Basic skel to effortlesly move around a yolov4-tiny training environment. Uses a needlessly beefy docker image, replace if you want to use a lighter one.
- Clone repo
- Link your data to
dataset
:ln -sf <your/dataset/path> dataset
- Run docker image with
bash docker_run.sh
- Run
make
to download and configure the net architecture - Train with
bash train.sh
- Test with
bash test.sh
# Symlink to avoid tree navigation through results
ln -s PyTorch_YOLOv4/runs/train train-runs
# Plot output from results.txt
import numpy as np
import matplotlib.pyplot as plt
res = np.loadtxt('results.txt', dtype=object)
ap50 = np.array(res[:,10], dtype=float)
map = np.array(res[:,11], dtype=float)
plt.plot(ap50)
plt.plot(map)
plt.savefig('results.png')
- WongKinYiu/PyTorch_YOLOv4/:
- Most code
- Cfg architecture base file