def clustering(entities, model, ratio, seed, verbose):
# Cluster embeddings (on the original space)
n_entities = len(entities)
if verbose:
print('Considering ' + str(n_entities) +
' entities from triple file...')
n_clusters = math.ceil(n_entities * ratio / 100)
if verbose:
print('Clustering with n_clusters = ' + str(n_clusters))
clustering_algorithm = KMeans(n_clusters=n_clusters,
n_init=50,
max_iter=500,
random_state=seed)
clusters = find_clusters(entities,
model,
clustering_algorithm,
mode='entity')
#pickle_cluster = "./temp/cluster_"+ str(ratio) +".pickle"
#with open(pickle_cluster, "wb") as fp: #Pickling
# pickle.dump(clusters, fp)
return clusters
def clustering(entities, model, rate, verbose):
# Cluster embeddings (on the original space)
n_entities = len(entities)
if verbose:
print('Considering ' + str(n_entities) + ' entities from triple file...')
n_clusters = math.ceil(n_entities*rate/100)
if verbose:
print('Clustering with n_clusters = '+str(n_clusters))
clustering_algorithm = KMeans(n_clusters=n_clusters, n_init=10, max_iter=300, random_state=0)
clusters = find_clusters(entities, model, clustering_algorithm, mode='entity')
return clusters
# Get the teams entities and their corresponding embeddings
triples_df = pd.DataFrame(X, columns=['s', 'p', 'o'])
teams = triples_df.s[triples_df.s.str.startswith('Team')].unique()
team_embeddings = dict(zip(teams, model.get_embeddings(teams)))
team_embeddings_array = np.array([i for i in team_embeddings.values()])
# Project embeddings into 2D space via PCA
embeddings_2d = PCA(n_components=2).fit_transform(team_embeddings_array)
# Cluster embeddings (on the original space)
clustering_algorithm = KMeans(n_clusters=6,
n_init=100,
max_iter=500,
random_state=0)
clusters = find_clusters(teams, model, clustering_algorithm, mode='entity')
# This function maps country to continent
def cn_to_ctn(country):
try:
original_name = ' '.join(re.findall('[A-Z][^A-Z]*', country[4:]))
return transformations.cn_to_ctn(original_name)
except KeyError:
return "unk"
plot_df = pd.DataFrame({
"teams": teams,
"embedding1": embeddings_2d[:, 0],
"embedding2": embeddings_2d[:, 1],
def clustering(entities, model):
# Cluster embeddings (on the original space)
n_entities = len(entities)
print('Considering ' + str(n_entities) + ' entities from triple file...')
# 25%
n_clusters = math.ceil(n_entities * 0.25)
print('Clustering with n_clusters = ' + str(n_clusters))
clustering_algorithm = KMeans(n_clusters=n_clusters,
n_init=10,
max_iter=300,
random_state=0)
cluster25 = find_clusters(entities,
model,
clustering_algorithm,
mode='entity')
# DF to File
cluster_df = pd.DataFrame({
"entities":
entities,
"cluster25":
"cluster25" + pd.Series(cluster25).astype(str)
})
print(cluster_df['cluster25'].value_counts())
print(cluster_df['cluster25'].value_counts().value_counts())
cluster_df.to_csv('./temp/cluster25.tsv',
sep='\t',
header=False,
index=False)
# 50%
n_clusters = math.ceil(n_entities * 0.5)
print('Clustering with n_clusters = ' + str(n_clusters))
clustering_algorithm = KMeans(n_clusters=n_clusters,
n_init=10,
max_iter=300,
random_state=0)
cluster50 = find_clusters(entities,
model,
clustering_algorithm,
mode='entity')
# DF to File
cluster_df = pd.DataFrame({
"entities":
entities,
"cluster50":
"cluster50" + pd.Series(cluster50).astype(str)
})
print(cluster_df['cluster50'].value_counts())
print(cluster_df['cluster50'].value_counts().value_counts())
cluster_df.to_csv('./temp/cluster50.tsv',
sep='\t',
header=False,
index=False)
# 75%
n_clusters = math.ceil(n_entities * 0.75)
print('Clustering with n_clusters = ' + str(n_clusters))
clustering_algorithm = KMeans(n_clusters=n_clusters,
n_init=10,
max_iter=300,
random_state=0)
cluster75 = find_clusters(entities,
model,
clustering_algorithm,
mode='entity')
# DF to File
cluster_df = pd.DataFrame({
"entities":
entities,
"cluster75":
"cluster75" + pd.Series(cluster75).astype(str)
})
print(cluster_df['cluster75'].value_counts())
print(cluster_df['cluster75'].value_counts().value_counts())
cluster_df.to_csv('./temp/cluster75.tsv',
sep='\t',
header=False,
index=False)