KeyError 'auto_reward_delay' on legacy pkl files

[ryyesterday]

Getting a KeyError on 'auto_reward_delay' when running validation functions on 2p6 change detection datasets. From https://github.com/AllenInstitute/visual_behavior_analysis/blob/master/visual_behavior/validation/qc.py#L40

As an example, see the pkl file here:

\allen\programs\braintv\production\visualbehavior\prod0\specimen_685152008\ophys_session_720323943\720323943_385255_20180716142658_stim.pkl

There may be other key errors - this is the first I saw. This occurs when running with 0.4.0.

[neuromusic]

I had no problems with this file. Note: 2P6 is running the legacy code, so the following works for me:

import pandas as pd
data = pd.read_pickle('/allen/programs/braintv/production/visualbehavior/prod0/specimen_685152008/ophys_session_720323943/720323943_385255_20180716142658_stim.pkl')
from visual_behavior.translator.foraging import data_to_change_detection_core
core_data = data_to_change_detection_core(data)

[ryyesterday]

I'm able to load the data. The problem occurs when running the validation functions.

Did you check that you have an 'auto_reward_delay' key in core_data['metadata']?

[neuromusic]

ohhhhhh

[neuromusic]

ok, fixed on branch 0.4.x (which is what you probably want to use until we've got ophys support stable on master)

you can also use the v0.4.1 tag

also: I believe that @nicain & @NileGraddis have some magic sauce setup to deploy tags as releases on a PyPi server somewhere?