Load custom yolov5 model on streamlit. AttributeError: Can't get attribute 'SPPF' on <module 'models.common'

[anhhaokudo]

Hi, I'm very new to yolo so I have a lot of fundamental questions, but ultimately I want to deploy a Streamlit webapp using a custom model.

I have been trying to follow https://github.com/xugaoxiang/yolov5-streamlit to deploy a streamlit webapp. The goal is to replace the model by my custom model.

My teammate trained a model and sent this new best.pt file to me. I replaced the yolov5s.pt by the best.pt file in the detect.py script and got the error

**2022-05-28 08:23:49.339 YOLOv5 🚀 39fe0ee torch 1.7.1 CPU

2022-05-28 08:23:49.556 Uncaught app exception

Traceback (most recent call last):

File "/home/appuser/venv/lib/python3.9/site-packages/streamlit/scriptrunner/script_runner.py", line 475, in _run_script

exec(code, module.dict)

File "/app/xx/main.py", line 95, in

detect(opt)

File "/app/xx/detect.py", line 37, in detect

model = attempt_load(weights, map_location=device) # load FP32 model

File "/app/xx/models/experimental.py", line 118, in attempt_load

ckpt = torch.load(w, map_location=map_location) # load

File "/home/appuser/venv/lib/python3.9/site-packages/torch/serialization.py", line 594, in load

return _load(opened_zipfile, map_location, pickle_module, **pickle_load_args)

File "/home/appuser/venv/lib/python3.9/site-packages/torch/serialization.py", line 853, in _load

result = unpickler.load()

AttributeError: Can't get attribute 'SPPF' on <module 'models.common' from xx**

I followed #5175 (comment) for troubleshooting but the error persists.

Thanks in advance if someone could guide me to the right direction.

[glenn-jocher]

@anhhaokudo your YOLOv5 code is out of date. To update:

• Git – git pull from within your yolov5/ directory or git clone https://github.com/ultralytics/yolov5 again
• PyTorch Hub – Force-reload model = torch.hub.load('ultralytics/yolov5', 'yolov5s', force_reload=True)
• Notebooks – View updated notebooks
• Docker – sudo docker pull ultralytics/yolov5:latest to update your image