Skip to content
This repository has been archived by the owner on Oct 9, 2023. It is now read-only.

Commit

Permalink
CHANGELOG.md
Browse files Browse the repository at this point in the history
  • Loading branch information
ethanwharris committed Aug 13, 2021
1 parent 38a978b commit cd3a806
Showing 2 changed files with 10 additions and 23 deletions.
10 changes: 10 additions & 0 deletions CHANGELOG.md
Original file line number Diff line number Diff line change
@@ -34,12 +34,20 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).

- Added Flash Zero, a zero code command line ML platform built with flash ([#611](https://github.com/PyTorchLightning/lightning-flash/pull/611))

- Added support for `.npy` and `.npz` files to `ImageClassificationData` and `AudioClassificationData` ([#651](https://github.com/PyTorchLightning/lightning-flash/pull/651))

- Added support for `from_csv` to the `AudioClassificationData` ([#651](https://github.com/PyTorchLightning/lightning-flash/pull/651))

- Added option to pass a `resolver` to the `from_csv` and `from_pandas` methods of `ImageClassificationData`, which is used to resolve filenames given IDs ([#651](https://github.com/PyTorchLightning/lightning-flash/pull/651))

### Changed

- Changed how pretrained flag works for loading weights for ImageClassifier task ([#560](https://github.com/PyTorchLightning/lightning-flash/pull/560))

- Removed bolts pretrained weights for SSL from ImageClassifier task ([#560](https://github.com/PyTorchLightning/lightning-flash/pull/560))

- Changed the behaviour of the `sampler` argument of the `DataModule` to take a `Sampler` type rather than instantiated object ([#651](https://github.com/PyTorchLightning/lightning-flash/pull/651))

### Fixed

- Fixed a bug where serve sanity checking would not be triggered using the latest PyTorchLightning version ([#493](https://github.com/PyTorchLightning/lightning-flash/pull/493))
@@ -50,6 +58,8 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).

- Fixed a bug where some tasks were not compatible with PyTorch 1.7 due to use of `torch.jit.isinstance` ([#611](https://github.com/PyTorchLightning/lightning-flash/pull/611))

- Fixed a bug where custom samplers would not be properly forwarded to the data loader ([#651](https://github.com/PyTorchLightning/lightning-flash/pull/651))

## [0.4.0] - 2021-06-22

### Added
23 changes: 0 additions & 23 deletions tests/image/classification/test_data.py
Original file line number Diff line number Diff line change
@@ -548,29 +548,6 @@ def test_from_csv_multi_target(multi_target_csv):
assert labels.shape == (2, 2)


# @pytest.fixture
# def bad_csv_multi_image(image_tmpdir):
# with open(image_tmpdir / "metadata.csv", "w") as csvfile:
# fieldnames = ["image", "target"]
# writer = csv.DictWriter(csvfile, fieldnames)
# writer.writeheader()
# writer.writerow({"image": "image", "target": "Ants"})
# return str(image_tmpdir / "metadata.csv")


# @pytest.mark.skipif(not _IMAGE_TESTING, reason="image libraries aren't installed.")
# def test_from_bad_csv_multi_image(bad_csv_multi_image):
# with pytest.raises(ValueError, match="Found multiple matches"):
# img_data = ImageClassificationData.from_csv(
# "image",
# ["target"],
# train_file=bad_csv_multi_image,
# batch_size=1,
# num_workers=0,
# )
# _ = next(iter(img_data.train_dataloader()))


@pytest.fixture
def bad_csv_no_image(image_tmpdir):
with open(image_tmpdir / "metadata.csv", "w") as csvfile:

0 comments on commit cd3a806

Please sign in to comment.