vu-minh/idr

Interactive Dimensionality Reduction

Interactive t-SNE

Build final elm-client:

cd idr-elm
elm make src/Main.elm --output=main.html

For demo (production), use this main.html:

• cp main.html ./data/
• run static server as the following notes.

Demo workflow in Built environment:

• Run python socket server (internal port 5000)

cd idr-server/tsnex
python tsnex_flask_socket.py 

• Run static server to serve the image and the final built output main.html: (port 8001)

cd idr-server
python -m http.server 8001

# goto lo

Demo workflow in DEV envirionment:

Locate into the root dir of the project

pwd # /home/vmvu/WorkspaceSync/IDR

• Run static server to serve the image: (port 8888)

cd idr-server/ # TODO re-check
python -m http.server 8888

• Run python socket server (internal port 5000)

cd idr-server/tsnex
python tsnex_flask_socket.py 

• Run client ELM app (web port 8000)

cd idr-elm
elm-reactor

Installation notes

1. Install elm and client packages

• Install elm
• Install used packages

cd idr-elm
elm package install --yes

# run elm-reactor in dev mode with auto reload
elm-reactor

=> go to http://localhost:8000/src/Main.elm

2. [Optional] Build client without debug

# compile elm to javascript and pack all in a sigle html file
cd idr-elm
elm-make src/Main.elm --output=main.html

# run a simple http server to serve main html file and assert
python -m http.server 8000

=> go to http://localhost:8000/main.html

3. Redis server:

   • Install, run service:

     sudo apt-get install redis-server
     sudo systemctl restart redis-server.service
     sudo systemctl enable redis-server.service # enable on system boot
     

   • Python lib: Redis-py

     pip install redis

   • Monitor:

     • MONITOR command.
     • Redis-stat
       • Run: java -jar redis-stat-0.4.14.jar --server
       • Host: http://localhost:63790/

4. Python websocker server:

   • Flask sockets

   • Money patching to: sklearn.manifold.t_sne._gradient_descent function.

   • Run server:

   cd idr-server/tsnex
   python tsnex_flash_socket.py
   

Algorithm

• Notes change in objective function of t-SNE