Dong

icon location Hefei,Anhui

Results 15 comments of Dong

Tensor("yolo_loss_sample_weights:0", shape=(?,), dtype=float32) must be from the same graph as Tensor("Mean:0", shape=(), dtype=float32)

> when i use my own dataset like VOC2007 format,this probolem occurs after first 50 epochs... Sir,have you solved that problem?I meet this bug too!Thanks in advance!

T7 file error on Windows 10

感谢！@lxfly2000

self._normalize() got NAN...

Bug description Sir, when I print upsampled_a in upsampled_a = self._normalize(self.hook_a, self.hook_a.ndim - 2) I got NAN in some value. Is there any bug? I set model to vgg19 target_layer1='features.35'...

hi, can you share the train/val/test raw images, better on BaiDuYun, you know, i can not download from google drive in China. thanks a lot.

请问有没有下载成功的，能不能把训练集上传百度云啊，下不下来，总会断。迅雷也不行。

demo error

那段代码加哪里？？就像Gavin666Github说的，在后边加没用。 目前我只能把所有出现.的地方换成_,也能运行，不过作者说的是什么意思，谁能解答一下

create_train.py

I think the author makes a slip in writing in create_train.py create_train.py saves the atom_map as 'atom',but we find it with 'ato' 118 line` h5f.create_dataset('ato',data=rep_atmosphere)`

train.py error

pytorch0.4改成 netG.load_state_dict(torch.load(opt.netG),strict=False)

你好，问下训练相关的问题

@wangmingrun1314 请问你的问题解决了吗

ValueError: Empty training data.

> 你好，自定义数据集（401张图片），报错：ValueError:Empty training data.修改过dataset里的shuffle参数，仍然报该错误，可以给指点一下嘛？ > 1/190 [..............................] - ETA: 21:51 - loss: 897.8262 - l1_loss: 169.9690 - l2_loss: 727.4587 - l1_p: 0.0769 - l1_r: 0.6667 - l2_p: 0.0000e+00 - l2_r:...

ValueError: Empty training data.

> 遇到同样的问题 https://github.com/zhen8838/K210_Yolo_framework/issues/26 我把训练集增加到20000+就好了。。