-
Notifications
You must be signed in to change notification settings - Fork 6
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
### Description Following #278, this PR reorganizes the exported BMZ README: - Put all configuration to the end - Fix title level of algorithm - Put data description first - Remove training section - Add a validation section - Fix link to documentation In addition, the API for the BMZ export has changed: - `data_description` is now mandatory - `model_version` is an optional parameter allowing to version models - `cover` is an optional parameter Users can now provide a path to a cover, no validation is done of that path. If no cover is provided, then we do something that is probably not optimized for multichannels: - For 2D images, the images (input and output) are split horizontally, or vertically if they have different height. If they have different width as well, then an error is raised. - For Z stack, the middle slice is selected. - For images with 2 channels, the channels are used as `blue` and `green` - For images with 3 channels, they are interpreted as RGB - For images with 4 or more channels, the first 4 channels are used. The pixel value are normalised and multiplied with the RGB vectors of 4 pre-defined colors, and all channels are summed into an RGB image. Note that that this is not a great way to represent scientific data, we should apply LUT to the grey scale channels and recompose overlays. Look up tables have been worked out for scientific figures and we could use those. If that is acceptable for now, it will be easy to replace it. ### Changes Made - **Added**: - `cover_factory.py` - helper scripts to inspect BMZ README and covers. - **Modified**: all BMZ related modules. ### Related Issues #278 #176 ### Breaking changes Any call to `careamist.export_bmz`. ### Additional Notes and Examples The results README looks like this: ```markdown # Noise2Void - CAREamics ## Data description Mydata ## Algorithm description: Noise2Void is a UNet-based self-supervised algorithm that uses blind-spot training to denoise images. In short, in every patches during training, random pixels are selected and their value replaced by a neighboring pixel value. The network is then trained to predict the original pixel value. The algorithm relies on the continuity of the signal (neighboring pixels have similar values) and the pixel-wise independence of the noise (the noise in a pixel is not correlated with the noise in neighboring pixels). ## Configuration Noise2Void was trained using CAREamics (version 0.1.0) using the following configuration: algorithm_config: algorithm: n2v loss: n2v lr_scheduler: name: ReduceLROnPlateau parameters: {} model: architecture: UNet conv_dims: 2 depth: 2 final_activation: None in_channels: 1 independent_channels: true n2v2: false num_channels_init: 32 num_classes: 1 optimizer: name: Adam parameters: lr: 0.0001 data_config: axes: YX batch_size: 2 data_type: array patch_size: - 64 - 64 transforms: - flip_x: true flip_y: true name: XYFlip p: 0.5 - name: XYRandomRotate90 p: 0.5 - masked_pixel_percentage: 0.2 name: N2VManipulate roi_size: 11 strategy: uniform struct_mask_axis: none struct_mask_span: 5 experiment_name: export_bmz_readme training_config: accumulate_grad_batches: 1 check_val_every_n_epoch: 1 checkpoint_callback: auto_insert_metric_name: false mode: min monitor: val_loss save_last: true save_top_k: 3 save_weights_only: false verbose: false enable_progress_bar: true gradient_clip_algorithm: norm max_steps: -1 num_epochs: 10 precision: '32' version: 0.1.0 ## Validation In order to validate the model, we encourage users to acquire a test dataset with ground-truth data. Additionally, inspecting the residual image (difference between input and predicted image) can be helpful to identify whether real signal is removed from the input image. ## References Krull, A., Buchholz, T.O. and Jug, F., 2019. "Noise2Void - Learning denoising from single noisy images". In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition (pp. 2129-2137). doi: 10.1109/cvpr.2019.00223 ## Links - [CAREamics repository](https://github.com/CAREamics/careamics) - [CAREamics documentation](https://careamics.github.io/) ``` --- **Please ensure your PR meets the following requirements:** - [x] Code builds and passes tests locally, including doctests - [x] New tests have been added (for bug fixes/features) - [x] Pre-commit passes - [ ] PR to the documentation exists (for bug fixes / features)
- Loading branch information
1 parent
6eb3627
commit 66e34f7
Showing
12 changed files
with
296 additions
and
52 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,29 @@ | ||
| #!/usr/bin/env python | ||
| """Export a README file for the bioimage model zoo.""" | ||
| from pathlib import Path | ||
|
|
||
| from careamics.config import create_n2v_configuration | ||
| from careamics.model_io.bioimage._readme_factory import readme_factory | ||
|
|
||
|
|
||
| def main(): | ||
| # create configuration | ||
| config = create_n2v_configuration( | ||
| experiment_name="export_bmz_readme", | ||
| data_type="array", | ||
| axes="YX", | ||
| patch_size=(64, 64), | ||
| batch_size=2, | ||
| num_epochs=10, | ||
| ) | ||
| # export README | ||
| readme_path = readme_factory( | ||
| config=config, careamics_version="0.1.0", data_description="Mydata" | ||
| ) | ||
|
|
||
| # copy file to __file__ | ||
| readme_path.rename(Path(__file__).parent / "README.md") | ||
|
|
||
|
|
||
| if __name__ == "__main__": | ||
| main() |
Empty file.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,171 @@ | ||
| """Convenience function to create covers for the BMZ.""" | ||
|
|
||
| from pathlib import Path | ||
|
|
||
| import numpy as np | ||
| from numpy.typing import NDArray | ||
| from PIL import Image | ||
|
|
||
| color_palette = np.array( | ||
| [ | ||
| np.array([255, 195, 0]), # grey | ||
| np.array([189, 226, 240]), | ||
| np.array([96, 60, 76]), | ||
| np.array([193, 225, 193]), | ||
| ] | ||
| ) | ||
|
|
||
|
|
||
| def _get_norm_slice(array: NDArray) -> NDArray: | ||
| """Get the normalized middle slice of a 4D or 5D array (SC(Z)YX). | ||
| Parameters | ||
| ---------- | ||
| array : NDArray | ||
| Array from which to get the middle slice. | ||
| Returns | ||
| ------- | ||
| NDArray | ||
| Normalized middle slice of the input array. | ||
| """ | ||
| if array.ndim not in (4, 5): | ||
| raise ValueError("Array must be 4D or 5D.") | ||
|
|
||
| channels = array.shape[1] > 1 | ||
| z_stack = array.ndim == 5 | ||
|
|
||
| # get slice | ||
| if z_stack: | ||
| array_slice = array[0, :, array.shape[2] // 2, ...] | ||
| else: | ||
| array_slice = array[0, ...] | ||
|
|
||
| # channels | ||
| if channels: | ||
| array_slice = np.moveaxis(array_slice, 0, -1) | ||
| else: | ||
| array_slice = array_slice[0, ...] | ||
|
|
||
| # normalize | ||
| array_slice = ( | ||
| 255 | ||
| * (array_slice - array_slice.min()) | ||
| / (array_slice.max() - array_slice.min()) | ||
| ) | ||
|
|
||
| return array_slice.astype(np.uint8) | ||
|
|
||
|
|
||
| def _four_channel_image(array: NDArray) -> Image: | ||
| """Convert 4-channel array to Image. | ||
| Parameters | ||
| ---------- | ||
| array : NDArray | ||
| Normalized array to convert. | ||
| Returns | ||
| ------- | ||
| Image | ||
| Converted array. | ||
| """ | ||
| colors = color_palette[np.newaxis, np.newaxis, :, :] | ||
| four_c_array = np.sum(array[..., :4, np.newaxis] * colors, axis=-2).astype(np.uint8) | ||
|
|
||
| return Image.fromarray(four_c_array).convert("RGB") | ||
|
|
||
|
|
||
| def _convert_to_image(original_shape: tuple[int, ...], array: NDArray) -> Image: | ||
| """Convert to Image. | ||
| Parameters | ||
| ---------- | ||
| original_shape : tuple | ||
| Original shape of the array. | ||
| array : NDArray | ||
| Normalized array to convert. | ||
| Returns | ||
| ------- | ||
| Image | ||
| Converted array. | ||
| """ | ||
| n_channels = original_shape[1] | ||
|
|
||
| if n_channels > 1: | ||
| if n_channels == 3: | ||
| return Image.fromarray(array).convert("RGB") | ||
| elif n_channels == 2: | ||
| # add an empty channel to the numpy array | ||
| array = np.concatenate([np.zeros_like(array[..., 0:1]), array], axis=-1) | ||
|
|
||
| return Image.fromarray(array).convert("RGB") | ||
| else: # more than 4 | ||
| return _four_channel_image(array[..., :4]) | ||
| else: | ||
| return Image.fromarray(array).convert("L").convert("RGB") | ||
|
|
||
|
|
||
| def create_cover(directory: Path, array_in: NDArray, array_out: NDArray) -> Path: | ||
| """Create a cover image from input and output arrays. | ||
| Input and output arrays are expected to be SC(Z)YX. For images with a Z | ||
| dimension, the middle slice is taken. | ||
| Parameters | ||
| ---------- | ||
| directory : Path | ||
| Directory in which to save the cover. | ||
| array_in : numpy.ndarray | ||
| Array from which to create the cover image. | ||
| array_out : numpy.ndarray | ||
| Array from which to create the cover image. | ||
| Returns | ||
| ------- | ||
| Path | ||
| Path to the saved cover image. | ||
| """ | ||
| # extract slice and normalize arrays | ||
| slice_in = _get_norm_slice(array_in) | ||
| slice_out = _get_norm_slice(array_out) | ||
|
|
||
| horizontal_split = slice_in.shape[-1] == slice_out.shape[-1] | ||
| if not horizontal_split: | ||
| if slice_in.shape[-2] != slice_out.shape[-2]: | ||
| raise ValueError("Input and output arrays have different shapes.") | ||
|
|
||
| # convert to Image | ||
| image_in = _convert_to_image(array_in.shape, slice_in) | ||
| image_out = _convert_to_image(array_out.shape, slice_out) | ||
|
|
||
| # split horizontally or vertically | ||
| if horizontal_split: | ||
| width = image_in.width // 2 | ||
|
|
||
| cover = Image.new("RGB", (image_in.width, image_in.height)) | ||
| cover.paste(image_in.crop((0, 0, width, image_in.height)), (0, 0)) | ||
| cover.paste( | ||
| image_out.crop( | ||
| (image_in.width - width, 0, image_in.width, image_in.height) | ||
| ), | ||
| (width, 0), | ||
| ) | ||
| else: | ||
| height = image_in.height // 2 | ||
|
|
||
| cover = Image.new("RGB", (image_in.width, image_in.height)) | ||
| cover.paste(image_in.crop((0, 0, image_in.width, height)), (0, 0)) | ||
| cover.paste( | ||
| image_out.crop( | ||
| (0, image_in.height - height, image_in.width, image_in.height) | ||
| ), | ||
| (0, height), | ||
| ) | ||
|
|
||
| # save | ||
| cover_path = directory / "cover.png" | ||
| cover.save(cover_path) | ||
|
|
||
| return cover_path |
Oops, something went wrong.