consolidate

src/depiction/image/multi_channel_image_concatenation.py

@@ -63,7 +63,7 @@ def relabel_combined_image(self, image: MultiChannelImage) -> MultiChannelImageC
         """
         # Ensure the new image has the same shape as the original combined image
         original_combined = self.get_combined_image()
-        if image.dimensions != original_combined.dimensions:
+        if (image.sizes["y"], image.sizes["x"]) != (original_combined.sizes["y"], original_combined.sizes["x"]):
             raise ValueError("The new image must have the same shape as the original combined image")
         labeled = image.append_channels(self._data.sel_channels(coords=["image_index"]))
         return MultiChannelImageConcatenation(data=labeled)

src/depiction/visualize/plot_image.py

@@ -84,7 +84,7 @@ def plot_channels_grid(
         n_cols = min(n_channels, n_per_row)
 
         # determine the figure size
-        im_width, im_height = self._image.dimensions
+        im_width, im_height = self._image.sizes["x"], self._image.sizes["y"]
         aspect_ratio = im_width / im_height
 
         # set up the grid

src/depiction_cluster_sandbox/workflow/proc/compute_image_umap_coefs.py

@@ -34,7 +34,7 @@ def compute_image_umap_coefs(
         input_image = retain_features(feature_selection=feature_selection, image=input_image)
 
     # compute the umap transformation into 2D
-    logger.info(f"Computing UMAP for input image with shape {input_image.dimensions}")
+    logger.info(f"Computing UMAP for input image {input_image}")
     umap = UMAP(n_components=2, n_jobs=n_jobs, random_state=random_state)
     values = umap.fit_transform(input_image.data_flat.values.T)