An error of input formate when converting

[nemoxb]

Anyone knows why this input format lead to a None type object?
Basically, I followed the documents from detectron2 official docs, but construct an input format like

x = torch.randn(3,224,224)
inputs = [{'image':x,'height':224,'width':224}]
was not working

I get these errors
File "/Library/Frameworks/Python.framework/Versions/3.9/lib/python3.9/site-packages/torch/nn/modules/module.py", line 1090, in _slow_forward
result = self.forward(*input, **kwargs)
File "/Users/nemo/Documents//VideoPose3D-main/inference/detectron2/modeling/meta_arch/rcnn.py", line 146, in forward
return self.inference(batched_inputs)
File "/Users/nemo/Documents/VideoPose3D-main/inference/detectron2/modeling/meta_arch/rcnn.py", line 199, in inference
images = self.preprocess_image(batched_inputs)
File "/Users/nemo/Documents/VideoPose3D-main/inference/detectron2/modeling/meta_arch/rcnn.py", line 224, in preprocess_image
images = [x["image"].to(self.device) for x in batched_inputs]
TypeError: 'NoneType' object is not iterable

from detectron2.config import get_cfg
from detectron2 import model_zoo
from detectron2.modeling import build_model
from detectron2.checkpoint import DetectionCheckpointer
import torch

cfg = get_cfg()
cfg.merge_from_file(model_zoo.get_config_file("COCO-Detection/faster_rcnn_R_101_FPN_3x.yaml"))
cfg.MODEL.DEVICE = "cpu"

model = build_model(cfg)
# optimizer = optim.SGD(model.parameters(), lr=0.001, momentum=0.9)
model.eval()
cfg.MODEL.WEIGHTS = model_zoo.get_checkpoint_url("COCO-Detection/faster_rcnn_R_101_FPN_3x.yaml")
DetectionCheckpointer(model).load(cfg.MODEL.WEIGHTS)

# predictor = DefaultPredictor(cfg)
x = torch.randn(3,224,224)
inputs = [{'image':x,'height':224,'width':224}]

torch.onnx.export(model,               # model being run
                  inputs,                         # model input (or a tuple for multiple inputs)
                  "model_t1.onnx",   # where to save the model (can be a file or file-like object)
                  export_params=True,        # store the trained parameter weights inside the model file
                  opset_version=10,          # the ONNX version to export the model to
                  do_constant_folding=True,  # whether to execute constant folding for optimization
                  input_names = ['input'],   # the model's input names
                  output_names = ['output']) # the model's output names