chore: Update tutorial for new BoxFilter

data_generator_tutorial.ipynb

@@ -318,7 +318,7 @@
  "outputs": [],
  "source": [
  "box_filter = BoxFilter(overlap_criterion='area',\n",
- " bounds=(0.4, 1.0))"
+ " overlap_bounds=(0.4, 1.0))"
  ]
  },
  {
@@ -327,6 +327,8 @@
  "source": [
  "This `BoxFilter` object can tell which bounding boxes meet the given overlap requirements with the image and which bounding boxes don't. The `overlap_criterion` determines how this overlap should be measured and the `bounds` determine within which range the overlap needs to be. In the example above, the overlap criterion is set to 'area', which means that a certain fraction of the area of a bounding box needs to overlap with the image. In our case, the `bounds` argument dictates that at least 0.4 and no more than 1.0 of the bounding box area needs to overlap with the image. There are two more overlap criteria to choose from: 'iou' and 'center_point'. 'center_point' determines a bounding box to be valid if its center point lies within the image (the `bounds` argument doesn't matter in this case). 'iou' determines a bounding box to be valid if the intersection over union (or Jaccard) similarity between the box and the image meets the given bounds.\n",
  "\n",
+ "`BoxFilter` can actually do a lot more than is described above: It can also check for and filter out degenerate boxes and it can filter out boxes that have less than a specified area. Both options are very useful but beyond the scope of this tutorial. The `BoxFilter` documentation contains all relevant details.\n",
+ "\n",
  "Let's pass this box filter to the `Translate` constructor and see what it does:"
  ]
  },