sfmig comments

Results 61 comments of


                                            sfmig

Update cell classification network

First steps (chat with @IgorTatarnikov) - [migrate to Keras 3.0](https://keras.io/guides/migrating_to_keras_3/#saving-a-model-in-the-tf-savedmodel-format) - verify accuracy of results (via tests) - verify no performance differences (via time/memory benchmarks)

Create a glossary of common terms

**Agreed terminology** - default (cartesian) coordinate system: - origin top left of the image, - x-axis is positive across columns (pointing right), - y-axis resulting from the [right hand rule](https://en.wikipedia.org/wiki/Right-hand_rule);...

Create a glossary of common terms

also include/link to the concept of [movement dataset](https://movement.neuroinformatics.dev/getting_started/movement_dataset.html)

Create a glossary of common terms

and the concept of bounding boxes (and tracked bounding boxes)

Implement LightningPoses losses for outlier detection

Currently we see this mostly for outliers detection, but we may want to additionally use this as a quality metric for predicted poses in the absence of ground truth.

Cluster imported pose data

The [ethome](https://github.com/benlansdell/ethome) project implements some clustering functionalities - may be a good source of inspo ✨

Cluster imported pose data

Somewhat related to #152 in the sense that both rely on computing PCA on pose data

Rethink definition of displacement vector

yes, I was aware of this option too 😁 My point is that it is a bit more intuitive to think displacement as t to t+1

Consider adding a distance and speed property to accessor

Make a method instead of a property. Nomenclature: `ds.get_cumulative_distance()` ?

Consider adding a distance and speed property to accessor

@niksirbi @lochhh seems like we can close this now?