Evaluation and full pipeline

[lizgzil]

---

Description

Quite a few changes in one PR - I'm sorry! A summary:

• Config changes and data loading extras
• Making READMEs more informative
• Evaluation script evaluate.py
• Trying to keep as much evaluation data out of the training datasets as possible (edits to prodigy_to_yolo.py)
• Special class to bring all the models together and output a dictionary of counts in predict_floorplan.py
• Refactor to visualise_image.py so the plotting alone can be used without also loading the data

Fixes #15 #14 #13

Instructions for Reviewer

There are two main things

1. Can you check evaluate.py is how you expect it to be and I've not made any errors
2. Try to run:

from asf_floorplan_interpreter.pipeline.predict_floorplan import FloorplanPredictor
img = 'outputs/figures/floorplan.png' # Local directory or a URL to an image file
fp = FloorplanPredictor(labels_to_predict = ["WINDOW", "DOOR","KITCHEN", "LIVING", "RESTROOM", "BEDROOM", "GARAGE"])
fp.load(local=True)
labels, label_counts = fp.predict_labels(img, conf_threshold=0)
fp.plot(img, labels, "outputs/figures/floorplan_prediction.png", plot_label=False)

Checklist:

• [ ] I have refactored my code out from notebooks/
• [ ] I have checked the code runs
• [ ] I have tested the code
• [ ] I have run pre-commit and addressed any issues not automatically fixed
• [ ] I have merged any new changes from dev
• [ ] I have documented the code
  • [ ] Major functions have docstrings
  • [ ] Appropriate information has been added to READMEs
• [ ] I have explained this PR above
• [ ] I have requested a code review

[lizgzil]

Thanks @cmbrennan002 ! the changes you suggested are in this commit https://github.com/nestauk/asf_floorplan_interpreter/pull/17/commits/4fc1e1c8477700ebbb154b0a78290b48ccd2de37

@sqr00t It might also be worth you checking over that commit to see whether any of the new changes are things you need to know about!