Phil 6.22.17

Research 6:45 – 7:45

• Thinking that the emergent agent properties might map to political theories. This is the best overview I’ve found: Political Theory/Ideologies of Government
• From Mindless Masses to Small Groups: Conceptualizing Collective Behavior in Crowd Modeling
• Self-organizing pedestrian movement
• Continuing writing. Getting close to a full outline

BRI 8:30 – 3:30

• Working on reading in files, then running permutations on cluster membership
• Because I can never remember how ass-backwards pandas.Dataframes are:

  def initDictSeries(rows = 3, cols = 5, offset=1, prefix ="doc_"):
      dict = {}
      row_names = []
      for j in range(cols):
          row_names.append("row_{0}".format(j))
      for i in range(rows):
          name = prefix+'{0}'.format(i)
          array = []
          for j in range(cols):
              array.append ((i+offset)*10 + j)
          #dict[name] = tf.Variable(np.random.rand(cols), tf.float32)
          dict[name] = array
      return pd.DataFrame(dict, index=row_names)
  
  df = initDictSeries()
  print("df = \n{0}".format(df))
  s = df.loc['row_1']
  print("\ns = df.loc['row_1'] = \n{0}".format(s))
  s['doc_1'] = 99
  print("modified = \n{0}".format(df))

• Yay! It looks like it’s working. Next step is to run it. Done. This is the compare of the two runs on the older data as it sits on the servers: 
• Here’s the same runs locally, using t-sne. This is more like what I was expecting to see: 
• Here’s the old vs. new (Left this cooking. These charts take forever to calculate)