For a more Specific guide, please refer to this repo. Thank Chalkeys for the detailed guide!
pip install -r requirements.txt
if you have a cuda capable gpu, you can running the following extra command
pip install torch==1.13.1+cu116 torchvision==0.14.1+cu116 torchaudio==0.13.1 --extra-index-url https://download.pytorch.org/whl/cu116
Add the following command
pip install --upgrade setuptools pip --user
pip install nvidia-pyindex
pip install --upgrade nvidia-tensorrt
pip install pycuda
If you can't install tensorrt in this way, you can look up this Nvidia guide
You have to transform the '.pt' model to '.trt' model by yourself, because the Tensorrt engines are environment specific. This repo may helpful: TensorRT-For-YOLO-Series
just run the main.py
file with the following command
python main.py
After a few seconds, the program will start to run. You can see Main Start
in the console.
Once you hold the right mouse button or the left mouse button (no matter you hold to aim or start shooting), the program will start to aim at the enemy.
You can change the settings in the args.py
file.
- model
- The default model is for Apex. However, you can train your own model using
train.py
, and switch the model using this setting. - There're several model in the "model" dir, you can choose one of them.
- The
.trt
models are for tensorRT, which is about 4 times faster than the.pt
models, but with the same accuracy. - Model speed:
8n>8s>8m
- Model accuracy:
8n<8s<8m
- The
- The default model is for Apex. However, you can train your own model using
- crop_size
- This setting determines the portion of the screen to be detected. Too high may cause difficulty in detecting little objects.
This program is only for educational purposes. I am not responsible for any damage caused by this program.