log.txt

I1122 14:39:02.817577 11005 net.cpp:192] conv4_bn needs backward computation.
I1122 14:39:02.817584 11005 net.cpp:192] conv4 needs backward computation.
I1122 14:39:02.817591 11005 net.cpp:192] pool3 needs backward computation.
I1122 14:39:02.817597 11005 net.cpp:192] relu3 needs backward computation.
I1122 14:39:02.817605 11005 net.cpp:192] conv3_bn needs backward computation.
I1122 14:39:02.817613 11005 net.cpp:192] conv3 needs backward computation.
I1122 14:39:02.817620 11005 net.cpp:192] pool2 needs backward computation.
I1122 14:39:02.817626 11005 net.cpp:192] relu2 needs backward computation.
I1122 14:39:02.817633 11005 net.cpp:192] conv2_bn needs backward computation.
I1122 14:39:02.817639 11005 net.cpp:192] conv2 needs backward computation.
I1122 14:39:02.817646 11005 net.cpp:192] pool1 needs backward computation.
I1122 14:39:02.817653 11005 net.cpp:192] relu1 needs backward computation.
I1122 14:39:02.817659 11005 net.cpp:192] conv1_bn needs backward computation.
I1122 14:39:02.817667 11005 net.cpp:192] conv1 needs backward computation.
I1122 14:39:02.817673 11005 net.cpp:194] norm does not need backward computation.
I1122 14:39:02.817680 11005 net.cpp:194] label_data_0_split does not need backward computation.
I1122 14:39:02.817688 11005 net.cpp:194] data does not need backward computation.
I1122 14:39:02.817694 11005 net.cpp:194] data does not need backward computation.
I1122 14:39:02.817701 11005 net.cpp:235] This network produces output accuracy
I1122 14:39:02.817708 11005 net.cpp:235] This network produces output loss
I1122 14:39:02.817715 11005 net.cpp:235] This network produces output per_class_accuracy
I1122 14:39:02.817735 11005 net.cpp:482] Collecting Learning Rate and Weight Decay.
I1122 14:39:02.817746 11005 net.cpp:247] Network initialization done.
I1122 14:39:02.817754 11005 net.cpp:248] Memory required for data: 599261200
I1122 14:39:02.817873 11005 solver.cpp:42] Solver scaffolding done.
I1122 14:39:02.817930 11005 solver.cpp:250] Solving segnet
I1122 14:39:02.817937 11005 solver.cpp:251] Learning Rate Policy: step
I1122 14:39:03.476389 11005 solver.cpp:214] Iteration 0, loss = 1.06872
I1122 14:39:03.476428 11005 solver.cpp:229]     Train net output #0: accuracy = 0.499298
I1122 14:39:03.476439 11005 solver.cpp:229]     Train net output #1: loss = 1.06872 (* 1 = 1.06872 loss)
I1122 14:39:03.476446 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.490004
I1122 14:39:03.476454 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.499593
I1122 14:39:03.476476 11005 solver.cpp:486] Iteration 0, lr = 1e-05
I1122 14:39:31.752205 11005 solver.cpp:214] Iteration 20, loss = 1.06384
I1122 14:39:31.752254 11005 solver.cpp:229]     Train net output #0: accuracy = 0.49865
I1122 14:39:31.752265 11005 solver.cpp:229]     Train net output #1: loss = 1.06384 (* 1 = 1.06384 loss)
I1122 14:39:31.752272 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.498846
I1122 14:39:31.752311 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.498081
I1122 14:39:31.752322 11005 solver.cpp:486] Iteration 20, lr = 1e-05
I1122 14:40:00.698418 11005 solver.cpp:214] Iteration 40, loss = 1.06438
I1122 14:40:00.698510 11005 solver.cpp:229]     Train net output #0: accuracy = 0.49799
I1122 14:40:00.698524 11005 solver.cpp:229]     Train net output #1: loss = 1.06438 (* 1 = 1.06438 loss)
I1122 14:40:00.698531 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.497755
I1122 14:40:00.698540 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.500095
I1122 14:40:00.698549 11005 solver.cpp:486] Iteration 40, lr = 1e-05
I1122 14:40:29.614552 11005 solver.cpp:214] Iteration 60, loss = 1.04793
I1122 14:40:29.614599 11005 solver.cpp:229]     Train net output #0: accuracy = 0.504002
I1122 14:40:29.614611 11005 solver.cpp:229]     Train net output #1: loss = 1.04793 (* 1 = 1.04793 loss)
I1122 14:40:29.614619 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.504203
I1122 14:40:29.614626 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.503943
I1122 14:40:29.614635 11005 solver.cpp:486] Iteration 60, lr = 1e-05
I1122 14:40:58.555565 11005 solver.cpp:214] Iteration 80, loss = 1.05809
I1122 14:40:58.555639 11005 solver.cpp:229]     Train net output #0: accuracy = 0.502617
I1122 14:40:58.555652 11005 solver.cpp:229]     Train net output #1: loss = 1.05809 (* 1 = 1.05809 loss)
I1122 14:40:58.555660 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.496437
I1122 14:40:58.555670 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.503433
I1122 14:40:58.555677 11005 solver.cpp:486] Iteration 80, lr = 1e-05
I1122 14:41:27.433779 11005 solver.cpp:214] Iteration 100, loss = 1.03838
I1122 14:41:27.433825 11005 solver.cpp:229]     Train net output #0: accuracy = 0.503174
I1122 14:41:27.433837 11005 solver.cpp:229]     Train net output #1: loss = 1.03838 (* 1 = 1.03838 loss)
I1122 14:41:27.433845 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.499236
I1122 14:41:27.433852 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.505504
I1122 14:41:27.433861 11005 solver.cpp:486] Iteration 100, lr = 1e-05
I1122 14:41:56.388880 11005 solver.cpp:214] Iteration 120, loss = 1.05377
I1122 14:41:56.389010 11005 solver.cpp:229]     Train net output #0: accuracy = 0.502213
I1122 14:41:56.389032 11005 solver.cpp:229]     Train net output #1: loss = 1.05377 (* 1 = 1.05377 loss)
I1122 14:41:56.389037 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.496299
I1122 14:41:56.389041 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.504178
I1122 14:41:56.389047 11005 solver.cpp:486] Iteration 120, lr = 1e-05
I1122 14:42:25.297181 11005 solver.cpp:214] Iteration 140, loss = 1.0412
I1122 14:42:25.297226 11005 solver.cpp:229]     Train net output #0: accuracy = 0.506565
I1122 14:42:25.297240 11005 solver.cpp:229]     Train net output #1: loss = 1.0412 (* 1 = 1.0412 loss)
I1122 14:42:25.297246 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.502634
I1122 14:42:25.297253 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.510479
I1122 14:42:25.297262 11005 solver.cpp:486] Iteration 140, lr = 1e-05
I1122 14:42:54.209035 11005 solver.cpp:214] Iteration 160, loss = 1.05713
I1122 14:42:54.209154 11005 solver.cpp:229]     Train net output #0: accuracy = 0.502514
I1122 14:42:54.209167 11005 solver.cpp:229]     Train net output #1: loss = 1.05713 (* 1 = 1.05713 loss)
I1122 14:42:54.209175 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.498977
I1122 14:42:54.209185 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.503612
I1122 14:42:54.209193 11005 solver.cpp:486] Iteration 160, lr = 1e-05
I1122 14:43:23.113016 11005 solver.cpp:214] Iteration 180, loss = 1.03555
I1122 14:43:23.113062 11005 solver.cpp:229]     Train net output #0: accuracy = 0.507942
I1122 14:43:23.113075 11005 solver.cpp:229]     Train net output #1: loss = 1.03555 (* 1 = 1.03555 loss)
I1122 14:43:23.113081 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.505849
I1122 14:43:23.113090 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.508657
I1122 14:43:23.113098 11005 solver.cpp:486] Iteration 180, lr = 1e-05
I1122 14:43:52.051172 11005 solver.cpp:214] Iteration 200, loss = 1.05083
I1122 14:43:52.051337 11005 solver.cpp:229]     Train net output #0: accuracy = 0.505226
I1122 14:43:52.051358 11005 solver.cpp:229]     Train net output #1: loss = 1.05083 (* 1 = 1.05083 loss)
I1122 14:43:52.051363 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.500357
I1122 14:43:52.051367 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.510127
I1122 14:43:52.051373 11005 solver.cpp:486] Iteration 200, lr = 1e-05
I1122 14:44:20.957080 11005 solver.cpp:214] Iteration 220, loss = 1.02279
I1122 14:44:20.957129 11005 solver.cpp:229]     Train net output #0: accuracy = 0.511902
I1122 14:44:20.957141 11005 solver.cpp:229]     Train net output #1: loss = 1.02279 (* 1 = 1.02279 loss)
I1122 14:44:20.957150 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.511161
I1122 14:44:20.957156 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.512159
I1122 14:44:20.957165 11005 solver.cpp:486] Iteration 220, lr = 1e-05
I1122 14:44:49.896126 11005 solver.cpp:214] Iteration 240, loss = 1.05018
I1122 14:44:49.896267 11005 solver.cpp:229]     Train net output #0: accuracy = 0.505913
I1122 14:44:49.896288 11005 solver.cpp:229]     Train net output #1: loss = 1.05018 (* 1 = 1.05018 loss)
I1122 14:44:49.896293 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.506299
I1122 14:44:49.896297 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.505907
I1122 14:44:49.896303 11005 solver.cpp:486] Iteration 240, lr = 1e-05
I1122 14:45:18.793390 11005 solver.cpp:214] Iteration 260, loss = 1.02073
I1122 14:45:18.793438 11005 solver.cpp:229]     Train net output #0: accuracy = 0.510315
I1122 14:45:18.793452 11005 solver.cpp:229]     Train net output #1: loss = 1.02073 (* 1 = 1.02073 loss)
I1122 14:45:18.793459 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.502699
I1122 14:45:18.793469 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.518069
I1122 14:45:18.793478 11005 solver.cpp:486] Iteration 260, lr = 1e-05
I1122 14:45:47.728776 11005 solver.cpp:214] Iteration 280, loss = 1.01173
I1122 14:45:47.728911 11005 solver.cpp:229]     Train net output #0: accuracy = 0.516296
I1122 14:45:47.728936 11005 solver.cpp:229]     Train net output #1: loss = 1.01173 (* 1 = 1.01173 loss)
I1122 14:45:47.728945 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.509372
I1122 14:45:47.728950 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.520151
I1122 14:45:47.728963 11005 solver.cpp:486] Iteration 280, lr = 1e-05
I1122 14:46:16.650317 11005 solver.cpp:214] Iteration 300, loss = 1.0572
I1122 14:46:16.650365 11005 solver.cpp:229]     Train net output #0: accuracy = 0.494808
I1122 14:46:16.650378 11005 solver.cpp:229]     Train net output #1: loss = 1.0572 (* 1 = 1.0572 loss)
I1122 14:46:16.650387 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.492227
I1122 14:46:16.650396 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.528214
I1122 14:46:16.650404 11005 solver.cpp:486] Iteration 300, lr = 1e-05
I1122 14:46:45.577258 11005 solver.cpp:214] Iteration 320, loss = 1.04277
I1122 14:46:45.577392 11005 solver.cpp:229]     Train net output #0: accuracy = 0.508183
I1122 14:46:45.577417 11005 solver.cpp:229]     Train net output #1: loss = 1.04277 (* 1 = 1.04277 loss)
I1122 14:46:45.577425 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.496504
I1122 14:46:45.577431 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.508993
I1122 14:46:45.577443 11005 solver.cpp:486] Iteration 320, lr = 1e-05
I1122 14:47:14.497674 11005 solver.cpp:214] Iteration 340, loss = 1.02284
I1122 14:47:14.497722 11005 solver.cpp:229]     Train net output #0: accuracy = 0.509922
I1122 14:47:14.497736 11005 solver.cpp:229]     Train net output #1: loss = 1.02284 (* 1 = 1.02284 loss)
I1122 14:47:14.497743 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.496112
I1122 14:47:14.497752 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.521929
I1122 14:47:14.497761 11005 solver.cpp:486] Iteration 340, lr = 1e-05
I1122 14:47:43.411173 11005 solver.cpp:214] Iteration 360, loss = 1.00121
I1122 14:47:43.411336 11005 solver.cpp:229]     Train net output #0: accuracy = 0.520084
I1122 14:47:43.411361 11005 solver.cpp:229]     Train net output #1: loss = 1.00121 (* 1 = 1.00121 loss)
I1122 14:47:43.411370 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.521488
I1122 14:47:43.411376 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.519734
I1122 14:47:43.411384 11005 solver.cpp:486] Iteration 360, lr = 1e-05
I1122 14:48:12.380188 11005 solver.cpp:214] Iteration 380, loss = 1.04309
I1122 14:48:12.380239 11005 solver.cpp:229]     Train net output #0: accuracy = 0.508312
I1122 14:48:12.380252 11005 solver.cpp:229]     Train net output #1: loss = 1.04309 (* 1 = 1.04309 loss)
I1122 14:48:12.380259 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.499938
I1122 14:48:12.380266 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.508578
I1122 14:48:12.380285 11005 solver.cpp:486] Iteration 380, lr = 1e-05
I1122 14:48:41.306768 11005 solver.cpp:214] Iteration 400, loss = 1.01678
I1122 14:48:41.306913 11005 solver.cpp:229]     Train net output #0: accuracy = 0.511887
I1122 14:48:41.306941 11005 solver.cpp:229]     Train net output #1: loss = 1.01678 (* 1 = 1.01678 loss)
I1122 14:48:41.306948 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.499641
I1122 14:48:41.306954 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.547384
I1122 14:48:41.306963 11005 solver.cpp:486] Iteration 400, lr = 1e-05
I1122 14:49:10.213568 11005 solver.cpp:214] Iteration 420, loss = 1.05626
I1122 14:49:10.213614 11005 solver.cpp:229]     Train net output #0: accuracy = 0.493088
I1122 14:49:10.213626 11005 solver.cpp:229]     Train net output #1: loss = 1.05626 (* 1 = 1.05626 loss)
I1122 14:49:10.213634 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.490231
I1122 14:49:10.213641 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.51876
I1122 14:49:10.213652 11005 solver.cpp:486] Iteration 420, lr = 1e-05
I1122 14:49:39.139946 11005 solver.cpp:214] Iteration 440, loss = 0.988768
I1122 14:49:39.140081 11005 solver.cpp:229]     Train net output #0: accuracy = 0.523777
I1122 14:49:39.140106 11005 solver.cpp:229]     Train net output #1: loss = 0.988768 (* 1 = 0.988768 loss)
I1122 14:49:39.140115 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.52437
I1122 14:49:39.140120 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.523605
I1122 14:49:39.140130 11005 solver.cpp:486] Iteration 440, lr = 1e-05
I1122 14:50:08.064565 11005 solver.cpp:214] Iteration 460, loss = 1.03653
I1122 14:50:08.064612 11005 solver.cpp:229]     Train net output #0: accuracy = 0.51096
I1122 14:50:08.064625 11005 solver.cpp:229]     Train net output #1: loss = 1.03653 (* 1 = 1.03653 loss)
I1122 14:50:08.064632 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.48807
I1122 14:50:08.064641 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.513984
I1122 14:50:08.064649 11005 solver.cpp:486] Iteration 460, lr = 1e-05
I1122 14:50:36.965035 11005 solver.cpp:214] Iteration 480, loss = 0.983061
I1122 14:50:36.965208 11005 solver.cpp:229]     Train net output #0: accuracy = 0.517948
I1122 14:50:36.965232 11005 solver.cpp:229]     Train net output #1: loss = 0.983061 (* 1 = 0.983061 loss)
I1122 14:50:36.965240 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.498261
I1122 14:50:36.965246 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.529599
I1122 14:50:36.965260 11005 solver.cpp:486] Iteration 480, lr = 1e-05
I1122 14:51:05.905592 11005 solver.cpp:214] Iteration 500, loss = 1.00935
I1122 14:51:05.905640 11005 solver.cpp:229]     Train net output #0: accuracy = 0.519306
I1122 14:51:05.905653 11005 solver.cpp:229]     Train net output #1: loss = 1.00935 (* 1 = 1.00935 loss)
I1122 14:51:05.905660 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.508167
I1122 14:51:05.905668 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.523008
I1122 14:51:05.905678 11005 solver.cpp:486] Iteration 500, lr = 1e-05
I1122 14:51:34.815644 11005 solver.cpp:214] Iteration 520, loss = 0.968033
I1122 14:51:34.815783 11005 solver.cpp:229]     Train net output #0: accuracy = 0.534496
I1122 14:51:34.815807 11005 solver.cpp:229]     Train net output #1: loss = 0.968033 (* 1 = 0.968033 loss)
I1122 14:51:34.815814 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.516871
I1122 14:51:34.815821 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.552044
I1122 14:51:34.815834 11005 solver.cpp:486] Iteration 520, lr = 1e-05
I1122 14:52:03.731465 11005 solver.cpp:214] Iteration 540, loss = 1.0223
I1122 14:52:03.731511 11005 solver.cpp:229]     Train net output #0: accuracy = 0.515362
I1122 14:52:03.731524 11005 solver.cpp:229]     Train net output #1: loss = 1.0223 (* 1 = 1.0223 loss)
I1122 14:52:03.731533 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.509156
I1122 14:52:03.731540 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.517288
I1122 14:52:03.731549 11005 solver.cpp:486] Iteration 540, lr = 1e-05
I1122 14:52:32.644445 11005 solver.cpp:214] Iteration 560, loss = 0.952894
I1122 14:52:32.644577 11005 solver.cpp:229]     Train net output #0: accuracy = 0.537037
I1122 14:52:32.644601 11005 solver.cpp:229]     Train net output #1: loss = 0.952894 (* 1 = 0.952894 loss)
I1122 14:52:32.644609 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.533816
I1122 14:52:32.644616 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.538137
I1122 14:52:32.644634 11005 solver.cpp:486] Iteration 560, lr = 1e-05
I1122 14:53:01.567126 11005 solver.cpp:214] Iteration 580, loss = 0.994493
I1122 14:53:01.567173 11005 solver.cpp:229]     Train net output #0: accuracy = 0.522861
I1122 14:53:01.567184 11005 solver.cpp:229]     Train net output #1: loss = 0.994493 (* 1 = 0.994493 loss)
I1122 14:53:01.567193 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.508653
I1122 14:53:01.567198 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.537164
I1122 14:53:01.567208 11005 solver.cpp:486] Iteration 580, lr = 1e-05
I1122 14:53:30.463860 11005 solver.cpp:214] Iteration 600, loss = 0.941316
I1122 14:53:30.463994 11005 solver.cpp:229]     Train net output #0: accuracy = 0.542694
I1122 14:53:30.464020 11005 solver.cpp:229]     Train net output #1: loss = 0.941316 (* 1 = 0.941316 loss)
I1122 14:53:30.464027 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.540903
I1122 14:53:30.464033 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.543315
I1122 14:53:30.464046 11005 solver.cpp:486] Iteration 600, lr = 1e-05
I1122 14:53:59.395335 11005 solver.cpp:214] Iteration 620, loss = 1.02613
I1122 14:53:59.395382 11005 solver.cpp:229]     Train net output #0: accuracy = 0.516705
I1122 14:53:59.395395 11005 solver.cpp:229]     Train net output #1: loss = 1.02613 (* 1 = 1.02613 loss)
I1122 14:53:59.395402 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.503618
I1122 14:53:59.395409 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.516894
I1122 14:53:59.395421 11005 solver.cpp:486] Iteration 620, lr = 1e-05
I1122 14:54:28.278941 11005 solver.cpp:214] Iteration 640, loss = 0.975315
I1122 14:54:28.279111 11005 solver.cpp:229]     Train net output #0: accuracy = 0.522125
I1122 14:54:28.279132 11005 solver.cpp:229]     Train net output #1: loss = 0.975315 (* 1 = 0.975315 loss)
I1122 14:54:28.279137 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.499992
I1122 14:54:28.279141 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.544661
I1122 14:54:28.279147 11005 solver.cpp:486] Iteration 640, lr = 1e-05
I1122 14:54:57.206048 11005 solver.cpp:214] Iteration 660, loss = 0.941213
I1122 14:54:57.206094 11005 solver.cpp:229]     Train net output #0: accuracy = 0.541481
I1122 14:54:57.206106 11005 solver.cpp:229]     Train net output #1: loss = 0.941213 (* 1 = 0.941213 loss)
I1122 14:54:57.206115 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.521117
I1122 14:54:57.206122 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.552816
I1122 14:54:57.206131 11005 solver.cpp:486] Iteration 660, lr = 1e-05
I1122 14:55:26.137435 11005 solver.cpp:214] Iteration 680, loss = 1.0468
I1122 14:55:26.137573 11005 solver.cpp:229]     Train net output #0: accuracy = 0.488594
I1122 14:55:26.137598 11005 solver.cpp:229]     Train net output #1: loss = 1.0468 (* 1 = 1.0468 loss)
I1122 14:55:26.137608 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.482151
I1122 14:55:26.137614 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.571983
I1122 14:55:26.137622 11005 solver.cpp:486] Iteration 680, lr = 1e-05
I1122 14:55:55.070721 11005 solver.cpp:214] Iteration 700, loss = 1.01949
I1122 14:55:55.070768 11005 solver.cpp:229]     Train net output #0: accuracy = 0.519115
I1122 14:55:55.070780 11005 solver.cpp:229]     Train net output #1: loss = 1.01949 (* 1 = 1.01949 loss)
I1122 14:55:55.070788 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.494801
I1122 14:55:55.070797 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.520804
I1122 14:55:55.070809 11005 solver.cpp:486] Iteration 700, lr = 1e-05
I1122 14:56:23.984040 11005 solver.cpp:214] Iteration 720, loss = 0.987833
I1122 14:56:23.984174 11005 solver.cpp:229]     Train net output #0: accuracy = 0.526329
I1122 14:56:23.984200 11005 solver.cpp:229]     Train net output #1: loss = 0.987833 (* 1 = 0.987833 loss)
I1122 14:56:23.984208 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.494349
I1122 14:56:23.984216 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.554135
I1122 14:56:23.984225 11005 solver.cpp:486] Iteration 720, lr = 1e-05
I1122 14:56:52.905498 11005 solver.cpp:214] Iteration 740, loss = 0.920269
I1122 14:56:52.905550 11005 solver.cpp:229]     Train net output #0: accuracy = 0.553783
I1122 14:56:52.905563 11005 solver.cpp:229]     Train net output #1: loss = 0.920269 (* 1 = 0.920269 loss)
I1122 14:56:52.905570 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.558519
I1122 14:56:52.905578 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.552603
I1122 14:56:52.905587 11005 solver.cpp:486] Iteration 740, lr = 1e-05
I1122 14:57:21.839640 11005 solver.cpp:214] Iteration 760, loss = 1.02188
I1122 14:57:21.839776 11005 solver.cpp:229]     Train net output #0: accuracy = 0.518917
I1122 14:57:21.839800 11005 solver.cpp:229]     Train net output #1: loss = 1.02188 (* 1 = 1.02188 loss)
I1122 14:57:21.839809 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.515212
I1122 14:57:21.839817 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.519035
I1122 14:57:21.839828 11005 solver.cpp:486] Iteration 760, lr = 1e-05
I1122 14:57:50.763090 11005 solver.cpp:214] Iteration 780, loss = 0.961693
I1122 14:57:50.763137 11005 solver.cpp:229]     Train net output #0: accuracy = 0.530777
I1122 14:57:50.763150 11005 solver.cpp:229]     Train net output #1: loss = 0.961693 (* 1 = 0.961693 loss)
I1122 14:57:50.763159 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.502114
I1122 14:57:50.763169 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.613864
I1122 14:57:50.763177 11005 solver.cpp:486] Iteration 780, lr = 1e-05
I1122 14:58:19.678887 11005 solver.cpp:214] Iteration 800, loss = 1.04947
I1122 14:58:19.679056 11005 solver.cpp:229]     Train net output #0: accuracy = 0.48642
I1122 14:58:19.679081 11005 solver.cpp:229]     Train net output #1: loss = 1.04947 (* 1 = 1.04947 loss)
I1122 14:58:19.679090 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.480565
I1122 14:58:19.679097 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.539024
I1122 14:58:19.679105 11005 solver.cpp:486] Iteration 800, lr = 1e-05
I1122 14:58:48.593041 11005 solver.cpp:214] Iteration 820, loss = 0.916568
I1122 14:58:48.593088 11005 solver.cpp:229]     Train net output #0: accuracy = 0.553894
I1122 14:58:48.593101 11005 solver.cpp:229]     Train net output #1: loss = 0.916568 (* 1 = 0.916568 loss)
I1122 14:58:48.593108 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.552383
I1122 14:58:48.593116 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.554333
I1122 14:58:48.593125 11005 solver.cpp:486] Iteration 820, lr = 1e-05
I1122 14:59:17.503304 11005 solver.cpp:214] Iteration 840, loss = 1.01921
I1122 14:59:17.503443 11005 solver.cpp:229]     Train net output #0: accuracy = 0.520634
I1122 14:59:17.503469 11005 solver.cpp:229]     Train net output #1: loss = 1.01921 (* 1 = 1.01921 loss)
I1122 14:59:17.503479 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.478068
I1122 14:59:17.503485 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.526258
I1122 14:59:17.503494 11005 solver.cpp:486] Iteration 840, lr = 1e-05
I1122 14:59:46.419699 11005 solver.cpp:214] Iteration 860, loss = 0.917841
I1122 14:59:46.419745 11005 solver.cpp:229]     Train net output #0: accuracy = 0.53907
I1122 14:59:46.419757 11005 solver.cpp:229]     Train net output #1: loss = 0.917841 (* 1 = 0.917841 loss)
I1122 14:59:46.419764 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.499451
I1122 14:59:46.419772 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.562517
I1122 14:59:46.419781 11005 solver.cpp:486] Iteration 860, lr = 1e-05
I1122 15:00:15.358508 11005 solver.cpp:214] Iteration 880, loss = 0.955243
I1122 15:00:15.358641 11005 solver.cpp:229]     Train net output #0: accuracy = 0.544678
I1122 15:00:15.358667 11005 solver.cpp:229]     Train net output #1: loss = 0.955243 (* 1 = 0.955243 loss)
I1122 15:00:15.358676 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.529045
I1122 15:00:15.358683 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.549872
I1122 15:00:15.358691 11005 solver.cpp:486] Iteration 880, lr = 1e-05
I1122 15:00:44.272840 11005 solver.cpp:214] Iteration 900, loss = 0.875857
I1122 15:00:44.272887 11005 solver.cpp:229]     Train net output #0: accuracy = 0.574261
I1122 15:00:44.272899 11005 solver.cpp:229]     Train net output #1: loss = 0.875857 (* 1 = 0.875857 loss)
I1122 15:00:44.272907 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.540246
I1122 15:00:44.272913 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.608126
I1122 15:00:44.272922 11005 solver.cpp:486] Iteration 900, lr = 1e-05
I1122 15:01:13.217528 11005 solver.cpp:214] Iteration 920, loss = 0.979696
I1122 15:01:13.217665 11005 solver.cpp:229]     Train net output #0: accuracy = 0.538021
I1122 15:01:13.217690 11005 solver.cpp:229]     Train net output #1: loss = 0.979696 (* 1 = 0.979696 loss)
I1122 15:01:13.217700 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.53077
I1122 15:01:13.217706 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.540271
I1122 15:01:13.217720 11005 solver.cpp:486] Iteration 920, lr = 1e-05
I1122 15:01:42.147079 11005 solver.cpp:214] Iteration 940, loss = 0.852329
I1122 15:01:42.147125 11005 solver.cpp:229]     Train net output #0: accuracy = 0.578682
I1122 15:01:42.147137 11005 solver.cpp:229]     Train net output #1: loss = 0.852329 (* 1 = 0.852329 loss)
I1122 15:01:42.147145 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.575653
I1122 15:01:42.147156 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.579717
I1122 15:01:42.147164 11005 solver.cpp:486] Iteration 940, lr = 1e-05
I1122 15:02:11.067209 11005 solver.cpp:214] Iteration 960, loss = 0.918802
I1122 15:02:11.067380 11005 solver.cpp:229]     Train net output #0: accuracy = 0.550404
I1122 15:02:11.067406 11005 solver.cpp:229]     Train net output #1: loss = 0.918802 (* 1 = 0.918802 loss)
I1122 15:02:11.067415 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.522637
I1122 15:02:11.067422 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.578355
I1122 15:02:11.067430 11005 solver.cpp:486] Iteration 960, lr = 1e-05
I1122 15:02:39.975908 11005 solver.cpp:214] Iteration 980, loss = 0.838539
I1122 15:02:39.975955 11005 solver.cpp:229]     Train net output #0: accuracy = 0.585312
I1122 15:02:39.975966 11005 solver.cpp:229]     Train net output #1: loss = 0.838539 (* 1 = 0.838539 loss)
I1122 15:02:39.975973 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.58211
I1122 15:02:39.975980 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.586423
I1122 15:02:39.975992 11005 solver.cpp:486] Iteration 980, lr = 1e-05
I1122 15:03:08.331473 11005 solver.cpp:361] Snapshotting to /home/gradescan/SegNet/Models/backup_basic/Training/segnet_basic_iter_1000.caffemodel
I1122 15:03:08.346374 11005 solver.cpp:369] Snapshotting solver state to /home/gradescan/SegNet/Models/backup_basic/Training/segnet_basic_iter_1000.solverstate
I1122 15:03:08.933265 11005 solver.cpp:214] Iteration 1000, loss = 1.00964
I1122 15:03:08.933310 11005 solver.cpp:229]     Train net output #0: accuracy = 0.529972
I1122 15:03:08.933321 11005 solver.cpp:229]     Train net output #1: loss = 1.00964 (* 1 = 1.00964 loss)
I1122 15:03:08.933329 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.515947
I1122 15:03:08.933336 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.530175
I1122 15:03:08.933349 11005 solver.cpp:486] Iteration 1000, lr = 1e-05
I1122 15:03:37.843001 11005 solver.cpp:214] Iteration 1020, loss = 0.91881
I1122 15:03:37.843050 11005 solver.cpp:229]     Train net output #0: accuracy = 0.541737
I1122 15:03:37.843063 11005 solver.cpp:229]     Train net output #1: loss = 0.91881 (* 1 = 0.91881 loss)
I1122 15:03:37.843070 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.501588
I1122 15:03:37.843077 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.582616
I1122 15:03:37.843086 11005 solver.cpp:486] Iteration 1020, lr = 1e-05
I1122 15:04:06.742977 11005 solver.cpp:214] Iteration 1040, loss = 0.853379
I1122 15:04:06.743059 11005 solver.cpp:229]     Train net output #0: accuracy = 0.579494
I1122 15:04:06.743072 11005 solver.cpp:229]     Train net output #1: loss = 0.853379 (* 1 = 0.853379 loss)
I1122 15:04:06.743080 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.544864
I1122 15:04:06.743089 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.598771
I1122 15:04:06.743098 11005 solver.cpp:486] Iteration 1040, lr = 1e-05
I1122 15:04:35.663641 11005 solver.cpp:214] Iteration 1060, loss = 1.03949
I1122 15:04:35.663689 11005 solver.cpp:229]     Train net output #0: accuracy = 0.480492
I1122 15:04:35.663702 11005 solver.cpp:229]     Train net output #1: loss = 1.03949 (* 1 = 1.03949 loss)
I1122 15:04:35.663710 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.46863
I1122 15:04:35.663719 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.633995
I1122 15:04:35.663728 11005 solver.cpp:486] Iteration 1060, lr = 1e-05
I1122 15:05:04.558969 11005 solver.cpp:214] Iteration 1080, loss = 1.00608
I1122 15:05:04.559077 11005 solver.cpp:229]     Train net output #0: accuracy = 0.535645
I1122 15:05:04.559092 11005 solver.cpp:229]     Train net output #1: loss = 1.00608 (* 1 = 1.00608 loss)
I1122 15:05:04.559099 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.491393
I1122 15:05:04.559106 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.538717
I1122 15:05:04.559119 11005 solver.cpp:486] Iteration 1080, lr = 1e-05
I1122 15:05:33.499739 11005 solver.cpp:214] Iteration 1100, loss = 0.955311
I1122 15:05:33.499788 11005 solver.cpp:229]     Train net output #0: accuracy = 0.547646
I1122 15:05:33.499799 11005 solver.cpp:229]     Train net output #1: loss = 0.955311 (* 1 = 0.955311 loss)
I1122 15:05:33.499807 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.492241
I1122 15:05:33.499814 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.595818
I1122 15:05:33.499824 11005 solver.cpp:486] Iteration 1100, lr = 1e-05
I1122 15:06:02.426502 11005 solver.cpp:214] Iteration 1120, loss = 0.824109
I1122 15:06:02.426579 11005 solver.cpp:229]     Train net output #0: accuracy = 0.599373
I1122 15:06:02.426592 11005 solver.cpp:229]     Train net output #1: loss = 0.824109 (* 1 = 0.824109 loss)
I1122 15:06:02.426599 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.608588
I1122 15:06:02.426606 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.597076
I1122 15:06:02.426614 11005 solver.cpp:486] Iteration 1120, lr = 1e-05
I1122 15:06:31.332609 11005 solver.cpp:214] Iteration 1140, loss = 1.0107
I1122 15:06:31.332664 11005 solver.cpp:229]     Train net output #0: accuracy = 0.531857
I1122 15:06:31.332676 11005 solver.cpp:229]     Train net output #1: loss = 1.0107 (* 1 = 1.0107 loss)
I1122 15:06:31.332684 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.522166
I1122 15:06:31.332690 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.532164
I1122 15:06:31.332701 11005 solver.cpp:486] Iteration 1140, lr = 1e-05
I1122 15:07:00.235925 11005 solver.cpp:214] Iteration 1160, loss = 0.893965
I1122 15:07:00.236028 11005 solver.cpp:229]     Train net output #0: accuracy = 0.558727
I1122 15:07:00.236042 11005 solver.cpp:229]     Train net output #1: loss = 0.893965 (* 1 = 0.893965 loss)
I1122 15:07:00.236048 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.50864
I1122 15:07:00.236057 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.703917
I1122 15:07:00.236065 11005 solver.cpp:486] Iteration 1160, lr = 1e-05
I1122 15:07:29.166470 11005 solver.cpp:214] Iteration 1180, loss = 1.04669
I1122 15:07:29.166517 11005 solver.cpp:229]     Train net output #0: accuracy = 0.47929
I1122 15:07:29.166530 11005 solver.cpp:229]     Train net output #1: loss = 1.04669 (* 1 = 1.04669 loss)
I1122 15:07:29.166538 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.469034
I1122 15:07:29.166545 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.571439
I1122 15:07:29.166554 11005 solver.cpp:486] Iteration 1180, lr = 1e-05
I1122 15:07:58.105350 11005 solver.cpp:214] Iteration 1200, loss = 0.826693
I1122 15:07:58.105479 11005 solver.cpp:229]     Train net output #0: accuracy = 0.598183
I1122 15:07:58.105501 11005 solver.cpp:229]     Train net output #1: loss = 0.826693 (* 1 = 0.826693 loss)
I1122 15:07:58.105506 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.597478
I1122 15:07:58.105509 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.598387
I1122 15:07:58.105515 11005 solver.cpp:486] Iteration 1200, lr = 1e-05
I1122 15:08:27.028475 11005 solver.cpp:214] Iteration 1220, loss = 1.01076
I1122 15:08:27.028525 11005 solver.cpp:229]     Train net output #0: accuracy = 0.533695
I1122 15:08:27.028538 11005 solver.cpp:229]     Train net output #1: loss = 1.01076 (* 1 = 1.01076 loss)
I1122 15:08:27.028545 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.47153
I1122 15:08:27.028555 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.541909
I1122 15:08:27.028563 11005 solver.cpp:486] Iteration 1220, lr = 1e-05
I1122 15:08:55.934972 11005 solver.cpp:214] Iteration 1240, loss = 0.8417
I1122 15:08:55.935077 11005 solver.cpp:229]     Train net output #0: accuracy = 0.570129
I1122 15:08:55.935091 11005 solver.cpp:229]     Train net output #1: loss = 0.8417 (* 1 = 0.8417 loss)
I1122 15:08:55.935098 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.502519
I1122 15:08:55.935106 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.610142
I1122 15:08:55.935117 11005 solver.cpp:486] Iteration 1240, lr = 1e-05
I1122 15:09:24.883695 11005 solver.cpp:214] Iteration 1260, loss = 0.885977
I1122 15:09:24.883744 11005 solver.cpp:229]     Train net output #0: accuracy = 0.57943
I1122 15:09:24.883756 11005 solver.cpp:229]     Train net output #1: loss = 0.885977 (* 1 = 0.885977 loss)
I1122 15:09:24.883764 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.564085
I1122 15:09:24.883774 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.584529
I1122 15:09:24.883781 11005 solver.cpp:486] Iteration 1260, lr = 1e-05
I1122 15:09:53.790976 11005 solver.cpp:214] Iteration 1280, loss = 0.76018
I1122 15:09:53.791091 11005 solver.cpp:229]     Train net output #0: accuracy = 0.631027
I1122 15:09:53.791112 11005 solver.cpp:229]     Train net output #1: loss = 0.76018 (* 1 = 0.76018 loss)
I1122 15:09:53.791117 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.572857
I1122 15:09:53.791121 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.688941
I1122 15:09:53.791126 11005 solver.cpp:486] Iteration 1280, lr = 1e-05
I1122 15:10:22.713475 11005 solver.cpp:214] Iteration 1300, loss = 0.921228
I1122 15:10:22.713523 11005 solver.cpp:229]     Train net output #0: accuracy = 0.572216
I1122 15:10:22.713536 11005 solver.cpp:229]     Train net output #1: loss = 0.921228 (* 1 = 0.921228 loss)
I1122 15:10:22.713544 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.573354
I1122 15:10:22.713551 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.571863
I1122 15:10:22.713562 11005 solver.cpp:486] Iteration 1300, lr = 1e-05
I1122 15:10:51.616813 11005 solver.cpp:214] Iteration 1320, loss = 0.73317
I1122 15:10:51.616932 11005 solver.cpp:229]     Train net output #0: accuracy = 0.637596
I1122 15:10:51.616960 11005 solver.cpp:229]     Train net output #1: loss = 0.73317 (* 1 = 0.73317 loss)
I1122 15:10:51.616966 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.635615
I1122 15:10:51.616974 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.638273
I1122 15:10:51.616982 11005 solver.cpp:486] Iteration 1320, lr = 1e-05
I1122 15:11:20.526033 11005 solver.cpp:214] Iteration 1340, loss = 0.8227
I1122 15:11:20.526082 11005 solver.cpp:229]     Train net output #0: accuracy = 0.589657
I1122 15:11:20.526094 11005 solver.cpp:229]     Train net output #1: loss = 0.8227 (* 1 = 0.8227 loss)
I1122 15:11:20.526103 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.54221
I1122 15:11:20.526109 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.63742
I1122 15:11:20.526118 11005 solver.cpp:486] Iteration 1340, lr = 1e-05
I1122 15:11:49.441898 11005 solver.cpp:214] Iteration 1360, loss = 0.713646
I1122 15:11:49.442028 11005 solver.cpp:229]     Train net output #0: accuracy = 0.646679
I1122 15:11:49.442052 11005 solver.cpp:229]     Train net output #1: loss = 0.713646 (* 1 = 0.713646 loss)
I1122 15:11:49.442060 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.642158
I1122 15:11:49.442070 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.648247
I1122 15:11:49.442080 11005 solver.cpp:486] Iteration 1360, lr = 1e-05
I1122 15:12:18.359225 11005 solver.cpp:214] Iteration 1380, loss = 1.00772
I1122 15:12:18.359273 11005 solver.cpp:229]     Train net output #0: accuracy = 0.550991
I1122 15:12:18.359287 11005 solver.cpp:229]     Train net output #1: loss = 1.00772 (* 1 = 1.00772 loss)
I1122 15:12:18.359293 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.537389
I1122 15:12:18.359303 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.551187
I1122 15:12:18.359313 11005 solver.cpp:486] Iteration 1380, lr = 1e-05
I1122 15:12:47.296095 11005 solver.cpp:214] Iteration 1400, loss = 0.854063
I1122 15:12:47.296257 11005 solver.cpp:229]     Train net output #0: accuracy = 0.569267
I1122 15:12:47.296283 11005 solver.cpp:229]     Train net output #1: loss = 0.854063 (* 1 = 0.854063 loss)
I1122 15:12:47.296291 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.505111
I1122 15:12:47.296303 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.634591
I1122 15:12:47.296311 11005 solver.cpp:486] Iteration 1400, lr = 1e-05
I1122 15:13:16.205652 11005 solver.cpp:214] Iteration 1420, loss = 0.753392
I1122 15:13:16.205700 11005 solver.cpp:229]     Train net output #0: accuracy = 0.62878
I1122 15:13:16.205713 11005 solver.cpp:229]     Train net output #1: loss = 0.753392 (* 1 = 0.753392 loss)
I1122 15:13:16.205720 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.575342
I1122 15:13:16.205730 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.658526
I1122 15:13:16.205739 11005 solver.cpp:486] Iteration 1420, lr = 1e-05
I1122 15:13:45.126734 11005 solver.cpp:214] Iteration 1440, loss = 1.04236
I1122 15:13:45.126816 11005 solver.cpp:229]     Train net output #0: accuracy = 0.471054
I1122 15:13:45.126829 11005 solver.cpp:229]     Train net output #1: loss = 1.04236 (* 1 = 1.04236 loss)
I1122 15:13:45.126837 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.452152
I1122 15:13:45.126843 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.715684
I1122 15:13:45.126854 11005 solver.cpp:486] Iteration 1440, lr = 1e-05
I1122 15:14:14.047806 11005 solver.cpp:214] Iteration 1460, loss = 1.01239
I1122 15:14:14.047852 11005 solver.cpp:229]     Train net output #0: accuracy = 0.554996
I1122 15:14:14.047863 11005 solver.cpp:229]     Train net output #1: loss = 1.01239 (* 1 = 1.01239 loss)
I1122 15:14:14.047873 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.484695
I1122 15:14:14.047883 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.559878
I1122 15:14:14.047894 11005 solver.cpp:486] Iteration 1460, lr = 1e-05
I1122 15:14:42.995906 11005 solver.cpp:214] Iteration 1480, loss = 0.9325
I1122 15:14:42.995980 11005 solver.cpp:229]     Train net output #0: accuracy = 0.574352
I1122 15:14:42.995992 11005 solver.cpp:229]     Train net output #1: loss = 0.9325 (* 1 = 0.9325 loss)
I1122 15:14:42.996000 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.493291
I1122 15:14:42.996006 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.644833
I1122 15:14:42.996017 11005 solver.cpp:486] Iteration 1480, lr = 1e-05
I1122 15:15:11.910950 11005 solver.cpp:214] Iteration 1500, loss = 0.719655
I1122 15:15:11.910998 11005 solver.cpp:229]     Train net output #0: accuracy = 0.656666
I1122 15:15:11.911010 11005 solver.cpp:229]     Train net output #1: loss = 0.719655 (* 1 = 0.719655 loss)
I1122 15:15:11.911017 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.671905
I1122 15:15:11.911023 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.652867
I1122 15:15:11.911032 11005 solver.cpp:486] Iteration 1500, lr = 1e-05
I1122 15:15:40.809674 11005 solver.cpp:214] Iteration 1520, loss = 1.02002
I1122 15:15:40.809749 11005 solver.cpp:229]     Train net output #0: accuracy = 0.547531
I1122 15:15:40.809762 11005 solver.cpp:229]     Train net output #1: loss = 1.02002 (* 1 = 1.02002 loss)
I1122 15:15:40.809769 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.552589
I1122 15:15:40.809777 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.547371
I1122 15:15:40.809784 11005 solver.cpp:486] Iteration 1520, lr = 1e-05
I1122 15:16:09.749729 11005 solver.cpp:214] Iteration 1540, loss = 0.818422
I1122 15:16:09.749778 11005 solver.cpp:229]     Train net output #0: accuracy = 0.592842
I1122 15:16:09.749789 11005 solver.cpp:229]     Train net output #1: loss = 0.818422 (* 1 = 0.818422 loss)
I1122 15:16:09.749796 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.52079
I1122 15:16:09.749804 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.801704
I1122 15:16:09.749811 11005 solver.cpp:486] Iteration 1540, lr = 1e-05
I1122 15:16:38.678692 11005 solver.cpp:214] Iteration 1560, loss = 1.05281
I1122 15:16:38.678799 11005 solver.cpp:229]     Train net output #0: accuracy = 0.469379
I1122 15:16:38.678817 11005 solver.cpp:229]     Train net output #1: loss = 1.05281 (* 1 = 1.05281 loss)
I1122 15:16:38.678825 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.453277
I1122 15:16:38.678831 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.614063
I1122 15:16:38.678843 11005 solver.cpp:486] Iteration 1560, lr = 1e-05
I1122 15:17:07.576664 11005 solver.cpp:214] Iteration 1580, loss = 0.727661
I1122 15:17:07.576712 11005 solver.cpp:229]     Train net output #0: accuracy = 0.653267
I1122 15:17:07.576724 11005 solver.cpp:229]     Train net output #1: loss = 0.727661 (* 1 = 0.727661 loss)
I1122 15:17:07.576731 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.653742
I1122 15:17:07.576737 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.653129
I1122 15:17:07.576746 11005 solver.cpp:486] Iteration 1580, lr = 1e-05
I1122 15:17:36.478775 11005 solver.cpp:214] Iteration 1600, loss = 1.01626
I1122 15:17:36.478853 11005 solver.cpp:229]     Train net output #0: accuracy = 0.549759
I1122 15:17:36.478868 11005 solver.cpp:229]     Train net output #1: loss = 1.01626 (* 1 = 1.01626 loss)
I1122 15:17:36.478874 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.464993
I1122 15:17:36.478880 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.560959
I1122 15:17:36.478889 11005 solver.cpp:486] Iteration 1600, lr = 1e-05
I1122 15:18:05.354699 11005 solver.cpp:214] Iteration 1620, loss = 0.762092
I1122 15:18:05.354746 11005 solver.cpp:229]     Train net output #0: accuracy = 0.609447
I1122 15:18:05.354758 11005 solver.cpp:229]     Train net output #1: loss = 0.762092 (* 1 = 0.762092 loss)
I1122 15:18:05.354765 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.508019
I1122 15:18:05.354773 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.669474
I1122 15:18:05.354784 11005 solver.cpp:486] Iteration 1620, lr = 1e-05
I1122 15:18:34.275059 11005 solver.cpp:214] Iteration 1640, loss = 0.808368
I1122 15:18:34.275192 11005 solver.cpp:229]     Train net output #0: accuracy = 0.62183
I1122 15:18:34.275213 11005 solver.cpp:229]     Train net output #1: loss = 0.808368 (* 1 = 0.808368 loss)
I1122 15:18:34.275218 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.611927
I1122 15:18:34.275221 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.625121
I1122 15:18:34.275226 11005 solver.cpp:486] Iteration 1640, lr = 1e-05
I1122 15:19:03.209727 11005 solver.cpp:214] Iteration 1660, loss = 0.638226
I1122 15:19:03.209775 11005 solver.cpp:229]     Train net output #0: accuracy = 0.696476
I1122 15:19:03.209787 11005 solver.cpp:229]     Train net output #1: loss = 0.638226 (* 1 = 0.638226 loss)
I1122 15:19:03.209795 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.614529
I1122 15:19:03.209805 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.778062
I1122 15:19:03.209813 11005 solver.cpp:486] Iteration 1660, lr = 1e-05
I1122 15:19:32.140157 11005 solver.cpp:214] Iteration 1680, loss = 0.845279
I1122 15:19:32.140233 11005 solver.cpp:229]     Train net output #0: accuracy = 0.619938
I1122 15:19:32.140245 11005 solver.cpp:229]     Train net output #1: loss = 0.845279 (* 1 = 0.845279 loss)
I1122 15:19:32.140254 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.64298
I1122 15:19:32.140260 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.612786
I1122 15:19:32.140269 11005 solver.cpp:486] Iteration 1680, lr = 1e-05
I1122 15:20:01.039692 11005 solver.cpp:214] Iteration 1700, loss = 0.612381
I1122 15:20:01.039739 11005 solver.cpp:229]     Train net output #0: accuracy = 0.703598
I1122 15:20:01.039751 11005 solver.cpp:229]     Train net output #1: loss = 0.612381 (* 1 = 0.612381 loss)
I1122 15:20:01.039758 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.701778
I1122 15:20:01.039764 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.70422
I1122 15:20:01.039774 11005 solver.cpp:486] Iteration 1700, lr = 1e-05
I1122 15:20:29.953929 11005 solver.cpp:214] Iteration 1720, loss = 0.720251
I1122 15:20:29.954038 11005 solver.cpp:229]     Train net output #0: accuracy = 0.639412
I1122 15:20:29.954052 11005 solver.cpp:229]     Train net output #1: loss = 0.720251 (* 1 = 0.720251 loss)
I1122 15:20:29.954061 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.568635
I1122 15:20:29.954071 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.710662
I1122 15:20:29.954078 11005 solver.cpp:486] Iteration 1720, lr = 1e-05
I1122 15:20:58.862046 11005 solver.cpp:214] Iteration 1740, loss = 0.588179
I1122 15:20:58.862093 11005 solver.cpp:229]     Train net output #0: accuracy = 0.715115
I1122 15:20:58.862105 11005 solver.cpp:229]     Train net output #1: loss = 0.588179 (* 1 = 0.588179 loss)
I1122 15:20:58.862112 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.70699
I1122 15:20:58.862118 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.717933
I1122 15:20:58.862128 11005 solver.cpp:486] Iteration 1740, lr = 1e-05
I1122 15:21:27.765861 11005 solver.cpp:214] Iteration 1760, loss = 1.02545
I1122 15:21:27.765974 11005 solver.cpp:229]     Train net output #0: accuracy = 0.571774
I1122 15:21:27.765995 11005 solver.cpp:229]     Train net output #1: loss = 1.02545 (* 1 = 1.02545 loss)
I1122 15:21:27.766000 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.569016
I1122 15:21:27.766002 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.571813
I1122 15:21:27.766008 11005 solver.cpp:486] Iteration 1760, lr = 1e-05
I1122 15:21:56.715473 11005 solver.cpp:214] Iteration 1780, loss = 0.789206
I1122 15:21:56.715522 11005 solver.cpp:229]     Train net output #0: accuracy = 0.602875
I1122 15:21:56.715534 11005 solver.cpp:229]     Train net output #1: loss = 0.789206 (* 1 = 0.789206 loss)
I1122 15:21:56.715543 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.51426
I1122 15:21:56.715553 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.693102
I1122 15:21:56.715564 11005 solver.cpp:486] Iteration 1780, lr = 1e-05
I1122 15:22:25.624506 11005 solver.cpp:214] Iteration 1800, loss = 0.656397
I1122 15:22:25.624583 11005 solver.cpp:229]     Train net output #0: accuracy = 0.682453
I1122 15:22:25.624595 11005 solver.cpp:229]     Train net output #1: loss = 0.656397 (* 1 = 0.656397 loss)
I1122 15:22:25.624603 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.611176
I1122 15:22:25.624613 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.722128
I1122 15:22:25.624621 11005 solver.cpp:486] Iteration 1800, lr = 1e-05
I1122 15:22:54.551554 11005 solver.cpp:214] Iteration 1820, loss = 1.06029
I1122 15:22:54.551601 11005 solver.cpp:229]     Train net output #0: accuracy = 0.461971
I1122 15:22:54.551614 11005 solver.cpp:229]     Train net output #1: loss = 1.06029 (* 1 = 1.06029 loss)
I1122 15:22:54.551622 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.435767
I1122 15:22:54.551632 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.801096
I1122 15:22:54.551641 11005 solver.cpp:486] Iteration 1820, lr = 1e-05
I1122 15:23:23.461148 11005 solver.cpp:214] Iteration 1840, loss = 1.04549
I1122 15:23:23.461313 11005 solver.cpp:229]     Train net output #0: accuracy = 0.574635
I1122 15:23:23.461338 11005 solver.cpp:229]     Train net output #1: loss = 1.04549 (* 1 = 1.04549 loss)
I1122 15:23:23.461346 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.478762
I1122 15:23:23.461354 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.581292
I1122 15:23:23.461364 11005 solver.cpp:486] Iteration 1840, lr = 1e-05
I1122 15:23:52.398573 11005 solver.cpp:214] Iteration 1860, loss = 0.924413
I1122 15:23:52.398622 11005 solver.cpp:229]     Train net output #0: accuracy = 0.599583
I1122 15:23:52.398634 11005 solver.cpp:229]     Train net output #1: loss = 0.924413 (* 1 = 0.924413 loss)
I1122 15:23:52.398641 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.498761
I1122 15:23:52.398648 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.687243
I1122 15:23:52.398658 11005 solver.cpp:486] Iteration 1860, lr = 1e-05
I1122 15:24:21.306133 11005 solver.cpp:214] Iteration 1880, loss = 0.629887
I1122 15:24:21.306275 11005 solver.cpp:229]     Train net output #0: accuracy = 0.711987
I1122 15:24:21.306300 11005 solver.cpp:229]     Train net output #1: loss = 0.629887 (* 1 = 0.629887 loss)
I1122 15:24:21.306309 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.727843
I1122 15:24:21.306316 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.708034
I1122 15:24:21.306325 11005 solver.cpp:486] Iteration 1880, lr = 1e-05
I1122 15:24:50.208884 11005 solver.cpp:214] Iteration 1900, loss = 1.05595
I1122 15:24:50.208933 11005 solver.cpp:229]     Train net output #0: accuracy = 0.56115
I1122 15:24:50.208946 11005 solver.cpp:229]     Train net output #1: loss = 1.05595 (* 1 = 1.05595 loss)
I1122 15:24:50.208953 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.57432
I1122 15:24:50.208963 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.560732
I1122 15:24:50.208972 11005 solver.cpp:486] Iteration 1900, lr = 1e-05
I1122 15:25:19.145054 11005 solver.cpp:214] Iteration 1920, loss = 0.749918
I1122 15:25:19.145187 11005 solver.cpp:229]     Train net output #0: accuracy = 0.623341
I1122 15:25:19.145210 11005 solver.cpp:229]     Train net output #1: loss = 0.749918 (* 1 = 0.749918 loss)
I1122 15:25:19.145215 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.53455
I1122 15:25:19.145218 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.880722
I1122 15:25:19.145225 11005 solver.cpp:486] Iteration 1920, lr = 1e-05
I1122 15:25:48.059375 11005 solver.cpp:214] Iteration 1940, loss = 1.07093
I1122 15:25:48.059422 11005 solver.cpp:229]     Train net output #0: accuracy = 0.463638
I1122 15:25:48.059434 11005 solver.cpp:229]     Train net output #1: loss = 1.07093 (* 1 = 1.07093 loss)
I1122 15:25:48.059443 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.441653
I1122 15:25:48.059453 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.661182
I1122 15:25:48.059464 11005 solver.cpp:486] Iteration 1940, lr = 1e-05
I1122 15:26:16.986161 11005 solver.cpp:214] Iteration 1960, loss = 0.637966
I1122 15:26:16.986237 11005 solver.cpp:229]     Train net output #0: accuracy = 0.709244
I1122 15:26:16.986248 11005 solver.cpp:229]     Train net output #1: loss = 0.637966 (* 1 = 0.637966 loss)
I1122 15:26:16.986255 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.708531
I1122 15:26:16.986263 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.709451
I1122 15:26:16.986273 11005 solver.cpp:486] Iteration 1960, lr = 1e-05
I1122 15:26:45.913492 11005 solver.cpp:214] Iteration 1980, loss = 1.03878
I1122 15:26:45.913538 11005 solver.cpp:229]     Train net output #0: accuracy = 0.561859
I1122 15:26:45.913550 11005 solver.cpp:229]     Train net output #1: loss = 1.03878 (* 1 = 1.03878 loss)
I1122 15:26:45.913558 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.460711
I1122 15:26:45.913568 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.575224
I1122 15:26:45.913576 11005 solver.cpp:486] Iteration 1980, lr = 1e-05
I1122 15:27:14.234902 11005 solver.cpp:361] Snapshotting to /home/gradescan/SegNet/Models/backup_basic/Training/segnet_basic_iter_2000.caffemodel
I1122 15:27:14.245158 11005 solver.cpp:369] Snapshotting solver state to /home/gradescan/SegNet/Models/backup_basic/Training/segnet_basic_iter_2000.solverstate
I1122 15:27:14.834743 11005 solver.cpp:214] Iteration 2000, loss = 0.694289
I1122 15:27:14.834786 11005 solver.cpp:229]     Train net output #0: accuracy = 0.648342
I1122 15:27:14.834799 11005 solver.cpp:229]     Train net output #1: loss = 0.694289 (* 1 = 0.694289 loss)
I1122 15:27:14.834810 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.517027
I1122 15:27:14.834817 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.726056
I1122 15:27:14.834831 11005 solver.cpp:486] Iteration 2000, lr = 1e-05
I1122 15:27:43.748435 11005 solver.cpp:214] Iteration 2020, loss = 0.735696
I1122 15:27:43.748482 11005 solver.cpp:229]     Train net output #0: accuracy = 0.662209
I1122 15:27:43.748494 11005 solver.cpp:229]     Train net output #1: loss = 0.735696 (* 1 = 0.735696 loss)
I1122 15:27:43.748502 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.660105
I1122 15:27:43.748507 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.662907
I1122 15:27:43.748517 11005 solver.cpp:486] Iteration 2020, lr = 1e-05
I1122 15:28:12.680444 11005 solver.cpp:214] Iteration 2040, loss = 0.533322
I1122 15:28:12.680572 11005 solver.cpp:229]     Train net output #0: accuracy = 0.753784
I1122 15:28:12.680593 11005 solver.cpp:229]     Train net output #1: loss = 0.533322 (* 1 = 0.533322 loss)
I1122 15:28:12.680598 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.654749
I1122 15:28:12.680601 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.852384
I1122 15:28:12.680606 11005 solver.cpp:486] Iteration 2040, lr = 1e-05
I1122 15:28:41.619139 11005 solver.cpp:214] Iteration 2060, loss = 0.763561
I1122 15:28:41.619187 11005 solver.cpp:229]     Train net output #0: accuracy = 0.668018
I1122 15:28:41.619199 11005 solver.cpp:229]     Train net output #1: loss = 0.763561 (* 1 = 0.763561 loss)
I1122 15:28:41.619206 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.720997
I1122 15:28:41.619213 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.651576
I1122 15:28:41.619222 11005 solver.cpp:486] Iteration 2060, lr = 1e-05
I1122 15:29:10.538887 11005 solver.cpp:214] Iteration 2080, loss = 0.513308
I1122 15:29:10.538961 11005 solver.cpp:229]     Train net output #0: accuracy = 0.762936
I1122 15:29:10.538975 11005 solver.cpp:229]     Train net output #1: loss = 0.513308 (* 1 = 0.513308 loss)
I1122 15:29:10.538981 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.759748
I1122 15:29:10.538988 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.764025
I1122 15:29:10.538996 11005 solver.cpp:486] Iteration 2080, lr = 1e-05
I1122 15:29:39.462870 11005 solver.cpp:214] Iteration 2100, loss = 0.629658
I1122 15:29:39.462916 11005 solver.cpp:229]     Train net output #0: accuracy = 0.688362
I1122 15:29:39.462929 11005 solver.cpp:229]     Train net output #1: loss = 0.629658 (* 1 = 0.629658 loss)
I1122 15:29:39.462936 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.597804
I1122 15:29:39.462946 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.779525
I1122 15:29:39.462954 11005 solver.cpp:486] Iteration 2100, lr = 1e-05
I1122 15:30:08.382310 11005 solver.cpp:214] Iteration 2120, loss = 0.483207
I1122 15:30:08.382446 11005 solver.cpp:229]     Train net output #0: accuracy = 0.778202
I1122 15:30:08.382470 11005 solver.cpp:229]     Train net output #1: loss = 0.483207 (* 1 = 0.483207 loss)
I1122 15:30:08.382478 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.762772
I1122 15:30:08.382484 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.783554
I1122 15:30:08.382493 11005 solver.cpp:486] Iteration 2120, lr = 1e-05
I1122 15:30:37.288358 11005 solver.cpp:214] Iteration 2140, loss = 1.05982
I1122 15:30:37.288408 11005 solver.cpp:229]     Train net output #0: accuracy = 0.589947
I1122 15:30:37.288421 11005 solver.cpp:229]     Train net output #1: loss = 1.05982 (* 1 = 1.05982 loss)
I1122 15:30:37.288429 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.605736
I1122 15:30:37.288439 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.589719
I1122 15:30:37.288446 11005 solver.cpp:486] Iteration 2140, lr = 1e-05
I1122 15:31:06.233793 11005 solver.cpp:214] Iteration 2160, loss = 0.735861
I1122 15:31:06.233901 11005 solver.cpp:229]     Train net output #0: accuracy = 0.635601
I1122 15:31:06.233916 11005 solver.cpp:229]     Train net output #1: loss = 0.735861 (* 1 = 0.735861 loss)
I1122 15:31:06.233923 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.528423
I1122 15:31:06.233932 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.74473
I1122 15:31:06.233940 11005 solver.cpp:486] Iteration 2160, lr = 1e-05
I1122 15:31:35.137053 11005 solver.cpp:214] Iteration 2180, loss = 0.580635
I1122 15:31:35.137102 11005 solver.cpp:229]     Train net output #0: accuracy = 0.726414
I1122 15:31:35.137114 11005 solver.cpp:229]     Train net output #1: loss = 0.580635 (* 1 = 0.580635 loss)
I1122 15:31:35.137121 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.646199
I1122 15:31:35.137130 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.771064
I1122 15:31:35.137138 11005 solver.cpp:486] Iteration 2180, lr = 1e-05
I1122 15:32:04.049667 11005 solver.cpp:214] Iteration 2200, loss = 1.08735
I1122 15:32:04.049808 11005 solver.cpp:229]     Train net output #0: accuracy = 0.459427
I1122 15:32:04.049834 11005 solver.cpp:229]     Train net output #1: loss = 1.08735 (* 1 = 1.08735 loss)
I1122 15:32:04.049841 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.428033
I1122 15:32:04.049849 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.865713
I1122 15:32:04.049862 11005 solver.cpp:486] Iteration 2200, lr = 1e-05
I1122 15:32:32.955325 11005 solver.cpp:214] Iteration 2220, loss = 1.09647
I1122 15:32:32.955373 11005 solver.cpp:229]     Train net output #0: accuracy = 0.587978
I1122 15:32:32.955384 11005 solver.cpp:229]     Train net output #1: loss = 1.09647 (* 1 = 1.09647 loss)
I1122 15:32:32.955392 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.466248
I1122 15:32:32.955401 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.596431
I1122 15:32:32.955412 11005 solver.cpp:486] Iteration 2220, lr = 1e-05
I1122 15:33:01.882738 11005 solver.cpp:214] Iteration 2240, loss = 0.922291
I1122 15:33:01.882814 11005 solver.cpp:229]     Train net output #0: accuracy = 0.621597
I1122 15:33:01.882828 11005 solver.cpp:229]     Train net output #1: loss = 0.922291 (* 1 = 0.922291 loss)
I1122 15:33:01.882839 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.510671
I1122 15:33:01.882846 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.718044
I1122 15:33:01.882855 11005 solver.cpp:486] Iteration 2240, lr = 1e-05
I1122 15:33:30.818325 11005 solver.cpp:214] Iteration 2260, loss = 0.566483
I1122 15:33:30.818372 11005 solver.cpp:229]     Train net output #0: accuracy = 0.75592
I1122 15:33:30.818384 11005 solver.cpp:229]     Train net output #1: loss = 0.566483 (* 1 = 0.566483 loss)
I1122 15:33:30.818392 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.774757
I1122 15:33:30.818397 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.751225
I1122 15:33:30.818406 11005 solver.cpp:486] Iteration 2260, lr = 1e-05
I1122 15:33:59.741935 11005 solver.cpp:214] Iteration 2280, loss = 1.10819
I1122 15:33:59.742010 11005 solver.cpp:229]     Train net output #0: accuracy = 0.567814
I1122 15:33:59.742024 11005 solver.cpp:229]     Train net output #1: loss = 1.10819 (* 1 = 1.10819 loss)
I1122 15:33:59.742032 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.58711
I1122 15:33:59.742040 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.567202
I1122 15:33:59.742049 11005 solver.cpp:486] Iteration 2280, lr = 1e-05
I1122 15:34:28.642221 11005 solver.cpp:214] Iteration 2300, loss = 0.694153
I1122 15:34:28.642269 11005 solver.cpp:229]     Train net output #0: accuracy = 0.64817
I1122 15:34:28.642282 11005 solver.cpp:229]     Train net output #1: loss = 0.694153 (* 1 = 0.694153 loss)
I1122 15:34:28.642289 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.551076
I1122 15:34:28.642297 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.929623
I1122 15:34:28.642305 11005 solver.cpp:486] Iteration 2300, lr = 1e-05
I1122 15:34:57.576838 11005 solver.cpp:214] Iteration 2320, loss = 1.09412
I1122 15:34:57.576943 11005 solver.cpp:229]     Train net output #0: accuracy = 0.459759
I1122 15:34:57.576957 11005 solver.cpp:229]     Train net output #1: loss = 1.09412 (* 1 = 1.09412 loss)
I1122 15:34:57.576966 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.433179
I1122 15:34:57.576972 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.698587
I1122 15:34:57.576980 11005 solver.cpp:486] Iteration 2320, lr = 1e-05
I1122 15:35:26.483737 11005 solver.cpp:214] Iteration 2340, loss = 0.571141
I1122 15:35:26.483785 11005 solver.cpp:229]     Train net output #0: accuracy = 0.75267
I1122 15:35:26.483798 11005 solver.cpp:229]     Train net output #1: loss = 0.571141 (* 1 = 0.571141 loss)
I1122 15:35:26.483805 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.749373
I1122 15:35:26.483814 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.753628
I1122 15:35:26.483822 11005 solver.cpp:486] Iteration 2340, lr = 1e-05
I1122 15:35:55.419081 11005 solver.cpp:214] Iteration 2360, loss = 1.06747
I1122 15:35:55.419209 11005 solver.cpp:229]     Train net output #0: accuracy = 0.570049
I1122 15:35:55.419229 11005 solver.cpp:229]     Train net output #1: loss = 1.06747 (* 1 = 1.06747 loss)
I1122 15:35:55.419234 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.463293
I1122 15:35:55.419239 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.584155
I1122 15:35:55.419244 11005 solver.cpp:486] Iteration 2360, lr = 1e-05
I1122 15:36:24.319002 11005 solver.cpp:214] Iteration 2380, loss = 0.643238
I1122 15:36:24.319048 11005 solver.cpp:229]     Train net output #0: accuracy = 0.680843
I1122 15:36:24.319061 11005 solver.cpp:229]     Train net output #1: loss = 0.643238 (* 1 = 0.643238 loss)
I1122 15:36:24.319069 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.531392
I1122 15:36:24.319077 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.76929
I1122 15:36:24.319088 11005 solver.cpp:486] Iteration 2380, lr = 1e-05
I1122 15:36:53.249456 11005 solver.cpp:214] Iteration 2400, loss = 0.676151
I1122 15:36:53.249529 11005 solver.cpp:229]     Train net output #0: accuracy = 0.695271
I1122 15:36:53.249542 11005 solver.cpp:229]     Train net output #1: loss = 0.676151 (* 1 = 0.676151 loss)
I1122 15:36:53.249549 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.697363
I1122 15:36:53.249557 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.694575
I1122 15:36:53.249565 11005 solver.cpp:486] Iteration 2400, lr = 1e-05
I1122 15:37:22.179039 11005 solver.cpp:214] Iteration 2420, loss = 0.456089
I1122 15:37:22.179086 11005 solver.cpp:229]     Train net output #0: accuracy = 0.794628
I1122 15:37:22.179098 11005 solver.cpp:229]     Train net output #1: loss = 0.456089 (* 1 = 0.456089 loss)
I1122 15:37:22.179106 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.687092
I1122 15:37:22.179113 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.901691
I1122 15:37:22.179122 11005 solver.cpp:486] Iteration 2420, lr = 1e-05
I1122 15:37:51.075827 11005 solver.cpp:214] Iteration 2440, loss = 0.692412
I1122 15:37:51.075938 11005 solver.cpp:229]     Train net output #0: accuracy = 0.705982
I1122 15:37:51.075953 11005 solver.cpp:229]     Train net output #1: loss = 0.692412 (* 1 = 0.692412 loss)
I1122 15:37:51.075960 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.781717
I1122 15:37:51.075969 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.682477
I1122 15:37:51.075978 11005 solver.cpp:486] Iteration 2440, lr = 1e-05
I1122 15:38:19.969665 11005 solver.cpp:214] Iteration 2460, loss = 0.445667
I1122 15:38:19.969713 11005 solver.cpp:229]     Train net output #0: accuracy = 0.805859
I1122 15:38:19.969725 11005 solver.cpp:229]     Train net output #1: loss = 0.445667 (* 1 = 0.445667 loss)
I1122 15:38:19.969732 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.80521
I1122 15:38:19.969739 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.80608
I1122 15:38:19.969748 11005 solver.cpp:486] Iteration 2460, lr = 1e-05
I1122 15:38:48.922511 11005 solver.cpp:214] Iteration 2480, loss = 0.5588
I1122 15:38:48.922590 11005 solver.cpp:229]     Train net output #0: accuracy = 0.727486
I1122 15:38:48.922602 11005 solver.cpp:229]     Train net output #1: loss = 0.558799 (* 1 = 0.558799 loss)
I1122 15:38:48.922610 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.626091
I1122 15:38:48.922616 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.829557
I1122 15:38:48.922626 11005 solver.cpp:486] Iteration 2480, lr = 1e-05
I1122 15:39:17.823187 11005 solver.cpp:214] Iteration 2500, loss = 0.409789
I1122 15:39:17.823231 11005 solver.cpp:229]     Train net output #0: accuracy = 0.823292
I1122 15:39:17.823243 11005 solver.cpp:229]     Train net output #1: loss = 0.409789 (* 1 = 0.409789 loss)
I1122 15:39:17.823251 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.80383
I1122 15:39:17.823258 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.830042
I1122 15:39:17.823267 11005 solver.cpp:486] Iteration 2500, lr = 1e-05
I1122 15:39:46.738936 11005 solver.cpp:214] Iteration 2520, loss = 1.09823
I1122 15:39:46.739063 11005 solver.cpp:229]     Train net output #0: accuracy = 0.599304
I1122 15:39:46.739084 11005 solver.cpp:229]     Train net output #1: loss = 1.09823 (* 1 = 1.09823 loss)
I1122 15:39:46.739089 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.627714
I1122 15:39:46.739092 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.598894
I1122 15:39:46.739099 11005 solver.cpp:486] Iteration 2520, lr = 1e-05
I1122 15:40:15.650358 11005 solver.cpp:214] Iteration 2540, loss = 0.698395
I1122 15:40:15.650406 11005 solver.cpp:229]     Train net output #0: accuracy = 0.663055
I1122 15:40:15.650418 11005 solver.cpp:229]     Train net output #1: loss = 0.698394 (* 1 = 0.698394 loss)
I1122 15:40:15.650426 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.545307
I1122 15:40:15.650434 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.782947
I1122 15:40:15.650444 11005 solver.cpp:486] Iteration 2540, lr = 1e-05
I1122 15:40:44.606662 11005 solver.cpp:214] Iteration 2560, loss = 0.525623
I1122 15:40:44.606792 11005 solver.cpp:229]     Train net output #0: accuracy = 0.758244
I1122 15:40:44.606820 11005 solver.cpp:229]     Train net output #1: loss = 0.525623 (* 1 = 0.525623 loss)
I1122 15:40:44.606827 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.678394
I1122 15:40:44.606829 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.80269
I1122 15:40:44.606835 11005 solver.cpp:486] Iteration 2560, lr = 1e-05
I1122 15:41:13.517115 11005 solver.cpp:214] Iteration 2580, loss = 1.11505
I1122 15:41:13.517163 11005 solver.cpp:229]     Train net output #0: accuracy = 0.461617
I1122 15:41:13.517177 11005 solver.cpp:229]     Train net output #1: loss = 1.11505 (* 1 = 1.11505 loss)
I1122 15:41:13.517185 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.42715
I1122 15:41:13.517199 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.907674
I1122 15:41:13.517207 11005 solver.cpp:486] Iteration 2580, lr = 1e-05
I1122 15:41:42.420253 11005 solver.cpp:214] Iteration 2600, loss = 1.15037
I1122 15:41:42.420382 11005 solver.cpp:229]     Train net output #0: accuracy = 0.59375
I1122 15:41:42.420395 11005 solver.cpp:229]     Train net output #1: loss = 1.15037 (* 1 = 1.15037 loss)
I1122 15:41:42.420403 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.454204
I1122 15:41:42.420410 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.60344
I1122 15:41:42.420418 11005 solver.cpp:486] Iteration 2600, lr = 1e-05
I1122 15:42:11.336453 11005 solver.cpp:214] Iteration 2620, loss = 0.913005
I1122 15:42:11.336499 11005 solver.cpp:229]     Train net output #0: accuracy = 0.641281
I1122 15:42:11.336511 11005 solver.cpp:229]     Train net output #1: loss = 0.913005 (* 1 = 0.913005 loss)
I1122 15:42:11.336519 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.531504
I1122 15:42:11.336525 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.736728
I1122 15:42:11.336534 11005 solver.cpp:486] Iteration 2620, lr = 1e-05
I1122 15:42:40.257264 11005 solver.cpp:214] Iteration 2640, loss = 0.529114
I1122 15:42:40.257403 11005 solver.cpp:229]     Train net output #0: accuracy = 0.78421
I1122 15:42:40.257424 11005 solver.cpp:229]     Train net output #1: loss = 0.529114 (* 1 = 0.529114 loss)
I1122 15:42:40.257429 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.803816
I1122 15:42:40.257432 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.779323
I1122 15:42:40.257438 11005 solver.cpp:486] Iteration 2640, lr = 1e-05
I1122 15:43:09.170604 11005 solver.cpp:214] Iteration 2660, loss = 1.16223
I1122 15:43:09.170660 11005 solver.cpp:229]     Train net output #0: accuracy = 0.571663
I1122 15:43:09.170673 11005 solver.cpp:229]     Train net output #1: loss = 1.16223 (* 1 = 1.16223 loss)
I1122 15:43:09.170681 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.599777
I1122 15:43:09.170691 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.570772
I1122 15:43:09.170701 11005 solver.cpp:486] Iteration 2660, lr = 1e-05
I1122 15:43:38.109215 11005 solver.cpp:214] Iteration 2680, loss = 0.649892
I1122 15:43:38.109351 11005 solver.cpp:229]     Train net output #0: accuracy = 0.66869
I1122 15:43:38.109372 11005 solver.cpp:229]     Train net output #1: loss = 0.649892 (* 1 = 0.649892 loss)
I1122 15:43:38.109377 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.569131
I1122 15:43:38.109380 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.957286
I1122 15:43:38.109386 11005 solver.cpp:486] Iteration 2680, lr = 1e-05
I1122 15:44:07.048812 11005 solver.cpp:214] Iteration 2700, loss = 1.11477
I1122 15:44:07.048861 11005 solver.cpp:229]     Train net output #0: accuracy = 0.459213
I1122 15:44:07.048872 11005 solver.cpp:229]     Train net output #1: loss = 1.11477 (* 1 = 1.11477 loss)
I1122 15:44:07.048880 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.42947
I1122 15:44:07.048888 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.726469
I1122 15:44:07.048897 11005 solver.cpp:486] Iteration 2700, lr = 1e-05
I1122 15:44:35.973354 11005 solver.cpp:214] Iteration 2720, loss = 0.52698
I1122 15:44:35.973495 11005 solver.cpp:229]     Train net output #0: accuracy = 0.781391
I1122 15:44:35.973529 11005 solver.cpp:229]     Train net output #1: loss = 0.52698 (* 1 = 0.52698 loss)
I1122 15:44:35.973548 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.774573
I1122 15:44:35.973562 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.783372
I1122 15:44:35.973578 11005 solver.cpp:486] Iteration 2720, lr = 1e-05
I1122 15:45:05.140769 11005 solver.cpp:214] Iteration 2740, loss = 1.09169
I1122 15:45:05.140818 11005 solver.cpp:229]     Train net output #0: accuracy = 0.575268
I1122 15:45:05.140831 11005 solver.cpp:229]     Train net output #1: loss = 1.09169 (* 1 = 1.09169 loss)
I1122 15:45:05.140840 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.46934
I1122 15:45:05.140848 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.589264
I1122 15:45:05.140857 11005 solver.cpp:486] Iteration 2740, lr = 1e-05
I1122 15:45:34.055796 11005 solver.cpp:214] Iteration 2760, loss = 0.605937
I1122 15:45:34.055966 11005 solver.cpp:229]     Train net output #0: accuracy = 0.70414
I1122 15:45:34.055989 11005 solver.cpp:229]     Train net output #1: loss = 0.605937 (* 1 = 0.605937 loss)
I1122 15:45:34.055994 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.545541
I1122 15:45:34.055996 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.798
I1122 15:45:34.056002 11005 solver.cpp:486] Iteration 2760, lr = 1e-05
I1122 15:46:02.993952 11005 solver.cpp:214] Iteration 2780, loss = 0.629938
I1122 15:46:02.994000 11005 solver.cpp:229]     Train net output #0: accuracy = 0.719334
I1122 15:46:02.994014 11005 solver.cpp:229]     Train net output #1: loss = 0.629938 (* 1 = 0.629938 loss)
I1122 15:46:02.994020 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.720932
I1122 15:46:02.994030 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.718802
I1122 15:46:02.994040 11005 solver.cpp:486] Iteration 2780, lr = 1e-05
I1122 15:46:31.907559 11005 solver.cpp:214] Iteration 2800, loss = 0.403338
I1122 15:46:31.907691 11005 solver.cpp:229]     Train net output #0: accuracy = 0.821754
I1122 15:46:31.907716 11005 solver.cpp:229]     Train net output #1: loss = 0.403338 (* 1 = 0.403338 loss)
I1122 15:46:31.907726 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.713395
I1122 15:46:31.907735 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.929637
I1122 15:46:31.907744 11005 solver.cpp:486] Iteration 2800, lr = 1e-05
I1122 15:47:00.829582 11005 solver.cpp:214] Iteration 2820, loss = 0.639616
I1122 15:47:00.829632 11005 solver.cpp:229]     Train net output #0: accuracy = 0.73222
I1122 15:47:00.829644 11005 solver.cpp:229]     Train net output #1: loss = 0.639616 (* 1 = 0.639616 loss)
I1122 15:47:00.829651 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.821546
I1122 15:47:00.829659 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.704496
I1122 15:47:00.829668 11005 solver.cpp:486] Iteration 2820, lr = 1e-05
I1122 15:47:29.745054 11005 solver.cpp:214] Iteration 2840, loss = 0.401696
I1122 15:47:29.745185 11005 solver.cpp:229]     Train net output #0: accuracy = 0.832336
I1122 15:47:29.745213 11005 solver.cpp:229]     Train net output #1: loss = 0.401696 (* 1 = 0.401696 loss)
I1122 15:47:29.745224 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.836756
I1122 15:47:29.745232 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.830826
I1122 15:47:29.745241 11005 solver.cpp:486] Iteration 2840, lr = 1e-05
I1122 15:47:58.678316 11005 solver.cpp:214] Iteration 2860, loss = 0.507238
I1122 15:47:58.678365 11005 solver.cpp:229]     Train net output #0: accuracy = 0.756615
I1122 15:47:58.678378 11005 solver.cpp:229]     Train net output #1: loss = 0.507238 (* 1 = 0.507238 loss)
I1122 15:47:58.678385 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.652417
I1122 15:47:58.678392 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.861508
I1122 15:47:58.678401 11005 solver.cpp:486] Iteration 2860, lr = 1e-05
I1122 15:48:27.590745 11005 solver.cpp:214] Iteration 2880, loss = 0.361183
I1122 15:48:27.590874 11005 solver.cpp:229]     Train net output #0: accuracy = 0.852962
I1122 15:48:27.590899 11005 solver.cpp:229]     Train net output #1: loss = 0.361183 (* 1 = 0.361183 loss)
I1122 15:48:27.590906 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.829944
I1122 15:48:27.590914 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.860947
I1122 15:48:27.590930 11005 solver.cpp:486] Iteration 2880, lr = 1e-05
I1122 15:48:56.549228 11005 solver.cpp:214] Iteration 2900, loss = 1.13249
I1122 15:48:56.549276 11005 solver.cpp:229]     Train net output #0: accuracy = 0.602947
I1122 15:48:56.549289 11005 solver.cpp:229]     Train net output #1: loss = 1.13249 (* 1 = 1.13249 loss)
I1122 15:48:56.549296 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.645403
I1122 15:48:56.549305 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.602334
I1122 15:48:56.549314 11005 solver.cpp:486] Iteration 2900, lr = 1e-05
I1122 15:49:25.463740 11005 solver.cpp:214] Iteration 2920, loss = 0.673653
I1122 15:49:25.463899 11005 solver.cpp:229]     Train net output #0: accuracy = 0.68166
I1122 15:49:25.463923 11005 solver.cpp:229]     Train net output #1: loss = 0.673653 (* 1 = 0.673653 loss)
I1122 15:49:25.463932 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.561684
I1122 15:49:25.463943 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.803819
I1122 15:49:25.463951 11005 solver.cpp:486] Iteration 2920, lr = 1e-05
I1122 15:49:54.345441 11005 solver.cpp:214] Iteration 2940, loss = 0.486935
I1122 15:49:54.345490 11005 solver.cpp:229]     Train net output #0: accuracy = 0.780136
I1122 15:49:54.345504 11005 solver.cpp:229]     Train net output #1: loss = 0.486935 (* 1 = 0.486935 loss)
I1122 15:49:54.345510 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.71107
I1122 15:49:54.345517 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.81858
I1122 15:49:54.345526 11005 solver.cpp:486] Iteration 2940, lr = 1e-05
I1122 15:50:22.380455 11005 solver.cpp:214] Iteration 2960, loss = 1.1355
I1122 15:50:22.380594 11005 solver.cpp:229]     Train net output #0: accuracy = 0.465954
I1122 15:50:22.380620 11005 solver.cpp:229]     Train net output #1: loss = 1.1355 (* 1 = 1.1355 loss)
I1122 15:50:22.380627 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.429989
I1122 15:50:22.380633 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.931394
I1122 15:50:22.380643 11005 solver.cpp:486] Iteration 2960, lr = 1e-05
I1122 15:50:51.274935 11005 solver.cpp:214] Iteration 2980, loss = 1.19513
I1122 15:50:51.274982 11005 solver.cpp:229]     Train net output #0: accuracy = 0.596027
I1122 15:50:51.274996 11005 solver.cpp:229]     Train net output #1: loss = 1.19512 (* 1 = 1.19512 loss)
I1122 15:50:51.275003 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.447565
I1122 15:50:51.275010 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.606336
I1122 15:50:51.275020 11005 solver.cpp:486] Iteration 2980, lr = 1e-05
I1122 15:51:19.679711 11005 solver.cpp:361] Snapshotting to /home/gradescan/SegNet/Models/backup_basic/Training/segnet_basic_iter_3000.caffemodel
I1122 15:51:19.693029 11005 solver.cpp:369] Snapshotting solver state to /home/gradescan/SegNet/Models/backup_basic/Training/segnet_basic_iter_3000.solverstate
I1122 15:51:20.204762 11005 solver.cpp:214] Iteration 3000, loss = 0.888425
I1122 15:51:20.204829 11005 solver.cpp:229]     Train net output #0: accuracy = 0.660671
I1122 15:51:20.204841 11005 solver.cpp:229]     Train net output #1: loss = 0.888425 (* 1 = 0.888425 loss)
I1122 15:51:20.204849 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.561819
I1122 15:51:20.204854 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.74662
I1122 15:51:20.204864 11005 solver.cpp:486] Iteration 3000, lr = 1e-05
I1122 15:51:49.145902 11005 solver.cpp:214] Iteration 3020, loss = 0.508919
I1122 15:51:49.145951 11005 solver.cpp:229]     Train net output #0: accuracy = 0.802635
I1122 15:51:49.145962 11005 solver.cpp:229]     Train net output #1: loss = 0.508918 (* 1 = 0.508918 loss)
I1122 15:51:49.145969 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.824195
I1122 15:51:49.145977 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.797261
I1122 15:51:49.145987 11005 solver.cpp:486] Iteration 3020, lr = 1e-05
I1122 15:52:18.057237 11005 solver.cpp:214] Iteration 3040, loss = 1.20567
I1122 15:52:18.057402 11005 solver.cpp:229]     Train net output #0: accuracy = 0.572128
I1122 15:52:18.057425 11005 solver.cpp:229]     Train net output #1: loss = 1.20567 (* 1 = 1.20567 loss)
I1122 15:52:18.057430 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.605489
I1122 15:52:18.057433 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.571071
I1122 15:52:18.057440 11005 solver.cpp:486] Iteration 3040, lr = 1e-05
I1122 15:52:46.994571 11005 solver.cpp:214] Iteration 3060, loss = 0.612739
I1122 15:52:46.994621 11005 solver.cpp:229]     Train net output #0: accuracy = 0.68718
I1122 15:52:46.994633 11005 solver.cpp:229]     Train net output #1: loss = 0.612739 (* 1 = 0.612739 loss)
I1122 15:52:46.994642 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.588894
I1122 15:52:46.994649 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.972084
I1122 15:52:46.994658 11005 solver.cpp:486] Iteration 3060, lr = 1e-05
I1122 15:53:15.904383 11005 solver.cpp:214] Iteration 3080, loss = 1.12893
I1122 15:53:15.904458 11005 solver.cpp:229]     Train net output #0: accuracy = 0.460819
I1122 15:53:15.904471 11005 solver.cpp:229]     Train net output #1: loss = 1.12893 (* 1 = 1.12893 loss)
I1122 15:53:15.904480 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.429266
I1122 15:53:15.904489 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.744334
I1122 15:53:15.904500 11005 solver.cpp:486] Iteration 3080, lr = 1e-05
I1122 15:53:44.832597 11005 solver.cpp:214] Iteration 3100, loss = 0.498883
I1122 15:53:44.832645 11005 solver.cpp:229]     Train net output #0: accuracy = 0.799213
I1122 15:53:44.832658 11005 solver.cpp:229]     Train net output #1: loss = 0.498883 (* 1 = 0.498883 loss)
I1122 15:53:44.832664 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.788537
I1122 15:53:44.832672 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.802315
I1122 15:53:44.832681 11005 solver.cpp:486] Iteration 3100, lr = 1e-05
I1122 15:54:13.750442 11005 solver.cpp:214] Iteration 3120, loss = 1.10659
I1122 15:54:13.750514 11005 solver.cpp:229]     Train net output #0: accuracy = 0.57819
I1122 15:54:13.750527 11005 solver.cpp:229]     Train net output #1: loss = 1.10659 (* 1 = 1.10659 loss)
I1122 15:54:13.750535 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.475747
I1122 15:54:13.750541 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.591725
I1122 15:54:13.750550 11005 solver.cpp:486] Iteration 3120, lr = 1e-05
I1122 15:54:42.680635 11005 solver.cpp:214] Iteration 3140, loss = 0.577625
I1122 15:54:42.680682 11005 solver.cpp:229]     Train net output #0: accuracy = 0.721794
I1122 15:54:42.680694 11005 solver.cpp:229]     Train net output #1: loss = 0.577625 (* 1 = 0.577625 loss)
I1122 15:54:42.680702 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.562912
I1122 15:54:42.680709 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.815822
I1122 15:54:42.680721 11005 solver.cpp:486] Iteration 3140, lr = 1e-05
I1122 15:55:11.609710 11005 solver.cpp:214] Iteration 3160, loss = 0.594362
I1122 15:55:11.609786 11005 solver.cpp:229]     Train net output #0: accuracy = 0.73867
I1122 15:55:11.609798 11005 solver.cpp:229]     Train net output #1: loss = 0.594362 (* 1 = 0.594362 loss)
I1122 15:55:11.609807 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.735692
I1122 15:55:11.609815 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.73966
I1122 15:55:11.609824 11005 solver.cpp:486] Iteration 3160, lr = 1e-05
I1122 15:55:40.503340 11005 solver.cpp:214] Iteration 3180, loss = 0.365441
I1122 15:55:40.503389 11005 solver.cpp:229]     Train net output #0: accuracy = 0.841034
I1122 15:55:40.503402 11005 solver.cpp:229]     Train net output #1: loss = 0.365441 (* 1 = 0.365441 loss)
I1122 15:55:40.503409 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.734698
I1122 15:55:40.503417 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.946902
I1122 15:55:40.503427 11005 solver.cpp:486] Iteration 3180, lr = 1e-05
I1122 15:56:09.410214 11005 solver.cpp:214] Iteration 3200, loss = 0.603971
I1122 15:56:09.410321 11005 solver.cpp:229]     Train net output #0: accuracy = 0.748745
I1122 15:56:09.410336 11005 solver.cpp:229]     Train net output #1: loss = 0.603971 (* 1 = 0.603971 loss)
I1122 15:56:09.410343 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.846478
I1122 15:56:09.410348 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.718412
I1122 15:56:09.410358 11005 solver.cpp:486] Iteration 3200, lr = 1e-05
I1122 15:56:38.314949 11005 solver.cpp:214] Iteration 3220, loss = 0.374411
I1122 15:56:38.314997 11005 solver.cpp:229]     Train net output #0: accuracy = 0.849277
I1122 15:56:38.315009 11005 solver.cpp:229]     Train net output #1: loss = 0.374411 (* 1 = 0.374411 loss)
I1122 15:56:38.315016 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.859809
I1122 15:56:38.315023 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.845679
I1122 15:56:38.315033 11005 solver.cpp:486] Iteration 3220, lr = 1e-05
I1122 15:57:07.223474 11005 solver.cpp:214] Iteration 3240, loss = 0.470014
I1122 15:57:07.223551 11005 solver.cpp:229]     Train net output #0: accuracy = 0.776703
I1122 15:57:07.223564 11005 solver.cpp:229]     Train net output #1: loss = 0.470014 (* 1 = 0.470014 loss)
I1122 15:57:07.223572 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.67529
I1122 15:57:07.223580 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.878793
I1122 15:57:07.223590 11005 solver.cpp:486] Iteration 3240, lr = 1e-05
I1122 15:57:36.131880 11005 solver.cpp:214] Iteration 3260, loss = 0.329166
I1122 15:57:36.131927 11005 solver.cpp:229]     Train net output #0: accuracy = 0.873188
I1122 15:57:36.131939 11005 solver.cpp:229]     Train net output #1: loss = 0.329166 (* 1 = 0.329166 loss)
I1122 15:57:36.131947 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.850799
I1122 15:57:36.131954 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.880954
I1122 15:57:36.131964 11005 solver.cpp:486] Iteration 3260, lr = 1e-05
I1122 15:58:05.072932 11005 solver.cpp:214] Iteration 3280, loss = 1.15811
I1122 15:58:05.073006 11005 solver.cpp:229]     Train net output #0: accuracy = 0.601227
I1122 15:58:05.073019 11005 solver.cpp:229]     Train net output #1: loss = 1.15811 (* 1 = 1.15811 loss)
I1122 15:58:05.073026 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.6513
I1122 15:58:05.073034 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.600504
I1122 15:58:05.073045 11005 solver.cpp:486] Iteration 3280, lr = 1e-05
I1122 15:58:33.977695 11005 solver.cpp:214] Iteration 3300, loss = 0.657628
I1122 15:58:33.977741 11005 solver.cpp:229]     Train net output #0: accuracy = 0.69418
I1122 15:58:33.977756 11005 solver.cpp:229]     Train net output #1: loss = 0.657628 (* 1 = 0.657628 loss)
I1122 15:58:33.977762 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.577018
I1122 15:58:33.977773 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.813473
I1122 15:58:33.977785 11005 solver.cpp:486] Iteration 3300, lr = 1e-05
I1122 15:59:02.888933 11005 solver.cpp:214] Iteration 3320, loss = 0.458267
I1122 15:59:02.889062 11005 solver.cpp:229]     Train net output #0: accuracy = 0.796585
I1122 15:59:02.889083 11005 solver.cpp:229]     Train net output #1: loss = 0.458267 (* 1 = 0.458267 loss)
I1122 15:59:02.889088 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.744407
I1122 15:59:02.889091 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.825629
I1122 15:59:02.889098 11005 solver.cpp:486] Iteration 3320, lr = 1e-05
I1122 15:59:31.823221 11005 solver.cpp:214] Iteration 3340, loss = 1.14666
I1122 15:59:31.823268 11005 solver.cpp:229]     Train net output #0: accuracy = 0.471722
I1122 15:59:31.823281 11005 solver.cpp:229]     Train net output #1: loss = 1.14666 (* 1 = 1.14666 loss)
I1122 15:59:31.823288 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.434855
I1122 15:59:31.823294 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.948838
I1122 15:59:31.823307 11005 solver.cpp:486] Iteration 3340, lr = 1e-05
I1122 16:00:00.753219 11005 solver.cpp:214] Iteration 3360, loss = 1.22858
I1122 16:00:00.753378 11005 solver.cpp:229]     Train net output #0: accuracy = 0.595688
I1122 16:00:00.753399 11005 solver.cpp:229]     Train net output #1: loss = 1.22858 (* 1 = 1.22858 loss)
I1122 16:00:00.753404 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.447154
I1122 16:00:00.753408 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.606002
I1122 16:00:00.753414 11005 solver.cpp:486] Iteration 3360, lr = 1e-05
I1122 16:00:29.670125 11005 solver.cpp:214] Iteration 3380, loss = 0.849168
I1122 16:00:29.670172 11005 solver.cpp:229]     Train net output #0: accuracy = 0.681004
I1122 16:00:29.670184 11005 solver.cpp:229]     Train net output #1: loss = 0.849168 (* 1 = 0.849168 loss)
I1122 16:00:29.670192 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.597859
I1122 16:00:29.670198 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.753295
I1122 16:00:29.670205 11005 solver.cpp:486] Iteration 3380, lr = 1e-05
I1122 16:00:58.583621 11005 solver.cpp:214] Iteration 3400, loss = 0.501115
I1122 16:00:58.583762 11005 solver.cpp:229]     Train net output #0: accuracy = 0.813747
I1122 16:00:58.583787 11005 solver.cpp:229]     Train net output #1: loss = 0.501115 (* 1 = 0.501115 loss)
I1122 16:00:58.583797 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.838916
I1122 16:00:58.583804 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.807473
I1122 16:00:58.583822 11005 solver.cpp:486] Iteration 3400, lr = 1e-05
I1122 16:01:27.506249 11005 solver.cpp:214] Iteration 3420, loss = 1.2394
I1122 16:01:27.506299 11005 solver.cpp:229]     Train net output #0: accuracy = 0.570049
I1122 16:01:27.506310 11005 solver.cpp:229]     Train net output #1: loss = 1.2394 (* 1 = 1.2394 loss)
I1122 16:01:27.506319 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.611201
I1122 16:01:27.506328 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.568745
I1122 16:01:27.506338 11005 solver.cpp:486] Iteration 3420, lr = 1e-05
I1122 16:01:56.428948 11005 solver.cpp:214] Iteration 3440, loss = 0.578743
I1122 16:01:56.429023 11005 solver.cpp:229]     Train net output #0: accuracy = 0.702179
I1122 16:01:56.429036 11005 solver.cpp:229]     Train net output #1: loss = 0.578742 (* 1 = 0.578742 loss)
I1122 16:01:56.429044 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.606123
I1122 16:01:56.429050 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.980621
I1122 16:01:56.429059 11005 solver.cpp:486] Iteration 3440, lr = 1e-05
I1122 16:02:25.330345 11005 solver.cpp:214] Iteration 3460, loss = 1.13665
I1122 16:02:25.330391 11005 solver.cpp:229]     Train net output #0: accuracy = 0.464848
I1122 16:02:25.330404 11005 solver.cpp:229]     Train net output #1: loss = 1.13665 (* 1 = 1.13665 loss)
I1122 16:02:25.330411 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.432539
I1122 16:02:25.330418 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.755152
I1122 16:02:25.330428 11005 solver.cpp:486] Iteration 3460, lr = 1e-05
I1122 16:02:54.278681 11005 solver.cpp:214] Iteration 3480, loss = 0.47998
I1122 16:02:54.278755 11005 solver.cpp:229]     Train net output #0: accuracy = 0.809067
I1122 16:02:54.278769 11005 solver.cpp:229]     Train net output #1: loss = 0.47998 (* 1 = 0.47998 loss)
I1122 16:02:54.278775 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.796519
I1122 16:02:54.278782 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.812712
I1122 16:02:54.278791 11005 solver.cpp:486] Iteration 3480, lr = 1e-05
I1122 16:03:23.196296 11005 solver.cpp:214] Iteration 3500, loss = 1.11116
I1122 16:03:23.196344 11005 solver.cpp:229]     Train net output #0: accuracy = 0.579971
I1122 16:03:23.196357 11005 solver.cpp:229]     Train net output #1: loss = 1.11116 (* 1 = 1.11116 loss)
I1122 16:03:23.196367 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.480192
I1122 16:03:23.196373 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.593155
I1122 16:03:23.196383 11005 solver.cpp:486] Iteration 3500, lr = 1e-05
I1122 16:03:52.109710 11005 solver.cpp:214] Iteration 3520, loss = 0.554245
I1122 16:03:52.109846 11005 solver.cpp:229]     Train net output #0: accuracy = 0.736385
I1122 16:03:52.109860 11005 solver.cpp:229]     Train net output #1: loss = 0.554245 (* 1 = 0.554245 loss)
I1122 16:03:52.109868 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.581946
I1122 16:03:52.109876 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.827784
I1122 16:03:52.109885 11005 solver.cpp:486] Iteration 3520, lr = 1e-05
I1122 16:04:21.026985 11005 solver.cpp:214] Iteration 3540, loss = 0.565983
I1122 16:04:21.027034 11005 solver.cpp:229]     Train net output #0: accuracy = 0.754635
I1122 16:04:21.027045 11005 solver.cpp:229]     Train net output #1: loss = 0.565983 (* 1 = 0.565983 loss)
I1122 16:04:21.027055 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.743752
I1122 16:04:21.027060 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.758251
I1122 16:04:21.027072 11005 solver.cpp:486] Iteration 3540, lr = 1e-05
I1122 16:04:49.946116 11005 solver.cpp:214] Iteration 3560, loss = 0.336473
I1122 16:04:49.946195 11005 solver.cpp:229]     Train net output #0: accuracy = 0.855518
I1122 16:04:49.946208 11005 solver.cpp:229]     Train net output #1: loss = 0.336473 (* 1 = 0.336473 loss)
I1122 16:04:49.946218 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.753668
I1122 16:04:49.946223 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.95692
I1122 16:04:49.946233 11005 solver.cpp:486] Iteration 3560, lr = 1e-05
I1122 16:05:18.875721 11005 solver.cpp:214] Iteration 3580, loss = 0.579322
I1122 16:05:18.875769 11005 solver.cpp:229]     Train net output #0: accuracy = 0.759647
I1122 16:05:18.875780 11005 solver.cpp:229]     Train net output #1: loss = 0.579321 (* 1 = 0.579321 loss)
I1122 16:05:18.875788 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.860893
I1122 16:05:18.875794 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.728225
I1122 16:05:18.875804 11005 solver.cpp:486] Iteration 3580, lr = 1e-05
I1122 16:05:47.790696 11005 solver.cpp:214] Iteration 3600, loss = 0.356356
I1122 16:05:47.790841 11005 solver.cpp:229]     Train net output #0: accuracy = 0.860828
I1122 16:05:47.790863 11005 solver.cpp:229]     Train net output #1: loss = 0.356356 (* 1 = 0.356356 loss)
I1122 16:05:47.790868 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.876796
I1122 16:05:47.790871 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.855373
I1122 16:05:47.790877 11005 solver.cpp:486] Iteration 3600, lr = 1e-05
I1122 16:06:16.703677 11005 solver.cpp:214] Iteration 3620, loss = 0.442676
I1122 16:06:16.703724 11005 solver.cpp:229]     Train net output #0: accuracy = 0.790283
I1122 16:06:16.703737 11005 solver.cpp:229]     Train net output #1: loss = 0.442676 (* 1 = 0.442676 loss)
I1122 16:06:16.703744 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.695806
I1122 16:06:16.703752 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.885391
I1122 16:06:16.703763 11005 solver.cpp:486] Iteration 3620, lr = 1e-05
I1122 16:06:45.618130 11005 solver.cpp:214] Iteration 3640, loss = 0.307646
I1122 16:06:45.618237 11005 solver.cpp:229]     Train net output #0: accuracy = 0.885551
I1122 16:06:45.618249 11005 solver.cpp:229]     Train net output #1: loss = 0.307646 (* 1 = 0.307646 loss)
I1122 16:06:45.618257 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.865996
I1122 16:06:45.618268 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.892335
I1122 16:06:45.618276 11005 solver.cpp:486] Iteration 3640, lr = 1e-05
I1122 16:07:14.530627 11005 solver.cpp:214] Iteration 3660, loss = 1.17662
I1122 16:07:14.530673 11005 solver.cpp:229]     Train net output #0: accuracy = 0.596954
I1122 16:07:14.530684 11005 solver.cpp:229]     Train net output #1: loss = 1.17662 (* 1 = 1.17662 loss)
I1122 16:07:14.530694 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.653712
I1122 16:07:14.530699 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.596135
I1122 16:07:14.530709 11005 solver.cpp:486] Iteration 3660, lr = 1e-05
I1122 16:07:43.452941 11005 solver.cpp:214] Iteration 3680, loss = 0.648795
I1122 16:07:43.453021 11005 solver.cpp:229]     Train net output #0: accuracy = 0.699963
I1122 16:07:43.453034 11005 solver.cpp:229]     Train net output #1: loss = 0.648795 (* 1 = 0.648795 loss)
I1122 16:07:43.453048 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.588837
I1122 16:07:43.453054 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.813111
I1122 16:07:43.453064 11005 solver.cpp:486] Iteration 3680, lr = 1e-05
I1122 16:08:12.366782 11005 solver.cpp:214] Iteration 3700, loss = 0.437265
I1122 16:08:12.366832 11005 solver.cpp:229]     Train net output #0: accuracy = 0.807381
I1122 16:08:12.366844 11005 solver.cpp:229]     Train net output #1: loss = 0.437264 (* 1 = 0.437264 loss)
I1122 16:08:12.366852 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.772795
I1122 16:08:12.366859 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.826632
I1122 16:08:12.366868 11005 solver.cpp:486] Iteration 3700, lr = 1e-05
I1122 16:08:41.295130 11005 solver.cpp:214] Iteration 3720, loss = 1.15179
I1122 16:08:41.295204 11005 solver.cpp:229]     Train net output #0: accuracy = 0.477196
I1122 16:08:41.295217 11005 solver.cpp:229]     Train net output #1: loss = 1.15179 (* 1 = 1.15179 loss)
I1122 16:08:41.295224 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.439971
I1122 16:08:41.295231 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.958943
I1122 16:08:41.295241 11005 solver.cpp:486] Iteration 3720, lr = 1e-05
I1122 16:09:10.213845 11005 solver.cpp:214] Iteration 3740, loss = 1.2531
I1122 16:09:10.213893 11005 solver.cpp:229]     Train net output #0: accuracy = 0.592403
I1122 16:09:10.213907 11005 solver.cpp:229]     Train net output #1: loss = 1.2531 (* 1 = 1.2531 loss)
I1122 16:09:10.213914 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.44827
I1122 16:09:10.213920 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.602412
I1122 16:09:10.213930 11005 solver.cpp:486] Iteration 3740, lr = 1e-05
I1122 16:09:35.741214 11005 solver.cpp:214] Iteration 3760, loss = 0.799041
I1122 16:09:35.741317 11005 solver.cpp:229]     Train net output #0: accuracy = 0.702126
I1122 16:09:35.741329 11005 solver.cpp:229]     Train net output #1: loss = 0.79904 (* 1 = 0.79904 loss)
I1122 16:09:35.741338 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.639354
I1122 16:09:35.741343 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.756704
I1122 16:09:35.741350 11005 solver.cpp:486] Iteration 3760, lr = 1e-05
I1122 16:09:49.921463 11005 solver.cpp:214] Iteration 3780, loss = 0.500026
I1122 16:09:49.921509 11005 solver.cpp:229]     Train net output #0: accuracy = 0.820236
I1122 16:09:49.921521 11005 solver.cpp:229]     Train net output #1: loss = 0.500025 (* 1 = 0.500025 loss)
I1122 16:09:49.921528 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.848952
I1122 16:09:49.921538 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.813078
I1122 16:09:49.921548 11005 solver.cpp:486] Iteration 3780, lr = 1e-05
I1122 16:10:09.207231 11005 solver.cpp:214] Iteration 3800, loss = 1.26479
I1122 16:10:09.207466 11005 solver.cpp:229]     Train net output #0: accuracy = 0.5676
I1122 16:10:09.207551 11005 solver.cpp:229]     Train net output #1: loss = 1.26479 (* 1 = 1.26479 loss)
I1122 16:10:09.207593 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.605489
I1122 16:10:09.207633 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.566399
I1122 16:10:09.207669 11005 solver.cpp:486] Iteration 3800, lr = 1e-05
I1122 16:10:23.394289 11005 solver.cpp:214] Iteration 3820, loss = 0.54835
I1122 16:10:23.394337 11005 solver.cpp:229]     Train net output #0: accuracy = 0.716278
I1122 16:10:23.394350 11005 solver.cpp:229]     Train net output #1: loss = 0.54835 (* 1 = 0.54835 loss)
I1122 16:10:23.394357 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.623367
I1122 16:10:23.394364 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.985603
I1122 16:10:23.394374 11005 solver.cpp:486] Iteration 3820, lr = 1e-05
I1122 16:10:37.590838 11005 solver.cpp:214] Iteration 3840, loss = 1.1393
I1122 16:10:37.590883 11005 solver.cpp:229]     Train net output #0: accuracy = 0.470291
I1122 16:10:37.590894 11005 solver.cpp:229]     Train net output #1: loss = 1.1393 (* 1 = 1.1393 loss)
I1122 16:10:37.590901 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.437588
I1122 16:10:37.590909 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.764141
I1122 16:10:37.590915 11005 solver.cpp:486] Iteration 3840, lr = 1e-05
I1122 16:10:51.800709 11005 solver.cpp:214] Iteration 3860, loss = 0.467676
I1122 16:10:51.800884 11005 solver.cpp:229]     Train net output #0: accuracy = 0.815636
I1122 16:10:51.800909 11005 solver.cpp:229]     Train net output #1: loss = 0.467675 (* 1 = 0.467675 loss)
I1122 16:10:51.800918 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.803518
I1122 16:10:51.800925 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.819156
I1122 16:10:51.800935 11005 solver.cpp:486] Iteration 3860, lr = 1e-05
I1122 16:11:05.943935 11005 solver.cpp:214] Iteration 3880, loss = 1.10664
I1122 16:11:05.943985 11005 solver.cpp:229]     Train net output #0: accuracy = 0.582684
I1122 16:11:05.943997 11005 solver.cpp:229]     Train net output #1: loss = 1.10664 (* 1 = 1.10664 loss)
I1122 16:11:05.944005 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.48686
I1122 16:11:05.944012 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.595344
I1122 16:11:05.944020 11005 solver.cpp:486] Iteration 3880, lr = 1e-05
I1122 16:11:20.113445 11005 solver.cpp:214] Iteration 3900, loss = 0.532621
I1122 16:11:20.113495 11005 solver.cpp:229]     Train net output #0: accuracy = 0.749271
I1122 16:11:20.113507 11005 solver.cpp:229]     Train net output #1: loss = 0.532621 (* 1 = 0.532621 loss)
I1122 16:11:20.113514 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.602077
I1122 16:11:20.113522 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.836383
I1122 16:11:20.113530 11005 solver.cpp:486] Iteration 3900, lr = 1e-05
I1122 16:11:34.290670 11005 solver.cpp:214] Iteration 3920, loss = 0.543264
I1122 16:11:34.290773 11005 solver.cpp:229]     Train net output #0: accuracy = 0.768532
I1122 16:11:34.290786 11005 solver.cpp:229]     Train net output #1: loss = 0.543263 (* 1 = 0.543263 loss)
I1122 16:11:34.290794 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.749579
I1122 16:11:34.290801 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.77483
I1122 16:11:34.290812 11005 solver.cpp:486] Iteration 3920, lr = 1e-05
I1122 16:11:48.471143 11005 solver.cpp:214] Iteration 3940, loss = 0.31329
I1122 16:11:48.471191 11005 solver.cpp:229]     Train net output #0: accuracy = 0.866905
I1122 16:11:48.471204 11005 solver.cpp:229]     Train net output #1: loss = 0.313289 (* 1 = 0.313289 loss)
I1122 16:11:48.471211 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.769527
I1122 16:11:48.471218 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.963855
I1122 16:11:48.471228 11005 solver.cpp:486] Iteration 3940, lr = 1e-05
I1122 16:12:02.666100 11005 solver.cpp:214] Iteration 3960, loss = 0.561434
I1122 16:12:02.666147 11005 solver.cpp:229]     Train net output #0: accuracy = 0.766697
I1122 16:12:02.666159 11005 solver.cpp:229]     Train net output #1: loss = 0.561434 (* 1 = 0.561434 loss)
I1122 16:12:02.666167 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.871265
I1122 16:12:02.666173 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.734243
I1122 16:12:02.666182 11005 solver.cpp:486] Iteration 3960, lr = 1e-05
I1122 16:12:16.850134 11005 solver.cpp:214] Iteration 3980, loss = 0.344601
I1122 16:12:16.850293 11005 solver.cpp:229]     Train net output #0: accuracy = 0.867111
I1122 16:12:16.850329 11005 solver.cpp:229]     Train net output #1: loss = 0.344601 (* 1 = 0.344601 loss)
I1122 16:12:16.850342 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.886697
I1122 16:12:16.850354 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.860419
I1122 16:12:16.850373 11005 solver.cpp:486] Iteration 3980, lr = 1e-05
I1122 16:12:35.905638 11005 solver.cpp:361] Snapshotting to /home/gradescan/SegNet/Models/backup_basic/Training/segnet_basic_iter_4000.caffemodel
I1122 16:12:35.915913 11005 solver.cpp:369] Snapshotting solver state to /home/gradescan/SegNet/Models/backup_basic/Training/segnet_basic_iter_4000.solverstate
I1122 16:12:36.146757 11005 solver.cpp:214] Iteration 4000, loss = 0.421174
I1122 16:12:36.146807 11005 solver.cpp:229]     Train net output #0: accuracy = 0.800411
I1122 16:12:36.146819 11005 solver.cpp:229]     Train net output #1: loss = 0.421174 (* 1 = 0.421174 loss)
I1122 16:12:36.146827 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.714352
I1122 16:12:36.146833 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.887045
I1122 16:12:36.146843 11005 solver.cpp:486] Iteration 4000, lr = 1e-05
I1122 16:12:50.312403 11005 solver.cpp:214] Iteration 4020, loss = 0.293063
I1122 16:12:50.312502 11005 solver.cpp:229]     Train net output #0: accuracy = 0.893326
I1122 16:12:50.312515 11005 solver.cpp:229]     Train net output #1: loss = 0.293063 (* 1 = 0.293063 loss)
I1122 16:12:50.312522 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.876735
I1122 16:12:50.312530 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.899081
I1122 16:12:50.312537 11005 solver.cpp:486] Iteration 4020, lr = 1e-05
I1122 16:13:08.322609 11005 solver.cpp:214] Iteration 4040, loss = 1.18976
I1122 16:13:08.322655 11005 solver.cpp:229]     Train net output #0: accuracy = 0.590523
I1122 16:13:08.322667 11005 solver.cpp:229]     Train net output #1: loss = 1.18976 (* 1 = 1.18976 loss)
I1122 16:13:08.322674 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.654516
I1122 16:13:08.322680 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.589599
I1122 16:13:08.322690 11005 solver.cpp:486] Iteration 4040, lr = 1e-05
I1122 16:13:23.778933 11005 solver.cpp:214] Iteration 4060, loss = 0.644468
I1122 16:13:23.779011 11005 solver.cpp:229]     Train net output #0: accuracy = 0.702438
I1122 16:13:23.779023 11005 solver.cpp:229]     Train net output #1: loss = 0.644468 (* 1 = 0.644468 loss)
I1122 16:13:23.779031 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.599369
I1122 16:13:23.779037 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.807383
I1122 16:13:23.779048 11005 solver.cpp:486] Iteration 4060, lr = 1e-05
I1122 16:13:37.953408 11005 solver.cpp:214] Iteration 4080, loss = 0.422975
I1122 16:13:37.953455 11005 solver.cpp:229]     Train net output #0: accuracy = 0.814911
I1122 16:13:37.953466 11005 solver.cpp:229]     Train net output #1: loss = 0.422975 (* 1 = 0.422975 loss)
I1122 16:13:37.953474 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.798792
I1122 16:13:37.953480 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.823883
I1122 16:13:37.953490 11005 solver.cpp:486] Iteration 4080, lr = 1e-05
I1122 16:13:52.104256 11005 solver.cpp:214] Iteration 4100, loss = 1.15202
I1122 16:13:52.104301 11005 solver.cpp:229]     Train net output #0: accuracy = 0.482677
I1122 16:13:52.104312 11005 solver.cpp:229]     Train net output #1: loss = 1.15202 (* 1 = 1.15202 loss)
I1122 16:13:52.104321 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.445408
I1122 16:13:52.104327 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.965006
I1122 16:13:52.104338 11005 solver.cpp:486] Iteration 4100, lr = 1e-05
I1122 16:14:06.282285 11005 solver.cpp:214] Iteration 4120, loss = 1.27091
I1122 16:14:06.282418 11005 solver.cpp:229]     Train net output #0: accuracy = 0.588531
I1122 16:14:06.282431 11005 solver.cpp:229]     Train net output #1: loss = 1.27091 (* 1 = 1.27091 loss)
I1122 16:14:06.282439 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.449739
I1122 16:14:06.282445 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.598169
I1122 16:14:06.282454 11005 solver.cpp:486] Iteration 4120, lr = 1e-05
I1122 16:14:20.444351 11005 solver.cpp:214] Iteration 4140, loss = 0.747256
I1122 16:14:20.444396 11005 solver.cpp:229]     Train net output #0: accuracy = 0.721581
I1122 16:14:20.444406 11005 solver.cpp:229]     Train net output #1: loss = 0.747255 (* 1 = 0.747255 loss)
I1122 16:14:20.444413 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.678625
I1122 16:14:20.444421 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.758929
I1122 16:14:20.444427 11005 solver.cpp:486] Iteration 4140, lr = 1e-05
I1122 16:14:34.631011 11005 solver.cpp:214] Iteration 4160, loss = 0.501405
I1122 16:14:34.631057 11005 solver.cpp:229]     Train net output #0: accuracy = 0.824753
I1122 16:14:34.631068 11005 solver.cpp:229]     Train net output #1: loss = 0.501405 (* 1 = 0.501405 loss)
I1122 16:14:34.631077 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.858282
I1122 16:14:34.631083 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.816395
I1122 16:14:34.631090 11005 solver.cpp:486] Iteration 4160, lr = 1e-05
I1122 16:14:48.837103 11005 solver.cpp:214] Iteration 4180, loss = 1.28468
I1122 16:14:48.837204 11005 solver.cpp:229]     Train net output #0: accuracy = 0.564865
I1122 16:14:48.837218 11005 solver.cpp:229]     Train net output #1: loss = 1.28468 (* 1 = 1.28468 loss)
I1122 16:14:48.837224 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.601018
I1122 16:14:48.837231 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.563719
I1122 16:14:48.837241 11005 solver.cpp:486] Iteration 4180, lr = 1e-05
I1122 16:15:03.026124 11005 solver.cpp:214] Iteration 4200, loss = 0.521423
I1122 16:15:03.026175 11005 solver.cpp:229]     Train net output #0: accuracy = 0.729515
I1122 16:15:03.026185 11005 solver.cpp:229]     Train net output #1: loss = 0.521423 (* 1 = 0.521423 loss)
I1122 16:15:03.026195 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.640134
I1122 16:15:03.026201 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.988608
I1122 16:15:03.026213 11005 solver.cpp:486] Iteration 4200, lr = 1e-05
I1122 16:15:17.198938 11005 solver.cpp:214] Iteration 4220, loss = 1.13972
I1122 16:15:17.198984 11005 solver.cpp:229]     Train net output #0: accuracy = 0.474682
I1122 16:15:17.198997 11005 solver.cpp:229]     Train net output #1: loss = 1.13972 (* 1 = 1.13972 loss)
I1122 16:15:17.199003 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.441823
I1122 16:15:17.199009 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.769931
I1122 16:15:17.199018 11005 solver.cpp:486] Iteration 4220, lr = 1e-05
I1122 16:15:31.378620 11005 solver.cpp:214] Iteration 4240, loss = 0.45922
I1122 16:15:31.378761 11005 solver.cpp:229]     Train net output #0: accuracy = 0.818584
I1122 16:15:31.378774 11005 solver.cpp:229]     Train net output #1: loss = 0.459219 (* 1 = 0.459219 loss)
I1122 16:15:31.378780 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.807179
I1122 16:15:31.378787 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.821898
I1122 16:15:31.378795 11005 solver.cpp:486] Iteration 4240, lr = 1e-05
I1122 16:15:45.533538 11005 solver.cpp:214] Iteration 4260, loss = 1.09667
I1122 16:15:45.533588 11005 solver.cpp:229]     Train net output #0: accuracy = 0.584846
I1122 16:15:45.533601 11005 solver.cpp:229]     Train net output #1: loss = 1.09667 (* 1 = 1.09667 loss)
I1122 16:15:45.533607 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.489835
I1122 16:15:45.533614 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.5974
I1122 16:15:45.533624 11005 solver.cpp:486] Iteration 4260, lr = 1e-05
I1122 16:15:59.716545 11005 solver.cpp:214] Iteration 4280, loss = 0.511664
I1122 16:15:59.716591 11005 solver.cpp:229]     Train net output #0: accuracy = 0.760868
I1122 16:15:59.716603 11005 solver.cpp:229]     Train net output #1: loss = 0.511664 (* 1 = 0.511664 loss)
I1122 16:15:59.716610 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.621274
I1122 16:15:59.716617 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.843481
I1122 16:15:59.716624 11005 solver.cpp:486] Iteration 4280, lr = 1e-05
I1122 16:16:13.890061 11005 solver.cpp:214] Iteration 4300, loss = 0.526007
I1122 16:16:13.890143 11005 solver.cpp:229]     Train net output #0: accuracy = 0.779938
I1122 16:16:13.890156 11005 solver.cpp:229]     Train net output #1: loss = 0.526007 (* 1 = 0.526007 loss)
I1122 16:16:13.890163 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.752715
I1122 16:16:13.890171 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.788984
I1122 16:16:13.890180 11005 solver.cpp:486] Iteration 4300, lr = 1e-05
I1122 16:16:28.059651 11005 solver.cpp:214] Iteration 4320, loss = 0.294561
I1122 16:16:28.059700 11005 solver.cpp:229]     Train net output #0: accuracy = 0.876606
I1122 16:16:28.059711 11005 solver.cpp:229]     Train net output #1: loss = 0.294561 (* 1 = 0.294561 loss)
I1122 16:16:28.059718 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.783749
I1122 16:16:28.059725 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.969055
I1122 16:16:28.059733 11005 solver.cpp:486] Iteration 4320, lr = 1e-05
I1122 16:16:42.242719 11005 solver.cpp:214] Iteration 4340, loss = 0.54577
I1122 16:16:42.242768 11005 solver.cpp:229]     Train net output #0: accuracy = 0.772923
I1122 16:16:42.242779 11005 solver.cpp:229]     Train net output #1: loss = 0.545769 (* 1 = 0.545769 loss)
I1122 16:16:42.242786 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.880961
I1122 16:16:42.242794 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.739392
I1122 16:16:42.242800 11005 solver.cpp:486] Iteration 4340, lr = 1e-05
I1122 16:16:56.428403 11005 solver.cpp:214] Iteration 4360, loss = 0.334421
I1122 16:16:56.428529 11005 solver.cpp:229]     Train net output #0: accuracy = 0.872742
I1122 16:16:56.428542 11005 solver.cpp:229]     Train net output #1: loss = 0.334421 (* 1 = 0.334421 loss)
I1122 16:16:56.428550 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.895625
I1122 16:16:56.428555 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.864923
I1122 16:16:56.428563 11005 solver.cpp:486] Iteration 4360, lr = 1e-05
I1122 16:17:10.649978 11005 solver.cpp:214] Iteration 4380, loss = 0.404839
I1122 16:17:10.650027 11005 solver.cpp:229]     Train net output #0: accuracy = 0.808189
I1122 16:17:10.650039 11005 solver.cpp:229]     Train net output #1: loss = 0.404839 (* 1 = 0.404839 loss)
I1122 16:17:10.650048 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.730572
I1122 16:17:10.650054 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.886325
I1122 16:17:10.650063 11005 solver.cpp:486] Iteration 4380, lr = 1e-05
I1122 16:17:24.828161 11005 solver.cpp:214] Iteration 4400, loss = 0.282416
I1122 16:17:24.828207 11005 solver.cpp:229]     Train net output #0: accuracy = 0.899048
I1122 16:17:24.828220 11005 solver.cpp:229]     Train net output #1: loss = 0.282415 (* 1 = 0.282415 loss)
I1122 16:17:24.828227 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.885
I1122 16:17:24.828233 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.903921
I1122 16:17:24.828243 11005 solver.cpp:486] Iteration 4400, lr = 1e-05
I1122 16:17:39.025329 11005 solver.cpp:214] Iteration 4420, loss = 1.1995
I1122 16:17:39.025573 11005 solver.cpp:229]     Train net output #0: accuracy = 0.584396
I1122 16:17:39.025656 11005 solver.cpp:229]     Train net output #1: loss = 1.1995 (* 1 = 1.1995 loss)
I1122 16:17:39.025697 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.651836
I1122 16:17:39.025735 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.583423
I1122 16:17:39.025774 11005 solver.cpp:486] Iteration 4420, lr = 1e-05
I1122 16:17:53.248710 11005 solver.cpp:214] Iteration 4440, loss = 0.643195
I1122 16:17:53.248759 11005 solver.cpp:229]     Train net output #0: accuracy = 0.701542
I1122 16:17:53.248770 11005 solver.cpp:229]     Train net output #1: loss = 0.643195 (* 1 = 0.643195 loss)
I1122 16:17:53.248777 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.606031
I1122 16:17:53.248783 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.798791
I1122 16:17:53.248790 11005 solver.cpp:486] Iteration 4440, lr = 1e-05
I1122 16:18:07.414623 11005 solver.cpp:214] Iteration 4460, loss = 0.412963
I1122 16:18:07.414670 11005 solver.cpp:229]     Train net output #0: accuracy = 0.819736
I1122 16:18:07.414682 11005 solver.cpp:229]     Train net output #1: loss = 0.412963 (* 1 = 0.412963 loss)
I1122 16:18:07.414690 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.819029
I1122 16:18:07.414695 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.82013
I1122 16:18:07.414703 11005 solver.cpp:486] Iteration 4460, lr = 1e-05
I1122 16:18:21.598014 11005 solver.cpp:214] Iteration 4480, loss = 1.15045
I1122 16:18:21.598096 11005 solver.cpp:229]     Train net output #0: accuracy = 0.487503
I1122 16:18:21.598109 11005 solver.cpp:229]     Train net output #1: loss = 1.15045 (* 1 = 1.15045 loss)
I1122 16:18:21.598116 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.450121
I1122 16:18:21.598124 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.971281
I1122 16:18:21.598131 11005 solver.cpp:486] Iteration 4480, lr = 1e-05
I1122 16:18:35.820703 11005 solver.cpp:214] Iteration 4500, loss = 1.28348
I1122 16:18:35.820749 11005 solver.cpp:229]     Train net output #0: accuracy = 0.58353
I1122 16:18:35.820760 11005 solver.cpp:229]     Train net output #1: loss = 1.28348 (* 1 = 1.28348 loss)
I1122 16:18:35.820768 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.454556
I1122 16:18:35.820775 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.592486
I1122 16:18:35.820783 11005 solver.cpp:486] Iteration 4500, lr = 1e-05
I1122 16:18:49.985987 11005 solver.cpp:214] Iteration 4520, loss = 0.698713
I1122 16:18:49.986035 11005 solver.cpp:229]     Train net output #0: accuracy = 0.738789
I1122 16:18:49.986047 11005 solver.cpp:229]     Train net output #1: loss = 0.698712 (* 1 = 0.698712 loss)
I1122 16:18:49.986054 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.712492
I1122 16:18:49.986060 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.761653
I1122 16:18:49.986071 11005 solver.cpp:486] Iteration 4520, lr = 1e-05
I1122 16:19:04.168011 11005 solver.cpp:214] Iteration 4540, loss = 0.503301
I1122 16:19:04.168241 11005 solver.cpp:229]     Train net output #0: accuracy = 0.828663
I1122 16:19:04.168301 11005 solver.cpp:229]     Train net output #1: loss = 0.503301 (* 1 = 0.503301 loss)
I1122 16:19:04.168329 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.865049
I1122 16:19:04.168351 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.819592
I1122 16:19:04.168375 11005 solver.cpp:486] Iteration 4540, lr = 1e-05
I1122 16:19:18.433367 11005 solver.cpp:214] Iteration 4560, loss = 1.29968
I1122 16:19:18.433429 11005 solver.cpp:229]     Train net output #0: accuracy = 0.561657
I1122 16:19:18.433441 11005 solver.cpp:229]     Train net output #1: loss = 1.29968 (* 1 = 1.29968 loss)
I1122 16:19:18.433449 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.592947
I1122 16:19:18.433454 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.560665
I1122 16:19:18.433461 11005 solver.cpp:486] Iteration 4560, lr = 1e-05
I1122 16:19:32.641777 11005 solver.cpp:214] Iteration 4580, loss = 0.497175
I1122 16:19:32.641839 11005 solver.cpp:229]     Train net output #0: accuracy = 0.741077
I1122 16:19:32.641849 11005 solver.cpp:229]     Train net output #1: loss = 0.497175 (* 1 = 0.497175 loss)
I1122 16:19:32.641857 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.654957
I1122 16:19:32.641865 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.99072
I1122 16:19:32.641875 11005 solver.cpp:486] Iteration 4580, lr = 1e-05
I1122 16:19:46.845898 11005 solver.cpp:214] Iteration 4600, loss = 1.13828
I1122 16:19:46.846014 11005 solver.cpp:229]     Train net output #0: accuracy = 0.478985
I1122 16:19:46.846026 11005 solver.cpp:229]     Train net output #1: loss = 1.13828 (* 1 = 1.13828 loss)
I1122 16:19:46.846035 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.446003
I1122 16:19:46.846040 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.77534
I1122 16:19:46.846047 11005 solver.cpp:486] Iteration 4600, lr = 1e-05
I1122 16:20:01.061800 11005 solver.cpp:214] Iteration 4620, loss = 0.452521
I1122 16:20:01.061849 11005 solver.cpp:229]     Train net output #0: accuracy = 0.821014
I1122 16:20:01.061866 11005 solver.cpp:229]     Train net output #1: loss = 0.452521 (* 1 = 0.452521 loss)
I1122 16:20:01.061879 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.809772
I1122 16:20:01.061892 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.82428
I1122 16:20:01.061904 11005 solver.cpp:486] Iteration 4620, lr = 1e-05
I1122 16:20:15.289831 11005 solver.cpp:214] Iteration 4640, loss = 1.08488
I1122 16:20:15.289878 11005 solver.cpp:229]     Train net output #0: accuracy = 0.587444
I1122 16:20:15.289896 11005 solver.cpp:229]     Train net output #1: loss = 1.08488 (* 1 = 1.08488 loss)
I1122 16:20:15.289909 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.495359
I1122 16:20:15.289921 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.599611
I1122 16:20:15.289933 11005 solver.cpp:486] Iteration 4640, lr = 1e-05
I1122 16:20:29.491852 11005 solver.cpp:214] Iteration 4660, loss = 0.491733
I1122 16:20:29.492024 11005 solver.cpp:229]     Train net output #0: accuracy = 0.772068
I1122 16:20:29.492064 11005 solver.cpp:229]     Train net output #1: loss = 0.491733 (* 1 = 0.491733 loss)
I1122 16:20:29.492076 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.639394
I1122 16:20:29.492086 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.850586
I1122 16:20:29.492100 11005 solver.cpp:486] Iteration 4660, lr = 1e-05
I1122 16:20:43.706450 11005 solver.cpp:214] Iteration 4680, loss = 0.513044
I1122 16:20:43.706502 11005 solver.cpp:229]     Train net output #0: accuracy = 0.790131
I1122 16:20:43.706513 11005 solver.cpp:229]     Train net output #1: loss = 0.513043 (* 1 = 0.513043 loss)
I1122 16:20:43.706521 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.756034
I1122 16:20:43.706527 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.801461
I1122 16:20:43.706533 11005 solver.cpp:486] Iteration 4680, lr = 1e-05
I1122 16:20:57.953009 11005 solver.cpp:214] Iteration 4700, loss = 0.279147
I1122 16:20:57.953058 11005 solver.cpp:229]     Train net output #0: accuracy = 0.884556
I1122 16:20:57.953068 11005 solver.cpp:229]     Train net output #1: loss = 0.279146 (* 1 = 0.279146 loss)
I1122 16:20:57.953075 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.795593
I1122 16:20:57.953081 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.973127
I1122 16:20:57.953088 11005 solver.cpp:486] Iteration 4700, lr = 1e-05
I1122 16:21:12.170542 11005 solver.cpp:214] Iteration 4720, loss = 0.531588
I1122 16:21:12.170637 11005 solver.cpp:229]     Train net output #0: accuracy = 0.778221
I1122 16:21:12.170650 11005 solver.cpp:229]     Train net output #1: loss = 0.531587 (* 1 = 0.531587 loss)
I1122 16:21:12.170657 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.887822
I1122 16:21:12.170663 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.744205
I1122 16:21:12.170670 11005 solver.cpp:486] Iteration 4720, lr = 1e-05
I1122 16:21:26.352427 11005 solver.cpp:214] Iteration 4740, loss = 0.324323
I1122 16:21:26.352478 11005 solver.cpp:229]     Train net output #0: accuracy = 0.877747
I1122 16:21:26.352488 11005 solver.cpp:229]     Train net output #1: loss = 0.324322 (* 1 = 0.324322 loss)
I1122 16:21:26.352495 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.902934
I1122 16:21:26.352501 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.86914
I1122 16:21:26.352509 11005 solver.cpp:486] Iteration 4740, lr = 1e-05
I1122 16:21:40.549754 11005 solver.cpp:214] Iteration 4760, loss = 0.390613
I1122 16:21:40.549796 11005 solver.cpp:229]     Train net output #0: accuracy = 0.814632
I1122 16:21:40.549808 11005 solver.cpp:229]     Train net output #1: loss = 0.390613 (* 1 = 0.390613 loss)
I1122 16:21:40.549814 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.7434
I1122 16:21:40.549821 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.886341
I1122 16:21:40.549829 11005 solver.cpp:486] Iteration 4760, lr = 1e-05
I1122 16:21:54.751965 11005 solver.cpp:214] Iteration 4780, loss = 0.273811
I1122 16:21:54.752068 11005 solver.cpp:229]     Train net output #0: accuracy = 0.903198
I1122 16:21:54.752080 11005 solver.cpp:229]     Train net output #1: loss = 0.27381 (* 1 = 0.27381 loss)
I1122 16:21:54.752087 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.893606
I1122 16:21:54.752094 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.906526
I1122 16:21:54.752101 11005 solver.cpp:486] Iteration 4780, lr = 1e-05
I1122 16:22:08.962337 11005 solver.cpp:214] Iteration 4800, loss = 1.20549
I1122 16:22:08.962388 11005 solver.cpp:229]     Train net output #0: accuracy = 0.579464
I1122 16:22:08.962399 11005 solver.cpp:229]     Train net output #1: loss = 1.20549 (* 1 = 1.20549 loss)
I1122 16:22:08.962406 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.640579
I1122 16:22:08.962412 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.578582
I1122 16:22:08.962421 11005 solver.cpp:486] Iteration 4800, lr = 1e-05
I1122 16:22:23.178048 11005 solver.cpp:214] Iteration 4820, loss = 0.642431
I1122 16:22:23.178097 11005 solver.cpp:229]     Train net output #0: accuracy = 0.701637
I1122 16:22:23.178108 11005 solver.cpp:229]     Train net output #1: loss = 0.642431 (* 1 = 0.642431 loss)
I1122 16:22:23.178115 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.614794
I1122 16:22:23.178122 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.790061
I1122 16:22:23.178130 11005 solver.cpp:486] Iteration 4820, lr = 1e-05
I1122 16:22:37.358420 11005 solver.cpp:214] Iteration 4840, loss = 0.405316
I1122 16:22:37.358489 11005 solver.cpp:229]     Train net output #0: accuracy = 0.823185
I1122 16:22:37.358501 11005 solver.cpp:229]     Train net output #1: loss = 0.405315 (* 1 = 0.405315 loss)
I1122 16:22:37.358507 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.835351
I1122 16:22:37.358515 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.816413
I1122 16:22:37.358520 11005 solver.cpp:486] Iteration 4840, lr = 1e-05
I1122 16:22:51.549998 11005 solver.cpp:214] Iteration 4860, loss = 1.14946
I1122 16:22:51.550043 11005 solver.cpp:229]     Train net output #0: accuracy = 0.490398
I1122 16:22:51.550055 11005 solver.cpp:229]     Train net output #1: loss = 1.14946 (* 1 = 1.14946 loss)
I1122 16:22:51.550061 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.45292
I1122 16:22:51.550068 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.975429
I1122 16:22:51.550076 11005 solver.cpp:486] Iteration 4860, lr = 1e-05
I1122 16:23:05.768579 11005 solver.cpp:214] Iteration 4880, loss = 1.29147
I1122 16:23:05.768623 11005 solver.cpp:229]     Train net output #0: accuracy = 0.578857
I1122 16:23:05.768635 11005 solver.cpp:229]     Train net output #1: loss = 1.29147 (* 1 = 1.29147 loss)
I1122 16:23:05.768642 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.456554
I1122 16:23:05.768649 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.58735
I1122 16:23:05.768656 11005 solver.cpp:486] Iteration 4880, lr = 1e-05
I1122 16:23:19.978709 11005 solver.cpp:214] Iteration 4900, loss = 0.65582
I1122 16:23:19.978941 11005 solver.cpp:229]     Train net output #0: accuracy = 0.752552
I1122 16:23:19.978971 11005 solver.cpp:229]     Train net output #1: loss = 0.655819 (* 1 = 0.655819 loss)
I1122 16:23:19.978981 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.740116
I1122 16:23:19.978991 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.763364
I1122 16:23:19.979001 11005 solver.cpp:486] Iteration 4900, lr = 1e-05
I1122 16:23:34.215591 11005 solver.cpp:214] Iteration 4920, loss = 0.504728
I1122 16:23:34.215657 11005 solver.cpp:229]     Train net output #0: accuracy = 0.831821
I1122 16:23:34.215672 11005 solver.cpp:229]     Train net output #1: loss = 0.504728 (* 1 = 0.504728 loss)
I1122 16:23:34.215679 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.871186
I1122 16:23:34.215685 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.822009
I1122 16:23:34.215695 11005 solver.cpp:486] Iteration 4920, lr = 1e-05
I1122 16:23:48.420640 11005 solver.cpp:214] Iteration 4940, loss = 1.3118
I1122 16:23:48.420709 11005 solver.cpp:229]     Train net output #0: accuracy = 0.559277
I1122 16:23:48.420722 11005 solver.cpp:229]     Train net output #1: loss = 1.3118 (* 1 = 1.3118 loss)
I1122 16:23:48.420729 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.58711
I1122 16:23:48.420737 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.558394
I1122 16:23:48.420743 11005 solver.cpp:486] Iteration 4940, lr = 1e-05
I1122 16:24:02.601451 11005 solver.cpp:214] Iteration 4960, loss = 0.477101
I1122 16:24:02.601642 11005 solver.cpp:229]     Train net output #0: accuracy = 0.750641
I1122 16:24:02.601670 11005 solver.cpp:229]     Train net output #1: loss = 0.477101 (* 1 = 0.477101 loss)
I1122 16:24:02.601677 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.66726
I1122 16:24:02.601687 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.992341
I1122 16:24:02.601696 11005 solver.cpp:486] Iteration 4960, lr = 1e-05
I1122 16:24:16.802534 11005 solver.cpp:214] Iteration 4980, loss = 1.13844
I1122 16:24:16.802587 11005 solver.cpp:229]     Train net output #0: accuracy = 0.484005
I1122 16:24:16.802599 11005 solver.cpp:229]     Train net output #1: loss = 1.13844 (* 1 = 1.13844 loss)
I1122 16:24:16.802606 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.451132
I1122 16:24:16.802613 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.779378
I1122 16:24:16.802620 11005 solver.cpp:486] Iteration 4980, lr = 1e-05
I1122 16:24:30.793202 11005 solver.cpp:361] Snapshotting to /home/gradescan/SegNet/Models/backup_basic/Training/segnet_basic_iter_5000.caffemodel
I1122 16:24:30.803319 11005 solver.cpp:369] Snapshotting solver state to /home/gradescan/SegNet/Models/backup_basic/Training/segnet_basic_iter_5000.solverstate
I1122 16:24:31.034348 11005 solver.cpp:214] Iteration 5000, loss = 0.446031
I1122 16:24:31.034394 11005 solver.cpp:229]     Train net output #0: accuracy = 0.823795
I1122 16:24:31.034405 11005 solver.cpp:229]     Train net output #1: loss = 0.446031 (* 1 = 0.446031 loss)
I1122 16:24:31.034412 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.814212
I1122 16:24:31.034420 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.826579
I1122 16:24:31.034428 11005 solver.cpp:486] Iteration 5000, lr = 1e-05
I1122 16:24:45.192045 11005 solver.cpp:214] Iteration 5020, loss = 1.07315
I1122 16:24:45.192263 11005 solver.cpp:229]     Train net output #0: accuracy = 0.589931
I1122 16:24:45.192301 11005 solver.cpp:229]     Train net output #1: loss = 1.07315 (* 1 = 1.07315 loss)
I1122 16:24:45.192309 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.495914
I1122 16:24:45.192315 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.602354
I1122 16:24:45.192325 11005 solver.cpp:486] Iteration 5020, lr = 1e-05
I1122 16:24:59.380650 11005 solver.cpp:214] Iteration 5040, loss = 0.471481
I1122 16:24:59.380700 11005 solver.cpp:229]     Train net output #0: accuracy = 0.783028
I1122 16:24:59.380713 11005 solver.cpp:229]     Train net output #1: loss = 0.471481 (* 1 = 0.471481 loss)
I1122 16:24:59.380720 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.653431
I1122 16:24:59.380727 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.859724
I1122 16:24:59.380733 11005 solver.cpp:486] Iteration 5040, lr = 1e-05
I1122 16:25:13.567522 11005 solver.cpp:214] Iteration 5060, loss = 0.502665
I1122 16:25:13.567570 11005 solver.cpp:229]     Train net output #0: accuracy = 0.798676
I1122 16:25:13.567582 11005 solver.cpp:229]     Train net output #1: loss = 0.502665 (* 1 = 0.502665 loss)
I1122 16:25:13.567589 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.756768
I1122 16:25:13.567595 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.812601
I1122 16:25:13.567602 11005 solver.cpp:486] Iteration 5060, lr = 1e-05
I1122 16:25:27.726243 11005 solver.cpp:214] Iteration 5080, loss = 0.267056
I1122 16:25:27.726322 11005 solver.cpp:229]     Train net output #0: accuracy = 0.891071
I1122 16:25:27.726336 11005 solver.cpp:229]     Train net output #1: loss = 0.267055 (* 1 = 0.267055 loss)
I1122 16:25:27.726343 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.805594
I1122 16:25:27.726349 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.976173
I1122 16:25:27.726359 11005 solver.cpp:486] Iteration 5080, lr = 1e-05
I1122 16:25:41.902026 11005 solver.cpp:214] Iteration 5100, loss = 0.517796
I1122 16:25:41.902076 11005 solver.cpp:229]     Train net output #0: accuracy = 0.783775
I1122 16:25:41.902086 11005 solver.cpp:229]     Train net output #1: loss = 0.517795 (* 1 = 0.517795 loss)
I1122 16:25:41.902093 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.892606
I1122 16:25:41.902101 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.749999
I1122 16:25:41.902109 11005 solver.cpp:486] Iteration 5100, lr = 1e-05
I1122 16:25:56.077263 11005 solver.cpp:214] Iteration 5120, loss = 0.313024
I1122 16:25:56.077312 11005 solver.cpp:229]     Train net output #0: accuracy = 0.883186
I1122 16:25:56.077324 11005 solver.cpp:229]     Train net output #1: loss = 0.313024 (* 1 = 0.313024 loss)
I1122 16:25:56.077332 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.90966
I1122 16:25:56.077337 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.874141
I1122 16:25:56.077344 11005 solver.cpp:486] Iteration 5120, lr = 1e-05
I1122 16:26:10.260749 11005 solver.cpp:214] Iteration 5140, loss = 0.378422
I1122 16:26:10.260917 11005 solver.cpp:229]     Train net output #0: accuracy = 0.820923
I1122 16:26:10.260939 11005 solver.cpp:229]     Train net output #1: loss = 0.378421 (* 1 = 0.378421 loss)
I1122 16:26:10.260954 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.754509
I1122 16:26:10.260958 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.88778
I1122 16:26:10.260964 11005 solver.cpp:486] Iteration 5140, lr = 1e-05
I1122 16:26:24.452898 11005 solver.cpp:214] Iteration 5160, loss = 0.26647
I1122 16:26:24.452949 11005 solver.cpp:229]     Train net output #0: accuracy = 0.906578
I1122 16:26:24.452961 11005 solver.cpp:229]     Train net output #1: loss = 0.26647 (* 1 = 0.26647 loss)
I1122 16:26:24.452975 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.899234
I1122 16:26:24.452981 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.909125
I1122 16:26:24.452989 11005 solver.cpp:486] Iteration 5160, lr = 1e-05
I1122 16:26:38.595134 11005 solver.cpp:214] Iteration 5180, loss = 1.20925
I1122 16:26:38.595185 11005 solver.cpp:229]     Train net output #0: accuracy = 0.57526
I1122 16:26:38.595196 11005 solver.cpp:229]     Train net output #1: loss = 1.20925 (* 1 = 1.20925 loss)
I1122 16:26:38.595203 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.632538
I1122 16:26:38.595209 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.574433
I1122 16:26:38.595217 11005 solver.cpp:486] Iteration 5180, lr = 1e-05
I1122 16:26:52.802423 11005 solver.cpp:214] Iteration 5200, loss = 0.64088
I1122 16:26:52.802592 11005 solver.cpp:229]     Train net output #0: accuracy = 0.701332
I1122 16:26:52.802628 11005 solver.cpp:229]     Train net output #1: loss = 0.640879 (* 1 = 0.640879 loss)
I1122 16:26:52.802637 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.620677
I1122 16:26:52.802650 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.783455
I1122 16:26:52.802661 11005 solver.cpp:486] Iteration 5200, lr = 1e-05
I1122 16:27:07.003962 11005 solver.cpp:214] Iteration 5220, loss = 0.398848
I1122 16:27:07.004010 11005 solver.cpp:229]     Train net output #0: accuracy = 0.82613
I1122 16:27:07.004022 11005 solver.cpp:229]     Train net output #1: loss = 0.398847 (* 1 = 0.398847 loss)
I1122 16:27:07.004030 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.848889
I1122 16:27:07.004036 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.813462
I1122 16:27:07.004043 11005 solver.cpp:486] Iteration 5220, lr = 1e-05
I1122 16:27:21.175216 11005 solver.cpp:214] Iteration 5240, loss = 1.14884
I1122 16:27:21.175262 11005 solver.cpp:229]     Train net output #0: accuracy = 0.493061
I1122 16:27:21.175274 11005 solver.cpp:229]     Train net output #1: loss = 1.14884 (* 1 = 1.14884 loss)
I1122 16:27:21.175282 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.455468
I1122 16:27:21.175287 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.979578
I1122 16:27:21.175295 11005 solver.cpp:486] Iteration 5240, lr = 1e-05
I1122 16:27:35.367724 11005 solver.cpp:214] Iteration 5260, loss = 1.29649
I1122 16:27:35.367822 11005 solver.cpp:229]     Train net output #0: accuracy = 0.575714
I1122 16:27:35.367835 11005 solver.cpp:229]     Train net output #1: loss = 1.29649 (* 1 = 1.29649 loss)
I1122 16:27:35.367842 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.463663
I1122 16:27:35.367848 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.583495
I1122 16:27:35.367856 11005 solver.cpp:486] Iteration 5260, lr = 1e-05
I1122 16:27:49.538547 11005 solver.cpp:214] Iteration 5280, loss = 0.62265
I1122 16:27:49.538595 11005 solver.cpp:229]     Train net output #0: accuracy = 0.762562
I1122 16:27:49.538607 11005 solver.cpp:229]     Train net output #1: loss = 0.622649 (* 1 = 0.622649 loss)
I1122 16:27:49.538614 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.760449
I1122 16:27:49.538621 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.764398
I1122 16:27:49.538628 11005 solver.cpp:486] Iteration 5280, lr = 1e-05
I1122 16:28:03.728560 11005 solver.cpp:214] Iteration 5300, loss = 0.503528
I1122 16:28:03.728605 11005 solver.cpp:229]     Train net output #0: accuracy = 0.835461
I1122 16:28:03.728617 11005 solver.cpp:229]     Train net output #1: loss = 0.503527 (* 1 = 0.503527 loss)
I1122 16:28:03.728624 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.87608
I1122 16:28:03.728631 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.825335
I1122 16:28:03.728638 11005 solver.cpp:486] Iteration 5300, lr = 1e-05
I1122 16:28:18.024587 11005 solver.cpp:214] Iteration 5320, loss = 1.32072
I1122 16:28:18.024817 11005 solver.cpp:229]     Train net output #0: accuracy = 0.557365
I1122 16:28:18.024866 11005 solver.cpp:229]     Train net output #1: loss = 1.32072 (* 1 = 1.32072 loss)
I1122 16:28:18.024888 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.57668
I1122 16:28:18.024906 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.556753
I1122 16:28:18.024943 11005 solver.cpp:486] Iteration 5320, lr = 1e-05
I1122 16:28:32.242583 11005 solver.cpp:214] Iteration 5340, loss = 0.460729
I1122 16:28:32.242631 11005 solver.cpp:229]     Train net output #0: accuracy = 0.759212
I1122 16:28:32.242645 11005 solver.cpp:229]     Train net output #1: loss = 0.460729 (* 1 = 0.460729 loss)
I1122 16:28:32.242650 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.678466
I1122 16:28:32.242657 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.993278
I1122 16:28:32.242665 11005 solver.cpp:486] Iteration 5340, lr = 1e-05
I1122 16:28:46.417194 11005 solver.cpp:214] Iteration 5360, loss = 1.13806
I1122 16:28:46.417244 11005 solver.cpp:229]     Train net output #0: accuracy = 0.487331
I1122 16:28:46.417256 11005 solver.cpp:229]     Train net output #1: loss = 1.13806 (* 1 = 1.13806 loss)
I1122 16:28:46.417263 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.454299
I1122 16:28:46.417270 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.784139
I1122 16:28:46.417278 11005 solver.cpp:486] Iteration 5360, lr = 1e-05
I1122 16:29:00.578910 11005 solver.cpp:214] Iteration 5380, loss = 0.439806
I1122 16:29:00.579020 11005 solver.cpp:229]     Train net output #0: accuracy = 0.825325
I1122 16:29:00.579033 11005 solver.cpp:229]     Train net output #1: loss = 0.439805 (* 1 = 0.439805 loss)
I1122 16:29:00.579041 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.816059
I1122 16:29:00.579046 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.828017
I1122 16:29:00.579054 11005 solver.cpp:486] Iteration 5380, lr = 1e-05
I1122 16:29:14.775457 11005 solver.cpp:214] Iteration 5400, loss = 1.06337
I1122 16:29:14.775506 11005 solver.cpp:229]     Train net output #0: accuracy = 0.592438
I1122 16:29:14.775518 11005 solver.cpp:229]     Train net output #1: loss = 1.06337 (* 1 = 1.06337 loss)
I1122 16:29:14.775526 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.5
I1122 16:29:14.775532 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.604651
I1122 16:29:14.775543 11005 solver.cpp:486] Iteration 5400, lr = 1e-05
I1122 16:29:28.975457 11005 solver.cpp:214] Iteration 5420, loss = 0.450691
I1122 16:29:28.975507 11005 solver.cpp:229]     Train net output #0: accuracy = 0.795013
I1122 16:29:28.975518 11005 solver.cpp:229]     Train net output #1: loss = 0.45069 (* 1 = 0.45069 loss)
I1122 16:29:28.975525 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.667077
I1122 16:29:28.975533 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.870727
I1122 16:29:28.975538 11005 solver.cpp:486] Iteration 5420, lr = 1e-05
I1122 16:29:43.135202 11005 solver.cpp:214] Iteration 5440, loss = 0.494917
I1122 16:29:43.135365 11005 solver.cpp:229]     Train net output #0: accuracy = 0.806385
I1122 16:29:43.135390 11005 solver.cpp:229]     Train net output #1: loss = 0.494917 (* 1 = 0.494917 loss)
I1122 16:29:43.135401 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.758236
I1122 16:29:43.135408 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.822384
I1122 16:29:43.135417 11005 solver.cpp:486] Iteration 5440, lr = 1e-05
I1122 16:29:57.307868 11005 solver.cpp:214] Iteration 5460, loss = 0.257164
I1122 16:29:57.307916 11005 solver.cpp:229]     Train net output #0: accuracy = 0.895317
I1122 16:29:57.307929 11005 solver.cpp:229]     Train net output #1: loss = 0.257164 (* 1 = 0.257164 loss)
I1122 16:29:57.307936 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.811803
I1122 16:29:57.307942 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.978464
I1122 16:29:57.307953 11005 solver.cpp:486] Iteration 5460, lr = 1e-05
I1122 16:30:11.496677 11005 solver.cpp:214] Iteration 5480, loss = 0.503453
I1122 16:30:11.496723 11005 solver.cpp:229]     Train net output #0: accuracy = 0.78994
I1122 16:30:11.496734 11005 solver.cpp:229]     Train net output #1: loss = 0.503453 (* 1 = 0.503453 loss)
I1122 16:30:11.496742 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.897679
I1122 16:30:11.496748 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.756502
I1122 16:30:11.496757 11005 solver.cpp:486] Iteration 5480, lr = 1e-05
I1122 16:30:25.659714 11005 solver.cpp:214] Iteration 5500, loss = 0.3002
I1122 16:30:25.659864 11005 solver.cpp:229]     Train net output #0: accuracy = 0.889214
I1122 16:30:25.659903 11005 solver.cpp:229]     Train net output #1: loss = 0.3002 (* 1 = 0.3002 loss)
I1122 16:30:25.659910 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.913645
I1122 16:30:25.659919 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.880866
I1122 16:30:25.659929 11005 solver.cpp:486] Iteration 5500, lr = 1e-05
I1122 16:30:39.827438 11005 solver.cpp:214] Iteration 5520, loss = 0.365679
I1122 16:30:39.827487 11005 solver.cpp:229]     Train net output #0: accuracy = 0.827499
I1122 16:30:39.827500 11005 solver.cpp:229]     Train net output #1: loss = 0.365679 (* 1 = 0.365679 loss)
I1122 16:30:39.827507 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.764303
I1122 16:30:39.827520 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.891117
I1122 16:30:39.827528 11005 solver.cpp:486] Iteration 5520, lr = 1e-05
I1122 16:30:54.023434 11005 solver.cpp:214] Iteration 5540, loss = 0.259698
I1122 16:30:54.023483 11005 solver.cpp:229]     Train net output #0: accuracy = 0.909969
I1122 16:30:54.023495 11005 solver.cpp:229]     Train net output #1: loss = 0.259697 (* 1 = 0.259697 loss)
I1122 16:30:54.023504 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.903707
I1122 16:30:54.023510 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.912141
I1122 16:30:54.023520 11005 solver.cpp:486] Iteration 5540, lr = 1e-05
I1122 16:31:08.190532 11005 solver.cpp:214] Iteration 5560, loss = 1.20947
I1122 16:31:08.190676 11005 solver.cpp:229]     Train net output #0: accuracy = 0.571983
I1122 16:31:08.190704 11005 solver.cpp:229]     Train net output #1: loss = 1.20947 (* 1 = 1.20947 loss)
I1122 16:31:08.190712 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.627982
I1122 16:31:08.190719 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.571175
I1122 16:31:08.190728 11005 solver.cpp:486] Iteration 5560, lr = 1e-05
I1122 16:31:22.364590 11005 solver.cpp:214] Iteration 5580, loss = 0.63895
I1122 16:31:22.364639 11005 solver.cpp:229]     Train net output #0: accuracy = 0.701996
I1122 16:31:22.364651 11005 solver.cpp:229]     Train net output #1: loss = 0.63895 (* 1 = 0.63895 loss)
I1122 16:31:22.364660 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.62606
I1122 16:31:22.364666 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.779313
I1122 16:31:22.364680 11005 solver.cpp:486] Iteration 5580, lr = 1e-05
I1122 16:31:36.550119 11005 solver.cpp:214] Iteration 5600, loss = 0.392737
I1122 16:31:36.550168 11005 solver.cpp:229]     Train net output #0: accuracy = 0.828182
I1122 16:31:36.550180 11005 solver.cpp:229]     Train net output #1: loss = 0.392737 (* 1 = 0.392737 loss)
I1122 16:31:36.550187 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.857722
I1122 16:31:36.550194 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.81174
I1122 16:31:36.550201 11005 solver.cpp:486] Iteration 5600, lr = 1e-05
I1122 16:31:50.742998 11005 solver.cpp:214] Iteration 5620, loss = 1.14768
I1122 16:31:50.743156 11005 solver.cpp:229]     Train net output #0: accuracy = 0.494278
I1122 16:31:50.743171 11005 solver.cpp:229]     Train net output #1: loss = 1.14768 (* 1 = 1.14768 loss)
I1122 16:31:50.743187 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.456512
I1122 16:31:50.743196 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.983035
I1122 16:31:50.743201 11005 solver.cpp:486] Iteration 5620, lr = 1e-05
I1122 16:32:04.927459 11005 solver.cpp:214] Iteration 5640, loss = 1.29915
I1122 16:32:04.927510 11005 solver.cpp:229]     Train net output #0: accuracy = 0.57275
I1122 16:32:04.927521 11005 solver.cpp:229]     Train net output #1: loss = 1.29915 (* 1 = 1.29915 loss)
I1122 16:32:04.927530 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.470771
I1122 16:32:04.927536 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.579831
I1122 16:32:04.927542 11005 solver.cpp:486] Iteration 5640, lr = 1e-05
I1122 16:32:19.095402 11005 solver.cpp:214] Iteration 5660, loss = 0.597315
I1122 16:32:19.095458 11005 solver.cpp:229]     Train net output #0: accuracy = 0.770138
I1122 16:32:19.095470 11005 solver.cpp:229]     Train net output #1: loss = 0.597315 (* 1 = 0.597315 loss)
I1122 16:32:19.095477 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.77614
I1122 16:32:19.095484 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.764919
I1122 16:32:19.095491 11005 solver.cpp:486] Iteration 5660, lr = 1e-05
I1122 16:32:33.297668 11005 solver.cpp:214] Iteration 5680, loss = 0.500058
I1122 16:32:33.297749 11005 solver.cpp:229]     Train net output #0: accuracy = 0.838463
I1122 16:32:33.297762 11005 solver.cpp:229]     Train net output #1: loss = 0.500058 (* 1 = 0.500058 loss)
I1122 16:32:33.297770 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.880592
I1122 16:32:33.297776 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.827961
I1122 16:32:33.297783 11005 solver.cpp:486] Iteration 5680, lr = 1e-05
I1122 16:32:47.497730 11005 solver.cpp:214] Iteration 5700, loss = 1.32677
I1122 16:32:47.497778 11005 solver.cpp:229]     Train net output #0: accuracy = 0.556473
I1122 16:32:47.497791 11005 solver.cpp:229]     Train net output #1: loss = 1.32677 (* 1 = 1.32677 loss)
I1122 16:32:47.497797 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.570222
I1122 16:32:47.497804 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.556037
I1122 16:32:47.497814 11005 solver.cpp:486] Iteration 5700, lr = 1e-05
I1122 16:33:01.704568 11005 solver.cpp:214] Iteration 5720, loss = 0.446936
I1122 16:33:01.704617 11005 solver.cpp:229]     Train net output #0: accuracy = 0.765053
I1122 16:33:01.704628 11005 solver.cpp:229]     Train net output #1: loss = 0.446936 (* 1 = 0.446936 loss)
I1122 16:33:01.704637 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.685874
I1122 16:33:01.704643 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.994572
I1122 16:33:01.704650 11005 solver.cpp:486] Iteration 5720, lr = 1e-05
I1122 16:33:15.992936 11005 solver.cpp:214] Iteration 5740, loss = 1.13693
I1122 16:33:15.993137 11005 solver.cpp:229]     Train net output #0: accuracy = 0.489811
I1122 16:33:15.993187 11005 solver.cpp:229]     Train net output #1: loss = 1.13693 (* 1 = 1.13693 loss)
I1122 16:33:15.993222 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.456486
I1122 16:33:15.993242 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.789243
I1122 16:33:15.993263 11005 solver.cpp:486] Iteration 5740, lr = 1e-05
I1122 16:33:30.185443 11005 solver.cpp:214] Iteration 5760, loss = 0.43383
I1122 16:33:30.185489 11005 solver.cpp:229]     Train net output #0: accuracy = 0.82748
I1122 16:33:30.185502 11005 solver.cpp:229]     Train net output #1: loss = 0.43383 (* 1 = 0.43383 loss)
I1122 16:33:30.185509 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.817821
I1122 16:33:30.185515 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.830286
I1122 16:33:30.185523 11005 solver.cpp:486] Iteration 5760, lr = 1e-05
I1122 16:33:44.377763 11005 solver.cpp:214] Iteration 5780, loss = 1.05514
I1122 16:33:44.377810 11005 solver.cpp:229]     Train net output #0: accuracy = 0.595001
I1122 16:33:44.377821 11005 solver.cpp:229]     Train net output #1: loss = 1.05514 (* 1 = 1.05514 loss)
I1122 16:33:44.377830 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.503922
I1122 16:33:44.377835 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.607035
I1122 16:33:44.377842 11005 solver.cpp:486] Iteration 5780, lr = 1e-05
I1122 16:34:02.033135 11005 solver.cpp:214] Iteration 5800, loss = 0.43087
I1122 16:34:02.033293 11005 solver.cpp:229]     Train net output #0: accuracy = 0.807144
I1122 16:34:02.033315 11005 solver.cpp:229]     Train net output #1: loss = 0.430869 (* 1 = 0.430869 loss)
I1122 16:34:02.033320 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.679636
I1122 16:34:02.033324 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.882605
I1122 16:34:02.033330 11005 solver.cpp:486] Iteration 5800, lr = 1e-05
I1122 16:34:20.420475 11005 solver.cpp:214] Iteration 5820, loss = 0.488382
I1122 16:34:20.420523 11005 solver.cpp:229]     Train net output #0: accuracy = 0.812508
I1122 16:34:20.420536 11005 solver.cpp:229]     Train net output #1: loss = 0.488381 (* 1 = 0.488381 loss)
I1122 16:34:20.420544 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.759628
I1122 16:34:20.420552 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.830079
I1122 16:34:20.420562 11005 solver.cpp:486] Iteration 5820, lr = 1e-05
I1122 16:34:38.859033 11005 solver.cpp:214] Iteration 5840, loss = 0.249188
I1122 16:34:38.859159 11005 solver.cpp:229]     Train net output #0: accuracy = 0.898544
I1122 16:34:38.859182 11005 solver.cpp:229]     Train net output #1: loss = 0.249188 (* 1 = 0.249188 loss)
I1122 16:34:38.859187 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.816696
I1122 16:34:38.859191 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.980032
I1122 16:34:38.859197 11005 solver.cpp:486] Iteration 5840, lr = 1e-05
I1122 16:34:57.300803 11005 solver.cpp:214] Iteration 5860, loss = 0.489761
I1122 16:34:57.300848 11005 solver.cpp:229]     Train net output #0: accuracy = 0.796078
I1122 16:34:57.300859 11005 solver.cpp:229]     Train net output #1: loss = 0.48976 (* 1 = 0.48976 loss)
I1122 16:34:57.300868 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.901271
I1122 16:34:57.300873 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.76343
I1122 16:34:57.300882 11005 solver.cpp:486] Iteration 5860, lr = 1e-05
I1122 16:35:14.693002 11005 solver.cpp:214] Iteration 5880, loss = 0.286183
I1122 16:35:14.693212 11005 solver.cpp:229]     Train net output #0: accuracy = 0.89719
I1122 16:35:14.693289 11005 solver.cpp:229]     Train net output #1: loss = 0.286183 (* 1 = 0.286183 loss)
I1122 16:35:14.693328 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.919831
I1122 16:35:14.693367 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.889454
I1122 16:35:14.693414 11005 solver.cpp:486] Iteration 5880, lr = 1e-05
I1122 16:35:29.064986 11005 solver.cpp:214] Iteration 5900, loss = 0.353585
I1122 16:35:29.065045 11005 solver.cpp:229]     Train net output #0: accuracy = 0.833645
I1122 16:35:29.065057 11005 solver.cpp:229]     Train net output #1: loss = 0.353585 (* 1 = 0.353585 loss)
I1122 16:35:29.065065 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.773253
I1122 16:35:29.065071 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.89444
I1122 16:35:29.065078 11005 solver.cpp:486] Iteration 5900, lr = 1e-05
I1122 16:35:43.431797 11005 solver.cpp:214] Iteration 5920, loss = 0.253013
I1122 16:35:43.431859 11005 solver.cpp:229]     Train net output #0: accuracy = 0.913326
I1122 16:35:43.431869 11005 solver.cpp:229]     Train net output #1: loss = 0.253012 (* 1 = 0.253012 loss)
I1122 16:35:43.431876 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.909306
I1122 16:35:43.431884 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.914721
I1122 16:35:43.431890 11005 solver.cpp:486] Iteration 5920, lr = 1e-05
I1122 16:35:57.727427 11005 solver.cpp:214] Iteration 5940, loss = 1.2064
I1122 16:35:57.727643 11005 solver.cpp:229]     Train net output #0: accuracy = 0.569687
I1122 16:35:57.727695 11005 solver.cpp:229]     Train net output #1: loss = 1.2064 (* 1 = 1.2064 loss)
I1122 16:35:57.727718 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.619137
I1122 16:35:57.727742 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.568973
I1122 16:35:57.727751 11005 solver.cpp:486] Iteration 5940, lr = 1e-05
I1122 16:36:11.940906 11005 solver.cpp:214] Iteration 5960, loss = 0.635569
I1122 16:36:11.940953 11005 solver.cpp:229]     Train net output #0: accuracy = 0.704411
I1122 16:36:11.940963 11005 solver.cpp:229]     Train net output #1: loss = 0.635569 (* 1 = 0.635569 loss)
I1122 16:36:11.940970 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.630249
I1122 16:36:11.940976 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.779921
I1122 16:36:11.940984 11005 solver.cpp:486] Iteration 5960, lr = 1e-05
I1122 16:36:26.197468 11005 solver.cpp:214] Iteration 5980, loss = 0.386345
I1122 16:36:26.197515 11005 solver.cpp:229]     Train net output #0: accuracy = 0.831516
I1122 16:36:26.197525 11005 solver.cpp:229]     Train net output #1: loss = 0.386345 (* 1 = 0.386345 loss)
I1122 16:36:26.197532 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.866374
I1122 16:36:26.197538 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.812114
I1122 16:36:26.197546 11005 solver.cpp:486] Iteration 5980, lr = 1e-05
I1122 16:36:40.197993 11005 solver.cpp:361] Snapshotting to /home/gradescan/SegNet/Models/backup_basic/Training/segnet_basic_iter_6000.caffemodel
I1122 16:36:40.210183 11005 solver.cpp:369] Snapshotting solver state to /home/gradescan/SegNet/Models/backup_basic/Training/segnet_basic_iter_6000.solverstate
I1122 16:36:40.442981 11005 solver.cpp:214] Iteration 6000, loss = 1.14684
I1122 16:36:40.443022 11005 solver.cpp:229]     Train net output #0: accuracy = 0.494797
I1122 16:36:40.443033 11005 solver.cpp:229]     Train net output #1: loss = 1.14684 (* 1 = 1.14684 loss)
I1122 16:36:40.443040 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.456943
I1122 16:36:40.443047 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.984683
I1122 16:36:40.443054 11005 solver.cpp:486] Iteration 6000, lr = 1e-05
I1122 16:36:54.673480 11005 solver.cpp:214] Iteration 6020, loss = 1.29656
I1122 16:36:54.673527 11005 solver.cpp:229]     Train net output #0: accuracy = 0.570442
I1122 16:36:54.673538 11005 solver.cpp:229]     Train net output #1: loss = 1.29656 (* 1 = 1.29656 loss)
I1122 16:36:54.673545 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.476999
I1122 16:36:54.673552 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.576931
I1122 16:36:54.673562 11005 solver.cpp:486] Iteration 6020, lr = 1e-05
I1122 16:37:08.908673 11005 solver.cpp:214] Iteration 6040, loss = 0.576447
I1122 16:37:08.908720 11005 solver.cpp:229]     Train net output #0: accuracy = 0.776363
I1122 16:37:08.908730 11005 solver.cpp:229]     Train net output #1: loss = 0.576446 (* 1 = 0.576446 loss)
I1122 16:37:08.908737 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.788468
I1122 16:37:08.908745 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.765839
I1122 16:37:08.908751 11005 solver.cpp:486] Iteration 6040, lr = 1e-05
I1122 16:37:23.132859 11005 solver.cpp:214] Iteration 6060, loss = 0.494603
I1122 16:37:23.132973 11005 solver.cpp:229]     Train net output #0: accuracy = 0.84129
I1122 16:37:23.132987 11005 solver.cpp:229]     Train net output #1: loss = 0.494602 (* 1 = 0.494602 loss)
I1122 16:37:23.132994 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.884626
I1122 16:37:23.133000 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.830487
I1122 16:37:23.133008 11005 solver.cpp:486] Iteration 6060, lr = 1e-05
I1122 16:37:37.313470 11005 solver.cpp:214] Iteration 6080, loss = 1.32884
I1122 16:37:37.313519 11005 solver.cpp:229]     Train net output #0: accuracy = 0.555664
I1122 16:37:37.313531 11005 solver.cpp:229]     Train net output #1: loss = 1.32884 (* 1 = 1.32884 loss)
I1122 16:37:37.313539 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.564138
I1122 16:37:37.313544 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.555395
I1122 16:37:37.313554 11005 solver.cpp:486] Iteration 6080, lr = 1e-05
I1122 16:37:51.508687 11005 solver.cpp:214] Iteration 6100, loss = 0.434441
I1122 16:37:51.508731 11005 solver.cpp:229]     Train net output #0: accuracy = 0.771095
I1122 16:37:51.508744 11005 solver.cpp:229]     Train net output #1: loss = 0.43444 (* 1 = 0.43444 loss)
I1122 16:37:51.508750 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.693858
I1122 16:37:51.508757 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.994988
I1122 16:37:51.508766 11005 solver.cpp:486] Iteration 6100, lr = 1e-05
I1122 16:38:05.730733 11005 solver.cpp:214] Iteration 6120, loss = 1.13635
I1122 16:38:05.730810 11005 solver.cpp:229]     Train net output #0: accuracy = 0.490982
I1122 16:38:05.730824 11005 solver.cpp:229]     Train net output #1: loss = 1.13635 (* 1 = 1.13635 loss)
I1122 16:38:05.730831 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.457254
I1122 16:38:05.730837 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.794043
I1122 16:38:05.730844 11005 solver.cpp:486] Iteration 6120, lr = 1e-05
I1122 16:38:19.966759 11005 solver.cpp:214] Iteration 6140, loss = 0.428738
I1122 16:38:19.966806 11005 solver.cpp:229]     Train net output #0: accuracy = 0.829109
I1122 16:38:19.966819 11005 solver.cpp:229]     Train net output #1: loss = 0.428737 (* 1 = 0.428737 loss)
I1122 16:38:19.966825 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.818855
I1122 16:38:19.966831 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.832088
I1122 16:38:19.966838 11005 solver.cpp:486] Iteration 6140, lr = 1e-05
I1122 16:38:35.520472 11005 solver.cpp:214] Iteration 6160, loss = 1.04933
I1122 16:38:35.520515 11005 solver.cpp:229]     Train net output #0: accuracy = 0.596302
I1122 16:38:35.520527 11005 solver.cpp:229]     Train net output #1: loss = 1.04933 (* 1 = 1.04933 loss)
I1122 16:38:35.520536 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.504805
I1122 16:38:35.520544 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.608391
I1122 16:38:35.520555 11005 solver.cpp:486] Iteration 6160, lr = 1e-05
I1122 16:38:53.970636 11005 solver.cpp:214] Iteration 6180, loss = 0.412865
I1122 16:38:53.970746 11005 solver.cpp:229]     Train net output #0: accuracy = 0.817894
I1122 16:38:53.970759 11005 solver.cpp:229]     Train net output #1: loss = 0.412864 (* 1 = 0.412864 loss)
I1122 16:38:53.970767 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.689209
I1122 16:38:53.970772 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.894051
I1122 16:38:53.970785 11005 solver.cpp:486] Iteration 6180, lr = 1e-05
I1122 16:39:12.416429 11005 solver.cpp:214] Iteration 6200, loss = 0.4834
I1122 16:39:12.416472 11005 solver.cpp:229]     Train net output #0: accuracy = 0.817879
I1122 16:39:12.416483 11005 solver.cpp:229]     Train net output #1: loss = 0.483399 (* 1 = 0.483399 loss)
I1122 16:39:12.416491 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.761356
I1122 16:39:12.416497 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.836661
I1122 16:39:12.416507 11005 solver.cpp:486] Iteration 6200, lr = 1e-05
I1122 16:39:30.843698 11005 solver.cpp:214] Iteration 6220, loss = 0.243047
I1122 16:39:30.843824 11005 solver.cpp:229]     Train net output #0: accuracy = 0.901772
I1122 16:39:30.843837 11005 solver.cpp:229]     Train net output #1: loss = 0.243047 (* 1 = 0.243047 loss)
I1122 16:39:30.843844 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.821552
I1122 16:39:30.843852 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.981638
I1122 16:39:30.843860 11005 solver.cpp:486] Iteration 6220, lr = 1e-05
I1122 16:39:49.268051 11005 solver.cpp:214] Iteration 6240, loss = 0.476434
I1122 16:39:49.268098 11005 solver.cpp:229]     Train net output #0: accuracy = 0.802917
I1122 16:39:49.268110 11005 solver.cpp:229]     Train net output #1: loss = 0.476433 (* 1 = 0.476433 loss)
I1122 16:39:49.268116 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.904895
I1122 16:39:49.268123 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.771268
I1122 16:39:49.268132 11005 solver.cpp:486] Iteration 6240, lr = 1e-05
I1122 16:40:04.554007 11005 solver.cpp:214] Iteration 6260, loss = 0.272558
I1122 16:40:04.554111 11005 solver.cpp:229]     Train net output #0: accuracy = 0.903759
I1122 16:40:04.554123 11005 solver.cpp:229]     Train net output #1: loss = 0.272557 (* 1 = 0.272557 loss)
I1122 16:40:04.554131 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.923381
I1122 16:40:04.554137 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.897055
I1122 16:40:04.554144 11005 solver.cpp:486] Iteration 6260, lr = 1e-05
I1122 16:40:18.784440 11005 solver.cpp:214] Iteration 6280, loss = 0.341443
I1122 16:40:18.784483 11005 solver.cpp:229]     Train net output #0: accuracy = 0.840714
I1122 16:40:18.784494 11005 solver.cpp:229]     Train net output #1: loss = 0.341442 (* 1 = 0.341442 loss)
I1122 16:40:18.784502 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.78199
I1122 16:40:18.784507 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.899829
I1122 16:40:18.784513 11005 solver.cpp:486] Iteration 6280, lr = 1e-05
I1122 16:40:32.972362 11005 solver.cpp:214] Iteration 6300, loss = 0.246382
I1122 16:40:32.972412 11005 solver.cpp:229]     Train net output #0: accuracy = 0.916523
I1122 16:40:32.972424 11005 solver.cpp:229]     Train net output #1: loss = 0.246381 (* 1 = 0.246381 loss)
I1122 16:40:32.972429 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.915261
I1122 16:40:32.972436 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.916961
I1122 16:40:32.972442 11005 solver.cpp:486] Iteration 6300, lr = 1e-05
I1122 16:40:47.200604 11005 solver.cpp:214] Iteration 6320, loss = 1.20007
I1122 16:40:47.200712 11005 solver.cpp:229]     Train net output #0: accuracy = 0.567688
I1122 16:40:47.200726 11005 solver.cpp:229]     Train net output #1: loss = 1.20007 (* 1 = 1.20007 loss)
I1122 16:40:47.200732 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.613508
I1122 16:40:47.200738 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.567026
I1122 16:40:47.200749 11005 solver.cpp:486] Iteration 6320, lr = 1e-05
I1122 16:41:01.379258 11005 solver.cpp:214] Iteration 6340, loss = 0.629799
I1122 16:41:01.379304 11005 solver.cpp:229]     Train net output #0: accuracy = 0.707844
I1122 16:41:01.379315 11005 solver.cpp:229]     Train net output #1: loss = 0.629798 (* 1 = 0.629798 loss)
I1122 16:41:01.379323 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.634249
I1122 16:41:01.379329 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.782778
I1122 16:41:01.379338 11005 solver.cpp:486] Iteration 6340, lr = 1e-05
I1122 16:41:15.594079 11005 solver.cpp:214] Iteration 6360, loss = 0.380381
I1122 16:41:15.594123 11005 solver.cpp:229]     Train net output #0: accuracy = 0.833683
I1122 16:41:15.594135 11005 solver.cpp:229]     Train net output #1: loss = 0.38038 (* 1 = 0.38038 loss)
I1122 16:41:15.594141 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.872625
I1122 16:41:15.594148 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.812007
I1122 16:41:15.594156 11005 solver.cpp:486] Iteration 6360, lr = 1e-05
I1122 16:41:29.828171 11005 solver.cpp:214] Iteration 6380, loss = 1.14525
I1122 16:41:29.828408 11005 solver.cpp:229]     Train net output #0: accuracy = 0.495918
I1122 16:41:29.828435 11005 solver.cpp:229]     Train net output #1: loss = 1.14525 (* 1 = 1.14525 loss)
I1122 16:41:29.828460 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.458086
I1122 16:41:29.828479 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.985534
I1122 16:41:29.828488 11005 solver.cpp:486] Iteration 6380, lr = 1e-05
I1122 16:41:44.069577 11005 solver.cpp:214] Iteration 6400, loss = 1.28987
I1122 16:41:44.069622 11005 solver.cpp:229]     Train net output #0: accuracy = 0.569912
I1122 16:41:44.069633 11005 solver.cpp:229]     Train net output #1: loss = 1.28987 (* 1 = 1.28987 loss)
I1122 16:41:44.069640 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.488925
I1122 16:41:44.069648 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.575536
I1122 16:41:44.069656 11005 solver.cpp:486] Iteration 6400, lr = 1e-05
I1122 16:41:58.312472 11005 solver.cpp:214] Iteration 6420, loss = 0.556087
I1122 16:41:58.312516 11005 solver.cpp:229]     Train net output #0: accuracy = 0.780418
I1122 16:41:58.312527 11005 solver.cpp:229]     Train net output #1: loss = 0.556086 (* 1 = 0.556086 loss)
I1122 16:41:58.312536 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.796293
I1122 16:41:58.312541 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.766616
I1122 16:41:58.312551 11005 solver.cpp:486] Iteration 6420, lr = 1e-05
I1122 16:42:12.541939 11005 solver.cpp:214] Iteration 6440, loss = 0.488159
I1122 16:42:12.542034 11005 solver.cpp:229]     Train net output #0: accuracy = 0.84457
I1122 16:42:12.542047 11005 solver.cpp:229]     Train net output #1: loss = 0.488159 (* 1 = 0.488159 loss)
I1122 16:42:12.542054 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.888698
I1122 16:42:12.542060 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.83357
I1122 16:42:12.542068 11005 solver.cpp:486] Iteration 6440, lr = 1e-05
I1122 16:42:26.881381 11005 solver.cpp:214] Iteration 6460, loss = 1.32804
I1122 16:42:26.881424 11005 solver.cpp:229]     Train net output #0: accuracy = 0.555748
I1122 16:42:26.881436 11005 solver.cpp:229]     Train net output #1: loss = 1.32804 (* 1 = 1.32804 loss)
I1122 16:42:26.881443 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.557432
I1122 16:42:26.881449 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.555695
I1122 16:42:26.881458 11005 solver.cpp:486] Iteration 6460, lr = 1e-05
I1122 16:42:41.079576 11005 solver.cpp:214] Iteration 6480, loss = 0.422328
I1122 16:42:41.079623 11005 solver.cpp:229]     Train net output #0: accuracy = 0.777515
I1122 16:42:41.079634 11005 solver.cpp:229]     Train net output #1: loss = 0.422327 (* 1 = 0.422327 loss)
I1122 16:42:41.079643 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.702359
I1122 16:42:41.079648 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.995375
I1122 16:42:41.079656 11005 solver.cpp:486] Iteration 6480, lr = 1e-05
I1122 16:42:55.273298 11005 solver.cpp:214] Iteration 6500, loss = 1.13486
I1122 16:42:55.273515 11005 solver.cpp:229]     Train net output #0: accuracy = 0.49242
I1122 16:42:55.273573 11005 solver.cpp:229]     Train net output #1: loss = 1.13486 (* 1 = 1.13486 loss)
I1122 16:42:55.273597 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.458038
I1122 16:42:55.273605 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.801356
I1122 16:42:55.273615 11005 solver.cpp:486] Iteration 6500, lr = 1e-05
I1122 16:43:09.476492 11005 solver.cpp:214] Iteration 6520, loss = 0.423178
I1122 16:43:09.476541 11005 solver.cpp:229]     Train net output #0: accuracy = 0.83115
I1122 16:43:09.476553 11005 solver.cpp:229]     Train net output #1: loss = 0.423178 (* 1 = 0.423178 loss)
I1122 16:43:09.476560 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.821295
I1122 16:43:09.476567 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.834013
I1122 16:43:09.476574 11005 solver.cpp:486] Iteration 6520, lr = 1e-05
I1122 16:43:23.728245 11005 solver.cpp:214] Iteration 6540, loss = 1.04452
I1122 16:43:23.728291 11005 solver.cpp:229]     Train net output #0: accuracy = 0.59803
I1122 16:43:23.728302 11005 solver.cpp:229]     Train net output #1: loss = 1.04452 (* 1 = 1.04452 loss)
I1122 16:43:23.728309 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.504641
I1122 16:43:23.728317 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.610369
I1122 16:43:23.728323 11005 solver.cpp:486] Iteration 6540, lr = 1e-05
I1122 16:43:38.003448 11005 solver.cpp:214] Iteration 6560, loss = 0.396136
I1122 16:43:38.003526 11005 solver.cpp:229]     Train net output #0: accuracy = 0.827785
I1122 16:43:38.003540 11005 solver.cpp:229]     Train net output #1: loss = 0.396135 (* 1 = 0.396135 loss)
I1122 16:43:38.003547 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.697571
I1122 16:43:38.003553 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.904848
I1122 16:43:38.003561 11005 solver.cpp:486] Iteration 6560, lr = 1e-05
I1122 16:43:52.248803 11005 solver.cpp:214] Iteration 6580, loss = 0.479422
I1122 16:43:52.248848 11005 solver.cpp:229]     Train net output #0: accuracy = 0.822041
I1122 16:43:52.248860 11005 solver.cpp:229]     Train net output #1: loss = 0.479421 (* 1 = 0.479421 loss)
I1122 16:43:52.248867 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.762305
I1122 16:43:52.248873 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.84189
I1122 16:43:52.248881 11005 solver.cpp:486] Iteration 6580, lr = 1e-05
I1122 16:44:06.470964 11005 solver.cpp:214] Iteration 6600, loss = 0.237791
I1122 16:44:06.471010 11005 solver.cpp:229]     Train net output #0: accuracy = 0.903721
I1122 16:44:06.471022 11005 solver.cpp:229]     Train net output #1: loss = 0.237791 (* 1 = 0.237791 loss)
I1122 16:44:06.471029 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.824281
I1122 16:44:06.471035 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.982811
I1122 16:44:06.471042 11005 solver.cpp:486] Iteration 6600, lr = 1e-05
I1122 16:44:20.668642 11005 solver.cpp:214] Iteration 6620, loss = 0.463071
I1122 16:44:20.668838 11005 solver.cpp:229]     Train net output #0: accuracy = 0.809361
I1122 16:44:20.668895 11005 solver.cpp:229]     Train net output #1: loss = 0.46307 (* 1 = 0.46307 loss)
I1122 16:44:20.668917 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.908744
I1122 16:44:20.668941 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.778516
I1122 16:44:20.668967 11005 solver.cpp:486] Iteration 6620, lr = 1e-05
I1122 16:44:34.890197 11005 solver.cpp:214] Iteration 6640, loss = 0.259697
I1122 16:44:34.890243 11005 solver.cpp:229]     Train net output #0: accuracy = 0.911285
I1122 16:44:34.890254 11005 solver.cpp:229]     Train net output #1: loss = 0.259697 (* 1 = 0.259697 loss)
I1122 16:44:34.890260 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.926841
I1122 16:44:34.890267 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.90597
I1122 16:44:34.890275 11005 solver.cpp:486] Iteration 6640, lr = 1e-05
I1122 16:44:49.116816 11005 solver.cpp:214] Iteration 6660, loss = 0.329557
I1122 16:44:49.116865 11005 solver.cpp:229]     Train net output #0: accuracy = 0.847973
I1122 16:44:49.116876 11005 solver.cpp:229]     Train net output #1: loss = 0.329557 (* 1 = 0.329557 loss)
I1122 16:44:49.116884 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.789572
I1122 16:44:49.116890 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.906764
I1122 16:44:49.116899 11005 solver.cpp:486] Iteration 6660, lr = 1e-05
I1122 16:45:03.539757 11005 solver.cpp:214] Iteration 6680, loss = 0.240705
I1122 16:45:03.539851 11005 solver.cpp:229]     Train net output #0: accuracy = 0.918983
I1122 16:45:03.539865 11005 solver.cpp:229]     Train net output #1: loss = 0.240705 (* 1 = 0.240705 loss)
I1122 16:45:03.539872 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.919719
I1122 16:45:03.539878 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.918728
I1122 16:45:03.539886 11005 solver.cpp:486] Iteration 6680, lr = 1e-05
I1122 16:45:21.918578 11005 solver.cpp:214] Iteration 6700, loss = 1.19065
I1122 16:45:21.918624 11005 solver.cpp:229]     Train net output #0: accuracy = 0.567417
I1122 16:45:21.918637 11005 solver.cpp:229]     Train net output #1: loss = 1.19065 (* 1 = 1.19065 loss)
I1122 16:45:21.918643 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.612168
I1122 16:45:21.918649 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.566771
I1122 16:45:21.918658 11005 solver.cpp:486] Iteration 6700, lr = 1e-05
I1122 16:45:40.402441 11005 solver.cpp:214] Iteration 6720, loss = 0.62081
I1122 16:45:40.402580 11005 solver.cpp:229]     Train net output #0: accuracy = 0.713696
I1122 16:45:40.402608 11005 solver.cpp:229]     Train net output #1: loss = 0.62081 (* 1 = 0.62081 loss)
I1122 16:45:40.402616 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.639066
I1122 16:45:40.402623 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.789684
I1122 16:45:40.402637 11005 solver.cpp:486] Iteration 6720, lr = 1e-05
I1122 16:45:58.798202 11005 solver.cpp:214] Iteration 6740, loss = 0.374245
I1122 16:45:58.798249 11005 solver.cpp:229]     Train net output #0: accuracy = 0.836086
I1122 16:45:58.798260 11005 solver.cpp:229]     Train net output #1: loss = 0.374244 (* 1 = 0.374244 loss)
I1122 16:45:58.798267 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.876487
I1122 16:45:58.798274 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.813598
I1122 16:45:58.798283 11005 solver.cpp:486] Iteration 6740, lr = 1e-05
I1122 16:46:17.214226 11005 solver.cpp:214] Iteration 6760, loss = 1.14353
I1122 16:46:17.214315 11005 solver.cpp:229]     Train net output #0: accuracy = 0.49617
I1122 16:46:17.214329 11005 solver.cpp:229]     Train net output #1: loss = 1.14353 (* 1 = 1.14353 loss)
I1122 16:46:17.214335 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.458287
I1122 16:46:17.214342 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.986438
I1122 16:46:17.214351 11005 solver.cpp:486] Iteration 6760, lr = 1e-05
I1122 16:46:33.691154 11005 solver.cpp:214] Iteration 6780, loss = 1.27896
I1122 16:46:33.691200 11005 solver.cpp:229]     Train net output #0: accuracy = 0.570122
I1122 16:46:33.691211 11005 solver.cpp:229]     Train net output #1: loss = 1.27896 (* 1 = 1.27896 loss)
I1122 16:46:33.691218 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.499853
I1122 16:46:33.691226 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.575001
I1122 16:46:33.691236 11005 solver.cpp:486] Iteration 6780, lr = 1e-05
I1122 16:46:47.926223 11005 solver.cpp:214] Iteration 6800, loss = 0.540028
I1122 16:46:47.926436 11005 solver.cpp:229]     Train net output #0: accuracy = 0.784542
I1122 16:46:47.926488 11005 solver.cpp:229]     Train net output #1: loss = 0.540027 (* 1 = 0.540027 loss)
I1122 16:46:47.926515 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.804142
I1122 16:46:47.926535 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.767501
I1122 16:46:47.926558 11005 solver.cpp:486] Iteration 6800, lr = 1e-05
I1122 16:47:02.150816 11005 solver.cpp:214] Iteration 6820, loss = 0.480547
I1122 16:47:02.150861 11005 solver.cpp:229]     Train net output #0: accuracy = 0.847099
I1122 16:47:02.150872 11005 solver.cpp:229]     Train net output #1: loss = 0.480547 (* 1 = 0.480547 loss)
I1122 16:47:02.150881 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.892368
I1122 16:47:02.150885 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.835815
I1122 16:47:02.150895 11005 solver.cpp:486] Iteration 6820, lr = 1e-05
I1122 16:47:16.354761 11005 solver.cpp:214] Iteration 6840, loss = 1.32477
I1122 16:47:16.354811 11005 solver.cpp:229]     Train net output #0: accuracy = 0.556389
I1122 16:47:16.354825 11005 solver.cpp:229]     Train net output #1: loss = 1.32477 (* 1 = 1.32477 loss)
I1122 16:47:16.354831 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.557184
I1122 16:47:16.354838 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.556364
I1122 16:47:16.354846 11005 solver.cpp:486] Iteration 6840, lr = 1e-05
I1122 16:47:30.557593 11005 solver.cpp:214] Iteration 6860, loss = 0.411726
I1122 16:47:30.557785 11005 solver.cpp:229]     Train net output #0: accuracy = 0.782295
I1122 16:47:30.557813 11005 solver.cpp:229]     Train net output #1: loss = 0.411725 (* 1 = 0.411725 loss)
I1122 16:47:30.557821 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.708593
I1122 16:47:30.557831 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.99594
I1122 16:47:30.557842 11005 solver.cpp:486] Iteration 6860, lr = 1e-05
I1122 16:47:44.787578 11005 solver.cpp:214] Iteration 6880, loss = 1.13349
I1122 16:47:44.787626 11005 solver.cpp:229]     Train net output #0: accuracy = 0.492847
I1122 16:47:44.787639 11005 solver.cpp:229]     Train net output #1: loss = 1.13349 (* 1 = 1.13349 loss)
I1122 16:47:44.787647 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.457762
I1122 16:47:44.787652 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.808098
I1122 16:47:44.787662 11005 solver.cpp:486] Iteration 6880, lr = 1e-05
I1122 16:47:59.009831 11005 solver.cpp:214] Iteration 6900, loss = 0.418805
I1122 16:47:59.009881 11005 solver.cpp:229]     Train net output #0: accuracy = 0.832787
I1122 16:47:59.009892 11005 solver.cpp:229]     Train net output #1: loss = 0.418804 (* 1 = 0.418804 loss)
I1122 16:47:59.009899 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.822905
I1122 16:47:59.009907 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.835657
I1122 16:47:59.009914 11005 solver.cpp:486] Iteration 6900, lr = 1e-05
I1122 16:48:13.208588 11005 solver.cpp:214] Iteration 6920, loss = 1.04098
I1122 16:48:13.208679 11005 solver.cpp:229]     Train net output #0: accuracy = 0.599136
I1122 16:48:13.208691 11005 solver.cpp:229]     Train net output #1: loss = 1.04097 (* 1 = 1.04097 loss)
I1122 16:48:13.208698 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.502125
I1122 16:48:13.208705 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.611954
I1122 16:48:13.208711 11005 solver.cpp:486] Iteration 6920, lr = 1e-05
I1122 16:48:27.443104 11005 solver.cpp:214] Iteration 6940, loss = 0.380512
I1122 16:48:27.443150 11005 solver.cpp:229]     Train net output #0: accuracy = 0.837715
I1122 16:48:27.443162 11005 solver.cpp:229]     Train net output #1: loss = 0.380511 (* 1 = 0.380511 loss)
I1122 16:48:27.443169 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.706395
I1122 16:48:27.443176 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.915431
I1122 16:48:27.443182 11005 solver.cpp:486] Iteration 6940, lr = 1e-05
I1122 16:48:41.639986 11005 solver.cpp:214] Iteration 6960, loss = 0.476045
I1122 16:48:41.640035 11005 solver.cpp:229]     Train net output #0: accuracy = 0.82579
I1122 16:48:41.640046 11005 solver.cpp:229]     Train net output #1: loss = 0.476044 (* 1 = 0.476044 loss)
I1122 16:48:41.640054 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.764293
I1122 16:48:41.640060 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.846225
I1122 16:48:41.640066 11005 solver.cpp:486] Iteration 6960, lr = 1e-05
I1122 16:48:55.889511 11005 solver.cpp:214] Iteration 6980, loss = 0.232994
I1122 16:48:55.889626 11005 solver.cpp:229]     Train net output #0: accuracy = 0.905785
I1122 16:48:55.889639 11005 solver.cpp:229]     Train net output #1: loss = 0.232993 (* 1 = 0.232993 loss)
I1122 16:48:55.889647 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.827401
I1122 16:48:55.889653 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.983823
I1122 16:48:55.889662 11005 solver.cpp:486] Iteration 6980, lr = 1e-05
I1122 16:49:09.887775 11005 solver.cpp:361] Snapshotting to /home/gradescan/SegNet/Models/backup_basic/Training/segnet_basic_iter_7000.caffemodel
I1122 16:49:09.898932 11005 solver.cpp:369] Snapshotting solver state to /home/gradescan/SegNet/Models/backup_basic/Training/segnet_basic_iter_7000.solverstate
I1122 16:49:10.131639 11005 solver.cpp:214] Iteration 7000, loss = 0.45067
I1122 16:49:10.131681 11005 solver.cpp:229]     Train net output #0: accuracy = 0.815678
I1122 16:49:10.131692 11005 solver.cpp:229]     Train net output #1: loss = 0.450669 (* 1 = 0.450669 loss)
I1122 16:49:10.131700 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.912303
I1122 16:49:10.131706 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.785689
I1122 16:49:10.131714 11005 solver.cpp:486] Iteration 7000, lr = 1e-05
I1122 16:49:24.373481 11005 solver.cpp:214] Iteration 7020, loss = 0.247887
I1122 16:49:24.373528 11005 solver.cpp:229]     Train net output #0: accuracy = 0.918148
I1122 16:49:24.373539 11005 solver.cpp:229]     Train net output #1: loss = 0.247886 (* 1 = 0.247886 loss)
I1122 16:49:24.373546 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.928968
I1122 16:49:24.373553 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.914451
I1122 16:49:24.373561 11005 solver.cpp:486] Iteration 7020, lr = 1e-05
I1122 16:49:38.582114 11005 solver.cpp:214] Iteration 7040, loss = 0.317368
I1122 16:49:38.582186 11005 solver.cpp:229]     Train net output #0: accuracy = 0.855904
I1122 16:49:38.582200 11005 solver.cpp:229]     Train net output #1: loss = 0.317368 (* 1 = 0.317368 loss)
I1122 16:49:38.582206 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.797647
I1122 16:49:38.582211 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.914549
I1122 16:49:38.582221 11005 solver.cpp:486] Iteration 7040, lr = 1e-05
I1122 16:49:52.782377 11005 solver.cpp:214] Iteration 7060, loss = 0.235025
I1122 16:49:52.782421 11005 solver.cpp:229]     Train net output #0: accuracy = 0.921078
I1122 16:49:52.782433 11005 solver.cpp:229]     Train net output #1: loss = 0.235024 (* 1 = 0.235024 loss)
I1122 16:49:52.782439 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.924222
I1122 16:49:52.782446 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.919987
I1122 16:49:52.782454 11005 solver.cpp:486] Iteration 7060, lr = 1e-05
I1122 16:50:06.978888 11005 solver.cpp:214] Iteration 7080, loss = 1.17787
I1122 16:50:06.978936 11005 solver.cpp:229]     Train net output #0: accuracy = 0.567055
I1122 16:50:06.978948 11005 solver.cpp:229]     Train net output #1: loss = 1.17787 (* 1 = 1.17787 loss)
I1122 16:50:06.978955 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.599035
I1122 16:50:06.978962 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.566593
I1122 16:50:06.978971 11005 solver.cpp:486] Iteration 7080, lr = 1e-05
I1122 16:50:21.170653 11005 solver.cpp:214] Iteration 7100, loss = 0.60964
I1122 16:50:21.170770 11005 solver.cpp:229]     Train net output #0: accuracy = 0.720737
I1122 16:50:21.170783 11005 solver.cpp:229]     Train net output #1: loss = 0.60964 (* 1 = 0.60964 loss)
I1122 16:50:21.170791 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.642778
I1122 16:50:21.170797 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.800115
I1122 16:50:21.170809 11005 solver.cpp:486] Iteration 7100, lr = 1e-05
I1122 16:50:35.371191 11005 solver.cpp:214] Iteration 7120, loss = 0.368397
I1122 16:50:35.371239 11005 solver.cpp:229]     Train net output #0: accuracy = 0.839062
I1122 16:50:35.371251 11005 solver.cpp:229]     Train net output #1: loss = 0.368397 (* 1 = 0.368397 loss)
I1122 16:50:35.371258 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.881223
I1122 16:50:35.371264 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.815593
I1122 16:50:35.371271 11005 solver.cpp:486] Iteration 7120, lr = 1e-05
I1122 16:50:49.599606 11005 solver.cpp:214] Iteration 7140, loss = 1.14185
I1122 16:50:49.599652 11005 solver.cpp:229]     Train net output #0: accuracy = 0.495678
I1122 16:50:49.599664 11005 solver.cpp:229]     Train net output #1: loss = 1.14185 (* 1 = 1.14185 loss)
I1122 16:50:49.599671 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.457671
I1122 16:50:49.599678 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.987555
I1122 16:50:49.599684 11005 solver.cpp:486] Iteration 7140, lr = 1e-05
I1122 16:51:03.799816 11005 solver.cpp:214] Iteration 7160, loss = 1.26363
I1122 16:51:03.800065 11005 solver.cpp:229]     Train net output #0: accuracy = 0.569378
I1122 16:51:03.800118 11005 solver.cpp:229]     Train net output #1: loss = 1.26362 (* 1 = 1.26362 loss)
I1122 16:51:03.800142 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.510781
I1122 16:51:03.800163 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.573447
I1122 16:51:03.800194 11005 solver.cpp:486] Iteration 7160, lr = 1e-05
I1122 16:51:17.986639 11005 solver.cpp:214] Iteration 7180, loss = 0.528802
I1122 16:51:17.986686 11005 solver.cpp:229]     Train net output #0: accuracy = 0.785881
I1122 16:51:17.986697 11005 solver.cpp:229]     Train net output #1: loss = 0.528801 (* 1 = 0.528801 loss)
I1122 16:51:17.986704 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.807751
I1122 16:51:17.986716 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.766866
I1122 16:51:17.986726 11005 solver.cpp:486] Iteration 7180, lr = 1e-05
I1122 16:51:32.213269 11005 solver.cpp:214] Iteration 7200, loss = 0.472202
I1122 16:51:32.213315 11005 solver.cpp:229]     Train net output #0: accuracy = 0.850052
I1122 16:51:32.213325 11005 solver.cpp:229]     Train net output #1: loss = 0.472201 (* 1 = 0.472201 loss)
I1122 16:51:32.213333 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.896115
I1122 16:51:32.213340 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.838569
I1122 16:51:32.213346 11005 solver.cpp:486] Iteration 7200, lr = 1e-05
I1122 16:51:46.424813 11005 solver.cpp:214] Iteration 7220, loss = 1.31837
I1122 16:51:46.424919 11005 solver.cpp:229]     Train net output #0: accuracy = 0.557137
I1122 16:51:46.424932 11005 solver.cpp:229]     Train net output #1: loss = 1.31837 (* 1 = 1.31837 loss)
I1122 16:51:46.424940 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.551844
I1122 16:51:46.424947 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.557304
I1122 16:51:46.424952 11005 solver.cpp:486] Iteration 7220, lr = 1e-05
I1122 16:52:00.644188 11005 solver.cpp:214] Iteration 7240, loss = 0.400837
I1122 16:52:00.644234 11005 solver.cpp:229]     Train net output #0: accuracy = 0.787739
I1122 16:52:00.644245 11005 solver.cpp:229]     Train net output #1: loss = 0.400836 (* 1 = 0.400836 loss)
I1122 16:52:00.644253 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.71575
I1122 16:52:00.644259 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.996416
I1122 16:52:00.644268 11005 solver.cpp:486] Iteration 7240, lr = 1e-05
I1122 16:52:14.853085 11005 solver.cpp:214] Iteration 7260, loss = 1.13136
I1122 16:52:14.853129 11005 solver.cpp:229]     Train net output #0: accuracy = 0.49202
I1122 16:52:14.853140 11005 solver.cpp:229]     Train net output #1: loss = 1.13136 (* 1 = 1.13136 loss)
I1122 16:52:14.853148 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.456257
I1122 16:52:14.853154 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.813355
I1122 16:52:14.853163 11005 solver.cpp:486] Iteration 7260, lr = 1e-05
I1122 16:52:29.079408 11005 solver.cpp:214] Iteration 7280, loss = 0.413876
I1122 16:52:29.079526 11005 solver.cpp:229]     Train net output #0: accuracy = 0.834892
I1122 16:52:29.079540 11005 solver.cpp:229]     Train net output #1: loss = 0.413876 (* 1 = 0.413876 loss)
I1122 16:52:29.079546 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.826041
I1122 16:52:29.079553 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.837464
I1122 16:52:29.079560 11005 solver.cpp:486] Iteration 7280, lr = 1e-05
I1122 16:52:43.282663 11005 solver.cpp:214] Iteration 7300, loss = 1.03787
I1122 16:52:43.282709 11005 solver.cpp:229]     Train net output #0: accuracy = 0.599396
I1122 16:52:43.282721 11005 solver.cpp:229]     Train net output #1: loss = 1.03787 (* 1 = 1.03787 loss)
I1122 16:52:43.282728 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.502157
I1122 16:52:43.282734 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.612244
I1122 16:52:43.282742 11005 solver.cpp:486] Iteration 7300, lr = 1e-05
I1122 16:52:57.476591 11005 solver.cpp:214] Iteration 7320, loss = 0.366118
I1122 16:52:57.476640 11005 solver.cpp:229]     Train net output #0: accuracy = 0.845764
I1122 16:52:57.476651 11005 solver.cpp:229]     Train net output #1: loss = 0.366117 (* 1 = 0.366117 loss)
I1122 16:52:57.476658 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.713547
I1122 16:52:57.476665 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.924012
I1122 16:52:57.476671 11005 solver.cpp:486] Iteration 7320, lr = 1e-05
I1122 16:53:11.681486 11005 solver.cpp:214] Iteration 7340, loss = 0.473881
I1122 16:53:11.681603 11005 solver.cpp:229]     Train net output #0: accuracy = 0.82872
I1122 16:53:11.681617 11005 solver.cpp:229]     Train net output #1: loss = 0.47388 (* 1 = 0.47388 loss)
I1122 16:53:11.681624 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.764645
I1122 16:53:11.681630 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.850012
I1122 16:53:11.681643 11005 solver.cpp:486] Iteration 7340, lr = 1e-05
I1122 16:53:25.863148 11005 solver.cpp:214] Iteration 7360, loss = 0.22935
I1122 16:53:25.863194 11005 solver.cpp:229]     Train net output #0: accuracy = 0.906944
I1122 16:53:25.863206 11005 solver.cpp:229]     Train net output #1: loss = 0.229349 (* 1 = 0.229349 loss)
I1122 16:53:25.863214 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.829083
I1122 16:53:25.863219 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.984463
I1122 16:53:25.863226 11005 solver.cpp:486] Iteration 7360, lr = 1e-05
I1122 16:53:40.107805 11005 solver.cpp:214] Iteration 7380, loss = 0.438047
I1122 16:53:40.107853 11005 solver.cpp:229]     Train net output #0: accuracy = 0.821075
I1122 16:53:40.107864 11005 solver.cpp:229]     Train net output #1: loss = 0.438046 (* 1 = 0.438046 loss)
I1122 16:53:40.107872 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.915122
I1122 16:53:40.107878 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.791887
I1122 16:53:40.107884 11005 solver.cpp:486] Iteration 7380, lr = 1e-05
I1122 16:53:54.320432 11005 solver.cpp:214] Iteration 7400, loss = 0.237765
I1122 16:53:54.320547 11005 solver.cpp:229]     Train net output #0: accuracy = 0.924362
I1122 16:53:54.320560 11005 solver.cpp:229]     Train net output #1: loss = 0.237765 (* 1 = 0.237765 loss)
I1122 16:53:54.320570 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.929822
I1122 16:53:54.320576 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.922497
I1122 16:53:54.320583 11005 solver.cpp:486] Iteration 7400, lr = 1e-05
I1122 16:54:08.613198 11005 solver.cpp:214] Iteration 7420, loss = 0.304989
I1122 16:54:08.613242 11005 solver.cpp:229]     Train net output #0: accuracy = 0.86375
I1122 16:54:08.613253 11005 solver.cpp:229]     Train net output #1: loss = 0.304988 (* 1 = 0.304988 loss)
I1122 16:54:08.613260 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.804871
I1122 16:54:08.613267 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.923023
I1122 16:54:08.613273 11005 solver.cpp:486] Iteration 7420, lr = 1e-05
I1122 16:54:22.956449 11005 solver.cpp:214] Iteration 7440, loss = 0.229068
I1122 16:54:22.956493 11005 solver.cpp:229]     Train net output #0: accuracy = 0.92416
I1122 16:54:22.956504 11005 solver.cpp:229]     Train net output #1: loss = 0.229068 (* 1 = 0.229068 loss)
I1122 16:54:22.956511 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.928947
I1122 16:54:22.956516 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.922499
I1122 16:54:22.956526 11005 solver.cpp:486] Iteration 7440, lr = 1e-05
I1122 16:54:37.285827 11005 solver.cpp:214] Iteration 7460, loss = 1.16174
I1122 16:54:37.285899 11005 solver.cpp:229]     Train net output #0: accuracy = 0.567318
I1122 16:54:37.285912 11005 solver.cpp:229]     Train net output #1: loss = 1.16174 (* 1 = 1.16174 loss)
I1122 16:54:37.285918 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.601715
I1122 16:54:37.285925 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.566821
I1122 16:54:37.285931 11005 solver.cpp:486] Iteration 7460, lr = 1e-05
I1122 16:54:51.954576 11005 solver.cpp:214] Iteration 7480, loss = 0.59576
I1122 16:54:51.954624 11005 solver.cpp:229]     Train net output #0: accuracy = 0.729725
I1122 16:54:51.954635 11005 solver.cpp:229]     Train net output #1: loss = 0.595759 (* 1 = 0.595759 loss)
I1122 16:54:51.954643 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.646922
I1122 16:54:51.954649 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.814035
I1122 16:54:51.954661 11005 solver.cpp:486] Iteration 7480, lr = 1e-05
I1122 16:55:09.947276 11005 solver.cpp:214] Iteration 7500, loss = 0.36238
I1122 16:55:09.947391 11005 solver.cpp:229]     Train net output #0: accuracy = 0.842216
I1122 16:55:09.947402 11005 solver.cpp:229]     Train net output #1: loss = 0.362379 (* 1 = 0.362379 loss)
I1122 16:55:09.947410 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.885555
I1122 16:55:09.947417 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.818093
I1122 16:55:09.947425 11005 solver.cpp:486] Iteration 7500, lr = 1e-05
I1122 16:55:27.894636 11005 solver.cpp:214] Iteration 7520, loss = 1.14109
I1122 16:55:27.894680 11005 solver.cpp:229]     Train net output #0: accuracy = 0.495274
I1122 16:55:27.894691 11005 solver.cpp:229]     Train net output #1: loss = 1.14109 (* 1 = 1.14109 loss)
I1122 16:55:27.894698 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.457136
I1122 16:55:27.894704 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.988832
I1122 16:55:27.894713 11005 solver.cpp:486] Iteration 7520, lr = 1e-05
I1122 16:55:45.891108 11005 solver.cpp:214] Iteration 7540, loss = 1.24456
I1122 16:55:45.891196 11005 solver.cpp:229]     Train net output #0: accuracy = 0.570744
I1122 16:55:45.891209 11005 solver.cpp:229]     Train net output #1: loss = 1.24456 (* 1 = 1.24456 loss)
I1122 16:55:45.891216 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.52776
I1122 16:55:45.891223 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.573728
I1122 16:55:45.891232 11005 solver.cpp:486] Iteration 7540, lr = 1e-05
I1122 16:56:03.865306 11005 solver.cpp:214] Iteration 7560, loss = 0.518521
I1122 16:56:03.865358 11005 solver.cpp:229]     Train net output #0: accuracy = 0.788136
I1122 16:56:03.865370 11005 solver.cpp:229]     Train net output #1: loss = 0.518521 (* 1 = 0.518521 loss)
I1122 16:56:03.865376 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.812082
I1122 16:56:03.865386 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.767315
I1122 16:56:03.865396 11005 solver.cpp:486] Iteration 7560, lr = 1e-05
I1122 16:56:21.789592 11005 solver.cpp:214] Iteration 7580, loss = 0.463826
I1122 16:56:21.789710 11005 solver.cpp:229]     Train net output #0: accuracy = 0.852112
I1122 16:56:21.789723 11005 solver.cpp:229]     Train net output #1: loss = 0.463825 (* 1 = 0.463825 loss)
I1122 16:56:21.789731 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.89774
I1122 16:56:21.789736 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.840738
I1122 16:56:21.789746 11005 solver.cpp:486] Iteration 7580, lr = 1e-05
I1122 16:56:36.268239 11005 solver.cpp:214] Iteration 7600, loss = 1.30857
I1122 16:56:36.268285 11005 solver.cpp:229]     Train net output #0: accuracy = 0.558403
I1122 16:56:36.268296 11005 solver.cpp:229]     Train net output #1: loss = 1.30857 (* 1 = 1.30857 loss)
I1122 16:56:36.268303 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.547125
I1122 16:56:36.268311 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.55876
I1122 16:56:36.268316 11005 solver.cpp:486] Iteration 7600, lr = 1e-05
I1122 16:56:50.469368 11005 solver.cpp:214] Iteration 7620, loss = 0.38995
I1122 16:56:50.469413 11005 solver.cpp:229]     Train net output #0: accuracy = 0.793755
I1122 16:56:50.469424 11005 solver.cpp:229]     Train net output #1: loss = 0.389949 (* 1 = 0.389949 loss)
I1122 16:56:50.469431 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.723769
I1122 16:56:50.469437 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.996624
I1122 16:56:50.469446 11005 solver.cpp:486] Iteration 7620, lr = 1e-05
I1122 16:57:04.708883 11005 solver.cpp:214] Iteration 7640, loss = 1.12995
I1122 16:57:04.709072 11005 solver.cpp:229]     Train net output #0: accuracy = 0.489647
I1122 16:57:04.709100 11005 solver.cpp:229]     Train net output #1: loss = 1.12995 (* 1 = 1.12995 loss)
I1122 16:57:04.709108 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.452921
I1122 16:57:04.709115 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.81964
I1122 16:57:04.709128 11005 solver.cpp:486] Iteration 7640, lr = 1e-05
I1122 16:57:18.919523 11005 solver.cpp:214] Iteration 7660, loss = 0.409677
I1122 16:57:18.919569 11005 solver.cpp:229]     Train net output #0: accuracy = 0.837299
I1122 16:57:18.919581 11005 solver.cpp:229]     Train net output #1: loss = 0.409676 (* 1 = 0.409676 loss)
I1122 16:57:18.919589 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.829294
I1122 16:57:18.919595 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.839625
I1122 16:57:18.919605 11005 solver.cpp:486] Iteration 7660, lr = 1e-05
I1122 16:57:33.104650 11005 solver.cpp:214] Iteration 7680, loss = 1.03571
I1122 16:57:33.104696 11005 solver.cpp:229]     Train net output #0: accuracy = 0.600555
I1122 16:57:33.104708 11005 solver.cpp:229]     Train net output #1: loss = 1.03571 (* 1 = 1.03571 loss)
I1122 16:57:33.104715 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.501765
I1122 16:57:33.104723 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.613608
I1122 16:57:33.104732 11005 solver.cpp:486] Iteration 7680, lr = 1e-05
I1122 16:57:47.304261 11005 solver.cpp:214] Iteration 7700, loss = 0.353534
I1122 16:57:47.304365 11005 solver.cpp:229]     Train net output #0: accuracy = 0.853672
I1122 16:57:47.304379 11005 solver.cpp:229]     Train net output #1: loss = 0.353533 (* 1 = 0.353533 loss)
I1122 16:57:47.304388 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.719929
I1122 16:57:47.304394 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.932822
I1122 16:57:47.304404 11005 solver.cpp:486] Iteration 7700, lr = 1e-05
I1122 16:58:01.509680 11005 solver.cpp:214] Iteration 7720, loss = 0.471797
I1122 16:58:01.509727 11005 solver.cpp:229]     Train net output #0: accuracy = 0.831131
I1122 16:58:01.509737 11005 solver.cpp:229]     Train net output #1: loss = 0.471796 (* 1 = 0.471796 loss)
I1122 16:58:01.509744 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.765715
I1122 16:58:01.509752 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.852868
I1122 16:58:01.509758 11005 solver.cpp:486] Iteration 7720, lr = 1e-05
I1122 16:58:15.750579 11005 solver.cpp:214] Iteration 7740, loss = 0.225827
I1122 16:58:15.750627 11005 solver.cpp:229]     Train net output #0: accuracy = 0.907677
I1122 16:58:15.750638 11005 solver.cpp:229]     Train net output #1: loss = 0.225826 (* 1 = 0.225826 loss)
I1122 16:58:15.750644 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.829886
I1122 16:58:15.750651 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.985125
I1122 16:58:15.750658 11005 solver.cpp:486] Iteration 7740, lr = 1e-05
I1122 16:58:29.966902 11005 solver.cpp:214] Iteration 7760, loss = 0.424554
I1122 16:58:29.967008 11005 solver.cpp:229]     Train net output #0: accuracy = 0.827663
I1122 16:58:29.967021 11005 solver.cpp:229]     Train net output #1: loss = 0.424554 (* 1 = 0.424554 loss)
I1122 16:58:29.967028 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.919374
I1122 16:58:29.967034 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.7992
I1122 16:58:29.967042 11005 solver.cpp:486] Iteration 7760, lr = 1e-05
I1122 16:58:44.171509 11005 solver.cpp:214] Iteration 7780, loss = 0.229633
I1122 16:58:44.171555 11005 solver.cpp:229]     Train net output #0: accuracy = 0.929577
I1122 16:58:44.171566 11005 solver.cpp:229]     Train net output #1: loss = 0.229632 (* 1 = 0.229632 loss)
I1122 16:58:44.171573 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.931545
I1122 16:58:44.171581 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.928904
I1122 16:58:44.171587 11005 solver.cpp:486] Iteration 7780, lr = 1e-05
I1122 16:58:58.380539 11005 solver.cpp:214] Iteration 7800, loss = 0.292596
I1122 16:58:58.380584 11005 solver.cpp:229]     Train net output #0: accuracy = 0.871227
I1122 16:58:58.380595 11005 solver.cpp:229]     Train net output #1: loss = 0.292595 (* 1 = 0.292595 loss)
I1122 16:58:58.380604 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.81189
I1122 16:58:58.380609 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.930961
I1122 16:58:58.380616 11005 solver.cpp:486] Iteration 7800, lr = 1e-05
I1122 16:59:12.598089 11005 solver.cpp:214] Iteration 7820, loss = 0.223464
I1122 16:59:12.598188 11005 solver.cpp:229]     Train net output #0: accuracy = 0.926464
I1122 16:59:12.598201 11005 solver.cpp:229]     Train net output #1: loss = 0.223463 (* 1 = 0.223463 loss)
I1122 16:59:12.598209 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.933716
I1122 16:59:12.598215 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.923948
I1122 16:59:12.598223 11005 solver.cpp:486] Iteration 7820, lr = 1e-05
I1122 16:59:26.823912 11005 solver.cpp:214] Iteration 7840, loss = 1.14542
I1122 16:59:26.823957 11005 solver.cpp:229]     Train net output #0: accuracy = 0.567581
I1122 16:59:26.823969 11005 solver.cpp:229]     Train net output #1: loss = 1.14542 (* 1 = 1.14542 loss)
I1122 16:59:26.823976 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.593675
I1122 16:59:26.823983 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.567204
I1122 16:59:26.823989 11005 solver.cpp:486] Iteration 7840, lr = 1e-05
I1122 16:59:41.032393 11005 solver.cpp:214] Iteration 7860, loss = 0.580651
I1122 16:59:41.032438 11005 solver.cpp:229]     Train net output #0: accuracy = 0.740471
I1122 16:59:41.032449 11005 solver.cpp:229]     Train net output #1: loss = 0.580651 (* 1 = 0.580651 loss)
I1122 16:59:41.032456 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.65189
I1122 16:59:41.032464 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.830664
I1122 16:59:41.032470 11005 solver.cpp:486] Iteration 7860, lr = 1e-05
I1122 16:59:55.249881 11005 solver.cpp:214] Iteration 7880, loss = 0.35561
I1122 16:59:55.249994 11005 solver.cpp:229]     Train net output #0: accuracy = 0.845467
I1122 16:59:55.250008 11005 solver.cpp:229]     Train net output #1: loss = 0.355609 (* 1 = 0.355609 loss)
I1122 16:59:55.250015 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.888584
I1122 16:59:55.250021 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.821466
I1122 16:59:55.250028 11005 solver.cpp:486] Iteration 7880, lr = 1e-05
I1122 17:00:09.458940 11005 solver.cpp:214] Iteration 7900, loss = 1.14073
I1122 17:00:09.458987 11005 solver.cpp:229]     Train net output #0: accuracy = 0.494564
I1122 17:00:09.458998 11005 solver.cpp:229]     Train net output #1: loss = 1.14073 (* 1 = 1.14073 loss)
I1122 17:00:09.459005 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.456351
I1122 17:00:09.459012 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.989097
I1122 17:00:09.459018 11005 solver.cpp:486] Iteration 7900, lr = 1e-05
I1122 17:00:23.676647 11005 solver.cpp:214] Iteration 7920, loss = 1.22357
I1122 17:00:23.676695 11005 solver.cpp:229]     Train net output #0: accuracy = 0.572155
I1122 17:00:23.676707 11005 solver.cpp:229]     Train net output #1: loss = 1.22357 (* 1 = 1.22357 loss)
I1122 17:00:23.676714 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.540744
I1122 17:00:23.676720 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.574336
I1122 17:00:23.676728 11005 solver.cpp:486] Iteration 7920, lr = 1e-05
I1122 17:00:37.879601 11005 solver.cpp:214] Iteration 7940, loss = 0.508241
I1122 17:00:37.879796 11005 solver.cpp:229]     Train net output #0: accuracy = 0.789772
I1122 17:00:37.879827 11005 solver.cpp:229]     Train net output #1: loss = 0.50824 (* 1 = 0.50824 loss)
I1122 17:00:37.879834 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.815699
I1122 17:00:37.879840 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.76723
I1122 17:00:37.879854 11005 solver.cpp:486] Iteration 7940, lr = 1e-05
I1122 17:00:52.114168 11005 solver.cpp:214] Iteration 7960, loss = 0.455741
I1122 17:00:52.114214 11005 solver.cpp:229]     Train net output #0: accuracy = 0.85503
I1122 17:00:52.114226 11005 solver.cpp:229]     Train net output #1: loss = 0.45574 (* 1 = 0.45574 loss)
I1122 17:00:52.114233 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.900015
I1122 17:00:52.114239 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.843816
I1122 17:00:52.114248 11005 solver.cpp:486] Iteration 7960, lr = 1e-05
I1122 17:01:06.335783 11005 solver.cpp:214] Iteration 7980, loss = 1.29886
I1122 17:01:06.335831 11005 solver.cpp:229]     Train net output #0: accuracy = 0.559956
I1122 17:01:06.335842 11005 solver.cpp:229]     Train net output #1: loss = 1.29886 (* 1 = 1.29886 loss)
I1122 17:01:06.335850 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.545635
I1122 17:01:06.335856 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.560409
I1122 17:01:06.335863 11005 solver.cpp:486] Iteration 7980, lr = 1e-05
I1122 17:01:20.323251 11005 solver.cpp:361] Snapshotting to /home/gradescan/SegNet/Models/backup_basic/Training/segnet_basic_iter_8000.caffemodel
I1122 17:01:20.333931 11005 solver.cpp:369] Snapshotting solver state to /home/gradescan/SegNet/Models/backup_basic/Training/segnet_basic_iter_8000.solverstate
I1122 17:01:20.566328 11005 solver.cpp:214] Iteration 8000, loss = 0.379898
I1122 17:01:20.566370 11005 solver.cpp:229]     Train net output #0: accuracy = 0.799221
I1122 17:01:20.566381 11005 solver.cpp:229]     Train net output #1: loss = 0.379897 (* 1 = 0.379897 loss)
I1122 17:01:20.566387 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.730993
I1122 17:01:20.566395 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.996996
I1122 17:01:20.566403 11005 solver.cpp:486] Iteration 8000, lr = 1e-05
I1122 17:01:34.739826 11005 solver.cpp:214] Iteration 8020, loss = 1.12944
I1122 17:01:34.739874 11005 solver.cpp:229]     Train net output #0: accuracy = 0.486557
I1122 17:01:34.739886 11005 solver.cpp:229]     Train net output #1: loss = 1.12944 (* 1 = 1.12944 loss)
I1122 17:01:34.739893 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.448398
I1122 17:01:34.739899 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.829429
I1122 17:01:34.739909 11005 solver.cpp:486] Iteration 8020, lr = 1e-05
I1122 17:01:48.934366 11005 solver.cpp:214] Iteration 8040, loss = 0.404894
I1122 17:01:48.934415 11005 solver.cpp:229]     Train net output #0: accuracy = 0.839199
I1122 17:01:48.934427 11005 solver.cpp:229]     Train net output #1: loss = 0.404893 (* 1 = 0.404893 loss)
I1122 17:01:48.934434 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.833548
I1122 17:01:48.934442 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.840841
I1122 17:01:48.934448 11005 solver.cpp:486] Iteration 8040, lr = 1e-05
I1122 17:02:03.127913 11005 solver.cpp:214] Iteration 8060, loss = 1.03485
I1122 17:02:03.128010 11005 solver.cpp:229]     Train net output #0: accuracy = 0.601555
I1122 17:02:03.128022 11005 solver.cpp:229]     Train net output #1: loss = 1.03485 (* 1 = 1.03485 loss)
I1122 17:02:03.128029 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.505197
I1122 17:02:03.128036 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.614286
I1122 17:02:03.128042 11005 solver.cpp:486] Iteration 8060, lr = 1e-05
I1122 17:02:17.352125 11005 solver.cpp:214] Iteration 8080, loss = 0.341693
I1122 17:02:17.352175 11005 solver.cpp:229]     Train net output #0: accuracy = 0.860115
I1122 17:02:17.352186 11005 solver.cpp:229]     Train net output #1: loss = 0.341692 (* 1 = 0.341692 loss)
I1122 17:02:17.352195 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.724936
I1122 17:02:17.352200 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.940115
I1122 17:02:17.352207 11005 solver.cpp:486] Iteration 8080, lr = 1e-05
I1122 17:02:31.570864 11005 solver.cpp:214] Iteration 8100, loss = 0.470259
I1122 17:02:31.570912 11005 solver.cpp:229]     Train net output #0: accuracy = 0.833744
I1122 17:02:31.570924 11005 solver.cpp:229]     Train net output #1: loss = 0.470259 (* 1 = 0.470259 loss)
I1122 17:02:31.570930 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.76671
I1122 17:02:31.570937 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.856019
I1122 17:02:31.570943 11005 solver.cpp:486] Iteration 8100, lr = 1e-05
I1122 17:02:45.736706 11005 solver.cpp:214] Iteration 8120, loss = 0.222375
I1122 17:02:45.736773 11005 solver.cpp:229]     Train net output #0: accuracy = 0.908978
I1122 17:02:45.736784 11005 solver.cpp:229]     Train net output #1: loss = 0.222374 (* 1 = 0.222374 loss)
I1122 17:02:45.736794 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.832333
I1122 17:02:45.736800 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.985285
I1122 17:02:45.736809 11005 solver.cpp:486] Iteration 8120, lr = 1e-05
I1122 17:02:59.956542 11005 solver.cpp:214] Iteration 8140, loss = 0.409626
I1122 17:02:59.956590 11005 solver.cpp:229]     Train net output #0: accuracy = 0.834332
I1122 17:02:59.956603 11005 solver.cpp:229]     Train net output #1: loss = 0.409625 (* 1 = 0.409625 loss)
I1122 17:02:59.956609 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.922708
I1122 17:02:59.956615 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.806903
I1122 17:02:59.956622 11005 solver.cpp:486] Iteration 8140, lr = 1e-05
I1122 17:03:14.176926 11005 solver.cpp:214] Iteration 8160, loss = 0.222834
I1122 17:03:14.176976 11005 solver.cpp:229]     Train net output #0: accuracy = 0.933426
I1122 17:03:14.176987 11005 solver.cpp:229]     Train net output #1: loss = 0.222833 (* 1 = 0.222833 loss)
I1122 17:03:14.176995 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.932474
I1122 17:03:14.177000 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.933751
I1122 17:03:14.177006 11005 solver.cpp:486] Iteration 8160, lr = 1e-05
I1122 17:03:28.379331 11005 solver.cpp:214] Iteration 8180, loss = 0.280163
I1122 17:03:28.379447 11005 solver.cpp:229]     Train net output #0: accuracy = 0.879654
I1122 17:03:28.379462 11005 solver.cpp:229]     Train net output #1: loss = 0.280163 (* 1 = 0.280163 loss)
I1122 17:03:28.379468 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.819539
I1122 17:03:28.379474 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.94017
I1122 17:03:28.379484 11005 solver.cpp:486] Iteration 8180, lr = 1e-05
I1122 17:03:42.599052 11005 solver.cpp:214] Iteration 8200, loss = 0.217341
I1122 17:03:42.599097 11005 solver.cpp:229]     Train net output #0: accuracy = 0.929386
I1122 17:03:42.599108 11005 solver.cpp:229]     Train net output #1: loss = 0.21734 (* 1 = 0.21734 loss)
I1122 17:03:42.599115 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.937656
I1122 17:03:42.599123 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.926517
I1122 17:03:42.599129 11005 solver.cpp:486] Iteration 8200, lr = 1e-05
I1122 17:03:56.809463 11005 solver.cpp:214] Iteration 8220, loss = 1.1277
I1122 17:03:56.809510 11005 solver.cpp:229]     Train net output #0: accuracy = 0.569798
I1122 17:03:56.809521 11005 solver.cpp:229]     Train net output #1: loss = 1.1277 (* 1 = 1.1277 loss)
I1122 17:03:56.809530 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.593675
I1122 17:03:56.809535 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.569453
I1122 17:03:56.809542 11005 solver.cpp:486] Iteration 8220, lr = 1e-05
I1122 17:04:11.035573 11005 solver.cpp:214] Iteration 8240, loss = 0.563627
I1122 17:04:11.035667 11005 solver.cpp:229]     Train net output #0: accuracy = 0.752392
I1122 17:04:11.035679 11005 solver.cpp:229]     Train net output #1: loss = 0.563627 (* 1 = 0.563627 loss)
I1122 17:04:11.035687 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.657886
I1122 17:04:11.035693 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.848618
I1122 17:04:11.035701 11005 solver.cpp:486] Iteration 8240, lr = 1e-05
I1122 17:04:25.245116 11005 solver.cpp:214] Iteration 8260, loss = 0.348754
I1122 17:04:25.245165 11005 solver.cpp:229]     Train net output #0: accuracy = 0.849918
I1122 17:04:25.245177 11005 solver.cpp:229]     Train net output #1: loss = 0.348753 (* 1 = 0.348753 loss)
I1122 17:04:25.245183 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.894238
I1122 17:04:25.245190 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.825249
I1122 17:04:25.245196 11005 solver.cpp:486] Iteration 8260, lr = 1e-05
I1122 17:04:39.457633 11005 solver.cpp:214] Iteration 8280, loss = 1.14126
I1122 17:04:39.457679 11005 solver.cpp:229]     Train net output #0: accuracy = 0.493477
I1122 17:04:39.457691 11005 solver.cpp:229]     Train net output #1: loss = 1.14126 (* 1 = 1.14126 loss)
I1122 17:04:39.457698 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.455176
I1122 17:04:39.457705 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.989151
I1122 17:04:39.457712 11005 solver.cpp:486] Iteration 8280, lr = 1e-05
I1122 17:04:53.669970 11005 solver.cpp:214] Iteration 8300, loss = 1.20165
I1122 17:04:53.670086 11005 solver.cpp:229]     Train net output #0: accuracy = 0.574596
I1122 17:04:53.670099 11005 solver.cpp:229]     Train net output #1: loss = 1.20165 (* 1 = 1.20165 loss)
I1122 17:04:53.670114 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.559133
I1122 17:04:53.670120 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.57567
I1122 17:04:53.670127 11005 solver.cpp:486] Iteration 8300, lr = 1e-05
I1122 17:05:07.880409 11005 solver.cpp:214] Iteration 8320, loss = 0.498309
I1122 17:05:07.880458 11005 solver.cpp:229]     Train net output #0: accuracy = 0.790955
I1122 17:05:07.880470 11005 solver.cpp:229]     Train net output #1: loss = 0.498308 (* 1 = 0.498308 loss)
I1122 17:05:07.880477 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.818537
I1122 17:05:07.880483 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.766973
I1122 17:05:07.880492 11005 solver.cpp:486] Iteration 8320, lr = 1e-05
I1122 17:05:22.087038 11005 solver.cpp:214] Iteration 8340, loss = 0.447349
I1122 17:05:22.087083 11005 solver.cpp:229]     Train net output #0: accuracy = 0.858448
I1122 17:05:22.087095 11005 solver.cpp:229]     Train net output #1: loss = 0.447348 (* 1 = 0.447348 loss)
I1122 17:05:22.087101 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.901411
I1122 17:05:22.087107 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.847738
I1122 17:05:22.087115 11005 solver.cpp:486] Iteration 8340, lr = 1e-05
I1122 17:05:36.298104 11005 solver.cpp:214] Iteration 8360, loss = 1.28639
I1122 17:05:36.298180 11005 solver.cpp:229]     Train net output #0: accuracy = 0.561882
I1122 17:05:36.298192 11005 solver.cpp:229]     Train net output #1: loss = 1.28639 (* 1 = 1.28639 loss)
I1122 17:05:36.298199 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.535701
I1122 17:05:36.298207 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.562712
I1122 17:05:36.298213 11005 solver.cpp:486] Iteration 8360, lr = 1e-05
I1122 17:05:50.495506 11005 solver.cpp:214] Iteration 8380, loss = 0.370467
I1122 17:05:50.495551 11005 solver.cpp:229]     Train net output #0: accuracy = 0.804916
I1122 17:05:50.495563 11005 solver.cpp:229]     Train net output #1: loss = 0.370467 (* 1 = 0.370467 loss)
I1122 17:05:50.495569 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.738571
I1122 17:05:50.495576 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.997234
I1122 17:05:50.495584 11005 solver.cpp:486] Iteration 8380, lr = 1e-05
I1122 17:06:04.696652 11005 solver.cpp:214] Iteration 8400, loss = 1.13041
I1122 17:06:04.696701 11005 solver.cpp:229]     Train net output #0: accuracy = 0.482914
I1122 17:06:04.696712 11005 solver.cpp:229]     Train net output #1: loss = 1.13041 (* 1 = 1.13041 loss)
I1122 17:06:04.696719 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.443336
I1122 17:06:04.696733 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.838533
I1122 17:06:04.696740 11005 solver.cpp:486] Iteration 8400, lr = 1e-05
I1122 17:06:18.927485 11005 solver.cpp:214] Iteration 8420, loss = 0.399927
I1122 17:06:18.927556 11005 solver.cpp:229]     Train net output #0: accuracy = 0.84211
I1122 17:06:18.927568 11005 solver.cpp:229]     Train net output #1: loss = 0.399927 (* 1 = 0.399927 loss)
I1122 17:06:18.927575 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.837039
I1122 17:06:18.927582 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.843583
I1122 17:06:18.927588 11005 solver.cpp:486] Iteration 8420, lr = 1e-05
I1122 17:06:33.129920 11005 solver.cpp:214] Iteration 8440, loss = 1.03438
I1122 17:06:33.129968 11005 solver.cpp:229]     Train net output #0: accuracy = 0.60215
I1122 17:06:33.129979 11005 solver.cpp:229]     Train net output #1: loss = 1.03438 (* 1 = 1.03438 loss)
I1122 17:06:33.129987 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.504118
I1122 17:06:33.129993 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.615103
I1122 17:06:33.129999 11005 solver.cpp:486] Iteration 8440, lr = 1e-05
I1122 17:06:47.331549 11005 solver.cpp:214] Iteration 8460, loss = 0.330444
I1122 17:06:47.331596 11005 solver.cpp:229]     Train net output #0: accuracy = 0.866375
I1122 17:06:47.331609 11005 solver.cpp:229]     Train net output #1: loss = 0.330443 (* 1 = 0.330443 loss)
I1122 17:06:47.331615 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.730569
I1122 17:06:47.331621 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.946746
I1122 17:06:47.331629 11005 solver.cpp:486] Iteration 8460, lr = 1e-05
I1122 17:07:01.502722 11005 solver.cpp:214] Iteration 8480, loss = 0.469461
I1122 17:07:01.502841 11005 solver.cpp:229]     Train net output #0: accuracy = 0.835785
I1122 17:07:01.502854 11005 solver.cpp:229]     Train net output #1: loss = 0.46946 (* 1 = 0.46946 loss)
I1122 17:07:01.502861 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.76726
I1122 17:07:01.502867 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.858555
I1122 17:07:01.502876 11005 solver.cpp:486] Iteration 8480, lr = 1e-05
I1122 17:07:15.725569 11005 solver.cpp:214] Iteration 8500, loss = 0.219333
I1122 17:07:15.725617 11005 solver.cpp:229]     Train net output #0: accuracy = 0.910328
I1122 17:07:15.725628 11005 solver.cpp:229]     Train net output #1: loss = 0.219332 (* 1 = 0.219332 loss)
I1122 17:07:15.725636 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.834573
I1122 17:07:15.725642 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.985749
I1122 17:07:15.725648 11005 solver.cpp:486] Iteration 8500, lr = 1e-05
I1122 17:07:29.943323 11005 solver.cpp:214] Iteration 8520, loss = 0.394497
I1122 17:07:29.943372 11005 solver.cpp:229]     Train net output #0: accuracy = 0.841579
I1122 17:07:29.943384 11005 solver.cpp:229]     Train net output #1: loss = 0.394496 (* 1 = 0.394496 loss)
I1122 17:07:29.943392 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.927701
I1122 17:07:29.943398 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.814851
I1122 17:07:29.943404 11005 solver.cpp:486] Iteration 8520, lr = 1e-05
I1122 17:07:44.123486 11005 solver.cpp:214] Iteration 8540, loss = 0.218011
I1122 17:07:44.123584 11005 solver.cpp:229]     Train net output #0: accuracy = 0.936333
I1122 17:07:44.123597 11005 solver.cpp:229]     Train net output #1: loss = 0.218011 (* 1 = 0.218011 loss)
I1122 17:07:44.123605 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.932429
I1122 17:07:44.123611 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.937667
I1122 17:07:44.123617 11005 solver.cpp:486] Iteration 8540, lr = 1e-05
I1122 17:07:58.334528 11005 solver.cpp:214] Iteration 8560, loss = 0.268816
I1122 17:07:58.334574 11005 solver.cpp:229]     Train net output #0: accuracy = 0.886868
I1122 17:07:58.334585 11005 solver.cpp:229]     Train net output #1: loss = 0.268816 (* 1 = 0.268816 loss)
I1122 17:07:58.334595 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.825653
I1122 17:07:58.334602 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.94849
I1122 17:07:58.334614 11005 solver.cpp:486] Iteration 8560, lr = 1e-05
I1122 17:08:12.550284 11005 solver.cpp:214] Iteration 8580, loss = 0.211209
I1122 17:08:12.550331 11005 solver.cpp:229]     Train net output #0: accuracy = 0.932236
I1122 17:08:12.550343 11005 solver.cpp:229]     Train net output #1: loss = 0.211208 (* 1 = 0.211208 loss)
I1122 17:08:12.550349 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.941596
I1122 17:08:12.550355 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.928989
I1122 17:08:12.550362 11005 solver.cpp:486] Iteration 8580, lr = 1e-05
I1122 17:08:26.748921 11005 solver.cpp:214] Iteration 8600, loss = 1.11039
I1122 17:08:26.749043 11005 solver.cpp:229]     Train net output #0: accuracy = 0.572086
I1122 17:08:26.749056 11005 solver.cpp:229]     Train net output #1: loss = 1.11039 (* 1 = 1.11039 loss)
I1122 17:08:26.749063 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.594211
I1122 17:08:26.749070 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.571767
I1122 17:08:26.749076 11005 solver.cpp:486] Iteration 8600, lr = 1e-05
I1122 17:08:40.974551 11005 solver.cpp:214] Iteration 8620, loss = 0.546025
I1122 17:08:40.974597 11005 solver.cpp:229]     Train net output #0: accuracy = 0.763512
I1122 17:08:40.974608 11005 solver.cpp:229]     Train net output #1: loss = 0.546024 (* 1 = 0.546024 loss)
I1122 17:08:40.974616 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.663088
I1122 17:08:40.974622 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.865763
I1122 17:08:40.974632 11005 solver.cpp:486] Iteration 8620, lr = 1e-05
I1122 17:08:55.174258 11005 solver.cpp:214] Iteration 8640, loss = 0.342195
I1122 17:08:55.174305 11005 solver.cpp:229]     Train net output #0: accuracy = 0.853439
I1122 17:08:55.174316 11005 solver.cpp:229]     Train net output #1: loss = 0.342194 (* 1 = 0.342194 loss)
I1122 17:08:55.174324 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.896628
I1122 17:08:55.174329 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.829399
I1122 17:08:55.174336 11005 solver.cpp:486] Iteration 8640, lr = 1e-05
I1122 17:09:09.395292 11005 solver.cpp:214] Iteration 8660, loss = 1.14305
I1122 17:09:09.395486 11005 solver.cpp:229]     Train net output #0: accuracy = 0.492012
I1122 17:09:09.395546 11005 solver.cpp:229]     Train net output #1: loss = 1.14305 (* 1 = 1.14305 loss)
I1122 17:09:09.395555 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.453586
I1122 17:09:09.395566 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.98931
I1122 17:09:09.395577 11005 solver.cpp:486] Iteration 8660, lr = 1e-05
I1122 17:09:23.626529 11005 solver.cpp:214] Iteration 8680, loss = 1.17868
I1122 17:09:23.626579 11005 solver.cpp:229]     Train net output #0: accuracy = 0.577419
I1122 17:09:23.626590 11005 solver.cpp:229]     Train net output #1: loss = 1.17868 (* 1 = 1.17868 loss)
I1122 17:09:23.626597 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.574819
I1122 17:09:23.626605 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.5776
I1122 17:09:23.626612 11005 solver.cpp:486] Iteration 8680, lr = 1e-05
I1122 17:09:37.837090 11005 solver.cpp:214] Iteration 8700, loss = 0.486631
I1122 17:09:37.837137 11005 solver.cpp:229]     Train net output #0: accuracy = 0.792099
I1122 17:09:37.837148 11005 solver.cpp:229]     Train net output #1: loss = 0.48663 (* 1 = 0.48663 loss)
I1122 17:09:37.837155 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.820472
I1122 17:09:37.837162 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.767429
I1122 17:09:37.837169 11005 solver.cpp:486] Iteration 8700, lr = 1e-05
I1122 17:09:52.050995 11005 solver.cpp:214] Iteration 8720, loss = 0.439754
I1122 17:09:52.051087 11005 solver.cpp:229]     Train net output #0: accuracy = 0.861782
I1122 17:09:52.051100 11005 solver.cpp:229]     Train net output #1: loss = 0.439754 (* 1 = 0.439754 loss)
I1122 17:09:52.051107 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.90252
I1122 17:09:52.051113 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.851627
I1122 17:09:52.051120 11005 solver.cpp:486] Iteration 8720, lr = 1e-05
I1122 17:10:06.261065 11005 solver.cpp:214] Iteration 8740, loss = 1.2715
I1122 17:10:06.261112 11005 solver.cpp:229]     Train net output #0: accuracy = 0.563862
I1122 17:10:06.261124 11005 solver.cpp:229]     Train net output #1: loss = 1.2715 (* 1 = 1.2715 loss)
I1122 17:10:06.261132 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.529244
I1122 17:10:06.261138 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.564959
I1122 17:10:06.261147 11005 solver.cpp:486] Iteration 8740, lr = 1e-05
I1122 17:10:20.472054 11005 solver.cpp:214] Iteration 8760, loss = 0.360932
I1122 17:10:20.472100 11005 solver.cpp:229]     Train net output #0: accuracy = 0.809605
I1122 17:10:20.472110 11005 solver.cpp:229]     Train net output #1: loss = 0.360931 (* 1 = 0.360931 loss)
I1122 17:10:20.472117 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.744882
I1122 17:10:20.472123 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.997219
I1122 17:10:20.472131 11005 solver.cpp:486] Iteration 8760, lr = 1e-05
I1122 17:10:34.702105 11005 solver.cpp:214] Iteration 8780, loss = 1.13237
I1122 17:10:34.702219 11005 solver.cpp:229]     Train net output #0: accuracy = 0.478687
I1122 17:10:34.702234 11005 solver.cpp:229]     Train net output #1: loss = 1.13237 (* 1 = 1.13237 loss)
I1122 17:10:34.702240 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.437842
I1122 17:10:34.702245 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.845694
I1122 17:10:34.702256 11005 solver.cpp:486] Iteration 8780, lr = 1e-05
I1122 17:10:48.927700 11005 solver.cpp:214] Iteration 8800, loss = 0.395414
I1122 17:10:48.927747 11005 solver.cpp:229]     Train net output #0: accuracy = 0.84483
I1122 17:10:48.927758 11005 solver.cpp:229]     Train net output #1: loss = 0.395413 (* 1 = 0.395413 loss)
I1122 17:10:48.927765 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.8407
I1122 17:10:48.927772 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.846029
I1122 17:10:48.927778 11005 solver.cpp:486] Iteration 8800, lr = 1e-05
I1122 17:11:03.127799 11005 solver.cpp:214] Iteration 8820, loss = 1.03471
I1122 17:11:03.127846 11005 solver.cpp:229]     Train net output #0: accuracy = 0.602375
I1122 17:11:03.127858 11005 solver.cpp:229]     Train net output #1: loss = 1.03471 (* 1 = 1.03471 loss)
I1122 17:11:03.127866 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.506537
I1122 17:11:03.127871 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.615038
I1122 17:11:03.127878 11005 solver.cpp:486] Iteration 8820, lr = 1e-05
I1122 17:11:17.339781 11005 solver.cpp:214] Iteration 8840, loss = 0.320228
I1122 17:11:17.339879 11005 solver.cpp:229]     Train net output #0: accuracy = 0.871555
I1122 17:11:17.339891 11005 solver.cpp:229]     Train net output #1: loss = 0.320227 (* 1 = 0.320227 loss)
I1122 17:11:17.339898 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.735392
I1122 17:11:17.339905 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.952138
I1122 17:11:17.339911 11005 solver.cpp:486] Iteration 8840, lr = 1e-05
I1122 17:11:31.588598 11005 solver.cpp:214] Iteration 8860, loss = 0.468502
I1122 17:11:31.588644 11005 solver.cpp:229]     Train net output #0: accuracy = 0.836777
I1122 17:11:31.588655 11005 solver.cpp:229]     Train net output #1: loss = 0.468501 (* 1 = 0.468501 loss)
I1122 17:11:31.588662 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.767765
I1122 17:11:31.588670 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.859709
I1122 17:11:31.588678 11005 solver.cpp:486] Iteration 8860, lr = 1e-05
I1122 17:11:45.800207 11005 solver.cpp:214] Iteration 8880, loss = 0.215817
I1122 17:11:45.800253 11005 solver.cpp:229]     Train net output #0: accuracy = 0.911121
I1122 17:11:45.800264 11005 solver.cpp:229]     Train net output #1: loss = 0.215816 (* 1 = 0.215816 loss)
I1122 17:11:45.800271 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.836202
I1122 17:11:45.800278 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.985711
I1122 17:11:45.800287 11005 solver.cpp:486] Iteration 8880, lr = 1e-05
I1122 17:12:00.007678 11005 solver.cpp:214] Iteration 8900, loss = 0.379197
I1122 17:12:00.007810 11005 solver.cpp:229]     Train net output #0: accuracy = 0.848011
I1122 17:12:00.007824 11005 solver.cpp:229]     Train net output #1: loss = 0.379196 (* 1 = 0.379196 loss)
I1122 17:12:00.007833 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.931936
I1122 17:12:00.007838 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.821964
I1122 17:12:00.007849 11005 solver.cpp:486] Iteration 8900, lr = 1e-05
I1122 17:12:14.202632 11005 solver.cpp:214] Iteration 8920, loss = 0.21472
I1122 17:12:14.202680 11005 solver.cpp:229]     Train net output #0: accuracy = 0.938484
I1122 17:12:14.202692 11005 solver.cpp:229]     Train net output #1: loss = 0.214719 (* 1 = 0.214719 loss)
I1122 17:12:14.202698 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.931125
I1122 17:12:14.202705 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.940999
I1122 17:12:14.202713 11005 solver.cpp:486] Iteration 8920, lr = 1e-05
I1122 17:12:30.329077 11005 solver.cpp:214] Iteration 8940, loss = 0.259434
I1122 17:12:30.329213 11005 solver.cpp:229]     Train net output #0: accuracy = 0.892845
I1122 17:12:30.329241 11005 solver.cpp:229]     Train net output #1: loss = 0.259433 (* 1 = 0.259433 loss)
I1122 17:12:30.329248 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.83147
I1122 17:12:30.329254 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.95463
I1122 17:12:30.329263 11005 solver.cpp:486] Iteration 8940, lr = 1e-05
I1122 17:12:44.526335 11005 solver.cpp:214] Iteration 8960, loss = 0.205476
I1122 17:12:44.526384 11005 solver.cpp:229]     Train net output #0: accuracy = 0.935234
I1122 17:12:44.526396 11005 solver.cpp:229]     Train net output #1: loss = 0.205476 (* 1 = 0.205476 loss)
I1122 17:12:44.526402 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.944899
I1122 17:12:44.526408 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.931881
I1122 17:12:44.526417 11005 solver.cpp:486] Iteration 8960, lr = 1e-05
I1122 17:12:58.737700 11005 solver.cpp:214] Iteration 8980, loss = 1.09481
I1122 17:12:58.737746 11005 solver.cpp:229]     Train net output #0: accuracy = 0.574844
I1122 17:12:58.737757 11005 solver.cpp:229]     Train net output #1: loss = 1.09481 (* 1 = 1.09481 loss)
I1122 17:12:58.737764 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.595819
I1122 17:12:58.737771 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.574542
I1122 17:12:58.737778 11005 solver.cpp:486] Iteration 8980, lr = 1e-05
I1122 17:13:12.716869 11005 solver.cpp:361] Snapshotting to /home/gradescan/SegNet/Models/backup_basic/Training/segnet_basic_iter_9000.caffemodel
I1122 17:13:12.729192 11005 solver.cpp:369] Snapshotting solver state to /home/gradescan/SegNet/Models/backup_basic/Training/segnet_basic_iter_9000.solverstate
I1122 17:13:12.961635 11005 solver.cpp:214] Iteration 9000, loss = 0.527354
I1122 17:13:12.961683 11005 solver.cpp:229]     Train net output #0: accuracy = 0.77544
I1122 17:13:12.961694 11005 solver.cpp:229]     Train net output #1: loss = 0.527353 (* 1 = 0.527353 loss)
I1122 17:13:12.961701 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.669356
I1122 17:13:12.961709 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.883455
I1122 17:13:12.961716 11005 solver.cpp:486] Iteration 9000, lr = 1e-05
I1122 17:13:27.160856 11005 solver.cpp:214] Iteration 9020, loss = 0.335413
I1122 17:13:27.160904 11005 solver.cpp:229]     Train net output #0: accuracy = 0.85754
I1122 17:13:27.160915 11005 solver.cpp:229]     Train net output #1: loss = 0.335412 (* 1 = 0.335412 loss)
I1122 17:13:27.160923 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.900191
I1122 17:13:27.160929 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.833799
I1122 17:13:27.160935 11005 solver.cpp:486] Iteration 9020, lr = 1e-05
I1122 17:13:41.384120 11005 solver.cpp:214] Iteration 9040, loss = 1.1458
I1122 17:13:41.384166 11005 solver.cpp:229]     Train net output #0: accuracy = 0.490704
I1122 17:13:41.384177 11005 solver.cpp:229]     Train net output #1: loss = 1.1458 (* 1 = 1.1458 loss)
I1122 17:13:41.384184 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.45218
I1122 17:13:41.384192 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.989257
I1122 17:13:41.384198 11005 solver.cpp:486] Iteration 9040, lr = 1e-05
I1122 17:13:55.587246 11005 solver.cpp:214] Iteration 9060, loss = 1.15535
I1122 17:13:55.587359 11005 solver.cpp:229]     Train net output #0: accuracy = 0.579624
I1122 17:13:55.587373 11005 solver.cpp:229]     Train net output #1: loss = 1.15535 (* 1 = 1.15535 loss)
I1122 17:13:55.587381 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.590447
I1122 17:13:55.587388 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.578873
I1122 17:13:55.587393 11005 solver.cpp:486] Iteration 9060, lr = 1e-05
I1122 17:14:09.836493 11005 solver.cpp:214] Iteration 9080, loss = 0.468409
I1122 17:14:09.836542 11005 solver.cpp:229]     Train net output #0: accuracy = 0.795994
I1122 17:14:09.836555 11005 solver.cpp:229]     Train net output #1: loss = 0.468408 (* 1 = 0.468408 loss)
I1122 17:14:09.836561 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.824943
I1122 17:14:09.836568 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.770824
I1122 17:14:09.836575 11005 solver.cpp:486] Iteration 9080, lr = 1e-05
I1122 17:14:24.015262 11005 solver.cpp:214] Iteration 9100, loss = 0.430802
I1122 17:14:24.015311 11005 solver.cpp:229]     Train net output #0: accuracy = 0.865887
I1122 17:14:24.015321 11005 solver.cpp:229]     Train net output #1: loss = 0.430801 (* 1 = 0.430801 loss)
I1122 17:14:24.015328 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.904317
I1122 17:14:24.015334 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.856307
I1122 17:14:24.015341 11005 solver.cpp:486] Iteration 9100, lr = 1e-05
I1122 17:14:38.215903 11005 solver.cpp:214] Iteration 9120, loss = 1.25896
I1122 17:14:38.216022 11005 solver.cpp:229]     Train net output #0: accuracy = 0.565617
I1122 17:14:38.216035 11005 solver.cpp:229]     Train net output #1: loss = 1.25896 (* 1 = 1.25896 loss)
I1122 17:14:38.216042 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.524773
I1122 17:14:38.216048 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.566911
I1122 17:14:38.216056 11005 solver.cpp:486] Iteration 9120, lr = 1e-05
I1122 17:14:52.789222 11005 solver.cpp:214] Iteration 9140, loss = 0.351544
I1122 17:14:52.789264 11005 solver.cpp:229]     Train net output #0: accuracy = 0.815525
I1122 17:14:52.789275 11005 solver.cpp:229]     Train net output #1: loss = 0.351544 (* 1 = 0.351544 loss)
I1122 17:14:52.789283 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.752804
I1122 17:14:52.789289 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.997338
I1122 17:14:52.789297 11005 solver.cpp:486] Iteration 9140, lr = 1e-05
I1122 17:15:07.938865 11005 solver.cpp:214] Iteration 9160, loss = 1.13616
I1122 17:15:07.938925 11005 solver.cpp:229]     Train net output #0: accuracy = 0.474533
I1122 17:15:07.938937 11005 solver.cpp:229]     Train net output #1: loss = 1.13616 (* 1 = 1.13616 loss)
I1122 17:15:07.938946 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.432475
I1122 17:15:07.938952 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.852436
I1122 17:15:07.938958 11005 solver.cpp:486] Iteration 9160, lr = 1e-05
I1122 17:15:22.190492 11005 solver.cpp:214] Iteration 9180, loss = 0.390344
I1122 17:15:22.190645 11005 solver.cpp:229]     Train net output #0: accuracy = 0.847946
I1122 17:15:22.190672 11005 solver.cpp:229]     Train net output #1: loss = 0.390344 (* 1 = 0.390344 loss)
I1122 17:15:22.190680 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.845106
I1122 17:15:22.190688 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.848771
I1122 17:15:22.190722 11005 solver.cpp:486] Iteration 9180, lr = 1e-05
I1122 17:15:36.421423 11005 solver.cpp:214] Iteration 9200, loss = 1.03493
I1122 17:15:36.421473 11005 solver.cpp:229]     Train net output #0: accuracy = 0.602947
I1122 17:15:36.421485 11005 solver.cpp:229]     Train net output #1: loss = 1.03493 (* 1 = 1.03493 loss)
I1122 17:15:36.421493 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.506145
I1122 17:15:36.421499 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.615737
I1122 17:15:36.421506 11005 solver.cpp:486] Iteration 9200, lr = 1e-05
I1122 17:15:50.632532 11005 solver.cpp:214] Iteration 9220, loss = 0.31049
I1122 17:15:50.632580 11005 solver.cpp:229]     Train net output #0: accuracy = 0.876637
I1122 17:15:50.632591 11005 solver.cpp:229]     Train net output #1: loss = 0.310489 (* 1 = 0.310489 loss)
I1122 17:15:50.632598 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.740881
I1122 17:15:50.632604 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.956978
I1122 17:15:50.632614 11005 solver.cpp:486] Iteration 9220, lr = 1e-05
I1122 17:16:04.853960 11005 solver.cpp:214] Iteration 9240, loss = 0.468334
I1122 17:16:04.854080 11005 solver.cpp:229]     Train net output #0: accuracy = 0.838089
I1122 17:16:04.854094 11005 solver.cpp:229]     Train net output #1: loss = 0.468333 (* 1 = 0.468333 loss)
I1122 17:16:04.854100 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.768468
I1122 17:16:04.854107 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.861223
I1122 17:16:04.854115 11005 solver.cpp:486] Iteration 9240, lr = 1e-05
I1122 17:16:19.061789 11005 solver.cpp:214] Iteration 9260, loss = 0.212745
I1122 17:16:19.061838 11005 solver.cpp:229]     Train net output #0: accuracy = 0.912258
I1122 17:16:19.061851 11005 solver.cpp:229]     Train net output #1: loss = 0.212745 (* 1 = 0.212745 loss)
I1122 17:16:19.061857 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.838083
I1122 17:16:19.061863 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.986107
I1122 17:16:19.061871 11005 solver.cpp:486] Iteration 9260, lr = 1e-05
I1122 17:16:33.283577 11005 solver.cpp:214] Iteration 9280, loss = 0.364006
I1122 17:16:33.283625 11005 solver.cpp:229]     Train net output #0: accuracy = 0.855003
I1122 17:16:33.283638 11005 solver.cpp:229]     Train net output #1: loss = 0.364005 (* 1 = 0.364005 loss)
I1122 17:16:33.283644 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.935802
I1122 17:16:33.283650 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.829927
I1122 17:16:33.283658 11005 solver.cpp:486] Iteration 9280, lr = 1e-05
I1122 17:16:47.500988 11005 solver.cpp:214] Iteration 9300, loss = 0.211256
I1122 17:16:47.501090 11005 solver.cpp:229]     Train net output #0: accuracy = 0.940243
I1122 17:16:47.501102 11005 solver.cpp:229]     Train net output #1: loss = 0.211256 (* 1 = 0.211256 loss)
I1122 17:16:47.501109 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.931185
I1122 17:16:47.501116 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.943338
I1122 17:16:47.501121 11005 solver.cpp:486] Iteration 9300, lr = 1e-05
I1122 17:17:01.719446 11005 solver.cpp:214] Iteration 9320, loss = 0.250669
I1122 17:17:01.719496 11005 solver.cpp:229]     Train net output #0: accuracy = 0.897892
I1122 17:17:01.719506 11005 solver.cpp:229]     Train net output #1: loss = 0.250668 (* 1 = 0.250668 loss)
I1122 17:17:01.719513 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.836002
I1122 17:17:01.719519 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.960195
I1122 17:17:01.719527 11005 solver.cpp:486] Iteration 9320, lr = 1e-05
I1122 17:17:15.934337 11005 solver.cpp:214] Iteration 9340, loss = 0.199813
I1122 17:17:15.934386 11005 solver.cpp:229]     Train net output #0: accuracy = 0.938068
I1122 17:17:15.934396 11005 solver.cpp:229]     Train net output #1: loss = 0.199812 (* 1 = 0.199812 loss)
I1122 17:17:15.934404 11005 solver.cpp:229]     Train net output #2: per_class_accuracy = 0.947743
I1122 17:17:15.934412 11005 solver.cpp:229]     Train net output #3: per_class_accuracy = 0.934712
I1122 17:17:15.934418 11005 solver.cpp:486] Iteration 9340, lr = 1e-05