demetoir / allgans Goto Github PK
View Code? Open in Web Editor NEW머신러닝 프레임워크를 활용한 비교사(Unsupervised) 학습 모델 구현 프로젝트
License: MIT License
머신러닝 프레임워크를 활용한 비교사(Unsupervised) 학습 모델 구현 프로젝트
License: MIT License
Please install our new product, Sonatype Lift with advanced features
add more dataset handler and helper class from http://deeplearning.net/datasets/
당 이슈 작성 시점까지 개발 완료된 비교사 학습 모델들의 결과물 출력
mnist, fashion-mnist 등의 dataset 적용된 결과를 출력하고 간단한 분석을 첨부하여 README
에 올릴 것
결과 이미지를 gif로 움짤 식으로 만들어주는 스크립트 구현
위에는 이미지 밑에는 iter 수 표시
we need some CLI tool for Instance Manager
https://github.com/demetoir/cli_tui_python
i think python-fire base CLI is just fit for ours
This issue lists Renovate updates and detected dependencies. Read the Dependency Dashboard docs to learn more.
These updates have all been created already. Click a checkbox below to force a retry/rebase of any.
These are blocked by an existing closed PR and will not be recreated unless you click a checkbox below.
setup.py
requests ==2.18.4
opencv-python ==3.4.13.47
scikit-learn ==0.19.1
scikit-image ==0.13.1
pillow ==5.0.0
pandas ==0.22.0
matplotlib ==2.2.5
update tutorial code at InstanceManager, Visualizer, data_handler
unlabeled dataset like LLD, apply AE or labeling..
"randomly connected sub neural net" (naming is not confirmed)
ref link : https://item4.github.io/2016-11-01/How-to-Write-a-Git-Commit-Message/
[한 일]
을 대문자로 표기[UPDATE] WTF is that?
[REFACTORING] 프로젝트 구조 개선
호출 모듈 간의 복잡성 개선을 위해 프로젝트 구조를 일부 변경함 (커밋에 대한 요약 50자 이하)
다른 예시)
Hotfix: setup.py의 경로 설정 기능이 bench시 적용되지 않음으로 비활성화
Modification: README를 markdown lint에 따라 일부 수정
foo.py에서 호출하는 모듈 dir_fast/bar.py와 env_set/path_flags.py의 복잡성 ~~~
(생략 가능)
해결된 부분에 대한 설명
(생략 가능)
커밋 내에 여러 가지 일을 포함하고 있으면
- 다음과 같이 개조식으로 서술
- 한 일 1
- 한 일 2
- ...
이슈 #123
파일 이력
modified: README.md
new file: setup.py
위와 같은 형태로 커밋 메시지 규칙을 설정하였으면 좋겠음.
의견 개진 바람.
while executing visualizer some fetching result may have overlap
overlapped data must fetch only once, because may cause performance issue, need to enhance
i think using cache for session fetching is answer
InstanceViewer
- web base InstanceViewer
DatasetManager
- manage dataset
decompose util functions to each file
- ex) numpy related util must move to util_numpy
after task, require test every source where util function used
from result file of print log visualizer to CSV format file
no need to chance code of visualizer,
analyzing script of visualizer and extract data to make CSV format file
ex)
visualizer code
...
self.log("data1: %s, data_name2%s"%(data_value1, data_value2))
....
visualizer's output file
data1 : 123, data2: 124
data1 : 123, data2: 34534
data1 : 123, data2: 1763773
data1 : 123, data2: 333
expect CSV file
data1, data2
123, 124
123, 34534
123, 1763773
123, 333
need auto detect tensorboard for both windows and linux
while generate new instance for model, readme.md file requires.
need to decide contents of readme.md
ex) author, model name, instance generated date, use dataset...
celebA dataset download script require
celebA dataset is big .
so this need better way implement
사용자 환경이 gpu가 없을수도 있으므로 tensorflow-gpu setup으로 설치 하지 않도록 해야함
readme.md 에서는 gpu 설치하는 링크를 걸어 놓고 dependency 수정해야함
while instance manager execute some part of code can be multiprocessing
like load checkpoint file, executing visualizer, write summary.
for better performance managing instance, need to enhance
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.