#Importieren Sie die erforderlichen Bibliotheken
import keras
from keras.datasets import mnist
import numpy as np
import pandas as pd
import sklearn
#Zeigen Sie Plotergebnisse im Notizbuch an, wenn Sie das Jupyter-Notizbuch verwenden
import matplotlib.pyplot as plt
%matplotlib inline
Using TensorFlow backend.
feature_dims = range(2, 12)
#Lesen Sie Daten mit der Keras-Funktion. Mische Daten und teile sie in Trainingsdaten und Trainingsdaten auf
(x_train, y_train), (x_test, y_test) = mnist.load_data()
#Konvertieren Sie 2D-Daten in numerische Werte
x_train = x_train.reshape(60000, 784)
x_test = x_test.reshape(10000, 784)
#Typkonvertierung
x_train = x_train.astype('float32')
x_test = x_test.astype('float32')
#Teilen Sie durch 255 als neue Variable
x_train /= 255
x_test /= 255
# one-Methode zur Hot-Codierung
from keras.utils.np_utils import to_categorical
#10 Klassen
num_classes = 10
y_train = y_train.astype('int32')
y_test = y_test.astype('int32')
labels = y_test
# one-hot encoding
y_train = to_categorical(y_train, num_classes)
y_test = to_categorical(y_test, num_classes)
def fitting(feature_dim, x_train, y_train, x_test, y_test):
#Verwenden Sie Adam zum Importieren und Optimieren der erforderlichen Bibliotheken
from keras.models import Sequential
from keras.layers import Dense, Dropout
from keras.optimizers import Adam
import gc
#Modellieren
model = Sequential()
model.add(Dense(512, activation='relu', input_shape=(784,)))
model.add(Dropout(0.2))
model.add(Dense(512, activation='relu'))
model.add(Dropout(0.2))
model.add(Dense(feature_dim, activation='relu')) #Es wurde eine Ebene hinzugefügt, die als Feature-Menge herausgenommen werden soll
model.add(Dense(10, activation='softmax'))
model.summary()
#Chargengröße, Anzahl der Epochen
batch_size = 128
epochs = 20
model.compile(loss='categorical_crossentropy',
optimizer=Adam(),
metrics=['accuracy'])
history = model.fit(x_train, y_train,
batch_size=batch_size,
epochs=epochs,
verbose=1,
validation_data=(x_test, y_test))
score = model.evaluate(x_test, y_test, verbose=0)
print('Test loss:', score[0])
print('Test accuracy:', score[1])
# #Accuracy
# print(history.history.keys())
# plt.plot(history.history['accuracy'])
# plt.plot(history.history['val_accuracy'])
# plt.title('model accuracy')
# plt.ylabel('accuracy')
# plt.xlabel('epoch')
# plt.legend(['train', 'test'], loc='upper left')
# plt.show()
# #loss
# plt.plot(history.history['loss'])
# plt.plot(history.history['val_loss'])
# plt.title('model loss')
# plt.ylabel('loss')
# plt.xlabel('epoch')
# plt.legend(['train', 'test'], loc='upper left')
# plt.show()
model.pop() #Entfernen Sie die Softmax-Ebene in der letzten Phase und verwenden Sie die Feature-Ebene als letzte Phase.
model.summary()
result = model.predict(x_test)
keras.backend.clear_session() #← Das ist
gc.collect()
from IPython.display import clear_output
clear_output()
return (history, model, result)
#model = fitting(10, x_train, y_train, x_test, y_test)
models = [None] * len(feature_dims)
histories = [None] * len(feature_dims)
results = [None] * len(feature_dims)
for i in range(len(feature_dims)):
(histories[i], models[i], results[i]) = fitting(feature_dims[i], x_train, y_train, x_test, y_test)
#model.save('model/mnist-10')
#model = keras.models.load_model('model/mnist-10')
#for i in range(len(feature_dims)):
# models[i].pop() #Entfernen Sie die Softmax-Ebene in der letzten Phase und verwenden Sie die Feature-Ebene als letzte Phase.
# models[i].summary()
#result = model.predict(x_test)
#results = [None] * len(feature_dims)
#for i in range(len(feature_dims)):
# keras.backend.clear_session()
# results[i] = models[i].predict(x_test)
def tsne(result):
#t-Dimensionsreduzierung mit SNE
from sklearn.manifold import TSNE
tsne = TSNE(n_components=2, random_state = 0, perplexity = 30, n_iter = 1000)
return tsne.fit_transform(result)
#tsne = tsne(result)
tsnes = [None] * len(feature_dims)
for i in range(len(feature_dims)):
tsnes[i] = tsne(results[i])
#df = pd.DataFrame(tsne, columns = ['x', 'y'])
#df['label'] = labels
def km(n_clusters, result):
# k-Cluster mittels
from sklearn.cluster import KMeans
return KMeans(n_clusters).fit_predict(result)
#km = km(10, result)
#df['km'] = km
kms = [None] * len(feature_dims)
for i in range(len(feature_dims)):
kms[i] = km(10, results[i])
def DBSCAN(n_clusters, result):
from sklearn.cluster import DBSCAN
db = DBSCAN(eps=0.2, min_samples=n_clusters).fit(result)
return db.labels_
#dbscan = DBSCAN(20, result)
#df['DBSCAN'] = dbscan
def hierarchy(result):
from scipy.cluster.hierarchy import linkage, dendrogram
result1 = linkage(result,
metric = 'braycurtis',
#metric = 'canberra',
#metric = 'chebyshev',
#metric = 'cityblock',
#metric = 'correlation',
#metric = 'cosine',
#metric = 'euclidean',
#metric = 'hamming',
#metric = 'jaccard',
#method= 'single')
method = 'average')
#method= 'complete')
#method='weighted')
return result1
#hierarchy = hierarchy(result)
#display(hierarchy)
#def cluster_visualization(x, y, label, cluster, method, n_clusters):
def cluster_visualization(x, y, label, cluster):
plt.figure(figsize = (30, 15))
plt.subplot(1,2,1)
plt.scatter(x, y, c=label)
# for i in range(10):
# tmp_df = df[df['label'] == i]
# plt.scatter(tmp_df['x'], tmp_df['y'], label=i)
# plt.legend(loc='upper left', bbox_to_anchor=(1,1))
plt.subplot(1,2,2)
plt.scatter(x, y, c=cluster)
# for i in range(n_clusters):
# tmp_df = df[df[method] == i]
# plt.scatter(tmp_df['x'], tmp_df['y'], label=i)
# plt.legend(loc='upper left', bbox_to_anchor=(1,1))
for i in range(len(feature_dims)):
cluster_visualization(tsnes[i][:,0], tsnes[i][:,1], labels, kms[i])
# https://qiita.com/mamika311/items/75c24f6892f85593f7e7
from sklearn.metrics.cluster import adjusted_rand_score
for i in range(len(feature_dims)):
print("dim:" + str(feature_dims[i]) + " RMI: " + str(adjusted_rand_score(labels, kms[i])))
dim:2 RMI: 0.36573507862590254
dim:3 RMI: 0.49974179932107105
dim:4 RMI: 0.6248257814760337
dim:5 RMI: 0.8225287029746797
dim:6 RMI: 0.8495039832620757
dim:7 RMI: 0.8417680081349097
dim:8 RMI: 0.8423268187793562
dim:9 RMI: 0.8450473012143238
dim:10 RMI: 0.836035505993697
dim:11 RMI: 0.8815919206871302
# https://scikit-learn.org/stable/modules/generated/sklearn.metrics.normalized_mutual_info_score.html
# https://qiita.com/kotap15/items/38289edfe822005e1e44
from sklearn.metrics import normalized_mutual_info_score
#display(normalized_mutual_info_score(labels, df['km']))
for i in range(len(feature_dims)):
print("dim:" + str(feature_dims[i]) + " NMI: " + str(normalized_mutual_info_score(labels, kms[i])))
dim:2 NMI: 0.5759443563915843
dim:3 NMI: 0.6735454178249051
dim:4 NMI: 0.7745736983918213
dim:5 NMI: 0.8626814016489588
dim:6 NMI: 0.8759626968874756
dim:7 NMI: 0.8766399602087444
dim:8 NMI: 0.8830520742914061
dim:9 NMI: 0.8706715369843739
dim:10 NMI: 0.8721342625213994
dim:11 NMI: 0.8992713472017846
def shilhouette(clusters, x_test):
from sklearn.metrics import silhouette_samples
from matplotlib import cm
plt.figure(figsize = (10, 10))
cluster_labels=np.unique(clusters)
n_clusters=cluster_labels.shape[0]
silhouette_vals=silhouette_samples(x_test,clusters,metric='euclidean')
y_ax_lower,y_ax_upper=0,0
yticks=[]
for i,c in enumerate(cluster_labels):
c_silhouette_vals=silhouette_vals[clusters==c]
print(len(c_silhouette_vals))
c_silhouette_vals.sort()
y_ax_upper +=len(c_silhouette_vals)
color=cm.jet(float(i)/n_clusters)
plt.barh(range(y_ax_lower,y_ax_upper),
c_silhouette_vals,
height=1.0,
edgecolor='none',
color=color
)
yticks.append((y_ax_lower+y_ax_upper)/2.)
y_ax_lower += len(c_silhouette_vals)
#Wenn der Silhouette-Koeffizient 1 ist, können Sie gut gruppieren.
#Wenn die Breite der Silhouette im Durchschnitt in Bezug auf die Anzahl der Cluster gleich ist, bedeutet dies, dass die gesamten Daten gleichmäßig aufgeteilt werden können.
#Diese Teilungsbreite=Eine mögliche Einstellmethode besteht darin, k so zu optimieren, dass die Breiten der Silhouette-Balken gleich sind und sich der Silhouette-Koeffizient 1 nähert..
#Zeichnen Sie eine Linie an der Durchschnittsposition
silhouette_avg=np.mean(silhouette_vals)
plt.axvline(silhouette_avg,color="red",linestyle="--")
plt.ylabel("Cluster")
plt.xlabel("Silhouette coefficient")
for i in range(len(feature_dims)):
shilhouette(kms[i], x_test)
1228
2862
388
406
760
398
304
1653
1194
807
1350
1053
620
969
1312
1081
1720
544
270
1081
633
753
2140
711
320
786
926
1060
888
1783
1076
842
998
740
938
1022
861
1265
1425
833
855
850
884
858
1007
1118
979
911
1180
1358
1132
860
936
962
923
835
844
1533
974
1001
891
784
1096
892
1040
910
1562
914
904
1007
890
947
1127
710
1018
902
960
902
1178
1366
1189
1002
1538
903
947
918
925
771
927
880
1133
914
923
950
986
890
954
1023
816
1411
Recommended Posts