文件列表:

ckpt (0, 2018-05-09)
config.py (1211, 2018-05-09)
create_tfrecords.py (2777, 2018-05-09)
darknet53_trainable.py (15796, 2018-05-09)
data_pipeline.py (2417, 2018-05-09)
draw_boxes.py (3356, 2018-05-09)
font (0, 2018-05-09)
font\FiraMono-Medium.otf (127344, 2018-05-09)
font\SIL Open Font License.txt (4434, 2018-05-09)
image (0, 2018-05-09)
image\bladerrunner.jpg (111967, 2018-05-09)
image\dog.jpg (163759, 2018-05-09)
image\yolov3_architecture.png (83420, 2018-05-09)
load_weights.c (1855, 2018-05-09)
predict_net.py (1653, 2018-05-09)
prediction.jpg (147066, 2018-05-09)
save_npz.py (4686, 2018-05-09)
test2007.txt (34664, 2018-05-09)
test_darknet53.py (1248, 2018-05-09)
test_mask.py (598, 2018-05-09)
test_movingavg.py (868, 2018-05-09)
test_upsampling.py (454, 2018-05-09)
test_yolo_head.py (683, 2018-05-09)
test_yolov3.py (413, 2018-05-09)
train_net.py (5063, 2018-05-09)
yolo_layer.py (15597, 2018-05-09)
yolo_top.py (8242, 2018-05-09)
yolov3-voc.cfg (8657, 2018-05-09)

# YOLOv3_tf Try to implement YOLOv3 with TensorFlow 1. transform darknet weight into npz: **save_npz.py** 2. darknet 52 conv layer: **darknet53_trainable.py** 3. yolo layer(loss & predict): **yolo_layer.py**

#### train (train_net.py) 1. for now, only train detector layer(50 epoch, batch size: 8) 2. froze feature extractor('darknet53.conv.74') 3. TODO: finetune end2end _traindata: VOC2007 trainval & VOC2012 trainval_ #### sample prediction (predict_net.py)


#### trained weights - [pan.baidu](https://pan.baidu.com/s/1fQ73S3HVoMeY8Tz5um12Mw) key: lbd4 - [Dropbox](https://pan.baidu.com/s/1fQ73S3HVoMeY8Tz5um12Mw) #### Acknowledgements & Reference - :fire: [Darknet](https://github.com/pjreddie/darknet) :fire: - :fire: [YAD2K](https://github.com/allanzelener/YAD2K) :fire: ##### Requirement 1. TensorFlow 1.* 2. easydict 3. PIL 4. numpy 5. matplotlib

近期下载者：

相关文件：

评论：[我要评论] [举报此文件]

收藏者：