Finetune YOLOv8 to detect shapes

[JanelleCai]

• using the shapes dataset (Create a shapes dataset #143), finetune a YOLOv8 model to detect the shapes
• if you are stuck on where to start, you can take a look at this file, which is what we used to finetune YOLOv8 to detect buoys. keep in mind, this notebook loads in the buoys dataset from roboflow - the main change you'll need to make is to instead load in the shapes dataset.
• you can decide if the model should detect both color + shape versus just the shape.
  • if the data includes all colors and shapes specified in Create a shapes dataset #143 , you can train the model to detect both color + shape (and this is probably the better option).
  • however, if the data is missing certain colors, you should train it to detect shape only (and then write another node that detects the color manually using HSV values).