Handling of 3D data

[adamrobcarter]

This is related to the section of the code that parses our "x, y, t" data

This line:

countoscope/countoscope/tools.py

Line 30 in a2b1ae7

elif np.shape(array)[1] == 4: # format if X Y Z N

is gonna cause problems, because we also use a format that is rows of "x, y, t, ID" (where ID is a unique integer for each particle in the dataset). I don't think we're using rows of "x, y, z, t" at any point at the moment, so maybe we don't need this 3D detection. Maybe we should formalise our format

[simongravelle]

We should definitely decide on what formats we are allowing.

In practice, it does not matter so much, as the converter is outside the code anyway. So we could have as many converters as we have file formats.

[simongravelle]

Ideally, we will need to have all the possible type of trajectory files within the "datasets" repository, so that we have a good overview of what we need to be able to import.