TypeError: RandomIoUCrop() requires input sample to contain tensor or PIL images and bounding boxes. Sample can also contain masks

Question

I am getting started on torch vision to train and evaluate object detection models. I am getting into issues and need some help. I am setting up basic steps to evaluate a pre-trained model on coco 2017 dataset. Here is my colab notebook. https://colab.research.google.com/drive/13nY17EEUx7IDsJCOmY9XEY9FjFX3E1pA?usp=drive_link

I am using this example tutorial from PyTorch to set up 'data loader' for coco dataset. https://pytorch.org/vision/main/auto_examples/transforms/plot_transforms_e2e.html#sphx-glr-auto-examples-transforms-plot-transforms-e2e-py

After the data loader steps, I added steps to train / evaluate using 'engine.py'.

Any help to resolve this issue will be appreciated. Or any suggestion on alternate ways.

Thanks,

Amit

I am trying to use coco 2017 dataset to train/evaluate an object detection model using PyTorch vision.