episode: 0 training return: tensor(-499.9670, device='cuda:0', grad_fn=<AddBackward0>)
episode: 1 training return: tensor(-454.2894, device='cuda:0', grad_fn=<AddBackward0>)
episode: 2 training return: tensor(-499.9711, device='cuda:0', grad_fn=<AddBackward0>)
episode: 3 training return: tensor(-490.4561, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 1 test_true_pfm: 1468.8059196510183
episode: 4 training return: tensor(-499.9428, device='cuda:0', grad_fn=<AddBackward0>)
episode: 5 training return: tensor(-499.9535, device='cuda:0', grad_fn=<AddBackward0>)
episode: 6 training return: tensor(-486.9932, device='cuda:0', grad_fn=<AddBackward0>)
episode: 7 training return: tensor(-499.9301, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 2 test_true_pfm: 642.3803843897998
episode: 8 training return: tensor(-499.9671, device='cuda:0', grad_fn=<AddBackward0>)
episode: 9 training return: tensor(-499.9492, device='cuda:0', grad_fn=<AddBackward0>)
episode: 10 training return: tensor(-499.9427, device='cuda:0', grad_fn=<AddBackward0>)
episode: 11 training return: tensor(-499.9569, device='cuda:0', grad_fn=<AddBackward0>)
epoch: 3 test_true_pfm: 2685.487124222984
