- Integrated from yolov7 official repo https://github.com/WongKinYiu/yolov7, fixed the issue on cuda core can't run on yolov7 model. Tested on conda environment with
python3.9, pytorch=1.11.0, cudatoolkit=1.13
MS COCO
| Model | Test Size | APtest | AP50test | AP75test | batch 1 fps | batch 32 average time |
|---|---|---|---|---|---|---|
| YOLOv7 | 640 | 51.4% | 69.7% | 55.9% | 161 fps | 2.8 ms |
| YOLOv7-X | 640 | 53.1% | 71.2% | 57.8% | 114 fps | 4.3 ms |
| YOLOv7-W6 | 1280 | 54.9% | 72.6% | 60.1% | 84 fps | 7.6 ms |
| YOLOv7-E6 | 1280 | 56.0% | 73.5% | 61.2% | 56 fps | 12.3 ms |
| YOLOv7-D6 | 1280 | 56.6% | 74.0% | 61.8% | 44 fps | 15.0 ms |
| YOLOv7-E6E | 1280 | 56.8% | 74.4% | 62.1% | 36 fps | 18.7 ms |
Conda environment -- Anaconda https://www.anaconda.com/
Python -- Python 3.9 installed with Anaconda
Select Installation Type : 'just me'
Anaconda Install Location : Anywhere you want, doesn't have to be on C drive
Advanved Installation Options :
Run git clone https://github.com/FlyerJB/YOLOv7-RoboMaster.git on your command prompt to some dir under C:/ drive or your OS drive to avoid Enviornment failure \
Open Conda Command Prompt with Admin Right
Cd into yolov7 dir with cd <where you clone your yolov7>
And craete Conda Environment
Conda Create -n <The Name You Like> Python3.9
Run command Conda activate <The Name You put from previous step>
Option 1 : Install yolov7 for training on CPU
pip install -r requirements.txt
Option 2 : Install yolov7 for training on RTX GPU
pip install -r requirement_nv_gpu.txt
Run python or python3 or py to run python
Run import torch
Run torch.cuda_is_available()
If it returns True, it means CUDA is successfully install on your device with Pytorch.
With GPU training
# train p6 models
python train.py --workers 1 --device 0 --batch-size 8 --epochs 50 --img 640 640 --data data/coco_custom.yaml --hyp data/hyp.scratch.custom.yaml --cfg cfg/training/yolov7-custom.yaml --name yolov7-custom-tut5-v1 --weights yolov7.ptWith CPU training
# train p6 models
python train.py --workers 1 --device CPU --batch-size 8 --epochs 50 --img 640 640 --data data/coco_custom.yaml --hyp data/hyp.scratch.custom.yaml --cfg cfg/training/yolov7-custom.yaml --name yolov7-custom-tut5-v1 --weights yolov7.pt