IDEAL_NPU

A Python module for machine learning

install

$ pip install IDEAL_NPU

PCN: A Portable clustering algorithm based on Compact Neighbors

A Python implementation of "A Portable Clustering Algorithm Based on Compact Neighbors for Face Tagging".

usage

import numpy as np
from sklearn.metrics.pairwise import euclidean_distances as EuDist2
from sklearn.metrics.cluster import fowlkes_mallows_score as fmi_f

from IDEAL_NPU import Funs
from IDEAL_NPU.cluster import PCN

# Data preparation
X, y_true, N, dim, c_true = Funs.load_Agg()
D_full = EuDist2(X, X, squared=True)
NN_full = np.argsort(D_full, axis=1)
knn = 33
NN = NN_full[:, 1:(knn+1)]
NND = Funs.matrix_index_take(D_full, NN)
for i in range(N):
    tmp_ind = np.lexsort((NN[i, :], NND[i, :]))
    NN[i, :] = NN[i, tmp_ind]

# Clustering
PCN_obj = PCN(NN, NND)
y_pred = PCN_obj.cluster()

# Metrics
pre = Funs.precision(y_true=y_true, y_pred=y_pred)
rec = Funs.recall(y_true=y_true, y_pred=y_pred)
f1 = 2 * pre * rec / (pre + rec)
fmi = fmi_f(y_true, y_pred)

print("{}".format(pre))
print("{}".format(f1))
print("{}".format(fmi))

EDG: An Efficient Density-based clustering incorporated with Graph partitioning

A Python implementation of "An Efficient Density-based Clustering Algorithm for Face Identification".

usage

import numpy as np
from sklearn.metrics.pairwise import euclidean_distances as EuDist2
from sklearn.metrics.cluster import fowlkes_mallows_score as fmi_f

from IDEAL_NPU import Funs
from IDEAL_NPU.cluster import EDG

# Data preparation
X, y_true, N, dim, c_true = Funs.load_Agg()
D_full = EuDist2(X, X, squared=True)
NN_full = np.argsort(D_full, axis=1)

# Clustering
knn_list = [5, 6, 7, 8, 9, 10]
Y = np.zeros((len(knn_list), N))
for i, knn in enumerate(knn_list):
    NN = NN_full[:, 1:(knn+1)]
    NND = Funs.matrix_index_take(D_full, NN)

    EDG_obj = EDG(NN, NND)
    y = EDG_obj.cluster()
    Y[i, :] = y

# Metrics
pre = np.array([Funs.precision(y_true=y_true, y_pred=y_pred) for y_pred in Y])
rec = np.array([Funs.recall(y_true=y_true, y_pred=y_pred) for y_pred in Y])
f1 = 2 * pre * rec / (pre + rec)
ind = np.argmax(f1)
fmi = fmi_f(y_true, Y[ind, :])

print("{}".format(pre[ind]))
print("{}".format(f1[ind]))
print("{}".format(fmi))

Contact

If you have any inquiries, please email me directly (shenfeipei@gmail.com).

License

MIT