"""
    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')