shekkizh/FCN.tensorflow issues and pull requests

#100 - Resource exhausted error

Issue - State: closed - Opened by yyong119 over 6 years ago - 1 comment

#100 - Resource exhausted error

Issue - State: closed - Opened by yyong119 over 6 years ago - 1 comment

#99 - replace VGG19 with resnet 50/101

Issue - State: open - Opened by arasharchor over 6 years ago

#99 - replace VGG19 with resnet 50/101

Issue - State: open - Opened by arasharchor over 6 years ago

#98 - mean_iu is not good enough

Issue - State: open - Opened by worstprogrammerCN over 6 years ago - 1 comment

#98 - mean_iu is not good enough

Issue - State: open - Opened by worstprogrammerCN over 6 years ago - 1 comment

#97 - Black photos after training

Issue - State: closed - Opened by SoulGuedria over 6 years ago

#97 - Black photos after training

Issue - State: closed - Opened by SoulGuedria over 6 years ago

#96 - Train_loss and Validation_loss didn't change

Issue - State: closed - Opened by suyanzhou626 over 6 years ago - 1 comment

#96 - Train_loss and Validation_loss didn't change

Issue - State: closed - Opened by suyanzhou626 over 6 years ago - 1 comment

#95 - ValueError: Cannot feed value of shape (0,) for Tensor 'input_image:0', which has shape '(?, 224, 224, 3)'

Issue - State: closed - Opened by MarkYangjiayi over 6 years ago - 6 comments

#95 - ValueError: Cannot feed value of shape (0,) for Tensor 'input_image:0', which has shape '(?, 224, 224, 3)'

Issue - State: closed - Opened by MarkYangjiayi over 6 years ago - 6 comments

#94 - Ignoring Conv5_4

Issue - State: open - Opened by aolsenjazz over 6 years ago - 3 comments

#94 - Ignoring Conv5_4

Issue - State: open - Opened by aolsenjazz over 6 years ago - 3 comments

#93 - Training summary

Issue - State: closed - Opened by ygjwd12345 over 6 years ago

#93 - Training summary

Issue - State: closed - Opened by ygjwd12345 over 6 years ago

#92 - The newcomer asked the gods how to fine-tune the model. It is what kind of output back-propagation or other operations？

Issue - State: closed - Opened by peppaJoeng over 6 years ago - 2 comments

#92 - The newcomer asked the gods how to fine-tune the model. It is what kind of output back-propagation or other operations？

Issue - State: closed - Opened by peppaJoeng over 6 years ago - 2 comments

#91 - "MITSceneParsing.pickle" Do I need to download this file?

Issue - State: closed - Opened by tanxjtu over 6 years ago - 3 comments

#91 - "MITSceneParsing.pickle" Do I need to download this file?

Issue - State: closed - Opened by tanxjtu over 6 years ago - 3 comments

#90 - Keyerror: 'normalization'

Issue - State: closed - Opened by rinawhale over 6 years ago - 1 comment

#90 - Keyerror: 'normalization'

Issue - State: closed - Opened by rinawhale over 6 years ago - 1 comment

#89 - For those who use windows and failed to generate pickle file

Issue - State: closed - Opened by geekboood over 6 years ago - 1 comment

#89 - For those who use windows and failed to generate pickle file

Issue - State: closed - Opened by geekboood over 6 years ago - 1 comment

#88 - Minor change requests and a Big question.

Issue - State: closed - Opened by ahnHeejune over 6 years ago - 5 comments

#88 - Minor change requests and a Big question.

Issue - State: closed - Opened by ahnHeejune over 6 years ago - 5 comments

#87 - MITSceneParsing.pickle Need to download it yourself?

Issue - State: closed - Opened by BDMLLang over 6 years ago - 4 comments

#87 - MITSceneParsing.pickle Need to download it yourself?

Issue - State: closed - Opened by BDMLLang over 6 years ago - 4 comments

#86 - 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?

Issue - State: closed - Opened by dongtuu over 6 years ago - 1 comment

#86 - 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?

Issue - State: closed - Opened by dongtuu over 6 years ago - 1 comment

#85 - New reader

Pull Request - State: closed - Opened by ChenhaoLiu-SeasPenn over 6 years ago

#85 - New reader

Pull Request - State: closed - Opened by ChenhaoLiu-SeasPenn over 6 years ago

#84 - How do you know you have finished training?

Issue - State: closed - Opened by WillSmisi over 6 years ago - 1 comment

#84 - How do you know you have finished training?

Issue - State: closed - Opened by WillSmisi over 6 years ago - 1 comment

#83 - Training on custom dataset

Issue - State: closed - Opened by rmmal over 6 years ago - 1 comment

#83 - Training on custom dataset

Issue - State: closed - Opened by rmmal over 6 years ago - 1 comment

#82 - Question regarding using custom dataset having 2 classes

Issue - State: closed - Opened by matvaibhav over 6 years ago - 5 comments

#82 - Question regarding using custom dataset having 2 classes

Issue - State: closed - Opened by matvaibhav over 6 years ago - 5 comments

#81 - Prediction Issue... Need your wisdom...

Issue - State: closed - Opened by bizcjh1027 almost 7 years ago - 1 comment

#81 - Prediction Issue... Need your wisdom...

Issue - State: closed - Opened by bizcjh1027 almost 7 years ago - 1 comment

#80 - wimdows

Issue - State: closed - Opened by CCTV-lulu almost 7 years ago - 1 comment

#80 - wimdows

Issue - State: closed - Opened by CCTV-lulu almost 7 years ago - 1 comment

#79 - I get nan during the inference stage using my own data ?

Issue - State: closed - Opened by Deephome almost 7 years ago - 1 comment

#79 - I get nan during the inference stage using my own data ?

Issue - State: closed - Opened by Deephome almost 7 years ago - 1 comment

#78 - MemoryError

Issue - State: closed - Opened by MandyMo almost 7 years ago - 1 comment

#78 - MemoryError

Issue - State: closed - Opened by MandyMo almost 7 years ago - 1 comment

#77 - When visualized,I got outputs with black photos ,no results. Why?

Issue - State: closed - Opened by SpadesQ almost 7 years ago - 7 comments

#77 - When visualized,I got outputs with black photos ,no results. Why?

Issue - State: closed - Opened by SpadesQ almost 7 years ago - 7 comments

#76 - How to test ?

Issue - State: closed - Opened by XZNWU almost 7 years ago - 6 comments

#76 - How to test ?

Issue - State: closed - Opened by XZNWU almost 7 years ago - 6 comments

#75 - % (np_val.shape, subfeed_t.name, str(subfeed_t.get_shape()))) ValueError: Cannot feed value of shape (224, 224) for Tensor 'conv2d_2_input:0', which has shape '(?, 224, 224, 3)'

Issue - State: closed - Opened by muradalqurishee almost 7 years ago - 10 comments

#75 - % (np_val.shape, subfeed_t.name, str(subfeed_t.get_shape()))) ValueError: Cannot feed value of shape (224, 224) for Tensor 'conv2d_2_input:0', which has shape '(?, 224, 224, 3)'

Issue - State: closed - Opened by muradalqurishee almost 7 years ago - 10 comments

#74 - Question: Could I ask why in the FCN structure.. it looks like using VGG to compute the feature first then input the feature to conv 3 times then deconv 3 times.. I think it might be look like VGG layers then deconv such number of layers

Issue - State: closed - Opened by wuqinhao almost 7 years ago - 1 comment

#74 - Question: Could I ask why in the FCN structure.. it looks like using VGG to compute the feature first then input the feature to conv 3 times then deconv 3 times.. I think it might be look like VGG layers then deconv such number of layers

Issue - State: closed - Opened by wuqinhao almost 7 years ago - 1 comment

#73 - why you read annotation only one channel,Not three channels??

Issue - State: closed - Opened by liuzhisheng1226 almost 7 years ago - 2 comments

#73 - why you read annotation only one channel,Not three channels??

Issue - State: closed - Opened by liuzhisheng1226 almost 7 years ago - 2 comments

#72 - Question: why use "np.transpose()" to change the weights when you initialize the weights of conv layers? when I read others code(https://github.com/MarvinTeichmann/tensorflow-fcn/blob/master/fcn8_vgg.py), I find they don't transpose the vgg16-weights. Note that the his vgg16-weights is equal with your used weights except the weights of conv1_1. So I feel confused whether it is necessary to transpose the vgg16-weights when initializing the weights of conv?

Issue - State: closed - Opened by OREAw almost 7 years ago - 1 comment

#72 - Question: why use "np.transpose()" to change the weights when you initialize the weights of conv layers? when I read others code(https://github.com/MarvinTeichmann/tensorflow-fcn/blob/master/fcn8_vgg.py), I find they don't transpose the vgg16-weights. Note that the his vgg16-weights is equal with your used weights except the weights of conv1_1. So I feel confused whether it is necessary to transpose the vgg16-weights when initializing the weights of conv?

Issue - State: closed - Opened by OREAw almost 7 years ago - 1 comment

#71 - image = tf.placeholder(tf.float32, shape=[None, IMAGE_SIZE, IMAGE_SIZE, 6]).

Issue - State: closed - Opened by MLjian almost 7 years ago - 2 comments

#71 - image = tf.placeholder(tf.float32, shape=[None, IMAGE_SIZE, IMAGE_SIZE, 6]).

Issue - State: closed - Opened by MLjian almost 7 years ago - 2 comments

#70 - Question: How to input my own picture and get the result?

Issue - State: closed - Opened by WANGYuSCU almost 7 years ago - 4 comments

#70 - Question: How to input my own picture and get the result?

Issue - State: closed - Opened by WANGYuSCU almost 7 years ago - 4 comments

#69 - Why set the NUM_OF_CLASSESS = 151

Issue - State: closed - Opened by MZ59 almost 7 years ago - 2 comments

#69 - Why set the NUM_OF_CLASSESS = 151

Issue - State: closed - Opened by MZ59 almost 7 years ago - 2 comments

#68 - different between annotation and valid_images

Issue - State: closed - Opened by engahmed1190 almost 7 years ago - 1 comment

#68 - different between annotation and valid_images

Issue - State: closed - Opened by engahmed1190 almost 7 years ago - 1 comment

#67 - FCN32, FCN16?

Issue - State: closed - Opened by lily10086 almost 7 years ago - 3 comments

#67 - FCN32, FCN16?

Issue - State: closed - Opened by lily10086 almost 7 years ago - 3 comments

#66 - why must download imagenet-vgg-verydeep-19.mat?

Issue - State: closed - Opened by lily10086 almost 7 years ago - 3 comments

#66 - why must download imagenet-vgg-verydeep-19.mat?

Issue - State: closed - Opened by lily10086 almost 7 years ago - 3 comments

#65 - How to load model one time but inference more time ?

Issue - State: closed - Opened by daicoolb about 7 years ago - 3 comments

#65 - How to load model one time but inference more time ?

Issue - State: closed - Opened by daicoolb about 7 years ago - 3 comments

#64 - Training accuracy 100%, validation accuracy half of it (near 50%); blank prediction image

Issue - State: closed - Opened by sanjeevan1234 about 7 years ago - 4 comments

#64 - Training accuracy 100%, validation accuracy half of it (near 50%); blank prediction image

Issue - State: closed - Opened by sanjeevan1234 about 7 years ago - 4 comments

#63 - unsupported pickle protocol 4

Issue - State: closed - Opened by ShiningRuby about 7 years ago - 1 comment

#63 - unsupported pickle protocol 4

Issue - State: closed - Opened by ShiningRuby about 7 years ago - 1 comment

#62 - run error

Issue - State: closed - Opened by haddis3 about 7 years ago - 4 comments

#62 - run error

Issue - State: closed - Opened by haddis3 about 7 years ago - 4 comments

#61 - Pretrained model

Issue - State: closed - Opened by Zumbalamambo about 7 years ago - 1 comment

#61 - Pretrained model

Issue - State: closed - Opened by Zumbalamambo about 7 years ago - 1 comment

#60 - need trained weights

Issue - State: closed - Opened by rnunziata about 7 years ago - 1 comment

#60 - need trained weights

Issue - State: closed - Opened by rnunziata about 7 years ago - 1 comment

#59 - Can you show me, pixel accuracy you retrieve when you evaluate your model in entire validate set ? thanks you so much!

Issue - State: closed - Opened by ghost about 7 years ago - 2 comments

#59 - Can you show me, pixel accuracy you retrieve when you evaluate your model in entire validate set ? thanks you so much!

Issue - State: closed - Opened by ghost about 7 years ago - 2 comments

#58 - put training and validation loss in the same graph in tensorboard

Pull Request - State: closed - Opened by txzhao about 7 years ago - 3 comments

#58 - put training and validation loss in the same graph in tensorboard

Pull Request - State: closed - Opened by txzhao about 7 years ago - 3 comments

#57 - up sampling phase structure?

Issue - State: closed - Opened by bhralzz about 7 years ago - 1 comment

#57 - up sampling phase structure?

Issue - State: closed - Opened by bhralzz about 7 years ago - 1 comment

#56 - A question in Image_Cmaped.ipynb

Issue - State: closed - Opened by fanhualta about 7 years ago - 2 comments

#56 - A question in Image_Cmaped.ipynb

Issue - State: closed - Opened by fanhualta about 7 years ago - 2 comments

#55 - Windows GPU 0.12version issue

Issue - State: closed - Opened by fanhualta over 7 years ago - 1 comment
Labels: duplicate

#55 - Windows GPU 0.12version issue

Issue - State: closed - Opened by fanhualta over 7 years ago - 1 comment
Labels: duplicate

#54 - Getting blank predicted images

Issue - State: closed - Opened by NitishMutha over 7 years ago - 13 comments

#54 - Getting blank predicted images

Issue - State: closed - Opened by NitishMutha over 7 years ago - 13 comments

#53 - How to test and demo?

Issue - State: closed - Opened by zhangzhizz over 7 years ago - 9 comments

#53 - How to test and demo?

Issue - State: closed - Opened by zhangzhizz over 7 years ago - 9 comments

#52 - Why are you padding on the sixth convolutional layer?!

Issue - State: open - Opened by TheRevanchist over 7 years ago - 2 comments
Labels: bug

#52 - Why are you padding on the sixth convolutional layer?!

Issue - State: open - Opened by TheRevanchist over 7 years ago - 2 comments
Labels: bug

#51 - ValueError: Cannot feed value of shape (0,) for Tensor 'input_image:0', which has shape '(?, 224, 224, 3)'

Issue - State: closed - Opened by jabgle over 7 years ago - 9 comments

#51 - ValueError: Cannot feed value of shape (0,) for Tensor 'input_image:0', which has shape '(?, 224, 224, 3)'

Issue - State: closed - Opened by jabgle over 7 years ago - 9 comments

GitHub / shekkizh/FCN.tensorflow issues and pull requests