Help in understanding caffe training log

[yxc...@gmail.com]

The following is a screenshot of caffe training. Just wondering, what is the difference between the loss in 

Iteration 162112 ...... loss = 1.78995 and

Train net output #0 ... loss = 1.5614.

Also, what does *1 in (*1 = 1.5614) mean?

Thanks.

I0415 11:02:32.611037  1521 sgd_solver.cpp:105] Iteration 162064, lr = 0.045

I0415 11:02:54.156378  1521 solver.cpp:219] Iteration 162080 (0.742637 iter/s, 21.5448s/16 iters), loss = 2.20358

I0415 11:02:54.156500  1521 solver.cpp:238]     Train net output #0: loss = 2.44907 (* 1 = 2.44907 loss)

I0415 11:02:54.156507  1521 sgd_solver.cpp:105] Iteration 162080, lr = 0.045

I0415 11:03:15.638656  1521 solver.cpp:219] Iteration 162096 (0.744822 iter/s, 21.4816s/16 iters), loss = 1.97801

I0415 11:03:15.638696  1521 solver.cpp:238]     Train net output #0: loss = 2.38628 (* 1 = 2.38628 loss)

I0415 11:03:15.638702  1521 sgd_solver.cpp:105] Iteration 162096, lr = 0.045

I0415 11:03:37.203069  1521 solver.cpp:219] Iteration 162112 (0.741982 iter/s, 21.5639s/16 iters), loss = 1.78995

I0415 11:03:37.203194  1521 solver.cpp:238]     Train net output #0: loss = 1.5614 (* 1 = 1.5614 loss)

I0415 11:03:37.203202  1521 sgd_solver.cpp:105] Iteration 162112, lr = 0.045