Stop generating at .npy when running Topostats #802
Labels
Comments
ns-rse
added a commit
that referenced
this issue
Feb 20, 2024
Closes #802 Arrays and data are now saved in HDF5 files (#790) and so `.npy` arrays are somewhat redundant. This PR leaves the `io.save_array()` function in place should interactive use require saving of arrays but removes its use from `topostats.processing.run_filters()` so that the `.npy` files are no longer saved to disk. Users wishing to access processed data should load it from the HDF5 formatted `.topostats` files that are saved during processing.
ns-rse
added a commit
that referenced
this issue
Mar 4, 2024
Closes #806 Whislt #806 was meant to address why no Matplotlib image was generated and tested/compared to a target image when plotting without axes or colorbar I discovered this was because the image is saved with `Images.save_array_figure()`. This in turn uses [`matplotlib.pyplot.imsave()`](https://matplotlib.org/stable/api/_as_gen/matplotlib.pyplot.imsave.html) to save the image and the test was using [`skimage.io.imread()`](https://scikit-image.org/docs/stable/api/skimage.io.html#skimage.io.imread) to read the image back as an array and test the `np.sum()` and `img.shape`. We no longer need to save arrays nor images as readable arrays (see #804 / #802) and so the need to save the image in this manner seemed redudant. The `Images.save_figure()` already has logic which excludes the axes and scale bars (see lines 321 of current changeset). Further the logic for deciding what to save within `Image.plot_and_save()` seemed overly complicated and if `Images.save_array_figure()` were being called `Images.plot_and_save()` did not return a `fig` (Matplotlib Image) that could be saved and tested using `pytest-mpl` extension. To this end I have... 1. Removed `Images.save_array_figure()`. 2. Tweaked the plotting options under `Images.savefig()` starting at line 321 to use a [tight layout](https://matplotlib.org/2.0.2/users/tight_layout_guide.html) which ensures there is no border (see note below). 3. Simplified the logic in `Images.plot_and_save()` controlling whether images are saved so that `ValueError` are raised if `Images.savefig == False` or `if Images.image_set in ["all", "core"] or self.core_set:` evaluates to `False`. Ensures images are explicitly closed to reduce memory usage. 4. Added tests for the raising of `ValueError` exceptions in `Images.plot_and_save()` 5. Tidied up some test names to be consistent (resulted in image name change). 6. Tidied up passing of `plotting_config` into these tests. 7. Added a `pytest-mpl` test for `test_plot_and_save_no_axes_no_colorbar()`, adding the target image as required by #806 Note on Borders The image generated by `pytest-mpl` against which comparisons are made _does_ have a border because it is saving the returned `fig` object itself. The actual generated image doesn't have a border (manual checks have been made).
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Is your feature request related to a problem?
A
.npyfile is generated for each processed image along side an image and an HDF5.topostatsfile. This is no longer necessary since the image data is all head in the.topostatsfile and the.npyfile just takes up extra space.Describe the solution you would like.
Stop the generation of the
.npyfile.Describe the alternatives you have considered.
Could also be made optional in the config file.
Additional context
No response
The text was updated successfully, but these errors were encountered: