ultralytics / yolov5

YOLOv5 🚀 in PyTorch > ONNX > CoreML > TFLite
https://docs.ultralytics.com
GNU Affero General Public License v3.0
49.83k stars 16.13k forks source link

pre-weight #5348

Closed yang-leon996 closed 2 years ago

yang-leon996 commented 2 years ago

❔Question

if I want to only load part of these pre-weight, what should I do ?such as I don't want to load the pre-weight of SPP. Thank you!

Additional context

github-actions[bot] commented 2 years ago

👋 Hello @yang-leon996, 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 screenshots and minimum viable code to reproduce your issue, otherwise we can not help you.

If this is a custom training ❓ Question, please provide as much information as possible, including dataset images, training logs, screenshots, and a public link to online W&B logging if available.

For business inquiries or professional support requests please visit https://ultralytics.com or email Glenn Jocher at glenn.jocher@ultralytics.com.

Requirements

Python>=3.6.0 with all requirements.txt installed including PyTorch>=1.7. To get started:

$ git clone https://github.com/ultralytics/yolov5
$ cd yolov5
$ pip install -r requirements.txt

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):

Status

CI CPU testing

If this badge is green, all YOLOv5 GitHub Actions Continuous Integration (CI) tests are currently passing. CI tests verify correct operation of YOLOv5 training (train.py), validation (val.py), inference (detect.py) and export (export.py) on MacOS, Windows, and Ubuntu every 24 hours and on every commit.

glenn-jocher commented 2 years ago

@yang-leon996 you can add any layers you want excluded to the exclude list: https://github.com/ultralytics/yolov5/blob/a4fece8c1480aed46a38a6344b403d79c81bd751/train.py#L119-L122

yang-leon996 commented 2 years ago

@yang-leon996 you can add any layers you want excluded to the exclude list:

https://github.com/ultralytics/yolov5/blob/a4fece8c1480aed46a38a6344b403d79c81bd751/train.py#L119-L122

yang-leon996 commented 2 years ago

@yang-leon996 you can add any layers you want excluded to the exclude list: https://github.com/ultralytics/yolov5/blob/a4fece8c1480aed46a38a6344b403d79c81bd751/train.py#L119-L122

yang-leon996 commented 2 years ago

@glenn-jocher exclude = ['anchor',"SPP"] if (cfg or hyp.get('anchors')) and not resume else [] is like this?

glenn-jocher commented 2 years ago

@yang-leon996 just print the module names and add whichever ones you'd like to to the exclude list.

yang-leon996 commented 2 years ago

Thank you very much for replying to me in your busy schedule, I will try it out!