Invalid dimensions errors on exported models · Issue #11129 · ultralytics/yolov5

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement . We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Search before asking

I have searched the YOLOv5 issues and discussions and found no similar questions.

Question

Hi there, love this project! I'm having some trouble using the exported models - but can't tell how to solve this.

I have trained a model with the resultant model in the root of the repo called "v3seaDrone.pt" (with an image size of 1920). this works fine with my test image "3.jpg" when called either with the python detect.py script or through the torch.hub.load method. I have exported this model to onnx and torchscript with the command:

python export.py --weights "v3seaDrone.pt" --img 1920 --include onnx torchscript

and these appear to work fine when I run them with the detect script:

python detect.py --weights "v3seaDrone.onnx" --imgsz 1920 --source "3.jpg"

however when I try to do the same thing with the hub load method:

    model = torch.hub.load(".", "custom",  path="v3seaDrone.onnx", source='local')
    results = model("3.jpg", size=1920)
    results.print()
I get the following error (I have installed onnx):
Loading v3seaDrone.onnx for ONNX Runtime inference...
Adding AutoShape...
Traceback (most recent call last):
  File "F:\ml\yolov5\detect_and_notify.py", line 21, in <module>
    results = model(img, size=1920)
  File "C:\Users\Me\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\torch\nn\modules\module.py", line 1194, in _call_impl
    return forward_call(*input, **kwargs)
  File "C:\Users\Me\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\torch\autograd\grad_mode.py", line 27, in decorate_context
    return func(*args, **kwargs)
  File "F:\ml\yolov5\.\models\common.py", line 704, in forward
    y = self.model(x, augment=augment)  # forward
  File "C:\Users\Me\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\torch\nn\modules\module.py", line 1194, in _call_impl
    return forward_call(*input, **kwargs)
  File "F:\ml\yolov5\.\models\common.py", line 523, in forward
    y = self.session.run(self.output_names, {self.session.get_inputs()[0].name: im})
  File "C:\Users\Me\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.10_qbz5n2kfra8p0\LocalCache\local-packages\Python310\site-packages\onnxruntime\capi\onnxruntime_inference_collection.py", line 200, in run
    return self._sess.run(output_names, input_feed, run_options)
onnxruntime.capi.onnxruntime_pybind11_state.InvalidArgument: [ONNXRuntimeError] : 2 : INVALID_ARGUMENT : Got invalid dimensions for input: images for the following indices
 index: 2 Got: 1472 Expected: 1920
 Please fix either the inputs or the model.
and with the torchscript version I get:
RuntimeError: The following operation failed in the TorchScript interpreter.
Traceback of TorchScript, serialized code (most recent call last):
  File "code/__torch__/models/yolo.py", line 101, in forward
    _51 = (_31).forward((_30).forward(act, _50, ), _40, )
    _52 = (_33).forward(_46, _48, _50, (_32).forward(_51, ), )
           ~~~~~~~~~~~~ <--- HERE
    return (_52,)
class Detect(Module):
  File "code/__torch__/models/yolo.py", line 134, in forward
    _25 = torch.split_with_sizes(torch.sigmoid(_24), [2, 2, 7], 4)
    xy, wh, conf, = _25
    _26 = torch.add(torch.mul(xy, CONSTANTS.c0), CONSTANTS.c1)
          ~~~~~~~~~ <--- HERE
    xy0 = torch.mul(_26, torch.select(CONSTANTS.c2, 0, 0))
    _27 = torch.pow(torch.mul(wh, CONSTANTS.c0), 2)
RuntimeError: The size of tensor a (184) must match the size of tensor b (240) at non-singleton dimension 2
Any ideas about what I might be doing wrong or how to fix this?
Additional
No response
          👋 Hello @tlong123, thank you for your interest in YOLOv5 🚀! Please visit our ⭐️ Tutorials to get started, where you can find quickstart guides for simple tasks like Custom Data Training all the way to advanced concepts like Hyperparameter Evolution.
If this is a 🐛 Bug Report, please provide a minimum reproducible example to help us debug it.
If this is a custom training ❓ Question, please provide as much information as possible, including dataset image examples and training logs, and verify you are following our Tips for Best Training Results.
Requirements
Python>=3.7.0 with all requirements.txt installed including PyTorch>=1.7. To get started:
git clone https://github.com/ultralytics/yolov5  # clone
cd yolov5
pip install -r requirements.txt  # install
Environments
YOLOv5 may be run in any of the following up-to-date verified environments (with all dependencies including CUDA/CUDNN, Python and PyTorch preinstalled):
Notebooks with free GPU:   
Google Cloud Deep Learning VM. See GCP Quickstart Guide
Amazon Deep Learning AMI. See AWS Quickstart Guide
Docker Image. See Docker Quickstart Guide 
Status
If this badge is green, all YOLOv5 GitHub Actions Continuous Integration (CI) tests are currently passing. CI tests verify correct operation of YOLOv5 training, validation, inference, export and benchmarks on MacOS, Windows, and Ubuntu every 24 hours and on every commit.
Introducing YOLOv8 🚀
We're excited to announce the launch of our latest state-of-the-art (SOTA) object detection model for 2023 - YOLOv8 🚀!
Designed to be fast, accurate, and easy to use, YOLOv8 is an ideal choice for a wide range of object detection, image segmentation and image classification tasks. With YOLOv8, you'll be able to quickly and accurately detect objects in real-time, streamline your workflows, and achieve new levels of accuracy in your projects.
Check out our YOLOv8 Docs for details and get started with:
pip install ultralytics