MILESTONE: Merging development into main 🎉

[KristinaUlicna]

PR contribution summary

Why is this PR useful / good for? Please describe the problem(s) you're trying to address.

• Major merger between development and main

List of proposed changes / linked issues & discussions

Among the already closed issues, this PR will close:

• ✅ Resolves [DEVELOPMENT] Utilise edge_attr in graph convolutions #220
• ✅ Resolves [DEVELOPMENT] Visualise the whole graph inference after each epoch #222
• ✅ Resolves [DEVELOPMENT] Generalise GCN for GNNClassifier object for training #244
• ✅ Resolves [DEVELOPMENT] How to set a global config with different font attributes in matplotlib? #247
• ✅ Resolves [FEATURE REQUEST] Consider using .yaml files for hyperparameter config #254
• ✅ Resolves [DEVELOPMENT] Transfer the PREDICTION probabilities into a dictionary  #255
• ✅ Resolves [FEATURE REQUEST] Log F1-score metric to Tensorboard log #257
• ✅ Resolves [FEATURE REQUEST] Define a configurable dictonary of colormaps throughout the pipeline #258
• ✅ Resolves [FEATURE REQUEST] Implement optimiser weight decay #261
• ✅ Resolves [FEATURE REQUEST] Pool all colour selections into styling.py script #263
• ✅ Resolves [DEVELOPMENT] Allow specifying variable number of graph conv & dense linear layers #281
• ✅ Resolves [FEATURE REQUEST] Output the summary of the model architecture for completeness #282
• ✅ Resolves [DEVELOPMENT] Store edge properties in Data object for training #284
• ✅ Resolves Add more sophisticated classifier model options #291
• ✅ Resolves [FEATURE REQUEST] Consider visualising the edge properties in the dataset  #294
• ✅ Resolves [BUG] Model architecture summary doesn't save up GNN layers by name #296
• ✅ Resolves [DEVELOPMENT] Allow feature / property extraction to be computed at run-time #297
• ✅ Resolves [DEVELOPMENT] Store edge properties in Data object for training #284
• ✅ Resolves [DEVELOPMENT] Write proper documentation for EdgePropertyCruncher #299
• ✅ Resolves [DEVELOPMENT] Compute edge properties #303
• ✅ Resolves [DEVELOPMENT] Define an EdgePropertyCruncher #304
• ✅ Resolves Clarify which properties are used in the training #310
• ✅ Resolves [BUG] Undefined roc_auc_score if only one class present in y_true #315
• ✅ Resolves [FEATURE REQUEST] Separate inference into a stand-alone module #318
• ✅ Resolves [DEVELOPMENT] Concatenate Laplacian-multiplied node features #330

What should a reviewer concentrate their feedback on?

Major merger, all changes have been reviewed when merged to development.

What type of PR is this? (check all applicable)

• 🪄 Feature
• 🐛 Bug fix
• 📝 Documentation / update README.md
• #️⃣ Documentation / code annotation
• 🧑‍💻 Code refactor / style
• 🔥 Performance Improvements
• ✅ Test

Added tests?

• 👍 yes
• 🙅 no, because they aren't needed
• 🙋 no, because I need some help

---

PR review summary

Describe what this PR does & how you reviewed the individual items, where needed:

Some helper checks to tick off:

• Focus on image annotation
• Focus on model training
• Could any optimization be applied?
• Is there any redundant code?
• Are there any spelling errors?

In conclusion, after my review, I'd like to:

• 🙋 ask some clarifying questions
• 🙅 suggest some specific changes

[review-notebook-app]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[marjanfamili]

• This branch runs without error for the following config file.
  config_hyperparams.txt

• All tests pass