Error in deploy yolov9-e custom weights

Project Type: Object detection
Operating System & Browser: Kaggle notebooks

I have deployed the yolov9 model trained with gelan-c.yaml.
But got errors when deploying the yolov9 model trained with gelan-e.yaml. or yolov9-e.yaml.
Is it allowed to deploy bigger models in Yolov9.

I found the following models are available

error message when I run
version.deploy(model_type=“yolov9”, model_path=f"{HOME}/yolov9/runs/train/exp20")

"File /opt/conda/lib/python3.10/site-packages/roboflow/core/version.py:531, in Version.deploy(self, model_type, model_path, filename)
526 with open(os.path.join(model_path, “opt.yaml”), “r”) as stream:
527 opts = yaml.safe_load(stream)
529 model_artifacts = {
530 “names”: class_names,
→ 531 “nc”: model[“model”].nc,
532 “args”: {
533 “imgsz”: opts[“imgsz”] if “imgsz” in opts else opts[“img_size”],
534 “batch”: opts[“batch_size”],
535 },
536 “model_type”: model_type,
537 }
538 if hasattr(model[“model”], “yaml”):
539 model_artifacts[“yaml”] = model[“model”].yaml

File /opt/conda/lib/python3.10/site-packages/torch/nn/modules/module.py:1695, in Module.getattr(self, name)
1693 if name in modules:
1694 return modules[name]
→ 1695 raise AttributeError(f"‘{type(self).name}’ object has no attribute ‘{name}’")

AttributeError: ‘DetectionModel’ object has no attribute ‘nc’"

if it is allowed, how can we deploy bigger models with Yolov9?
(The trained gelan-e model seems different from )
I have checked that model[“model”] has no attribute “nc”