機械学習　拡散シミュレーション　確率計算

mimu 
NUM = len(df_mem_info.index)
T_NUM = len(df_mem_info.columns) - 1

count_active = 0
count_active_to_inactive = 0
 #消滅確率 
for t in range(T_NUM):
  for i in range(NUM):
    if (df_mem_info.iloc[i][t] == 1):
      count_active_to_inactive += 1
      if (df_mem_info.iloc[i][t + 1] == 0):
        count_active += 1
estimated_percent_disapparence = count_active / count_active_to_inactive

count_link = 0
count_link_to_active = 0
count_link_temp = 0
 #拡散確率 

for t in range(T_NUM):
  df_link_t = df_mem_info[df_mem_info[str(t)] == 1]
  temp_flg_count = np.zeros(NUM)
  for i in range(len(df_link_t.index)):
    index_i = int(df_link_t.index[i].replace('Node', ''))
    df_link_temp = df_mem_links[df_mem_links['Node' + str(index_i)] == 1]
    for j in range(len(df_link_temp.index)):
      index_j = int(df_link_temp.index[j].replace('Node', ''))
      if (df_mem_info.iloc[index_j][t] == 0):
        if (temp_flg_count[index_j] == 0):
          count_link += 1
        if (df_mem_info.iloc[index_j][t + 1] == 1):
          if(temp_flg_count[index_j] == 0):
            temp_flg_count[index_j] = 1
            count_link_to_active += 1
estimated_percent_percolation = count_link_to_active / count_link

拡散の方、なんじゃこりゃ( ﾟДﾟ)ﾊｧ?　
難しい(´;ω;｀)


この記事が気に入ったらサポートをしてみませんか？