fix random.shuffle (#1204)

fluid/face_detection/image_util.py

@@ -79,9 +79,9 @@ def bbox_area(src_bbox):
 
 
 def generate_sample(sampler, image_width, image_height):
-    scale = random.uniform(sampler.min_scale, sampler.max_scale)
-    aspect_ratio = random.uniform(sampler.min_aspect_ratio,
-                                  sampler.max_aspect_ratio)
+    scale = np.random.uniform(sampler.min_scale, sampler.max_scale)
+    aspect_ratio = np.random.uniform(sampler.min_aspect_ratio,
+                                     sampler.max_aspect_ratio)
     aspect_ratio = max(aspect_ratio, (scale**2.0))
     aspect_ratio = min(aspect_ratio, 1 / (scale**2.0))
 
@@ -97,8 +97,8 @@ def generate_sample(sampler, image_width, image_height):
 
     xmin_bound = 1 - bbox_width
     ymin_bound = 1 - bbox_height
-    xmin = random.uniform(0, xmin_bound)
-    ymin = random.uniform(0, ymin_bound)
+    xmin = np.random.uniform(0, xmin_bound)
+    ymin = np.random.uniform(0, ymin_bound)
     xmax = xmin + bbox_width
     ymax = ymin + bbox_height
     sampled_bbox = bbox(xmin, ymin, xmax, ymax)
@@ -141,25 +141,29 @@ def data_anchor_sampling(sampler, bbox_labels, image_width, image_height,
         min_resize_val = scale_array[rand_idx_size] / 2.0
         max_resize_val = min(2.0 * scale_array[rand_idx_size],
                              2 * math.sqrt(wid * hei))
-        scale_choose = random.uniform(min_resize_val, max_resize_val)
+        scale_choose = np.random.uniform(min_resize_val, max_resize_val)
         sample_bbox_size = wid * resize_width / scale_choose
 
         w_off_orig = 0.0
         h_off_orig = 0.0
         if sample_bbox_size < max(image_height, image_width):
             if wid <= sample_bbox_size:
-                w_off_orig = random.uniform(xmin + wid - sample_bbox_size, xmin)
+                w_off_orig = np.random.uniform(xmin + wid - sample_bbox_size,
+                                               xmin)
             else:
-                w_off_orig = random.uniform(xmin, xmin + wid - sample_bbox_size)
+                w_off_orig = np.random.uniform(xmin,
+                                               xmin + wid - sample_bbox_size)
 
             if hei <= sample_bbox_size:
-                h_off_orig = random.uniform(ymin + hei - sample_bbox_size, ymin)
+                h_off_orig = np.random.uniform(ymin + hei - sample_bbox_size,
+                                               ymin)
             else:
-                h_off_orig = random.uniform(ymin, ymin + hei - sample_bbox_size)
+                h_off_orig = np.random.uniform(ymin,
+                                               ymin + hei - sample_bbox_size)
 
         else:
-            w_off_orig = random.uniform(image_width - sample_bbox_size, 0.0)
-            h_off_orig = random.uniform(image_height - sample_bbox_size, 0.0)
+            w_off_orig = np.random.uniform(image_width - sample_bbox_size, 0.0)
+            h_off_orig = np.random.uniform(image_height - sample_bbox_size, 0.0)
 
         w_off_orig = math.floor(w_off_orig)
         h_off_orig = math.floor(h_off_orig)
@@ -418,36 +422,36 @@ def crop_image_sampling(img, bbox_labels, sample_bbox, image_width,
 
 
 def random_brightness(img, settings):
-    prob = random.uniform(0, 1)
+    prob = np.random.uniform(0, 1)
     if prob < settings.brightness_prob:
-        delta = random.uniform(-settings.brightness_delta,
-                               settings.brightness_delta) + 1
+        delta = np.random.uniform(-settings.brightness_delta,
+                                  settings.brightness_delta) + 1
         img = ImageEnhance.Brightness(img).enhance(delta)
     return img
 
 
 def random_contrast(img, settings):
-    prob = random.uniform(0, 1)
+    prob = np.random.uniform(0, 1)
     if prob < settings.contrast_prob:
-        delta = random.uniform(-settings.contrast_delta,
-                               settings.contrast_delta) + 1
+        delta = np.random.uniform(-settings.contrast_delta,
+                                  settings.contrast_delta) + 1
         img = ImageEnhance.Contrast(img).enhance(delta)
     return img
 
 
 def random_saturation(img, settings):
-    prob = random.uniform(0, 1)
+    prob = np.random.uniform(0, 1)
     if prob < settings.saturation_prob:
-        delta = random.uniform(-settings.saturation_delta,
-                               settings.saturation_delta) + 1
+        delta = np.random.uniform(-settings.saturation_delta,
+                                  settings.saturation_delta) + 1
         img = ImageEnhance.Color(img).enhance(delta)
     return img
 
 
 def random_hue(img, settings):
-    prob = random.uniform(0, 1)
+    prob = np.random.uniform(0, 1)
     if prob < settings.hue_prob:
-        delta = random.uniform(-settings.hue_delta, settings.hue_delta)
+        delta = np.random.uniform(-settings.hue_delta, settings.hue_delta)
         img_hsv = np.array(img.convert('HSV'))
         img_hsv[:, :, 0] = img_hsv[:, :, 0] + delta
         img = Image.fromarray(img_hsv, mode='HSV').convert('RGB')
@@ -455,7 +459,7 @@ def random_hue(img, settings):
 
 
 def distort_image(img, settings):
-    prob = random.uniform(0, 1)
+    prob = np.random.uniform(0, 1)
     # Apply different distort order
     if prob > 0.5:
         img = random_brightness(img, settings)
@@ -471,14 +475,14 @@ def distort_image(img, settings):
 
 
 def expand_image(img, bbox_labels, img_width, img_height, settings):
-    prob = random.uniform(0, 1)
+    prob = np.random.uniform(0, 1)
     if prob < settings.expand_prob:
         if settings.expand_max_ratio - 1 >= 0.01:
-            expand_ratio = random.uniform(1, settings.expand_max_ratio)
+            expand_ratio = np.random.uniform(1, settings.expand_max_ratio)
             height = int(img_height * expand_ratio)
             width = int(img_width * expand_ratio)
-            h_off = math.floor(random.uniform(0, height - img_height))
-            w_off = math.floor(random.uniform(0, width - img_width))
+            h_off = math.floor(np.random.uniform(0, height - img_height))
+            w_off = math.floor(np.random.uniform(0, width - img_width))
             expand_bbox = bbox(-w_off / img_width, -h_off / img_height,
                                (width - w_off) / img_width,
                                (height - h_off) / img_height)

fluid/face_detection/reader.py

@@ -18,7 +18,6 @@ from __future__ import print_function
 
 import image_util
 from paddle.utils.image_util import *
-import random
 from PIL import Image
 from PIL import ImageDraw
 import numpy as np
@@ -98,7 +97,7 @@ def preprocess(img, bbox_labels, mode, settings, image_path):
         # sampling
         batch_sampler = []
 
-        prob = random.uniform(0., 1.)
+        prob = np.random.uniform(0., 1.)
         if prob > settings.data_anchor_sampling_prob:
             scale_array = np.array([16, 32, 64, 128, 256, 512])
             batch_sampler.append(
@@ -109,7 +108,7 @@ def preprocess(img, bbox_labels, mode, settings, image_path):
                 settings.resize_width, settings.resize_height)
             img = np.array(img)
             if len(sampled_bbox) > 0:
-                idx = int(random.uniform(0, len(sampled_bbox)))
+                idx = int(np.random.uniform(0, len(sampled_bbox)))
                 img, sampled_labels = image_util.crop_image_sampling(
                     img, bbox_labels, sampled_bbox[idx], img_width, img_height,
                     settings.resize_width, settings.resize_height,
@@ -140,7 +139,7 @@ def preprocess(img, bbox_labels, mode, settings, image_path):
 
             img = np.array(img)
             if len(sampled_bbox) > 0:
-                idx = int(random.uniform(0, len(sampled_bbox)))
+                idx = int(np.random.uniform(0, len(sampled_bbox)))
                 img, sampled_labels = image_util.crop_image(
                     img, bbox_labels, sampled_bbox[idx], img_width, img_height,
                     settings.resize_width, settings.resize_height,
@@ -153,7 +152,7 @@ def preprocess(img, bbox_labels, mode, settings, image_path):
     img = np.array(img)
 
     if mode == 'train':
-        mirror = int(random.uniform(0, 2))
+        mirror = int(np.random.uniform(0, 2))
         if mirror == 1:
             img = img[:, ::-1, :]
             for i in six.moves.xrange(len(sampled_labels)):
@@ -225,7 +224,7 @@ def train_generator(settings, file_list, batch_size, shuffle=True):
     file_dict = load_file_list(file_list)
     while True:
         if shuffle:
-            random.shuffle(file_dict)
+            np.random.shuffle(file_dict)
         images, face_boxes, head_boxes, label_ids = [], [], [], []
         label_offs = [0]