Question

for k in pref_n_cluster.pred_numb:
    af = AffinityPropagation(preference=k, affinity='precomputed').fit(X)
    labels = af.labels_
    n_clusters = len(np.unique(labels))
    score = silhouette_score(frechet, labels)

    print("Preference: {0}, cluster: {2}, Silhouette score: {1}".format(k,score,n_clusters))

是否可以将其保存在DataFrame中？我知道，与此相关的问题很多，但所有这些问题都有一个属于该词典的变量。但我不。就我而言，它可以与print()

一起使用

我尝试过：

place = []
...
place = k,score,n_clusters

但这没用

Answer 1

创建元组列表并传递给DataFrame cosntructor：

L = []
for k in pref_n_cluster.pred_numb:
    af = AffinityPropagation(preference=k, affinity='precomputed').fit(X)
    labels = af.labels_
    n_clusters = len(np.unique(labels))
    score = silhouette_score(frechet, labels)
    L.append((k,score,n_clusters))

df = pd.DataFrame(L, columns = ['k','score','n_clusters'])

另一个想法是创建词典列表：

L1 = []
for k in pref_n_cluster.pred_numb:
    af = AffinityPropagation(preference=k, affinity='precomputed').fit(X)
    labels = af.labels_
    n_clusters = len(np.unique(labels))
    score = silhouette_score(frechet, labels)
    L.append({'k':k,'score':score,'n_clusters':n_clusters})

df = pd.DataFrame(L1)

将输出从字典存储到数据框

1 个答案: