更改

添加290字节 、 2024年6月6日 (星期四)
第577行: 第577行:  
==计算EI的源代码==
 
==计算EI的源代码==
   −
这是计算一个马尔科夫概率转移矩阵的Python源代码。输入tpm为一个满足行归一化条件的马尔科夫概率转移矩阵,返回的ei_all为其EI值。
+
这是计算一个马尔科夫概率转移矩阵的Python源代码。输入tpm为一个满足行归一化条件的马尔科夫概率转移矩阵,返回的ei_all为其EI值,其余两项分别为确定性和简并性。
 
python:<syntaxhighlight lang="python3">
 
python:<syntaxhighlight lang="python3">
 
def tpm_ei(tpm, log_base = 2):
 
def tpm_ei(tpm, log_base = 2):
  # marginal distribution of y given x ~ Unifrom Dist
+
    # marginal distribution of y given x ~ Unifrom Dist
  puy = tpm.sum(axis=0)
+
    puy = tpm.sum(axis=0)
  n = tpm.shape[0]
+
    n = tpm.shape[0]
  # replace 0 to a small positive number to avoid log error
+
   
  eps = 1E-10
+
    # replace 0 to a small positive number to avoid log error
  tpm_e = np.where(tpm==0, eps, tpm)
+
    eps = 1E-10
  puy_e = np.where(tpm==0, eps, puy)
+
    tpm_e = np.where(tpm==0, eps, tpm)
 
+
    puy_e = np.where(tpm==0, eps, puy)
  # calculate EI of specific x
+
   
  ei_x = (np.log2(n * tpm_e / puy_e) / np.log2(log_base)  * tpm).sum(axis=1)
+
    # calculate EI of specific x
 
+
    ei_x = (np.log2(n * tpm_e / puy_e) / np.log2(log_base)  * tpm).sum(axis=1)
  # calculate total EI
+
   
  ei_all = ei_x.mean()
+
    # calculate det and deg
  return ei_all
+
    det = np.log2(n) + (tpm * np.log2(tpm_e)).sum(axis=1).mean(axis=0)
 +
    deg = np.log2(n) + (tpm.mean(axis=0) * np.log2(tpm_e.mean(axis=0))).sum()
 +
   
 +
    # calculate total EI
 +
    ei_all = ei_x.mean()
 +
    return ei_all,det/np.log2(log_base),deg/np.log2(log_base)
 
</syntaxhighlight>
 
</syntaxhighlight>
  
1,177

个编辑