tensorflow-yolov3 icon indicating copy to clipboard operation
tensorflow-yolov3 copied to clipboard

🔥 TensorFlow Code for technical report: "YOLOv3: An Incremental Improvement"

Results 114 tensorflow-yolov3 issues
Sort by recently updated
recently updated
newest added

提供的convert_weight.py,显示"input/input_data", "pred_sbbox/concat_2", "pred_mbbox/concat_2", "pred_lbbox/concat_2不在图中

I am trying to convert your model into .dlc format which is Qualcomm Snapdragon (SNPE) format. I gave the output node names "`pred_sbbox/concat_2, pred_mbbox/concat_2, pred_lbbox/concat_2"` However its says that `ERROR_TF_NODE_NOT_FOUND_IN_GRAPH:...

你好,请问 readme给出的88.3%的精度,是使用预训练模型对voc进行微调后,在voc2012验证集上得到的结果么?

The model works well until I want to train large-object branch only! In training phase, I freeze medium-object branch and small-object branch (set "trainable=False"), and I only calculate loss using...

博主你好 我用的YunYang1994的代码来跑的yolov3代码 跑出来的模型是原本模型的3倍 ,这个是怎么回事?检查代码那些感觉没检查出问题

运行map中的main.py时出现的问题,有大佬知道怎么解决吗?

你好,作者,非常感谢提供tf的yolo,项目很好,但我还是有一个问题:请问如何在训练了一段时间中断了后继续加载之前的模型继续训练呢,目前跑完了设置的epoch,但loss还是比较大,所以想继续接着之前的训练,望回复! 万分感谢!

['yolov3/yolo-v3/Conv_6/weights', TensorShape([Dimension(1), Dimension(1), Dimension(1024), Dimension(255)])] ['conv_lbbox/weight', TensorShape([Dimension(1), Dimension(1), Dimension(1024), Dimension(75)])] Traceback (most recent call last): File "convert_weight.py", line 74, in raise RuntimeError

作者你好,我的数据集里有多张图片,且图片尺寸是不一样的,有的尺寸是1551*512,有的是1192*512,以前使用darknet可以放在一起训练,将darknet的.cfg文件配置为“‘’”width=1280 height=512”。 使用tensorflow-yolov3训练时,报错: => Now it starts to train YOLOV3 from scratch ... 0%| | 0/2 [00:00