Alessio Buccino

@chrishalcrow @zm711 I pushed an update that removes the need to compute `template_similarity`, which is counter-intuituve. In fact, this is only needed for cross-correlograms, but not for auto-correlogram plots. Tested...

Thanks! One simple option is to make the `InjectDriftingTemplatesRecording` not JSON serializable, so it will be saved to pkl at all instances

@cwindolf you can just add this at the end of the init: ``` self._serializability["json"] = False ```

@h-mayorquin maybe we just need to extend the `SIJsonEncoder` to support `Templates`?

@cwindolf no we don't need to inherit from `BaseExtractor`, because that's for a `Recording` or `Sorting` object. @h-mayorquin let's discuss about this tomorrow!

@cwindolf can you test this one #3130? With this PR the provenance is saved to pkl if the recording is not JSON serializable (and this is the case for then...

@cwindolf you can also do this! (better API) ``` rec1.get_parent().drifting_templates.templates_array_moved.shape ```

Hi Gabe! This has been on our Todo list for a while! I'll try to add the DBC probes over the summer :) Cheers Alessio

Unfortunately not yet..We have a scheduled bug fix release by the end of next week and we'll aim to tackle this then! Sorry for the wait!

Fixed in #287