#visualize.visualize_image_matrix_grayscale(image_matrix, "out_image_original_gray.png")
#for i in range(10):
network.visualize_image(image_matrix_rgb, "original_rgb")
network.visualize_image(image_matrix, "original_greyscale")
hopfield = Hopfield(image_matrix, alpha=alpha, strmost=strmost)
while True:
# recompute strmost
_new_strmost = 1.0 * (step * batchsize - strmost_increase_after) / \
(strmost_increase_until - strmost_increase_after) * \
(strmost_final - strmost) + strmost
if _new_strmost > strmost:
hopfield.strmost = _new_strmost
#hopfield.recompute_random()
hopfield.recompute_random_batch(batchsize)
if (step) % visualize_image_after == 0:
print("{}: strmost={:.1f}, image visualization...".format(
step, hopfield.strmost))
new_image_matrix = hopfield.get_recomputed_image_matrix()
#visualize.visualize_image_matrix_grayscale(new_image_matrix, "out_image.png")
network.visualize_image(new_image_matrix, "result")
print("Visualized")
step += 1
_new_strmost = 1.0 * (step * batchsize - strmost_increase_after) / \
(strmost_increase_until - strmost_increase_after) * \
(strmost_final - strmost) + strmost
if _new_strmost > strmost:
hopfield.strmost = _new_strmost
print("{}: strmost={:.1f}".format(step * batchsize, hopfield.strmost))
t = time.time()
hopfield.recompute_random_batch(batchsize)
print("Batch took {:.2f}s".format(time.time() - t))
t = time.time()
network.visualize_image(hopfield.get_recomputed_image_matrix(), "result")
network.visualize_image(hopfield.get_recomputed_image_matrix_mean(),
"result_mean")
network.visualize_image(hopfield.get_recomputed_image_matrix_valid(),
"result_valid")
network.visualize_image(hopfield.get_recomputed_image_matrix_var(), "var")
for i in range(len(b)):
network.visualize_image(
hopfield.get_recomputed_image_matrix_one_color(i),
f"result_colors/result_color_{i}")
print(hopfield.V[10, 10])
print("Visualisation took {:.2f}s".format(time.time() - t))
step += 1
