TypeError when using TensorBoard with Matterport - Mask R-CNN

Question

I am using this tutorial for object detection using the matterport mask r-cnn repo, except I'm doing cats rather than kangaroos, and I would like to use TensorBoard with it, but I get this error every time I run it.

TypeError: Value passed to parameter 'values' has DataType bool not in list of allowed values: float32, float64, int32, uint8, int16, int8, int64, bfloat16, uint16, float16, uint32, uint64

I'm following this guide and checked this question and this other question, as well as going through the code to figure out where the issue was coming from (there are 11 levels to the traceback). The thing causing the issue seems to be "input_gt_masks_1:0" but I can't figure out how to change that.

%load_ext tensorboard
!rm -rf ./logs/
config = CatConfig()
config.display()
model = MaskRCNN(mode='training', model_dir='./', config=config)
logdir = os.path.join("logs", datetime.datetime.now().strftime("%Y%m%d-%H%M%S"))
tensorboard_callback = tf.keras.callbacks.TensorBoard(logdir, histogram_freq=1)

This is the code I've been running. It's in Google Colab if that changes anything.