Beispiel #1
0
def add_diff_rot90(g_outmap):
    g_out = g_outmap["output"]
    grid_idx = g_outmap['grid_idx']
    z_rot90 = g_outmap['z_rot90']
    alphas = binary_mask(grid_idx, black=0.5, ignore=1.0, white=0.5)
    bw_mask = binary_mask(grid_idx, black=0., ignore=0, white=0.5)
    combined = (alphas * g_out) + bw_mask
    return rotate_by_multiple_of_90(combined, z_rot90)
Beispiel #2
0
def add_diff_rot90(g_outmap):
    g_out = g_outmap["output"]
    grid_idx = g_outmap['grid_idx']
    z_rot90 = g_outmap['z_rot90']
    alphas = binary_mask(grid_idx, black=0.5, ignore=1.0,  white=0.5)
    bw_mask = binary_mask(grid_idx, black=0., ignore=0,  white=0.5)
    combined = (alphas * g_out) + bw_mask
    return rotate_by_multiple_of_90(combined, z_rot90)
Beispiel #3
0
 def reconstruct(g_outmap):
     g_out = g_outmap["output"]
     grid_idx = g_outmap["grid_idx"]
     z_rot90 = g_outmap['z_rot90']
     alphas = binary_mask(grid_idx, black=variation_weight,
                          ignore=1.0, white=variation_weight)
     m = theano.gradient.disconnected_grad(g_out[:, :1])
     v = g_out[:, 1:]
     combined = v  # T.clip(m + alphas*v, 0., 1.)
     return rotate_by_multiple_of_90(combined, z_rot90)
Beispiel #4
0
 def reconstruct(g_outmap):
     g_out = g_outmap["output"]
     grid_idx = g_outmap["grid_idx"]
     z_rot90 = g_outmap['z_rot90']
     alphas = binary_mask(grid_idx,
                          black=variation_weight,
                          ignore=1.0,
                          white=variation_weight)
     m = theano.gradient.disconnected_grad(g_out[:, :1])
     v = g_out[:, 1:]
     combined = v  # T.clip(m + alphas*v, 0., 1.)
     return rotate_by_multiple_of_90(combined, z_rot90)
Beispiel #5
0
def test_util_rotate_by_multiple_of_90(batch):
    n = len(batch)
    th_batch = K.variable(batch)
    rots = K.variable(np.array([0, 1, 2, 3]))
    rotated = rotate_by_multiple_of_90(th_batch, rots).eval()
    for i in range(n):
        plt.subplot(131)
        plt.imshow(batch[i, 0])
        plt.subplot(132)
        plt.imshow(rotated[i, 0])
        plt.subplot(133)
        plt.imshow(np.rot90(batch[i, 0], k=i))
        plt_save_and_maybe_show("utils/rotate_{}.png".format(i))

    assert rotated.shape == batch.shape
    for i in range(n):
        assert (rotated[i, 0] == np.rot90(batch[i, 0], k=i)).all(), i
Beispiel #6
0
def test_util_rotate_by_multiple_of_90_missing_rots(batch):
    th_batch = K.variable(batch)
    rots = K.variable(np.array([0, 0, 2, 2]))
    rotated = rotate_by_multiple_of_90(th_batch, rots).eval()
    assert rotated.shape == batch.shape