K-means 클러스터링

@@ -75,7 +75,7 @@ def mobile_net(input_shape):
         
     x = mobile_net_block(x, 1024, 2)
     x = mobile_net_block(x, 1024)
-    x = cbam_block(x)
+    #x = cbam_block(x)
     
     x = GlobalAvgPool2D()(x)
     
@@ -153,4 +153,4 @@ if __name__ == "__main__":
     # predict sample
     predict(model, X_to_predict, y_to_predict)
     
-    model.save('C:/Users/nokh9/Desktop/mobile_net2.h5')
\ No newline at end of file
+    model.save('C:/Users/nokh9/Desktop/수업/캡스톤디자인2/소스코드/mobile_net3.h5')
\ No newline at end of file

+"""
+    VGG16 모델과 K-Means Clustering을 이용하여 소리에 대한 데이터를 군집화 후 분
+"""
+from keras.preprocessing import image
+from keras.applications.vgg16 import VGG16
+from keras.applications.vgg16 import preprocess_input
+import numpy as np
+from sklearn.cluster import KMeans
+import os, shutil, glob, os.path
+from PIL import Image as pil_image
+
+image.LOAD_TRUNCATED_IMAGES = True 
+model = VGG16(weights='imagenet', include_top=False)
+
+imdir = 'PATH WHERE YOUR FILE IS'
+targetdir = 'PATH WHERE YOUT FILE SAVE'
+number_clusters = 2
+
+# Loop over files and get features
+filelist = glob.glob(os.path.join(imdir, '*.png'))
+#filelist.sort()
+featurelist = []
+
+for i, imagepath in enumerate(filelist):
+    print("    Status: %s / %s" %(i, len(filelist)), end="\r")
+    img = image.load_img(imagepath, target_size=(370, 470))
+    img_data = image.img_to_array(img)
+    img_data = np.expand_dims(img_data, axis=0)
+    img_data = preprocess_input(img_data)
+    features = np.array(model.predict(img_data))
+    featurelist.append(features.flatten())
+
+    if i % 10:
+      print(i)
+
+# K-Means 클러스터링 
+kmeans = KMeans(n_clusters=number_clusters, random_state=0).fit(np.array(featurelist))
+
+# Copy with cluster name
+print("\n")
+for i, m in enumerate(kmeans.labels_):
+    print("    Copy: %s / %s" %(i, len(kmeans.labels_)), end="\r")
+    shutil.copy(filelist[i], targetdir + str(m) + "_" + str(i) + ".jpg")
+
+    if(i % 100) == 0:
+      print(str(i)+'is finished')
\ No newline at end of file