commit 90396074f5666f56176e1eac3a7f4260ccdba039 Author: Tom Ritter tom@ritter.vg Date: Tue Jul 5 12:50:26 2016 -0500

Make the graph's y-axis the standard deviation instead of the min/max. --- graphs.py | 38 +++++++++++++++++++++++++++----------- parseOldConsensuses.py | 5 ++++- 2 files changed, 31 insertions(+), 12 deletions(-)

diff --git a/graphs.py b/graphs.py index c45c58e..8b55c15 100755 --- a/graphs.py +++ b/graphs.py @@ -282,6 +282,7 @@ class GraphWriter: return d3_dsv.csvParse(text); }).then(function(data) {

+ // For each of the configured graphs for(g in GRAPHS_TO_GENERATE) { graph = GRAPHS_TO_GENERATE[g]; @@ -291,30 +292,48 @@ class GraphWriter: else data_subset = data

+ // Calculate the Graph Boundaries ----------------------------------------- min = 10000; max = 0; + total = 0; + count = 0; for(d in data_subset) { for(b in BWAUTHS) { data_subset[d][BWAUTHS[b]] = Number(data_subset[d][BWAUTHS[b]]); - if(data_subset[d][BWAUTHS[b]] < min && data_subset[d][BWAUTHS[b]] > BWAUTH_LOGICAL_MIN) { - min = data_subset[d][BWAUTHS[b]]; - } - if(data_subset[d][BWAUTHS[b]] > max) { - max = data_subset[d][BWAUTHS[b]]; - } + var x = data_subset[d][BWAUTHS[b]]; + if(x < min && x > BWAUTH_LOGICAL_MIN) + min = x; + if(x > max) + max = x; + + total += x; + count++; } } - console.log("Data Length: " + data_subset.length + " Y-Axis Min: " + min + " Max: " + max); + avg = total / count; + sumvariance = 0; + for(d in data_subset) + { + for(b in BWAUTHS) + { + var x = data_subset[d][BWAUTHS[b]]; + sumvariance += (x - avg) * (x - avg); + } + } + variance = sumvariance / count; + stddev = Math.sqrt(variance); + console.log("Data Length: " + data_subset.length + " Y-Axis Min: " + min + " Max: " + max + " Avg: " + avg + " Var: " + variance + " StdDev: " + stddev);

+ // Create the Graph ----------------------------------------- var x = d3.scaleTime() .domain([new Date(Number(data_subset[0].date)), new Date(Number(data_subset[data_subset.length-1].date))]) .range([0, WIDTH]) ;

var y = d3.scaleLinear() - .domain([min, max]) + .domain([avg-(stddev), avg+(stddev)]) .range([HEIGHT, 0]);

var lines = [] @@ -401,9 +420,6 @@ if __name__ == '__main__': v = pickle.load(open('votes.p', 'rb')) g.set_votes(v)

- import pdb - pdb.set_trace() - CONFIG = stem.util.conf.config_dict('consensus', { 'ignored_authorities': [], 'bandwidth_authorities': [], diff --git a/parseOldConsensuses.py b/parseOldConsensuses.py index e45fd27..834c386 100755 --- a/parseOldConsensuses.py +++ b/parseOldConsensuses.py @@ -20,6 +20,9 @@ import stem.util.enum from stem import Flag from stem.util.lru_cache import lru_cache

+def get_dirauths_in_tables(): + return "faravahar, gabelmoo, dizum, moria1, urras, maatuska, longclaw, tor26, dannenberg, turtles".split(", ") +

def get_dirauth_from_filename(filename): key = filename.split('-') @@ -64,7 +67,7 @@ def get_time_from_filename(filename): return voteTime

def main(dir): - dirAuths = "faravahar, gabelmoo, dizum, moria1, urras, maatuska, longclaw, tor26, dannenberg, turtles".split(", ") + dirAuths = get_dirauths_in_tables() dbc = sqlite3.connect(os.path.join('data', 'historical.db'))

dirauth_columns = ""