash so that the MD5 doesn't change even if the dataset is later updated.

Ok. Updated the dataset URL to use a fixed commit hash to ensure stable MD5

adamjstewart · 2025-06-19T09:20:04Z

docs/api/datasets/non_geo_datasets.csv

@@ -55,6 +55,7 @@ Dataset,Task,Source,License,# Samples,# Classes,Size (px),Resolution (m),Bands
 `SkyScript`_,IC,"NAIP, orthophotos, Planet SkySat, Sentinel-2, Landsat 8--9",MIT,5.2M,-,100--1000,0.1--30,RGB
 `So2Sat`_,C,Sentinel-1/2,"CC-BY-4.0","400,673",17,32x32,10,"SAR, MSI"
 `SODA`_,OD,Aerial,"CC-BY-NC-4.0","2513",9,"~2700x~4800","RGB"
+`Solar Plants Brazil`_,S,Aerial,"CC-BY-NC-4.0","272",2,256x256,10,"RGB + NIR"


License now matches on https://huggingface.co/datasets/FederCO23/solar-plants-brazil

adamjstewart · 2025-06-19T09:22:39Z

tests/datasets/test_solar_plants_brazil.py

+        sample = dataset[0].copy()
+        sample['prediction'] = sample['mask'].clone()


I don't think a copy or clone is actually needed is it?

You're right, since prediction is not being modified, both .clone() and .copy() are unnecessary and can be removed.
For reference, I had initially followed a similar pattern used in other datasets like cloud_cover and chesapeake, which use .clone() in their test_plot() methods."

adamjstewart · 2025-06-19T09:24:05Z

tests/datasets/test_solar_plants_brazil.py

+        sample = dataset[0]
+        assert torch.all(sample['image'] > 0)
+
+    def test_download_called(


I don't think we test this for any other datasets, it's better to test that it's possible to "download" files from the tests/data directory to the test directory. See all of the other dataset tests.

Ok. Removed test_download_called and added test_already_downloaded and test_not_downloaded instead, following the structure used in other datasets.

adamjstewart · 2025-06-19T09:24:25Z

tests/datasets/test_solar_plants_brazil.py

+
+        assert called['triggered']
+
+    def test_missing_dataset_triggers_error(self, tmp_path: Path) -> None:


Isn't this identical to test_missing_dataset_raises?

Removed test_missing_dataset_triggers_error (duplicate of test_not_downloaded) and renamed for consistency with other datasets

adamjstewart · 2025-06-19T09:33:21Z

torchgeo/datasets/solar_plants_brazil.py

+        Returns:
+            None


adamjstewart · 2025-06-19T09:34:06Z

torchgeo/datasets/solar_plants_brazil.py

+        if len(self.image_paths) == 0:
+            raise DatasetNotFoundError(self)


Isn't this checked by _verify?

I’ve now removed the if len(self.image_paths) == 0 check from __init__() to align with the datasets pattern.
That said, I originally added this check to catch cases where the folder exists but is empty (ex: failed unzip). In that situation, _verify() would silently pass, but indexing into the dataset would fail later on in a less clear way.
I transfer that check to _verify()

adamjstewart · 2025-06-19T09:34:20Z

torchgeo/datasets/solar_plants_brazil.py

+        Returns:
+            None


adamjstewart · 2025-06-19T09:34:39Z

torchgeo/datasets/solar_plants_brazil.py

+        """Retrieve an image-mask pair by index.
+
+        Args:
+            index (int): Index of the sample to retrieve.


Suggested change

index (int): Index of the sample to retrieve.

index: Index of the sample to retrieve.

Already gotten from type hints

Ok, this has been addressed

adamjstewart · 2025-06-19T09:36:21Z

torchgeo/datasets/solar_plants_brazil.py

+            dict: A dictionary with the following keys:
+                - 'image': A float32 tensor of shape (C, H, W)
+                - 'mask': A long tensor of shape (1, H, W), containing binary labels


This doesn't seem to format correctly, maybe try:

Suggested change

dict: A dictionary with the following keys:

- 'image': A float32 tensor of shape (C, H, W)

- 'mask': A long tensor of shape (1, H, W), containing binary labels

A dictionary with the following keys:

- 'image': A float32 tensor of shape (C, H, W)

- 'mask': A long tensor of shape (1, H, W), containing binary labels

or just copy-n-paste the same description used by the other 125+ builtin datasets.

Got it. I'll simplify the docstring to match the style used by the other datasets. Thanks for the tip!

adamjstewart · 2025-06-19T09:38:33Z

torchgeo/datamodules/solar_plants_brazil.py

+
+    This datamodule wraps the SolarPlantsBrazil dataset, which contains
+    predefined train/val/test splits. This design ensures spatial separation
+    between samples by solar plant, preventing data leakage during training.


.. versionadded:: 0.8

Ok, included 🙌

…tion

…azil

…ad and transform logic

…PlantsBrazil - Deleted as requested - Created for trainer config - Registered solar_plants_brazil in - Updated dummy dataset and dataset test to match 256x256 input requirement

…ary test parametrization

…h other datasets

…akage

- Return mask as [H, W] to match loss function expectations - Updated test_getitem assertion to reflect correct shape - Trainer test now passes with correct mask format

- Enforced cross-platform compatibility using os.path.join in tests. - Used Literal typing for 'split' argument to improve static type safety. - Silenced mypy type check for the 'invalid split' test using # type: ignore[arg-type]. - Fully updated all docstrings in dataset and datamodule files to comply with standards (Args and Returns included).

…ring in datasets/solar_plants_brazil.py (__getitem__ method)

…rchGeo datasets

… remove batch transfer method

…d URL and _verify logic

github-actions bot added documentation Improvements or additions to documentation datasets Geospatial or benchmark datasets testing Continuous integration testing labels May 24, 2025

robmarkcole reviewed May 25, 2025

View reviewed changes

torchgeo/datasets/solar_plants_brazil.py Outdated Show resolved Hide resolved

robmarkcole reviewed May 25, 2025

View reviewed changes

torchgeo/datasets/solar_plants_brazil.py Show resolved Hide resolved

adamjstewart added this to the 0.8.0 milestone May 25, 2025

adamjstewart requested changes May 25, 2025

View reviewed changes

github-actions bot added the datamodules PyTorch Lightning datamodules label May 26, 2025

FederCO23 requested a review from adamjstewart May 27, 2025 21:26

FederCO23 force-pushed the main branch from 5d3ec4c to 46c03dc Compare May 31, 2025 18:16

adamjstewart requested changes Jun 4, 2025

View reviewed changes

FederCO23 force-pushed the main branch from 46c03dc to 6a19e7a Compare June 8, 2025 04:04

adamjstewart self-assigned this Jun 15, 2025

adamjstewart requested changes Jun 19, 2025

View reviewed changes

adamjstewart reviewed Jun 19, 2025

View reviewed changes

FederCO23 added 4 commits June 20, 2025 17:57

Add SolarPlantsBrazil dataset class with download and parsing logic

abe97b8

Register SolarPlantsBrazil dataset in __init__.py

b7a0fa5

Document SolarPlantsBrazil in non_geo_datasets.csv

be4d4bb

Add unit tests for SolarPlantsBrazil dataset

7d61775

FederCO23 added 29 commits June 20, 2025 17:57

Add SolarPlantsBrazil dataset class with docstring and download logic

0f6971f

Add unit tests for SolarPlantsBrazil dataset

1dc4a11

Add SolarPlantBrazil to dataset documentation and catalog

b3123c1

Add SolarPlantsBrazilDataModules including normalization and augmenta…

1252e62

…tion

Add tests and dummy data for SolarPlantsBrazil dataset

d08cd2f

Update dummy dataset to include val and test splits for SolarPlantsBr…

5509194

…azil

Add SolarPlantsBrazilDataModule tests and documentation entry

f8a2cfe

Complete test coverage for SolarPlantsBrazil dataset including downlo…

4236ec7

…ad and transform logic

Add unit test to cover ndim==3 edge case in on_after_batch_transfer

bfa61d3

finalize SolarPlantsBrazil dataset with pytest coverage and lint fixes

dfbc368

docs: update dataset display name to 'Solar Plants Brazil'

211941b

Refactor: Remove old datamodule test, add full trainer test for Solar…

0b2596f

…PlantsBrazil - Deleted as requested - Created for trainer config - Registered solar_plants_brazil in - Updated dummy dataset and dataset test to match 256x256 input requirement

Use os.path.join for cross-platform path handling and remove unnecess…

19d03e1

…ary test parametrization

Move download and error handling tests into class for consistency wit…

a0830c8

…h other datasets

Replace mean and std values with training split only to avoid data le…

3414209

…akage

Remove custom setup and use base NonGeoDataModule logic

324457c

Remove extra channel dim in mask output

a9a7d7b

- Return mask as [H, W] to match loss function expectations - Updated test_getitem assertion to reflect correct shape - Trainer test now passes with correct mask format

docs: fix docstring formatting for correct HTML rendering

78a8a95

docs: move DatasetNotFoundError to __init__ docstring

1948115

Add Literal type for split and cross-platform paths in tests

5eb6676

Fix Prettier formatting in solar_plants_brazil.yaml and invalid docst…

d4a3143

…ring in datasets/solar_plants_brazil.py (__getitem__ method)

Fix docstring formatting in __getitem__ after 'ruff check .'

da63d20

Add optional checksum validation to the dataset

af655f0

Update dataset URL to use fixed commit hash and stable MD5

b338f89

Refactor tests: remove duplicates, simplify plot, align with other To…

8f8c52f

…rchGeo datasets

Update docstring with dataset link, remove unused Returns section and…

535112d

… remove batch transfer method

Refactor SolarPlantsBrazil dataset and datamodule docstrings, downloa…

115d9eb

…d URL and _verify logic

Simplify test_plot by removing unnecessary .copy() and .clone() calls

84e344e

FederCO23 force-pushed the main branch from db53f48 to 84e344e Compare June 20, 2025 21:05


		"""

		url = 'https://huggingface.co/datasets/FederCO23/solar-plants-brazil/resolve/main/solarplantsbrazil.zip'

		sample = dataset[0].copy()
		sample['prediction'] = sample['mask'].clone()


		assert called['triggered']

		def test_missing_dataset_triggers_error(self, tmp_path: Path) -> None:

		if len(self.image_paths) == 0:
		raise DatasetNotFoundError(self)

	index (int): Index of the sample to retrieve.
	index: Index of the sample to retrieve.

Add SolarPlantsBrazil dataset for photovoltaic panel detection (semantic segmentation) #2797

Are you sure you want to change the base?

Add SolarPlantsBrazil dataset for photovoltaic panel detection (semantic segmentation) #2797

Uh oh!

Conversation

FederCO23 commented May 24, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Changes introduced:

Example output from plot():

References:

Uh oh!

FederCO23 commented May 24, 2025

Uh oh!

FederCO23 commented May 24, 2025 via email

Uh oh!

FederCO23 commented May 24, 2025 via email

Uh oh!

Uh oh!

Uh oh!

robmarkcole commented May 25, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

adamjstewart left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

FederCO23 commented May 27, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

robmarkcole commented May 28, 2025

Uh oh!

robmarkcole commented May 30, 2025

Uh oh!

FederCO23 commented May 31, 2025

Uh oh!

adamjstewart commented May 31, 2025

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

FederCO23 commented Jun 8, 2025

Uh oh!

adamjstewart commented Jun 15, 2025

Uh oh!

adamjstewart left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

FederCO23 commented May 24, 2025 •

edited

Loading

Example output from `plot()`:

robmarkcole commented May 25, 2025 •

edited

Loading

FederCO23 commented May 27, 2025 •

edited

Loading