keras-faster-rcnn icon indicating copy to clipboard operation
keras-faster-rcnn copied to clipboard

keras实现faster rcnn,end2end训练、预测; 持续更新中,见todo... ;欢迎试用、关注并反馈问题

Results 41 keras-faster-rcnn issues
Sort by recently updated
recently updated
newest added

您好,想问一下如果想预测出测试集图片后把所有的预测结果图片保存到一个文件夹,不是随机展示,该怎么改呢,希望您能回复,非常感激您

运行启动gpu后 `2020-02-22 10:32:07.920229: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1326] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 10023 MB memory) -> physical GPU (device: 0, name: GeForce RTX 2080 Ti, pci bus id: 0000:68:00.0, compute capability: 7.5)`...

用我自己的数据集,训练了150轮,预测的时候总是没有框生成,请问这是为什么呀?

請問樓主有去試試將epoch降低能更好的解決overfitting問題嗎

首先感谢的作者的努力! 想问下您训练不带gtbox的图片,因为阳性目标与比较相似,想单独训练下背景图片。

您好,我想问一下坐标归一化是什么意思

h:[-37.42, -34.72, -28.17, -18.79, -16.83, 3.0, 25.87, 37.58, 43.39] w:[44.99, 31.9, -28.96, 8.64, 1.92, -19.27, -31.65, 38.65, -38.1] ratio:[-0.83, -1.09, 0.97, -2.17, -8.77, -0.16, -0.82, 0.97, -1.14] ious:{1: inf} mean...

想求问如何才能输出acc和val acc

File "D:\pythonfile\keras-faster-rcnn-master\keras-faster-rcnn-master\faster_rcnn\utils\model_utils.py", line 79, in compile keras_model.add_metric(loss, aggregation='mean', name=name) AttributeError: 'Model' object has no attribute 'add_metric'