file arrangement 2

feature_model.py

@@ -64,6 +64,12 @@ def make_point_space(im_LR, im_GX, im_GY, im_GZ, patchNumber, w, point_space, MA
  return point_space, patchNumber
 
 
+def init_buckets(Q_TOTAL):
+ patchS = [[] for j in range(C.Q_TOTAL)]
+ xS = [[] for j in range(C.Q_TOTAL)]
+ return patchS, xS
+
+
 def k_means_modeling(quantization):
 
  with open('./arrays/qua', 'rb') as p:

filter_func.py

@@ -34,16 +34,6 @@ def dog_sharpener(input, sigma=0.85, alpha=1.414, r=15, ksize=(3,3,3)):
  return output
 
 
-def add_weight(im1, w1, im2, w2, b):
- return im1 * w1 + im2 * w2 + b
-
-
-def clip_image(im):
- clip_value = np.sort(im.ravel())[int(np.prod(im.shape) * 0.999)]
- im = np.clip(im, 0, clip_value)
- return im
-
-
 @njit(parallel=True)
 def ct_descriptor(im):
  H, W, D = im.shape

test.py

@@ -1,26 +1,22 @@
 import os
-import glob
 import time
 
-import numpy as np
 import cupy as cp
-from numba import jit, njit, prange
-
 import nibabel as nib
+import numpy as np
+
+from numba import jit, njit, prange
+from skimage.metrics import peak_signal_noise_ratio, structural_similarity
 
 import filter_constant as C
 
 from crop_black import *
-from filter_constant import *
+from feature_model import *
 from filter_func import *
-from get_lr import *
+from kmeans_vector import KMeans_Vector
 from matrix_compute import *
-from train import load_kmeans_model, get_features
+from preprocessing import *
 from util import *
-from kmeans_vector import KMeans_Vector
-from feature_model import *
-
-from skimage.metrics import peak_signal_noise_ratio, structural_similarity
 
 
 def make_image(im_LR, im_GX, im_GY, im_GZ, w, kmeans, h):