Precision: [tensor(0.4586, device='cuda:0'), tensor(0.4523, device='cuda:0'), tensor(0.4520, device='cuda:0'), tensor(0.4557, device='cuda:0'), tensor(0.4565, device='cuda:0'), tensor(0.4588, device='cuda:0'), tensor(0.4538, device='cuda:0'), tensor(0.4547, device='cuda:0'), tensor(0.4439, device='cuda:0'), tensor(0.4513, device='cuda:0')]
Output distance: [tensor(19.2307, device='cuda:0'), tensor(19.2609, device='cuda:0'), tensor(19.2633, device='cuda:0'), tensor(19.2444, device='cuda:0'), tensor(19.2417, device='cuda:0'), tensor(19.2294, device='cuda:0'), tensor(19.2548, device='cuda:0'), tensor(19.2497, device='cuda:0'), tensor(19.3026, device='cuda:0'), tensor(19.2669, device='cuda:0')]
Prediction loss: [tensor(104.6701, device='cuda:0'), tensor(104.3147, device='cuda:0'), tensor(104.5527, device='cuda:0'), tensor(104.5420, device='cuda:0'), tensor(104.5420, device='cuda:0'), tensor(104.9266, device='cuda:0'), tensor(105.2451, device='cuda:0'), tensor(104.5575, device='cuda:0'), tensor(104.4942, device='cuda:0'), tensor(104.4341, device='cuda:0')]
Others: [{'iter_num': 5, 'num_positive': tensor(16390, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 7, 'num_positive': tensor(16340, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 5, 'num_positive': tensor(16412, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 7, 'num_positive': tensor(16347, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 5, 'num_positive': tensor(16445, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 7, 'num_positive': tensor(16388, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 7, 'num_positive': tensor(16428, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 7, 'num_positive': tensor(16382, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 7, 'num_positive': tensor(16342, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}, {'iter_num': 7, 'num_positive': tensor(16408, device='cuda:0'), 'num_positive_true': tensor(125872, device='cuda:0')}]
Compressed training loss: [tensor(0.0001, device='cuda:0'), tensor(0.0001, device='cuda:0'), tensor(0.0001, device='cuda:0'), tensor(0.0001, device='cuda:0'), tensor(0.0001, device='cuda:0'), tensor(0.0001, device='cuda:0'), tensor(0.0001, device='cuda:0'), tensor(0.0001, device='cuda:0'), tensor(0.0001, device='cuda:0'), tensor(0.0001, device='cuda:0')]
Training loss: 0
Prediction time: [datetime.timedelta(seconds=10, microseconds=883789), datetime.timedelta(seconds=14, microseconds=250469), datetime.timedelta(seconds=10, microseconds=866973), datetime.timedelta(seconds=14, microseconds=216895), datetime.timedelta(seconds=10, microseconds=849574), datetime.timedelta(seconds=14, microseconds=266479), datetime.timedelta(seconds=14, microseconds=166682), datetime.timedelta(seconds=14, microseconds=299925), datetime.timedelta(seconds=14, microseconds=166498), datetime.timedelta(seconds=14, microseconds=383222)]
Phi time: [datetime.timedelta(seconds=99, microseconds=681957), datetime.timedelta(seconds=99, microseconds=354027), datetime.timedelta(seconds=99, microseconds=443624), datetime.timedelta(seconds=99, microseconds=484873), datetime.timedelta(seconds=99, microseconds=432851), datetime.timedelta(seconds=99, microseconds=366098), datetime.timedelta(seconds=99, microseconds=314969), datetime.timedelta(seconds=99, microseconds=565777), datetime.timedelta(seconds=99, microseconds=360937), datetime.timedelta(seconds=99, microseconds=476027)]
