Commits

Author Commit Message Labels Comments Date
rgrp
[hgignore][s]: add an hgignore file.
rgrp
[econ/patterns/nber][m]: work over last 2 days to produced "animations" showing technology flows over time (based on original work from 6 months ago). * nber/analyze.py: improvement to ui * new make_animations method (to make gifs from static flows_subcat_* * support for plots of cumulative flows * improve option management and filenaming for FlowPlotter (so now easy to make lots of different plots without files being overwritten) * introduce optparse in __main__ * nber/docs/flows.txt: full write up (what was posted on rufusp…
rgrp
[econ/patterns][l]: major overhaul/extensions of 'proposal' (should be renamed ...). * Rewrite almost all sections a bit * Turn examples from stubs into proper sections * analyze.py: generate a latex table of subcats and do a few other minor fixes.
rgrp
[econ/patterns/nber][s]: (old: last week) minor improvements to plotting to prepare for posting online.
rgrp
[econ/patterns/nber][m]: implement pca analysis using subcat areas of cited patents as location vectors.
rgrp
[econ/patterns/nber][m]: (old: 1m+) code to convert citation graphs to matrices for pca processing (did not really work ...).
rgrp
[econ/patterns/nber][s]: fixes to convert_to_db.py to make it better/more robust resulting from loading again on new laptop last week.
rgrp
[econ/patterns/nber][m]: flow graphs done including 'animation' over time.
rgrp
[econ/patterns/nber][m]: major speedups in extracting flows data.
rgrp
[econ/patterns/nber][s]: (addendum to r1431) missed out test unit from previous commit.
rgrp
[econ/patterns/nber][m]: work to extract flow stats. Did it the simple way but too slow. About to move to quicker way (and will remove old slow way) and therefore committing now.
rgrp
[econ/patterns/nber][m]: work from couple of days ago to do more analysis of patent data (getting ready for graph stuff).
rgrp
[econ/pattern/nber][s]: (addendum to prev. commit r1427) missed out rename of test file.
rgrp
[econ/pattern/nber][m]: create a new analyze.py to extract data from patent db + various other updates. * nber/nxdrawdemo.py: demo drawing graphs using networkx/graphviz * nber/analyse.py -> nber/dirichlet.py (rename to avoid confusion with analyze.py) * nber/analyze.py: new file to do db extraction * nber/convert_to_db.py: add in Patent and Citation object * nber/networkx_tryout.py: graph subcat 13 using graphviz algorithms ...
rgrp
[econ/patterns/nber][m]: got script to load patent data into db fully functional (+ converted to sqlalchemy).
rgrp
[econ/patterns/nber][s]: fix up analyse code to use mdp and pass all tests. (NB: plenty more work to understand, and use, mdp properly).
rgrp
[econ/patterns][m]: mainly work on data analysis scripts to start investigating degree distributions again. Also: * Quite a few more items to list in the bib file * Some minor additions to sources.txt
rgrp
[econ/patterns]: (some old) move over (not-in-svn) citation directory and merge in material from README.txt etc into patterns stuff. Citation was almost exactly the same and contained in terms of data the citec material. This has now been moved to ./citec within ./patterns (though not yet added to svn).
rgrp
[econ/patterns][m]: add in some more existing files (sources.txt, ideas.txt) plus a new main README.txt and a couple of fixes.
rgrp
[econ/patterns][s]: move all nber patent stuff out of trunk and into trunk/nber.