Prof. Henry Kautz at University of Rochester's Research Group Repository
This is the code for the web server used to visualize TwitterHealth. This includes both a backend server used to collect and classify tweets, and a frontend for serving pages. Readme
Here is the code used to train an SVM by using N-grams from already labeled data: Readme
If you have a social graph you would like to visualize : Readme
How to configure and then autoscale servers on AWS Readme
Ruby code for (1) Downloading an entire follows graph and (2) using Twitter's REST and Streaming APIs Readme
-
There are two database files missing from this repository because of github's policy changes. I will add links to them ASAP.
-
Each folder in this repository should be a standalone application for now. In the future, each project should be self contained in its own repo, and submoduled here.
-
Each repository should have it's own README with instructions.
-
The goal for each project is to have 3 commands to build a working environment.
configure # should look through any system and find all dependencies for each machine
make # should generate appropriate makefile
make install # should compile and install all components necessary
make run # should start all the services necessary for the given software.
Render heatmap of twitter flights:
cd ./heatmap-flights
python heatmap.py -p twitter_1M.coors.txt -r 5 -o NYC_z14_1M_blue.png -G gradient-adam-blue.png --osm -e 40.8393,-74.2635,40.5,-73.6558 -z 14
Compile gmtk-2012-12-07_2148 on MAC mountain lion x64
# on MacOSX
./configure CFLAGS=-m64 CXXFLAGS=-m64 LDFLAGS=-m64 && make && make install
# on linux, just
./configure prefix=.../bin && make && make install
to compile with DBN structure visualization (gmtkViz), run this first (throws compile errors though):
# you'll need to have brew installed first (http://brew.sh/)
brew install wxmac --devel
which wx-config # To check the version.