class AP(object):
def __init__(self, damping=.5, max_iter=200, convergence_iter=15,
copy=True, preference=None, affinity='euclidean',
verbose=False):
"""
:param damping:
:param max_iter:
:param convergence_iter:
:param copy:
:param preference:
:param affinity:
:param verbose:
"""
self.model = AffinityPropagation(damping=damping,
max_iter=max_iter,
convergence_iter=convergence_iter,
copy=copy,
preference=preference,
affinity=affinity,
verbose=verbose)
def fit(self, x, y=None):
self.model.fit(X=x, y=y)
def fit_predict(self, x, y=None):
return self.model.fit_predict(X=x, y=y)
def get_params(self, deep=True):
return self.model.get_params(deep=deep)
def predict(self, x):
return self.model.predict(X=x)
def set_params(self, **params):
self.model.set_params(**params)
def get_attributes(self):
cluster_centers = self.model.cluster_centers_
cluster_centers_indices = self.model.cluster_centers_indices_
labels = self.model.labels_
affinity_matrix = self.model.affinity_matrix_
n_iter = self.model.n_iter_
return cluster_centers, cluster_centers_indices, labels, affinity_matrix, n_iter
print "Calculating correlation matrix between gene sets... This will be used as similarity matrix for Affinity Propagation"
#matrix_corr = np.corrcoef(df_reconstituted_genesets, rowvar=0) # NUMPY approach | slighty faster than Pandas
matrix_corr = df_reconstituted_genesets.corr(
method='pearson', min_periods=df_reconstituted_genesets.shape[0]
) # PANDAS approach | Note that "min_periods=df_reconstituted_genesets_t.shape[0]" should not be needed, but it ensures that no "NaN" values are present to give weird results
print "Dimension of correlation matrix: [{} x {}]".format(
matrix_corr.shape[0], matrix_corr.shape[1])
################## Running AP ##################
#sklearn.cluster.AffinityPropagation(damping=0.5, max_iter=200, convergence_iter=15, copy=True, preference=None, affinity='euclidean', verbose=False)
af_obj = AffinityPropagation(
affinity='precomputed', max_iter=10000,
convergence_iter=1000) # using almost only default parameters
print "Affinity Propagation parameters:"
for param, val in af_obj.get_params().items():
print "\t{}: {}".format(param, val)
print "Perfoming Affinity Propagation.."
af = af_obj.fit(matrix_corr)
n_iter = af.n_iter_
print "Affinity Propagation done"
print "Number of iterations used: {}".format(n_iter)
### Saving labels and centers
cluster_centers_indices = af.cluster_centers_indices_ # array, shape (n_clusters, n_features) | cluster center (boolean)s ("exemplars")
# cluster_centers_indices take on values in the range {0...n_samples-1}
labels = af.labels_ # array, shape (n_samples,) | Get the "labels"/assignments of each data point to a cluster index
# labels take on values in the range: {0...n_clusters-1}
### Display some stats
n_clusters = len(cluster_centers_indices)
# http://pandas.pydata.org/pandas-docs/stable/computation.html#correlation
# df.corr() # Compute pairwise correlation of columns, excluding NA/null values
#matrix_corr = df_reconstituted_genesets.corr(method='pearson', min_periods=df_reconstituted_genesets.shape[0]) # --> Note that "min_periods=df_reconstituted_genesets_t.shape[0]" should not be needed, but it ensures that no "NaN" values are present to give weird results
### METHOD IN USE
print "Calculating correlation matrix between gene sets... This will be used as similarity matrix for Affinity Propagation"
#matrix_corr = np.corrcoef(df_reconstituted_genesets, rowvar=0) # NUMPY approach | slighty faster than Pandas
matrix_corr = df_reconstituted_genesets.corr(method='pearson', min_periods=df_reconstituted_genesets.shape[0]) # PANDAS approach | Note that "min_periods=df_reconstituted_genesets_t.shape[0]" should not be needed, but it ensures that no "NaN" values are present to give weird results
print "Dimension of correlation matrix: [{} x {}]".format(matrix_corr.shape[0], matrix_corr.shape[1])
################## Running AP ##################
#sklearn.cluster.AffinityPropagation(damping=0.5, max_iter=200, convergence_iter=15, copy=True, preference=None, affinity='euclidean', verbose=False)
af_obj = AffinityPropagation(affinity = 'precomputed', max_iter=10000, convergence_iter=1000) # using almost only default parameters
print "Affinity Propagation parameters:"
for param, val in af_obj.get_params().items():
print "\t{}: {}".format(param, val)
print "Perfoming Affinity Propagation.."
af = af_obj.fit(matrix_corr)
n_iter = af.n_iter_
print "Affinity Propagation done"
print "Number of iterations used: {}".format(n_iter)
### Saving labels and centers
cluster_centers_indices = af.cluster_centers_indices_ # array, shape (n_clusters, n_features) | cluster center (boolean)s ("exemplars")
# cluster_centers_indices take on values in the range {0...n_samples-1}
labels = af.labels_ # array, shape (n_samples,) | Get the "labels"/assignments of each data point to a cluster index
# labels take on values in the range: {0...n_clusters-1}
class AP(object):
def __init__(self,
damping=.5,
max_iter=200,
convergence_iter=15,
copy=True,
preference=None,
affinity='euclidean',
verbose=False,
random_state='warn'):
"""
Parameters
----------
damping : TYPE, optional
阻尼系数 0.5~1 之间
DESCRIPTION. The default is .5.
max_iter : TYPE, optional
最大迭代次数
DESCRIPTION. The default is 200.
convergence_iter : TYPE, optional
停止收敛的估计簇数没有变化的迭代数
DESCRIPTION. The default is 15.
copy : TYPE, optional
复制输入数据 True
DESCRIPTION. The default is True.
preference : TYPE, optional
DESCRIPTION. The default is None.
affinity : TYPE, optional
{"euclidean","precomputed"}
欧氏距离 与与计算
DESCRIPTION. The default is 'euclidean'.
verbose : TYPE, optional
DESCRIPTION. The default is False.
random_state : TYPE, optional
DESCRIPTION. The default is 'warn'.
Returns
-------
None.
"""
self.ap_cluster = AffinityPropagation(
damping=damping,
max_iter=max_iter,
convergence_iter=convergence_iter,
copy=copy,
preference=preference,
affinity=affinity,
verbose=verbose,
random_state=random_state)
def fit(self, x, y=None):
self.ap_cluster.fit(X=x, y=y)
def fit_predict(self, x, y=None):
return self.ap_cluster.fit_predict(X=x, y=y)
def get_params(self, deep=True):
return self.ap_cluster.get_params(deep=deep)
def set_params(self, params):
self.ap_cluster.set_params(**params)
def predict(self, x):
return self.ap_cluster.predict(X=x)
def get_cluster_centers_indices(self):
return self.ap_cluster.cluster_centers_indices_
def get_cluster_centers(self):
return self.ap_cluster.cluster_centers_
def get_labels(self):
return self.ap_cluster.labels_
def get_affinity_matrix(self):
return self.ap_cluster.affinity_matrix_
def get_n_iter(self):
return self.ap_cluster.n_iter_
x = m83_data['x'][final_data]
y = m83_data['y'][final_data]
id_ = m83_data['id_'][final_data]
X = np.vstack([colour1, colour2, colour3, colour4, colour5, colour6]).T #, colour4, colour5, colour6, colour7]).T
similarity = pairwise_distances(X)
##############################################################################
# Compute Affinity Propagation
pref = -len(X)*0.1
damp = 0.95
af = AffinityPropagation(preference=pref, damping=damp).fit(similarity)
cluster_centers_indices = af.cluster_centers_indices_
labels = af.labels_
print cluster_centers_indices
print af.get_params()
n_clusters_ = len(cluster_centers_indices)
print('Estimated number of clusters: %d' % n_clusters_)
print "objects: {}".format(len(colour1))
print("Silhouette Coefficient: %0.3f"
% metrics.silhouette_score(X, labels))
##############################################################################
# Plot result
import matplotlib.pyplot as plt
from itertools import cycle
#plt.close('all')
fig = plt.figure()
ax = fig.add_subplot(111)
[row['Plataforma'], row['Genero']], clf), axis=1)
teste = teste.drop_duplicates()
pd.DataFrame(data=teste, columns=cols).to_csv(
r'dados/teste_final.csv', sep=',', index=False)
timeEnd = time.strftime("%H:%M:%S")
# Validando modelo
pureza = getPureza(teste)
entropia = getEntropia(teste)
# Calculando resultados
colunas = ['Versão', 'Pureza', 'Entropia', 'Começo',
'Término', 'Qtde Clusters', 'Parâmetros do cluster']
qtdeCluster = teste['Agrupamento'].nunique()
result = pd.DataFrame(data=[], columns=colunas)
result = result.append(pd.DataFrame(
data=[[p, pureza, entropia, timeStart, timeEnd, qtdeCluster, str(clf.get_params())]], columns=colunas))
# avaliacao = pd.DataFrame(data=[], columns={'Escolhas de jogos', 'Recomendações', 'Curtida'})
print("Término do processamento: ", time.strftime("%H:%M:%S"))
# Rotas
# Info da api
@app.route('/api/probe', methods=['GET'])
def probe():
return result.to_json(orient='records'), 200
# Busca de jogos
@app.route('/search/<name>', methods=['GET'])
def search(name):
lista = []
for i, row in teste[teste['Nome'].str.contains(name, case=False)].iterrows():
x = m83_data['x'][final_data]
y = m83_data['y'][final_data]
id_ = m83_data['id_'][final_data]
X = np.vstack([colour1, colour2, colour3, colour4, colour5,
colour6]).T #, colour4, colour5, colour6, colour7]).T
similarity = pairwise_distances(X)
##############################################################################
# Compute Affinity Propagation
pref = -len(X) * 0.1
damp = 0.95
af = AffinityPropagation(preference=pref, damping=damp).fit(similarity)
cluster_centers_indices = af.cluster_centers_indices_
labels = af.labels_
print cluster_centers_indices
print af.get_params()
n_clusters_ = len(cluster_centers_indices)
print('Estimated number of clusters: %d' % n_clusters_)
print "objects: {}".format(len(colour1))
print("Silhouette Coefficient: %0.3f" % metrics.silhouette_score(X, labels))
##############################################################################
# Plot result
import matplotlib.pyplot as plt
from itertools import cycle
#plt.close('all')
fig = plt.figure()
ax = fig.add_subplot(111)
