인공지능
Implementation of YOLO v3 object detector in TensorFlow
https://github.com/YunYang1994/tensorflow-yolov3
2. Quick start
* error1)
>> requirements.txt 수정
numpy==1.17.0
Pillow==5.3.0
scipy==1.1.0
tensorflow-gpu==1.14.0
wget==3.2
seaborn==0.9.0
- checkpoint
* error2)
>> install CUDA 10.0
: pip install tf-nightly
* error3)
>> pip install easydict
- get some .pb files & run the demo script
image_demo.py 결과
video_demo.py 결과
3. Train on my own dataset
3.1 Train VOC dataset
- 디렉토리 구성
- edit ./core/config.py
(2) train from COCO weights(recommend):
* error4)
>> pip install tqdm
- 실행
- test and evaluate
__C.TEST.WEIGHT_FILE = "./checkpoint/yolov3_test_loss=44.4873.ckpt-1"
>> 수정
evaluate.py 결과
>> 해결해야 함.
'인공지능' 카테고리의 다른 글
[PBL] 데이터셋 라벨링 (0) | 2019.10.01 |
---|---|
[labeling] Yolo_mark로 라벨링 (1) | 2019.09.04 |
Violence Detection 분석 (0) | 2019.08.09 |
Violence Detection 실습 (0) | 2019.08.05 |
Tensorflow Object Detection API with custom training data (0) | 2019.07.05 |
댓글