Fixes for GPU and MultiLoss

[jdb78]

Description

This PR fixes issues with training on GPU and MultiLoss

Checklist

• Linked issues (if existing)
• Amended changelog for large changes (and added myself there as contributor)
• Added/modified tests
• Used pre-commit hooks when committing to ensure that code is compliant with hooks. Install hooks with pre-commit install.
  To run hooks independent of commit, execute pre-commit run --all-files

Make sure to have fun coding!

[codecov-commenter]

Codecov Report

Merging #908 (022147c) into master (af9e1d3) will decrease coverage by 0.65%.
The diff coverage is 79.31%.

@@            Coverage Diff             @@
##           master     #908      +/-   ##
==========================================
- Coverage   89.88%   89.22%   -0.66%     
==========================================
  Files          26       26              
  Lines        4200     4224      +24     
==========================================
- Hits         3775     3769       -6     
- Misses        425      455      +30     

Flag	Coverage Δ
cpu	89.22% <79.31%> (-0.66%)	⬇️
pytest	89.22% <79.31%> (-0.66%)	⬇️

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

Impacted Files	Coverage Δ
...ing/models/temporal_fusion_transformer/__init__.py	99.56% <ø> (ø)
pytorch_forecasting/metrics.py	87.32% <77.77%> (-4.65%)	⬇️
pytorch_forecasting/data/encoders.py	88.30% <100.00%> (-0.30%)	⬇️
pytorch_forecasting/models/base_model.py	88.82% <100.00%> (ø)
pytorch_forecasting/models/nn/rnn.py	87.34% <0.00%> (-5.07%)	⬇️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update af9e1d3...022147c. Read the comment docs.