acc_train_loss, nb_train_samples = 0, 0
for mazes, policies in task.policy_batches(split="train"):
- order = random_order(input, task.height * task.width)
+ order = random_order(mazes, task.height * task.width)
x = shuffle(mazes, order)
x = gpt(mygpt.BracketedSequence(x), mode=args.oneshot_input, order=order).x
output_gpt = shuffle(x, order, reorder=True)
acc_test_loss, nb_test_samples = 0, 0
for mazes, policies in task.policy_batches(split="test"):
- order = random_order(input, task.height * task.width)
+ order = random_order(mazes, task.height * task.width)
x = shuffle(mazes, order)
x = gpt(mygpt.BracketedSequence(x), mode=args.oneshot_input, order=order).x
output_gpt = shuffle(x, order, reorder=True)
# -------------------
mazes = task.test_input[:32, : task.height * task.width]
policies = task.test_policies[:32]
- order = random_order(input, task.height * task.width)
+ order = random_order(mazes, task.height * task.width)
x = shuffle(mazes, order)
x = gpt(mygpt.BracketedSequence(x), mode=args.oneshot_input, order=order).x
output_gpt = shuffle(x, order, reorder=True)