shekkizh FCN.tensorflow issues

shekkizh / FCN.tensorflow

Tensorflow implementation of Fully Convolutional Networks for Semantic Segmentation (http://fcn.berkeleyvision.org)

MIT License

1.25k stars 529 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Implementing GradCam

#130 varungupta31 opened 2 years ago
1
Logging and Visualizing Training Metrics on Tensorboard

#129 varungupta31 opened 2 years ago
0
ValueError: Unknown mat file type, version 110, 116

#128 varungupta31 closed 2 years ago
2
i made a reimplement with tensorflow2

#127 YuhangJi opened 2 years ago
3
'NoneType' object is not subscriptable

#126 wangxb-code opened 3 years ago
0
Can I use FCN on Windows 10?

#125 lgw9705 opened 3 years ago
0
ValueError: Cannot feed value of shape (2, 227, 227, 1, 4) for Tensor 'annotation:0', which has shape '(?, 227, 227, 1)'

#124 Rundong4026 opened 4 years ago
2
Is anyone could share the trained ckpt???thanks!!!!

#123 huizhilei opened 4 years ago
0
Fix/visualize t1ce

#122 Pigrabbit closed 4 years ago
0
What is output node's name?

#121 YuRimNam closed 4 years ago
1
Training problem

#120 dxw1997 closed 4 years ago
1
how to train my own dataset?

#119 kunyu17 opened 4 years ago
0
Train the model for face parsing problem

#118 myusernamesxy opened 4 years ago
2
Fixed link to HTTPS

#117 Jakobu5 closed 2 years ago
0
KeyError: 'normalization'

#116 sixisir opened 4 years ago
1
论文中是说最后输入1000张特征图，取1000张特征图中的各像素最大值，为啥这个代码要用3通道呢

#115 jiashihuigithub opened 5 years ago
2
为什么没有把pool5的结果和最后结果上采样做差值，我看代码只做了pool4和pool3的结果

#114 jiashihuigithub closed 5 years ago
1
change on FCN.py

#113 yeLer opened 5 years ago
0
is fusion operation the same as the paper?

#112 fortunechen opened 5 years ago
0
run FCN.py , path wrong

#111 cena001plus opened 5 years ago
1
Expected string passed to parameter 'tag' of op 'ImageSummary', got 'input_image_1' of type 'str' instead.

#110 cena001plus closed 5 years ago
1
Update read_MITSceneParsingData.py

#109 wsh540677278 opened 5 years ago
0
Questions about training

#108 lensea closed 5 years ago
0
Why not implement batch normalization?

#107 MarkYangjiayi opened 5 years ago
0
loss=3.x

#106 ShifengJin opened 5 years ago
0
how to test my own pictures and get the result?

#105 Fu7388 opened 5 years ago
8
The test code

#104 LCJHust opened 5 years ago
1
I want to use another database: pascal voc2012.how to change the code?

#103 mingweiY opened 5 years ago
3
The loss function of my network can be quickly reduced to 3, but basically it will not be reduced afterwards. And when I test on an image I can't get a correct pred

#102 Bigwode opened 6 years ago
4
training for 6 bands data

#101 summer17s opened 6 years ago
1
Resource exhausted error

#100 yyong119 closed 6 years ago
1
replace VGG19 with resnet 50/101

#99 arasharchor opened 6 years ago
0
mean_iu is not good enough

#98 worstprogrammerCN opened 6 years ago
1
Black photos after training

#97 SoulGuedria closed 6 years ago
0
Train_loss and Validation_loss didn't change

#96 suyanzhou626 closed 6 years ago
1
ValueError: Cannot feed value of shape (0,) for Tensor 'input_image:0', which has shape '(?, 224, 224, 3)'

#95 MarkYangjiayi closed 6 years ago
6
Ignoring Conv5_4

#94 aolsenjazz opened 6 years ago
3
Training summary

#93 ygjwd12345 closed 6 years ago
0
The newcomer asked the gods how to fine-tune the model. It is what kind of output back-propagation or other operations？

#92 peppaJoeng closed 6 years ago
2
"MITSceneParsing.pickle" Do I need to download this file?

#91 tanxjtu closed 6 years ago
3
Keyerror: 'normalization'

#90 rinawhale closed 6 years ago
1
For those who use windows and failed to generate pickle file

#89 geekboood closed 6 years ago
1
Minor change requests and a Big question.

#88 ahnHeejune closed 6 years ago
5
MITSceneParsing.pickle Need to download it yourself?

#87 BDMLLang closed 6 years ago
4
I want to test FCN with my own image but lack of time and GPU for train, is there any way to get trained weights?

#86 dongtuu closed 6 years ago
1
New reader

#85 ChenhaoLiu-SeasPenn closed 6 years ago
0
How do you know you have finished training?

#84 WillSmisi closed 6 years ago
1
Training on custom dataset

#83 rmmal closed 6 years ago
1
Question regarding using custom dataset having 2 classes

#82 matvaibhav closed 6 years ago
5
Prediction Issue... Need your wisdom...

#81 bizcjh1027 closed 6 years ago
1