yu-wu / one-example-person-reid Goto Github PK
View Code? Open in Web Editor NEWCode for TIP2019 Progressive Learning for Person Re-Identification with One Example
License: MIT License
Code for TIP2019 Progressive Learning for Person Re-Identification with One Example
License: MIT License
Hello, thank you for sharing, but I run the error after following the readme file.
NotADirectoryError: [Errno 20] Not a directory: '/home/Ensoleille/One-Example-Person-ReID-master/data/market1501/train/0002_c1s1_000451_03.jpg'.I hope you can tell me how to reproduce smoothly.
I'm sorry to bother you.I have run your code several times, mAP only reaches 8%, and top-1 only reaches 20% at most. Otherwise,the results are different when each time I run the code. How can I achieve the effect in your paper and keep the results at the same?
for example:
dataset market1501:the fixed_layer = False
dataset MARS: the fixed_layer = True
Hi, Thanks for your awesome work. I read the paper and found baseline in market1501 is rank1=26.0, mAP=9.0. In my opinion, this is trained using 751 images with 751 ids. But I just try, and I cannot get the baseline result. My rank1=8.6, mAP=3.0. I want to know how to reproduce the baseline result using this code.
Thanks for your many great works!
I have a question about how to get the static strategy work,such as the DGM+IDE ,which rank-1=36.8 and map=16.9 ?
Many thanks for you !
Hi,I want to replace the resnet50 with other networks,but every time I switch to the source code for another network, the program runs up a lot of memory after it shows" create dataloader for Training with batch_size 16", and then the code stops running with showing nothing.I don't know why and what should I do if I want to change the network?
Thank you for your great work!But i found the exloss cannot converge when training, the optimizer is adam, lr=3e-4
Hellow!For video datasets, such as Mars, I would like to ask whether a tracklet is similar to a single frame image,? And whether all the frames in a tracklet are input into the network at the same time?
Hello,When i train the model ,i will always get this problem?Is anyone have solved it?
Traceback (most recent call last):
File "run.py", line 123, in
main(parser.parse_args())
File "run.py", line 87, in main
eug.train(new_train_data, unselected_data, step, loss=args.loss, epochs=args.epochs, step_size=args.step_size, init_lr=0.1) if step != resume_step else eug.resume(ckpt_file, step)
File "/home/xutian/One-Example-Person-ReID-master/reid/eug.py", line 92, in train
return self.softmax_train(train_data, unselected_data, step, epochs, step_size, init_lr, dropout, loss)
File "/home/xutian/One-Example-Person-ReID-master/reid/eug.py", line 135, in softmax_train
u_dataloader = self.get_dataloader(unselected_data, training=True, is_ulabeled=True)
File "/home/xutian/One-Example-Person-ReID-master/reid/eug.py", line 82, in get_dataloader
shuffle=training, pin_memory=True, drop_last=training)
File "/home/xutian/tools/anaconda3/envs/freedom/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 802, in init
sampler = RandomSampler(dataset)
File "/home/xutian/tools/anaconda3/envs/freedom/lib/python3.6/site-packages/torch/utils/data/sampler.py", line 64, in init
"value, but got num_samples={}".format(self.num_samples))
ValueError: num_samples should be a positive integeral value, but got num_samples=0
Hello author, how do you get 'market1501_init_-1.0.pickle','duke_init_-1.0.pickle' in file examples? Plase help me! Please! Thank you very much!
Hello author, I am recreating your paper. I want to ask how to use the Pycharm editor to run the run.sh file. I am a beginner. I want to adjust the parameters in the code because my GPU memory is not enough. What can I do? Thank you very much.
Hi, I am trying your code on the duke dataset but I am getting the following error.
/usr/local/lib/python3.6/dist-packages/sklearn/externals/joblib/externals/cloudpickle/cloudpickle.py:47: DeprecationWarning: the imp module is deprecated in favour of importlib; see the module's documentation for alternative uses
import imp
Files already downloaded and verified
Duke dataset loaded
subset | # ids | # tracklets
---------------------------
train | 702 | 16522
query | 702 | 2228
gallery | 1110 | 17661
labeled | N/A | 702
unlabeled | N/A | 15820
Create new one-shot split and save it to ./examples/duke_init_-1.pickle
Runing: EF=10%, step 0: Nums_to_be_select 1582 Ritio Logs-dir 0.1
create dataloader for Training with batch_size 16
create dataloader for Training with batch_size 16
Traceback (most recent call last):
File "run.py", line 124, in <module>
main(parser.parse_args())
File "run.py", line 88, in main
eug.train(new_train_data, unselected_data, step, loss=args.loss, epochs=args.epochs, step_size=args.step_size, init_lr=0.1) if step != resume_step else eug.resume(ckpt_file, step)
File "/home/zamra/PYreps/Person_ReId/One-Example-Person-ReID/my_reid/eug.py", line 92, in train
return self.softmax_train(train_data, unselected_data, step, epochs, step_size, init_lr, dropout, loss)
File "/home/zamra/PYreps/Person_ReId/One-Example-Person-ReID/my_reid/eug.py", line 142, in softmax_train
trainer.train(epoch, s_dataloader, u_dataloader, optimizer, use_unselcted_data, print_freq=len(s_dataloader)//2)
File "/home/zamra/PYreps/Person_ReId/One-Example-Person-ReID/my_reid/trainers.py", line 67, in train
u_loss, u_prec1 = self._forward(u_inputs, u_targets, 'u')
File "/home/zamra/PYreps/Person_ReId/One-Example-Person-ReID/my_reid/trainers.py", line 114, in _forward
u_loss, outputs = self.u_criterion(u_feats, targets)
File "/home/zamra/.local/lib/python3.6/site-packages/torch/nn/modules/module.py", line 489, in __call__
result = self.forward(*input, **kwargs)
File "/home/zamra/PYreps/Person_ReId/One-Example-Person-ReID/my_reid/exclusive_loss.py", line 36, in forward
outputs = Exclusive(self.V)(inputs, targets) * self.t
File "/home/zamra/PYreps/Person_ReId/One-Example-Person-ReID/my_reid/exclusive_loss.py", line 15, in forward
outputs = inputs.mm(self.V.t())
RuntimeError: size mismatch, m1: [16 x 2048], m2: [1024 x 15820] at /pytorch/aten/src/THC/generic/THCTensorMathBlas.cu:266
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.