Add condition_labels as an argument
#18
Conversation
| if condition_labels is None: | ||
| condition_labels = np.array([f"condition_{i}" for i in range(number_of_conditions)], dtype="U") |
There was a problem hiding this comment.
why generate if they don't exist? Why not just make them optional?
There was a problem hiding this comment.
I'd rather not add data if it is redundant and does not include any new information
There was a problem hiding this comment.
Assuming people want a mock with data if they specify num_conditions > 0
There was a problem hiding this comment.
I think I'd really rather not. It's auto-magic, which is hard for users to predict, and it is duplication of data.
There was a problem hiding this comment.
Sure. Don't feel very strong about this. Removing.
There was a problem hiding this comment.
But your comment of duplication of data I don't understand. What is being duplicated?
There was a problem hiding this comment.
"condition 0" is at index 0
"condition 1" is at index 1
etc.
I guess it's not really duplicating since that data does not already exist, but it's adding data that is not really telling us anything we couldn't already infer. I think it's more meaningful to be able to tell whether conditions are labeled or not.
There was a problem hiding this comment.
Makes sense, I was just wondering if I missed something. I eliminated this behavior for the mock in the last commit.
…' into add_condition_label
* test for condition-less case working * tests for multiple conditions * documentation * ruff * typos * Update spec/ndx-binned-spikes.extensions.yaml * timestamps to event timestamps * ruff and spelling * Add `condition_labels` as an argument (#18) * add condition labels * Update src/pynwb/ndx_binned_spikes/__init__.py * Update spec/ndx-binned-spikes.extensions.yaml * remove automatic creation of labels in the mock * typo on the spec generation --------- Co-authored-by: Ben Dichter <[email protected]> --------- Co-authored-by: Ben Dichter <[email protected]>
A request from @bendichter in #17.
Separated and with #17 as its base to simplify the discussion.