visionlearninggroup / dance Goto Github PK
View Code? Open in Web Editor NEWrepository for Universal Domain Adaptation through Self-supervision
License: MIT License
repository for Universal Domain Adaptation through Self-supervision
License: MIT License
Hey! I was trying to run this code on my machine, but I am not able to create the conda environment due to package errors.
This is the error I am greeted with when I try to create the environment. Could you provide me with a yaml file for this environment or give me some tips to resolve this error.
Thank you,
`
conda create --name dance --file requirements.txt
Collecting package metadata (current_repodata.json): done
Solving environment: failed with repodata from current_repodata.json, will retry with next repodata source.
Collecting package metadata (repodata.json): done
Solving environment: failedPackagesNotFoundError: The following packages are not available from current channels:
Current channels:
To search for alternate channels that may provide the conda package you're
looking for, navigate to
https://anaconda.org
and use the search bar at the top of the page.
Dear authors,
Thanks for the code release. I was curious about the fact that accuracy of method DANN (in closed set for office 31) reported in the paper is much higher than accuracies reported for it in other papers like CDAN. So I wanted to know what architecture was used for the classifier and discriminator network along with other hyperparams for your DANN implementation.
Thanks
Hi, thanks very much for submitting the good work. I have some simple questions about code in train_dance.py:
in 188-189 lines : ### We do not use memory features present in mini-batch
feat_mat[:, index_t] = -1 / conf.model.temp
I know the meaning is calculating the similarity between min-batch and min-batch with current features not memory features, but what's the meaning about -1 / conf.model.temp?
in 195-196 lines: loss_nc = conf.train.eta * entropy(torch.cat([out_t, feat_mat, feat_mat2], 1))
I can't understand what's the effect of direct connection of feat_mat and feat_mat2. why not put the feat_mat2 into the proper index position in feat_mat, as we know, the index of feat_t in different iteration is not same.
Thanks very much and hope to get your reply
Thank you for sharing your codes. I have a question.
Regarding Table 4, Office(10/0/11) seems to have a typo.
Which is correct, Office(10/10/11) or Office(10/0/21)?
In the page of 3, in section 3, the authors have said that "In addition, we utilize domain-specific" batch normalization to ..."; However, I do not found that the code is implemented using domain-specific batch normalization. So I am a little curious about that? Can you explain why?
Hello, Thanks for your interesting and inspiring work~
Recently, [1] and [2] proposed a new metric, harmonic mean of Known ACC and Unknown ACC, to better evaluate the capability on identifying known samples and unknown samples.
Will you consider reporting the results with this metric? Or could you release the model trained with the proposed method(DANCE) ?.
[1] Learning to Detect Open Classes for Universal Domain Adaptation
[2] On the Effectiveness of Image Rotation for Open Set Domain Adaptation
Hi, thanks very much for your wonderful work about UniDA.
I have a question about the test Function in eval.py:
I don’t know if I misunderstood the code. Thanks very much and hope to get your reply.
After carefully checking the training code, I have a little question about the feat_mat2
in loss_nc
.
feat_mat2
help the performance of neighborhood clustering?I see when computing the loss_nc
the entropy was fed with the concatenation of three values which are out_t
, feat_mat
and feat_mat2
. It turns out that the out_t
and feat_mat
denote the feat_mat2
has the same influence with feat_mat
. But it only works in the mini batch data, does it? Is there any reasonable explanation about it? Thanks in advance.
In txt file, I find paths of some images from /research/masaito/, what the difference between images in /research/masaito/ and data/
removed
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.