NVlabs/DG-Net issues and pull requests

#80 - why detach the gradient of appearance codes?

Issue - State: closed - Opened by SDret over 1 year ago - 4 comments

#79 - test_2label报错 yaml.load()缺少 loader

Issue - State: closed - Opened by taqta over 1 year ago - 2 comments

#78 - online feeding and identity supervision

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

#77 - appearance code and structure code

Issue - State: closed - Opened by Wooks-git almost 2 years ago - 1 comment

#76 - how does two different networks extract appearance and clothing separately?

Issue - State: closed - Opened by lakshaydulani over 2 years ago

#75 - IndexError:Dimension out of range(excepted to be in range of [-1,0],but got 1)

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

#74 - 关于FID计算

Issue - State: closed - Opened by zichihuanian almost 3 years ago - 4 comments

#73 - 关于inception score

Issue - State: closed - Opened by zichihuanian almost 3 years ago - 1 comment

#72 - when I run train.py, print there error message. Can you help to solve the issue? Thank you !

Issue - State: open - Opened by yuanqs over 3 years ago - 1 comment

#71 - Out of memory error while training the model

Issue - State: open - Opened by Haneesha123 over 3 years ago - 1 comment

#70 - test_folder

Issue - State: closed - Opened by bixiaopeng0 over 3 years ago - 1 comment

#69 - How to use DG-Net for vehicle reid?

Issue - State: open - Opened by twj1999 over 3 years ago - 2 comments

#68 - Doubt about dis_opt and gen_opt

Issue - State: open - Opened by songbo0925 over 3 years ago - 2 comments

#67 - in trainer.py /gen_update(), i cannot understand "encoder is tuned, input is fixed" and "encoder is fixed, input is tuned"

Issue - State: closed - Opened by wdf19961118 almost 4 years ago - 1 comment

#66 - How to test accuracy.

Issue - State: open - Opened by LiMxStar almost 4 years ago - 1 comment

#65 - 如何采用训练好的模型去生成图片

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

#64 - DG-Net_Trainer类的forward函数的一个疑问

Issue - State: closed - Opened by RuriApoka about 4 years ago - 1 comment

#63 - doubt about a loss in the ‘gen_update’

Issue - State: closed - Opened by 5Yesterday about 4 years ago - 2 comments

#62 - Change the data set to train

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

#61 - Train fp16 interrupt

Issue - State: closed - Opened by 5Yesterday about 4 years ago - 3 comments

#60 - Failed to converge

Issue - State: open - Opened by iyu-Fang about 4 years ago - 9 comments

#59 - A question about setting f=f.detach() in ft_netAB

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

#58 - issue with 'net_last.pth'

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

#57 - 基于teacher model训练，使用数据集MSMT，方便提供下config嘛

Issue - State: open - Opened by wqq19930507 about 4 years ago - 9 comments

#56 - 请问这个GEN只针对market数据吗

Issue - State: open - Opened by guanyonglai about 4 years ago - 6 comments

#55 - A question about the ”appearance code”

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

#54 - Question about alpha

Issue - State: closed - Opened by exityan about 4 years ago - 1 comment

#53 - A question about the loss (6) in the paper

Issue - State: open - Opened by Xiaofu233 about 4 years ago - 1 comment

#52 - Can you share the trained models on the dataset MSMT17 and the dataset MSMT17?

Issue - State: closed - Opened by Be-yourself-zxy about 4 years ago - 2 comments

#51 - Can you share the trained models on the dataset DukeMTMC-reID?

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

#50 - how to train a teacher model

Issue - State: closed - Opened by ganyibo over 4 years ago - 1 comment

#49 - Reconstruction id embedding loss after detach

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

#48 - Interpreting the results of the score in evaluate_gpu.py

Issue - State: closed - Opened by ksuhartono97 over 4 years ago - 2 comments

#47 - From my training logs, the TEACHER is always 0. Is that normal?

Issue - State: open - Opened by y122302868 over 4 years ago - 3 comments

#46 - What should I do if I want to switch to another dataset, such as Duke and MSMT17?Does the teacher model need to be retrained?

Issue - State: closed - Opened by Be-yourself-zxy over 4 years ago - 5 comments

#45 - ModuleNotFoundError: No module named 'torch.utils.hipify'

Issue - State: open - Opened by Dtappledoghuati over 4 years ago - 6 comments

#44 - The training settings and teacher models of training in DukeMTMC and MSMT17.

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

#43 - How to change image name output in 'test_folder.py'

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

#42 - How is it made sure that the image returned from train_loader_a and train_loader_b is of different identity?

Issue - State: closed - Opened by chaitrasj over 4 years ago - 1 comment

#41 - using feature matrix to calculate loss?

Issue - State: closed - Opened by yian2271368 over 4 years ago - 4 comments

#40 - Some questions about the detach of the f in the ft_netAB

Issue - State: closed - Opened by Liu-1994 over 4 years ago - 6 comments

#39 - Some questions about the model structure and the reappearance of the results.

Issue - State: open - Opened by Liu-1994 over 4 years ago - 4 comments

#38 - did you resize MSMT17 images to the same size while you training them?

Issue - State: closed - Opened by yian2271368 over 4 years ago - 8 comments

#37 - Some questions about the appearance space and the structure space

Issue - State: open - Opened by SY-Xuan over 4 years ago - 8 comments

#36 - Why the "val" data-set is subset of "train" data-set?

Issue - State: open - Opened by bhooshan-supe-gmail over 4 years ago - 12 comments

#35 - Regarding sampling code for visualisation of images

Issue - State: open - Opened by asthaverma13 over 4 years ago - 2 comments

#34 - Run DG-Net on Nvidia Jetson Nano

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

#33 - someone use this repository in industrial circles？

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

#32 - When I use another dataset，The generated data is not as much as the original data

Issue - State: open - Opened by CWF-999 almost 5 years ago - 1 comment

#31 - size mismatch

Issue - State: closed - Opened by ghost almost 5 years ago - 7 comments

#30 - how can i test your model to try self-identity generation

Issue - State: open - Opened by superCman almost 5 years ago - 3 comments

#29 - What does "off_gan_bg1" and "off_gan_id1"mean in visual_data/test_folder.py?

Issue - State: closed - Opened by sherlockedlee almost 5 years ago - 1 comment

#28 - Model&configs of MSMT17 dataset

Issue - State: closed - Opened by SikaStar about 5 years ago - 2 comments

#27 - why tuple index out of range?

Issue - State: closed - Opened by Zouhanyu123 about 5 years ago

#26 - I got so low rank1

Issue - State: open - Opened by jinyl777 about 5 years ago - 10 comments

#25 - Generate my own image?

Issue - State: closed - Opened by kayakaracakaya about 5 years ago - 1 comment

#24 - when multi-gpu traing, tensorboard not working.

Issue - State: open - Opened by walalala233 about 5 years ago - 2 comments

#23 - Regarding Transfer Learning from market to duke

Issue - State: closed - Opened by asthaverma13 about 5 years ago - 2 comments

#22 - How to move background to Appearance Space/code?

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

#21 - Evaluation

Issue - State: closed - Opened by LilySys about 5 years ago - 2 comments

#20 - data sets generated with DG-net

Issue - State: closed - Opened by 2110317008 about 5 years ago - 2 comments

#19 - How can I test your model?

Issue - State: closed - Opened by dasmehdix about 5 years ago - 1 comment

#18 - why did you use log-softmax in student-net, and softmax in teacher-net?

Issue - State: closed - Opened by maliho0803 over 5 years ago - 2 comments

#17 - what is the accuracy of the teacher model on market1501? if the teacher model will influence the final results of DG-net?

Issue - State: closed - Opened by maliho0803 over 5 years ago - 1 comment

#16 - Network structure

Issue - State: closed - Opened by CuongNguyen218 over 5 years ago - 4 comments

#15 - Supplementary of paper

Issue - State: closed - Opened by CuongNguyen218 over 5 years ago - 3 comments

#14 - about gradients for appearance encoder (id_a).

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

#13 - torch.utils.serialization.load_lua was removed in pytorch v1.0

Issue - State: closed - Opened by cclauss over 5 years ago - 1 comment

#12 - from torch.nn.utils.spectral_norm import SpectralNorm

Pull Request - State: closed - Opened by cclauss over 5 years ago

#11 - basic question about fp16

Issue - State: open - Opened by Adorablepet over 5 years ago - 30 comments

#10 - Question about 'alpha' in evaluate_gpu.py

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

#9 - basic information about predict_label

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

#8 - question about gan type

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

#7 - One question about generating images

Issue - State: closed - Opened by muzishen over 5 years ago - 4 comments

#6 - What is the meaning of "a,b" in trainer?

Issue - State: closed - Opened by jackie840129 over 5 years ago - 4 comments

#5 - multi-gpu training

Issue - State: open - Opened by tau-yihouxiang over 5 years ago - 13 comments

#4 - Cross-dataset testing?

Issue - State: closed - Opened by ghost over 5 years ago - 7 comments

#3 - Multiple GPU Training

Issue - State: closed - Opened by Phi-C over 5 years ago - 3 comments

#2 - Teacher Model

Issue - State: closed - Opened by tau-yihouxiang over 5 years ago

#1 - Fix readme typo

Pull Request - State: closed - Opened by erjanmx over 5 years ago - 1 comment

GitHub / NVlabs/DG-Net issues and pull requests