NVlabs/SPADE issues and pull requests

#174 - Training the model for Image to Image translation without label_nc and instance maps.

Issue - State: open - Opened by VikasOjha666 about 1 year ago

#173 - fix(sec): upgrade torch to 1.13.1

Pull Request - State: open - Opened by chncaption over 1 year ago

#172 - Online demo app no longer on nvidia site?

Issue - State: open - Opened by BlakePR over 1 year ago

#171 - any pretrained models on ade20k or cityscapes with use_vae?

Issue - State: open - Opened by fido20160817 over 1 year ago

#169 - add --tf_log , program can not run

Issue - State: open - Opened by dp1556235341 over 1 year ago - 1 comment

#168 - GauGAN2 Paper

Issue - State: open - Opened by yozhikoff over 2 years ago - 1 comment

#167 - Issue a black background is added to the transparent image

Issue - State: open - Opened by Ghaleb-alnakhlani over 2 years ago - 1 comment

#166 - how to prepare the custom datasets?

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

#165 - errors with custom dataset

Issue - State: open - Opened by emoebel over 2 years ago - 3 comments

#164 - upadte some code to normalize data to prevent array subscripts from c…

Pull Request - State: closed - Opened by Larry-zx over 2 years ago

#163 - IndexError with custom dataset

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

#162 - Extend to instance segmentation

Issue - State: open - Opened by INF800 almost 3 years ago

#161 - Public API

Issue - State: open - Opened by ezzcodeezzlife almost 3 years ago

#160 - CUDA Error??

Issue - State: open - Opened by edwardcho almost 3 years ago - 4 comments

#159 - How "label_nc" in the setting works?

Issue - State: open - Opened by KentaItakura almost 3 years ago - 2 comments

#158 - Image to Image translation like pix2pix

Pull Request - State: open - Opened by iamrishab almost 3 years ago - 3 comments

#157 - How many categories have you predicted for flickr data using deeplab?

Issue - State: open - Opened by zhouwy19 almost 3 years ago

#156 - Choice between multi-scale discriminator and progrssive growing of GANs ?

Issue - State: open - Opened by leo27945875 almost 3 years ago

#155 - Main

Pull Request - State: open - Opened by sjinu96 almost 3 years ago - 1 comment

#154 - Converting to onnx

Issue - State: open - Opened by cjenkins5614 about 3 years ago - 2 comments

#153 - RuntimeError: cublas runtime error : an access to GPU memory space failed at

Issue - State: open - Opened by WNingup about 3 years ago

#152 - Two update_learning_rate with same name

Issue - State: open - Opened by Zhiozhio about 3 years ago

#151 - CUDA error: device-side assert triggered

Issue - State: closed - Opened by Devicharith over 3 years ago - 1 comment

#150 - Update README.md

Pull Request - State: closed - Opened by chaficii over 3 years ago

#149 - How can I run this using RGB label as input ?

Issue - State: closed - Opened by DWCTOD over 3 years ago - 4 comments

#148 - test.py how to use Style Images,I have some Style Images,but I dont know how to use.

Issue - State: closed - Opened by husiluanxiang over 3 years ago - 2 comments

#147 - continue training ?

Issue - State: closed - Opened by ranch-hands over 3 years ago

#146 - update as_grey argument of skimage.io.imread to as_gray

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

#145 - Synchronized-BatchNorm-Pytorch

Issue - State: closed - Opened by HyunjunGil over 3 years ago - 1 comment

#144 - Python test.py cannot be Successful opened

Issue - State: closed - Opened by tociv05 over 3 years ago - 2 comments

#143 - Flickr Landscape Segmentation map

Issue - State: closed - Opened by DezeZhao over 3 years ago - 1 comment

#142 - there is no models.networks.sync_batchnorm

Issue - State: closed - Opened by oliverck over 3 years ago

#141 - Using '--mode custom' to train my own dataset but failed and the error information is "RuntimeError: Given groups=1, weight of size [1024, 14, 3, 3], expected input[1, 16, 8, 8] to have 14 channels, but got 16 channels instead"

Issue - State: closed - Opened by 25th-engineer over 3 years ago - 1 comment

#140 - 这玩意儿能用来搞pix2pix那种图像翻译吗？怎么配置？

Issue - State: closed - Opened by 25th-engineer over 3 years ago

#139 - --z_dim hard-coded in encoder

Issue - State: closed - Opened by kampelmuehler over 3 years ago - 1 comment

#138 - Segfault when --label_nc is incorrect

Issue - State: closed - Opened by kampelmuehler over 3 years ago - 1 comment

#137 - --label_nc and --contain_dontcare_label

Issue - State: open - Opened by sashi2295 over 3 years ago - 2 comments

#136 - Is the val_img folder in datasets a test set for training?

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

#135 - Segmentation Fault in Pytorch 1.4

Issue - State: closed - Opened by crockwell almost 4 years ago

#134 - Question about the image encoder

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

#133 - throw errors when test my own model

Issue - State: closed - Opened by jmssxgp almost 4 years ago - 1 comment

#132 - "__scale_shortside()" gives strange results

Issue - State: open - Opened by crrrr30 about 4 years ago

#131 - Replace scipy.misc.toimage to fix NVlabs#98

Pull Request - State: closed - Opened by cpvannier about 4 years ago

#130 - question about compute_discriminator_loss

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

#129 - Question about using Spectral Normalization

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

#128 - How to do example guided image generation, as shown in Fig.1 of the paper?

Issue - State: open - Opened by Ha0Tang about 4 years ago - 3 comments

#127 - Failure when testing my own trained model

Issue - State: closed - Opened by firasguediri about 4 years ago - 3 comments

#126 - why train throws runtime error?

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

#125 - Add run on repl.it badge to README

Pull Request - State: open - Opened by HeroToxins about 4 years ago

#124 - The compute of mIOU

Issue - State: open - Opened by fuyou123 about 4 years ago

#123 - train mode was not turned on by default during training

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

#122 - May the Gan hinge loss be negative?

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

#121 - pix2pixHD++ for image to image translation

Issue - State: open - Opened by BEpresent about 4 years ago - 2 comments

#120 - license related issue - open source research code using SPADE as code base

Issue - State: closed - Opened by YingdaXia about 4 years ago

#119 - how to set to train paired images, image to iamge

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

#118 - Index out of bounds error for pix2pix custom dataset, with simple fix

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

#117 - A typo in coco_generate_instance_map.py.

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

#116 - Errors when training model with scale_width option and multi-gpu

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

#115 - Cannot load pretrained model with --use_vae

Issue - State: open - Opened by keerthan2 over 4 years ago - 1 comment

#114 - name 'SynchronizedBatchNorm2d' is not defined

Issue - State: open - Opened by happy-yu0531 over 4 years ago - 2 comments

#113 - Mismatch of Paper SPADE Equation and Code

Issue - State: closed - Opened by mbbrodie over 4 years ago - 2 comments

#112 - Image translation with SPADE?

Issue - State: open - Opened by kex243 over 4 years ago - 9 comments

#111 - Gradient on netG is computed when updating discriminator

Issue - State: closed - Opened by htzheng over 4 years ago - 2 comments

#110 - Trying to upload landscape image creates CORS error on playground

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

#109 - cast BoolTensor to ByteTensor; compatible with Pytorch 1.4

Pull Request - State: open - Opened by lmxyy over 4 years ago

#108 - Non reponsive website formed

Issue - State: closed - Opened by ArvindAROO over 4 years ago - 2 comments

#107 - Expected object of scalar type Byte but got scalar type Bool for argument

Issue - State: open - Opened by atopheim over 4 years ago - 4 comments

#106 - Expected training time on COCO?

Issue - State: open - Opened by Ha0Tang over 4 years ago

#105 - Possibility of making SPADE work on Panoptic Annotations.

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

#104 - Doubt regarding implementation of SPADE normalization

Issue - State: closed - Opened by rajatongit almost 5 years ago - 2 comments

#103 - Using Custom Dataset option with Pretrained Models

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

#102 - Why turn gradient on for discriminator input?

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

#101 - Question regarding generating Infrared images from Visual images

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

#100 - Pix Accuracy=93

Issue - State: open - Opened by zkchen95 almost 5 years ago

#99 - Fixes #98

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

#98 - AttributeError: module 'scipy.misc' has no attribute 'toimage'

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

#97 - Do not use CUDA for calculating VGG loss when gpu_ids=-1

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

#96 - Good training output, but Poor Testing Quality, and Input label become All Green?

Issue - State: closed - Opened by aldhosutra almost 5 years ago - 3 comments

#95 - Do not use CUDA for vgg when gpu_ids is set to -1 (not using GPU)

Pull Request - State: closed - Opened by legitmaxwu almost 5 years ago

#94 - training without convergence

Issue - State: open - Opened by YoungSharp almost 5 years ago - 2 comments

#93 - why pretrained model need the path for the real images__

Issue - State: open - Opened by SaidaSaad almost 5 years ago - 2 comments

#92 - Generate RGB Images

Issue - State: open - Opened by SaidaSaad almost 5 years ago - 5 comments

#91 - Are the discriminators for the pretrained networks available?

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

#90 - Can you provide the results of your expriment?

Issue - State: open - Opened by liby001 almost 5 years ago

#89 - (bug fix) data type incompatibility between boolean values while creating instance_edge_map

Pull Request - State: open - Opened by vdhyani96 almost 5 years ago

#88 - Different from the published paper

Issue - State: open - Opened by ThisIsIsaac almost 5 years ago - 2 comments

#87 - Why not generate instance map for ADE20k?

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

#86 - RuntimeError: Expected object of scalar type Byte but got scalar type Bool

Issue - State: open - Opened by vdhyani96 about 5 years ago - 2 comments

#84 - out = normalized*(1+gamma)+beta

Issue - State: closed - Opened by Dominoer about 5 years ago - 4 comments

#74 - Why nef option is not used in Encoder?

Issue - State: open - Opened by kekke-gk about 5 years ago - 1 comment

#72 - Some errors when training the SPADE over COCO-stuff dataset

Issue - State: open - Opened by ShihuaHuang95 over 5 years ago - 1 comment

#68 - Launch train with custom dataset - Invalid index in scatter at ..

Issue - State: open - Opened by Luke035 over 5 years ago - 17 comments

#66 - SPADE for RGB to RGB translation task?

Issue - State: open - Opened by aayushARM over 5 years ago - 6 comments

#64 - Resume training - Save money using Spot Instances

Issue - State: open - Opened by jstoeffler over 5 years ago - 3 comments

#62 - Bug in architecture.py file

Issue - State: open - Opened by araffiee1989 over 5 years ago - 1 comment

#57 - Cudnn error when using SPADE

Issue - State: open - Opened by emanueledelsozzo over 5 years ago - 14 comments

GitHub / NVlabs/SPADE issues and pull requests