kailiang-zhong / descn Goto Github PK
View Code? Open in Web Editor NEWImplementation of paper DESCN, which is accepted in SIGKDD 2022.
License: BSD 2-Clause "Simplified" License
descn's People
Contributors
Stargazers
Watchers
Forkers
fengtong-xiao apple-jack code-wd beijinggao amano-ginji broadcastair yafeiwei nike-adidas senwang98 channingbeayu majie666 rejae arleffdescn's Issues
question about hydra
@kailiang-zhong
hello, I met one env problem when using your code.
python main.py \
./conf4models/lzd_real_data/DESCN.txt \
1 \
./data/real_bin_set_full.5.train.npz \
./data/real_bin_set_full.5.test.npz
But I get the folling hydra problem, which I have speed half day to fix it but failed.
main.py:859: UserWarning:
The version_base parameter is not specified.
Please specify a compatability version level, or None.
Will assume defaults for version 1.1
@hydra.main(config_path='./conf', config_name='conf_lzd_real_bin_v0_full.yaml')
2023-10-12 00:16:57,287 - DEBUG - Setting JobRuntime:name=UNKNOWN_NAME
2023-10-12 00:16:57,287 - DEBUG - Setting JobRuntime:name=main
Traceback (most recent call last):
File "main.py", line 903, in <module>
main()
File "/home/dell/miniconda3/envs/pt1.12/lib/python3.7/site-packages/hydra/main.py", line 99, in decorated_main
config_name=config_name,
File "/home/dell/miniconda3/envs/pt1.12/lib/python3.7/site-packages/hydra/_internal/utils.py", line 401, in _run_hydra
overrides=overrides,
File "/home/dell/miniconda3/envs/pt1.12/lib/python3.7/site-packages/hydra/_internal/utils.py", line 458, in _run_app
lambda: hydra.run(
File "/home/dell/miniconda3/envs/pt1.12/lib/python3.7/site-packages/hydra/_internal/utils.py", line 223, in run_and_report
raise ex
File "/home/dell/miniconda3/envs/pt1.12/lib/python3.7/site-packages/hydra/_internal/utils.py", line 220, in run_and_report
return func()
File "/home/dell/miniconda3/envs/pt1.12/lib/python3.7/site-packages/hydra/_internal/utils.py", line 461, in <lambda>
overrides=overrides,
File "/home/dell/miniconda3/envs/pt1.12/lib/python3.7/site-packages/hydra/_internal/hydra.py", line 109, in run
run_mode=RunMode.RUN,
File "/home/dell/miniconda3/envs/pt1.12/lib/python3.7/site-packages/hydra/_internal/hydra.py", line 599, in compose_config
validate_sweep_overrides=validate_sweep_overrides,
File "/home/dell/miniconda3/envs/pt1.12/lib/python3.7/site-packages/hydra/_internal/config_loader_impl.py", line 147, in load_configuration
validate_sweep_overrides=validate_sweep_overrides,
File "/home/dell/miniconda3/envs/pt1.12/lib/python3.7/site-packages/hydra/_internal/config_loader_impl.py", line 245, in _load_configuration_impl
config_name, overrides
File "/home/dell/miniconda3/envs/pt1.12/lib/python3.7/site-packages/hydra/_internal/config_loader_impl.py", line 228, in _parse_overrides_and_create_caching_repo
parsed_overrides = parser.parse_overrides(overrides=overrides)
File "/home/dell/miniconda3/envs/pt1.12/lib/python3.7/site-packages/hydra/core/override_parser/overrides_parser.py", line 100, in parse_overrides
) from e.__cause__
hydra.errors.OverrideParseException: LexerNoViableAltException: ./conf4models/lzd_real_data/DESCN.txt
^
See https://hydra.cc/docs/1.2/advanced/override_grammar/basic for details
Have you meet this problem before?
为什么auuc波动这么大呢？
作者，您好，我运行了仓库中的代码，发现了AUUC波动很大，跟你贴出的jupyter中的出现的现象是一样的，请问造成这个现象的原因有哪些呢。
另外就是发现x-network比descn要好，根据jupyter中运行descn的超参数配置，AUUC为：0.031541，可能的原因在哪呢。
About the ACIC_2019 dataset.
Thanks for the code! 💯 But I encountered a problem.
In this git repo and the ACIC original website, only the data generator is provided.
In the data generator`s code, it is said that we could find the data.csv from "Epilepsy dataset from UCI repository http://archive.ics.uci.edu/ml/datasets/Epileptic+Seizure+Recognition".
But I found that this dataset is removed from the UCI repo. Would you mind providing the "data.csv" that used in this paper? Because getting the data.csv from different sources might not yield the same dataset that is used in this paper.
I would really appreciate some help~ : )
Some questions about the codes
Hi, thanks for the open source.
But I have some questions about the codes:
-
On line 659 in the main.py file, the total loss includes 8 sub-losses. But tr_loss and cr_loss are removed explicitly in Equation 7 of the paper, and the CFR regularization term (i.e., imb_dist_loss) is not used as described in Section 3.3.
-
In the parameter configuration file (i.e., conf_lzd_real_bin_v0_full.yaml), I find that the weight parameters corresponding to the proposed 5 new sub-losses are all set to 0 (i.e., prpsy_w, escvr1_w, escvr0_w, mu0hat_w and mu1hat_w).
It may be necessary to double check whether it is consistent with the best experiment.
Best regards
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.