Use concurrent batch pipeline for ~30x speed up

[dfulu]

Add the PVNet concurrent datapipe to the concurrent batch creation and backtest scripts. The new data is already in ocf_datapipes and is about 30x faster than the old datapipe for creating concurrent batches

[codecov]

Codecov Report

Attention: Patch coverage is 0% with 1 line in your changes missing coverage. Please review.

Please upload report for BASE (main@4473d33). Learn more about missing BASE report.

Files	Patch %	Lines
pvnet/models/base_model.py	0.00%	1 Missing ⚠️

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #236   +/-   ##
=======================================
  Coverage        ?   58.48%           
=======================================
  Files           ?       29           
  Lines           ?     1879           
  Branches        ?        0           
=======================================
  Hits            ?     1099           
  Misses          ?      780           
  Partials        ?        0           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[zakwatts]

Looks good!

[zakwatts]

Might be worth getting @AUdaltsova and @Sukh-P to review also as they use datapipes quite a bit

[Sukh-P]

Great work adding this in and getting the speedup! Just checking has this been reran and tested locally?

[dfulu]

Great work adding this in and getting the speedup! Just checking has this been reran and tested locally?

Yeh, it has been tested locally on a small set of examples. I'm going to leave it open for now since I am making concurrent batches to train on and Zak is doing a backtest. Just in case more bugs come out of the woodwork during those