Matt Hancock

I'm saying that in the non-notebook setting, we can play fast and loose with making things too big since the window manager automatically rescales things to fit in the available...

There's no such thing as "nodule" in the LIDC dataset, i.e., there's no field that indicates Annotation 1 and Annotation 2 actually refer to the same physical nodule. However, there...

There's no reason that all the annotations were annotated on the same number of slices. `mask[:,:,2]` is attempting to get the third slice of the mask, and apparently for annotation...

Sorry, I'm not sure what you're asking. It sounds like the cluster annotations function already gives your what you want. It returns a list of list of list of annotations....

`cluster_annotations` is a member function of the Scan class: https://pylidc.github.io/scan.html#pylidc.Scan.cluster_annotations 1. Retrieve a Scan instance (e.g., querying for a certain patient ID) 2. Run `cluster_annotations` on the scan. This groups...

```python nodules = scan.cluster_annotations() nodule1 = nodules[0] annotation3_nodule1 = nodule1[2] ... mask = annotation3_nodule1.boolean_mask() ... ``` etc

> This looks like a duplicate of #481, could you confirm please? Yep; looks like we're talking about the same underlying issue.

You're repeatedly computing `scan.to_volume()` and `ann.boolean_mask()` in your augmentations pipeline. `to_volume()` reads all the DICOM from disk and converts it to a numpy volume. `ann.boolean_mask()` does ray-casting for each pixel...

Closing as this is quite old now

This PR has 3K lines of code changed. I don't think that was your intent.