Fix flaky test_gluon_data_vision.test_random_transforms

tests/python/unittest/test_gluon_data_vision.py

@@ -42,12 +42,12 @@ def test_to_tensor():
     out_nd = transforms.ToTensor()(nd.array(data_in, dtype='uint8'))
     assert_almost_equal(out_nd.asnumpy(), np.transpose(
                         data_in.astype(dtype=np.float32) / 255.0, (0, 3, 1, 2)))
-    
+
     # Invalid Input
     invalid_data_in = nd.random.uniform(0, 255, (5, 5, 300, 300, 3)).astype(dtype=np.uint8)
     transformer = transforms.ToTensor()
     assertRaises(MXNetError, transformer, invalid_data_in)
-    
+
     # Bounds (0->0, 255->1)
     data_in = np.zeros((10, 20, 3)).astype(dtype=np.uint8)
     out_nd = transforms.ToTensor()(nd.array(data_in, dtype='uint8'))
@@ -126,7 +126,7 @@ def _test_resize_with_diff_type(dtype):
         assertRaises(MXNetError, invalid_transform, data_in)
 
     for dtype in ['uint8', 'float32', 'float64']:
-        _test_resize_with_diff_type(dtype)    
+        _test_resize_with_diff_type(dtype)
 
 
 @with_seed()
@@ -159,7 +159,7 @@ def _test_crop_resize_with_diff_type(dtype):
         # test with resize height and width should be greater than 0
         transformer = transforms.CropResize(0, 0, 100, 50, (-25, 25), 1)
         assertRaises(MXNetError, transformer, data_in)
-        # test height and width should be greater than 0 
+        # test height and width should be greater than 0
         transformer = transforms.CropResize(0, 0, -100, -50)
         assertRaises(MXNetError, transformer, data_in)
         # test cropped area is bigger than input data
@@ -168,7 +168,7 @@ def _test_crop_resize_with_diff_type(dtype):
         assertRaises(MXNetError, transformer, data_bath_in)
 
     for dtype in ['uint8', 'float32', 'float64']:
-        _test_crop_resize_with_diff_type(dtype)  
+        _test_crop_resize_with_diff_type(dtype)
 
     # test nd.image.crop backward
     def test_crop_backward(test_nd_arr, TestCase):
@@ -288,18 +288,18 @@ def test_random_rotation():
 @with_seed()
 def test_random_transforms():
     from mxnet.gluon.data.vision import transforms
-    
+
     tmp_t = transforms.Compose([transforms.Resize(300), transforms.RandomResizedCrop(224)])
     transform = transforms.Compose([transforms.RandomApply(tmp_t, 0.5)])
 
     img = mx.nd.ones((10, 10, 3), dtype='uint8')
-    iteration = 1000
+    iteration = 2000
     num_apply = 0
     for _ in range(iteration):
         out = transform(img)
         if out.shape[0] == 224:
             num_apply += 1
-    assert_almost_equal(num_apply/float(iteration), 0.5, 0.1)
+    assert_almost_equal(np.array(num_apply/float(iteration)), np.array(0.5), rtol=0.1, atol=1e-2)
 
 
 if __name__ == '__main__':