cszn/DnCNN issues and pull requests

#102 - params_updates中adam算法的参数更新好像有误

Issue - State: open - Opened by yihaodong1 about 1 year ago

#101 - 提供训练集

Issue - State: open - Opened by c1h2e3n4y5u6n7 about 1 year ago

#100 - The Wrong when I using the \DnCNN\TrainingCodes\DnCNN_TrainingCodes_v1.1\Demo_Train_model_64_25_Res_Bnorm_Adam.m

Issue - State: open - Opened by yihaodong1 about 1 year ago

#99 - The wrng

Issue - State: closed - Opened by yihaodong1 about 1 year ago

#98 - About image shape debug

Issue - State: open - Opened by darkkkcc almost 2 years ago - 2 comments

#97 - About Testing

Issue - State: open - Opened by saqibnaziir over 2 years ago

#96 - Repetitive Dataset and DataLoader Creation in PyTorch version

Issue - State: open - Opened by sptom over 2 years ago

#95 - Lower depth yields same performance

Issue - State: open - Opened by Youyoun over 2 years ago

#94 - If I want to input my own image for deep learning,but i don't need add to Gassian noise level,what should I do?

Issue - State: open - Opened by LIN-CHENG-YUEH almost 3 years ago - 1 comment

#93 - Sigma=45 model in .h5

Issue - State: open - Opened by nehagour over 3 years ago

#92 - Update data_generator.py

Pull Request - State: open - Opened by Gaochenyin over 3 years ago

#91 - main_test.py

Issue - State: open - Opened by s0966066980 over 3 years ago

#90 - Are there any files missing?Such as vl_simplenn_move.m and vl_argparse.m

Issue - State: open - Opened by lsfhuihuiff almost 4 years ago

#89 - vl_argparse(opts, varargin);No function or variable is defined—— 'vl_argparse'。And I can't find matlab code about this function

Issue - State: open - Opened by lsfhuihuiff almost 4 years ago

#88 - Data Preparation for DnCNN3

Issue - State: open - Opened by supriyaarun27 almost 4 years ago

#87 - ERROR: Could not find a version that satisfies the requirement torch==0.4.1 ERROR: No matching distribution found for torch==0.4.1

Issue - State: open - Opened by muhammad-aamir94 about 4 years ago - 1 comment

#86 - Batch Normalization gives worse perfromence

Issue - State: open - Opened by Sumching about 4 years ago

#85 - bug in usage of OpenCV resize function

Issue - State: open - Opened by hailihu about 4 years ago

#84 - How to resolve this bug? could you please tell me?maybe I solved above problem,but I met the follow problem :

Issue - State: open - Opened by 191578010 about 4 years ago

#83 - Version of DnCNN that is compatible with tensorflow 2.0

Issue - State: open - Opened by tamireiny about 4 years ago

#82 - Size of input images

Issue - State: closed - Opened by Nipl0d over 4 years ago

#81 - color dcnn model's act_mode is different in train and test phase (pytorch version)

Issue - State: open - Opened by lionzhnn over 4 years ago

#80 - About training data creating

Issue - State: open - Opened by Kaede93 almost 5 years ago - 1 comment

#79 - Color Images input as test in keras version

Issue - State: open - Opened by Alirezak2n almost 5 years ago

#78 - Add bibtex syntax highlighting to readme

Pull Request - State: open - Opened by HappyFacade almost 5 years ago

#77 - 有训练好的Pytorch的彩色去噪模型吗

Issue - State: closed - Opened by lionzhnn about 5 years ago

#76 - Average PSNR vs epochs - denoised images with spots?

Issue - State: open - Opened by betty97 about 5 years ago

#75 - it seems that there are some preblems in the pytorch implement of this code!

Issue - State: open - Opened by cscss about 5 years ago - 3 comments

#74 - Discrepancy in the parameters used for batch normalisation between pytorch and keras

Issue - State: open - Opened by zaccharieramzi about 5 years ago

#73 - When training in pytorch with SGD, loss is inf , how to do with it?

Issue - State: open - Opened by gzz1996 about 5 years ago

#72 - pytorch training code sucks

Issue - State: closed - Opened by oneTaken about 5 years ago

#71 - Loss function in Keras code

Issue - State: open - Opened by KewJieLong over 5 years ago - 1 comment

#70 - 把损失函数换成mse

Issue - State: closed - Opened by ssywzzc over 5 years ago

#69 - Incompatible Image shape error

Issue - State: closed - Opened by ghost over 5 years ago - 2 comments

#68 - can pytorch version>0.4 run this pytorch code?

Issue - State: open - Opened by johsnows over 5 years ago - 1 comment

#67 - Pytorch Implementation Loss function Line 163 (main_train.py)

Issue - State: open - Opened by dagarshali over 5 years ago - 1 comment

#66 - About simultaneous denoise and super resolution task

Issue - State: open - Opened by bittergourd1224 over 5 years ago

#65 - There is an error in the Keras code

Issue - State: closed - Opened by DieGlory over 5 years ago

#64 - [keras version]The Labels in the matlab code are noises, while the labels in the keras code are clean data. Is the code wrong or am I wrong?

Issue - State: closed - Opened by wangweiwei104 over 5 years ago

#63 - Is the code on line 23 in the main_train.py file[DnCNN-keras] correct?

Issue - State: closed - Opened by wangweiwei104 over 5 years ago - 6 comments

#62 - Time till loss convergence

Issue - State: open - Opened by AnasShahzad1996 over 5 years ago

#61 - License ?

Issue - State: open - Opened by titsitits over 5 years ago - 1 comment

#60 - Kernel Initialization

Issue - State: open - Opened by riteshkumarumassedu over 5 years ago - 2 comments

#59 - data/Train400

Issue - State: open - Opened by HugoGuadalupe over 5 years ago - 1 comment

#58 - Train the net by my own (clean-noisy)image pairs

Issue - State: open - Opened by hujiayu712 over 5 years ago - 1 comment

#57 - Couldn't run main_train.py (keras) due to an error caused by datagenerator.py

Issue - State: closed - Opened by wlf114 almost 6 years ago - 1 comment

#56 - pretrained model of CDnCNN-B (pytorch version)

Issue - State: open - Opened by HappyExploring almost 6 years ago - 1 comment

#55 - pertained model of sigma-25

Issue - State: open - Opened by Gurpremm almost 6 years ago - 2 comments

#54 - pretrained model of sigma-50

Issue - State: open - Opened by Jee-King almost 6 years ago - 1 comment

#53 - Keras-Version: in Model-Definition DnCNN() the parameter use_bnorm is faulty?

Issue - State: open - Opened by mullanptr almost 6 years ago - 2 comments

#52 - CBSD432 dataset for CDnCNN-B

Issue - State: open - Opened by Shakarim94 about 6 years ago

#51 - hello， do you have test model for RGB image？

Issue - State: closed - Opened by chensongkui about 6 years ago - 1 comment

#50 - Hello, I wonder if you have tested your code using multi-gpu with matconvnet?

Issue - State: closed - Opened by XSLXANDY about 6 years ago - 1 comment

#49 - Pytorch model doesn't produce reported PSNR scores.

Issue - State: closed - Opened by CodeR57 over 6 years ago - 3 comments

#48 - add scripts

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

#47 - Is there anyone who has implemented this code? And made his own changes?

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

#46 - error in main_train.py

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

#45 - Struggle with PyTorch version code

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

#44 - loss function in DnCNN-Keras

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

#43 - Error in Demo_test_DnCNN_C (line 57)

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

#42 - Where can I get train400 dataset?

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

#41 - strange output from the DnCNN denosing

Issue - State: closed - Opened by Allen1207 over 6 years ago

#40 - BSD68 RGB dataset

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

#39 - Quesitons about the training process.

Issue - State: closed - Opened by hyf6661669 over 6 years ago

#38 - Demo_test_DnCNN.m

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

#37 - how to fine-tuned your model with new data sets

Issue - State: closed - Opened by msiraj83 almost 7 years ago - 1 comment

#36 - Question about learning rate decay

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

#35 - Colored images after JPEG deblocking

Issue - State: closed - Opened by perdigas91 almost 7 years ago

#34 - 不加BN层，在BSD68测试集上效果反而更加好

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

#33 - the data in the training data set can be more than 255?

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

#32 - Forward:out of memory

Issue - State: closed - Opened by my1347 about 7 years ago - 3 comments

#31 - receptive field size

Issue - State: closed - Opened by yangisme about 7 years ago

#30 - Overfitting problem

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

#29 - Denoise the same picture with different nets

Issue - State: closed - Opened by sdlpkxd about 7 years ago

#28 - MATLAB JPEG encoder?

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

#27 - Can I ask a question about sigma setting of blind Gaussian denoising

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

#26 - Hello

Issue - State: closed - Opened by chenjian123321 about 7 years ago - 1 comment

#25 - sorry can i ask your one more question?

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

#24 - can i ask you specipic option about uploaded models?

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

#23 - model with sigma 5 or 1

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

#22 - Question about learning rate for bias.

Issue - State: closed - Opened by ngcthuong over 7 years ago - 1 comment

#21 - can i ask you about last layer?

Issue - State: closed - Opened by sooyeonshin over 7 years ago - 1 comment

#20 - Get back the clean image

Issue - State: closed - Opened by yangisme over 7 years ago - 12 comments

#19 - Understanding vl_nnloss.m

Issue - State: closed - Opened by ngcthuong over 7 years ago - 4 comments

#18 - data_size is NAN

Issue - State: closed - Opened by ghost over 7 years ago - 23 comments

#17 - Does it require GPU to train this network?

Issue - State: closed - Opened by doantientai over 7 years ago - 1 comment

#16 - error 0 in training

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

#15 - MLP code for sigma = 50

Issue - State: closed - Opened by saeed-anwar over 7 years ago - 1 comment

#14 - BN layers in given model

Issue - State: closed - Opened by ghost over 7 years ago - 12 comments

#13 - Training with 980Ti data size NAN

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

#12 - what's exactly the convlution implement?can you offer a reference?

Issue - State: closed - Opened by BigJerry over 7 years ago - 1 comment

#11 - Reference to non-existent field 'dilate', matconvnet 1.0-beta25

Issue - State: closed - Opened by ngcthuong over 7 years ago - 8 comments

#10 - why your models trained already are so small,though they have a large depth? they're very memory-friendly,I'm wondering how you get there

Issue - State: closed - Opened by BigJerry over 7 years ago

#9 - Ease of use changes

Pull Request - State: open - Opened by kevinzying over 7 years ago - 1 comment

#8 - Why you comment most of your ‘net = vl_simplenn_tidy(net);’

Issue - State: closed - Opened by tendence over 7 years ago - 2 comments

#7 - Blind Color Denoiser Model

Issue - State: closed - Opened by hazirbas almost 8 years ago - 3 comments

#6 - About the "bnorm" layers

Issue - State: closed - Opened by wbhu almost 8 years ago - 1 comment

#5 - About the usage of DnCNN with matconvnet-1.0-beta24 on linux

Issue - State: closed - Opened by ZhangJingrong almost 8 years ago - 1 comment

#3 - Are you going to port it to python?

Issue - State: closed - Opened by MariasStory almost 8 years ago - 9 comments

#1 - vl_simplenn_move.m not found.

Issue - State: closed - Opened by leonardltk over 8 years ago - 3 comments

GitHub / cszn/DnCNN issues and pull requests