[BUG] fixes correct passing of index in v2

[phoeenniixx]

Fixes #2367

Adds a new prediction_start_idx to the collate_fn that is returned when return_info has index.

[phoeenniixx]

Has to remove index from the cell in the tutorial as previously it was just printing y[1] (a random placeholder, see here). We still need to add some thing like x_to_index of v1, to make this index actually useful. But for now, it is similar to v1, matching group with decoder_idx (see x_to_index).

But this breaks what the cell printed before, as now preds["index"] is a 2d tensor and not 1d. So, i just removed it, it makes no sense there.

[review-notebook-app]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[phoeenniixx]

The big diff in the tutorial is because I had to rerun the whole nb locally to update the outputs :)

[codecov]

Codecov Report

❌ Patch coverage is 84.61538% with 2 lines in your changes missing coverage. Please review.
⚠️ Please upload report for BASE (main@062aeb4). Learn more about missing BASE report.

Files with missing lines	Patch %	Lines
pytorch_forecasting/callbacks/predict.py	81.81%	2 Missing ⚠️

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #2369   +/-   ##
=======================================
  Coverage        ?   88.12%           
=======================================
  Files           ?      196           
  Lines           ?    11061           
  Branches        ?        0           
=======================================
  Hits            ?     9747           
  Misses          ?     1314           
  Partials        ?        0           

Flag	Coverage Δ
cpu	88.12% <84.61%> (?)
pytest	88.12% <84.61%> (?)

Flags with carried forward coverage won't be shown. Click here to find out more.

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

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[fkiraly]

The nb looks substantially different, can you kindly explain?

[phoeenniixx]

The nb looks substantially different, can you kindly explain?

As the nb had wrong indexes printed (which were just y[1]), I had to rerun the whole nb to remove that wrong values.