checkpoints/shallow010_naslarge_2b8b_wSAM/train_log.txt

[1] loss: 3.794, train acc: 11.314 test acc: 19.980  28.649 s
[2] loss: 2.973, train acc: 25.248 test acc: 29.210  28.386 s
[3] loss: 2.512, train acc: 34.380 test acc: 37.360  28.325 s
[4] loss: 2.216, train acc: 40.850 test acc: 42.340  28.356 s
[5] loss: 2.014, train acc: 45.374 test acc: 45.960  28.438 s
[6] loss: 1.861, train acc: 49.168 test acc: 47.990  28.279 s
[7] loss: 1.739, train acc: 52.020 test acc: 50.610  28.384 s
[8] loss: 1.649, train acc: 54.222 test acc: 52.230  28.304 s
[9] loss: 1.562, train acc: 56.226 test acc: 53.370  28.357 s
[10] loss: 1.498, train acc: 58.022 test acc: 53.630  28.309 s
[11] loss: 1.444, train acc: 58.826 test acc: 55.240  29.686 s
[12] loss: 1.386, train acc: 60.702 test acc: 55.650  28.376 s
[13] loss: 1.343, train acc: 61.484 test acc: 56.510  28.304 s
[14] loss: 1.300, train acc: 62.678 test acc: 57.170  28.336 s
[15] loss: 1.260, train acc: 63.882 test acc: 57.760  28.336 s
[16] loss: 1.222, train acc: 64.736 test acc: 57.530  28.324 s
[17] loss: 1.187, train acc: 65.578 test acc: 57.460  28.353 s
[18] loss: 1.159, train acc: 66.304 test acc: 58.260  28.386 s
[19] loss: 1.134, train acc: 66.972 test acc: 59.360  28.291 s
[20] loss: 1.100, train acc: 67.888 test acc: 58.510  28.352 s
[21] loss: 1.080, train acc: 68.236 test acc: 59.560  28.336 s
[22] loss: 1.053, train acc: 68.996 test acc: 60.050  28.282 s
[23] loss: 1.024, train acc: 69.864 test acc: 59.990  28.355 s
[24] loss: 1.010, train acc: 70.018 test acc: 59.690  28.445 s
[25] loss: 0.990, train acc: 70.714 test acc: 61.140  28.411 s
[26] loss: 0.972, train acc: 71.212 test acc: 61.370  28.295 s
[27] loss: 0.948, train acc: 71.904 test acc: 60.460  28.393 s
[28] loss: 0.929, train acc: 72.322 test acc: 60.280  28.316 s
[29] loss: 0.910, train acc: 72.824 test acc: 60.750  28.273 s
[30] loss: 0.890, train acc: 73.380 test acc: 61.100  28.303 s
[31] loss: 0.878, train acc: 73.954 test acc: 60.840  28.488 s
[32] loss: 0.860, train acc: 74.326 test acc: 60.730  28.333 s
[33] loss: 0.847, train acc: 74.520 test acc: 62.230  28.375 s
[34] loss: 0.831, train acc: 74.982 test acc: 61.950  28.350 s
[35] loss: 0.821, train acc: 75.254 test acc: 61.950  28.284 s
[36] loss: 0.801, train acc: 75.954 test acc: 61.720  29.320 s
[37] loss: 0.792, train acc: 76.140 test acc: 61.990  28.403 s
[38] loss: 0.774, train acc: 76.660 test acc: 62.140  28.331 s
[39] loss: 0.764, train acc: 76.916 test acc: 61.880  28.367 s
[40] loss: 0.749, train acc: 77.284 test acc: 62.350  28.329 s
[41] loss: 0.740, train acc: 77.480 test acc: 62.680  28.290 s
[42] loss: 0.722, train acc: 78.000 test acc: 63.190  28.259 s
[43] loss: 0.716, train acc: 78.110 test acc: 63.250  28.446 s
[44] loss: 0.708, train acc: 78.464 test acc: 61.970  28.400 s
[45] loss: 0.692, train acc: 78.902 test acc: 62.720  28.325 s
[46] loss: 0.684, train acc: 79.270 test acc: 62.810  28.268 s
[47] loss: 0.675, train acc: 79.278 test acc: 62.310  29.411 s
[48] loss: 0.667, train acc: 79.626 test acc: 62.110  28.411 s
[49] loss: 0.647, train acc: 80.156 test acc: 62.520  28.336 s
[50] loss: 0.648, train acc: 79.978 test acc: 62.570  29.862 s
[51] loss: 0.631, train acc: 80.598 test acc: 62.670  28.317 s
[52] loss: 0.626, train acc: 80.732 test acc: 62.530  28.365 s
[53] loss: 0.619, train acc: 80.730 test acc: 63.110  28.362 s
[54] loss: 0.613, train acc: 81.146 test acc: 62.790  28.430 s
[55] loss: 0.605, train acc: 81.116 test acc: 63.030  28.423 s
[56] loss: 0.594, train acc: 81.756 test acc: 63.140  28.416 s
[57] loss: 0.584, train acc: 82.130 test acc: 62.950  28.383 s
[58] loss: 0.578, train acc: 82.098 test acc: 62.990  28.345 s
[59] loss: 0.574, train acc: 82.260 test acc: 62.930  28.286 s
[60] loss: 0.559, train acc: 82.526 test acc: 63.010  28.465 s
[61] loss: 0.549, train acc: 82.688 test acc: 62.730  28.250 s
[62] loss: 0.541, train acc: 83.058 test acc: 62.950  28.412 s
[63] loss: 0.538, train acc: 83.222 test acc: 63.390  28.378 s
[64] loss: 0.532, train acc: 83.452 test acc: 63.100  28.360 s
[65] loss: 0.529, train acc: 83.350 test acc: 62.970  28.408 s
[66] loss: 0.516, train acc: 83.780 test acc: 63.590  28.377 s
[67] loss: 0.510, train acc: 84.062 test acc: 63.590  28.390 s
[68] loss: 0.505, train acc: 84.186 test acc: 62.820  28.325 s
[69] loss: 0.503, train acc: 84.272 test acc: 63.420  28.433 s
[70] loss: 0.496, train acc: 84.420 test acc: 63.630  28.348 s
[71] loss: 0.490, train acc: 84.562 test acc: 63.520  29.374 s
[72] loss: 0.478, train acc: 84.770 test acc: 63.390  28.367 s
[73] loss: 0.474, train acc: 85.018 test acc: 63.410  28.372 s
[74] loss: 0.472, train acc: 85.144 test acc: 63.630  28.389 s
[75] loss: 0.468, train acc: 85.160 test acc: 63.540  28.444 s
[76] loss: 0.464, train acc: 85.244 test acc: 62.660  28.221 s
[77] loss: 0.457, train acc: 85.578 test acc: 63.180  28.342 s
[78] loss: 0.451, train acc: 85.774 test acc: 63.340  28.383 s
[79] loss: 0.449, train acc: 85.718 test acc: 63.030  28.377 s
[80] loss: 0.441, train acc: 86.142 test acc: 63.500  28.411 s
[81] loss: 0.438, train acc: 86.044 test acc: 63.730  28.400 s
[82] loss: 0.428, train acc: 86.570 test acc: 62.710  28.281 s
[83] loss: 0.430, train acc: 86.322 test acc: 63.390  28.343 s
[84] loss: 0.421, train acc: 86.714 test acc: 63.760  28.409 s
[85] loss: 0.415, train acc: 86.826 test acc: 62.490  29.667 s
[86] loss: 0.410, train acc: 87.032 test acc: 62.560  28.359 s
[87] loss: 0.407, train acc: 87.194 test acc: 63.650  28.362 s
[88] loss: 0.407, train acc: 87.148 test acc: 62.980  28.469 s
[89] loss: 0.397, train acc: 87.180 test acc: 63.800  28.305 s
[90] loss: 0.397, train acc: 87.224 test acc: 63.320  28.356 s
[91] loss: 0.381, train acc: 87.956 test acc: 63.420  28.373 s
[92] loss: 0.384, train acc: 87.844 test acc: 63.080  28.319 s
[93] loss: 0.385, train acc: 87.690 test acc: 62.900  28.307 s
[94] loss: 0.379, train acc: 87.868 test acc: 62.320  28.421 s
[95] loss: 0.374, train acc: 88.070 test acc: 63.080  28.350 s
[96] loss: 0.370, train acc: 88.200 test acc: 63.380  28.353 s
[97] loss: 0.361, train acc: 88.494 test acc: 63.780  28.391 s
[98] loss: 0.361, train acc: 88.472 test acc: 63.530  28.384 s
[99] loss: 0.355, train acc: 88.702 test acc: 62.670  28.379 s
[100] loss: 0.357, train acc: 88.516 test acc: 63.270  28.477 s
[1] loss: 0.271, train acc: 91.810 test acc: 65.350  28.323 s
[2] loss: 0.247, train acc: 92.710 test acc: 64.830  28.374 s
[3] loss: 0.241, train acc: 92.850 test acc: 65.590  28.389 s
[4] loss: 0.233, train acc: 93.202 test acc: 65.190  28.360 s
[5] loss: 0.226, train acc: 93.504 test acc: 65.020  28.330 s
[6] loss: 0.221, train acc: 93.620 test acc: 65.150  29.502 s
[7] loss: 0.222, train acc: 93.612 test acc: 64.840  28.418 s
[8] loss: 0.221, train acc: 93.684 test acc: 65.070  28.350 s
[9] loss: 0.216, train acc: 93.858 test acc: 64.960  28.376 s
[10] loss: 0.213, train acc: 93.892 test acc: 64.780  28.359 s
[1] loss: 0.196, train acc: 94.640 test acc: 65.170  28.452 s
[2] loss: 0.190, train acc: 94.756 test acc: 65.180  28.363 s
[3] loss: 0.188, train acc: 94.910 test acc: 65.030  28.463 s
[4] loss: 0.185, train acc: 95.040 test acc: 65.150  28.360 s
[5] loss: 0.185, train acc: 94.954 test acc: 65.210  28.269 s
[6] loss: 0.185, train acc: 95.046 test acc: 65.580  28.365 s
[7] loss: 0.183, train acc: 95.106 test acc: 65.330  28.365 s
[8] loss: 0.182, train acc: 95.074 test acc: 65.210  28.348 s
[9] loss: 0.180, train acc: 95.168 test acc: 65.340  28.460 s
[10] loss: 0.182, train acc: 94.968 test acc: 65.350  28.263 s
[11] loss: 0.179, train acc: 95.218 test acc: 65.320  28.304 s
[12] loss: 0.179, train acc: 95.206 test acc: 65.260  29.589 s
[13] loss: 0.176, train acc: 95.276 test acc: 65.360  29.665 s
[14] loss: 0.175, train acc: 95.280 test acc: 65.290  28.355 s
[15] loss: 0.173, train acc: 95.452 test acc: 65.450  28.395 s
[16] loss: 0.177, train acc: 95.278 test acc: 65.130  28.421 s
[17] loss: 0.175, train acc: 95.400 test acc: 65.310  28.305 s
[18] loss: 0.174, train acc: 95.342 test acc: 65.300  28.328 s
[19] loss: 0.170, train acc: 95.636 test acc: 65.380  28.278 s
[20] loss: 0.174, train acc: 95.384 test acc: 65.510  28.317 s
[21] loss: 0.169, train acc: 95.412 test acc: 65.530  28.333 s
[22] loss: 0.169, train acc: 95.556 test acc: 65.390  28.421 s
[23] loss: 0.173, train acc: 95.312 test acc: 65.310  28.247 s
[24] loss: 0.169, train acc: 95.442 test acc: 65.440  28.473 s
[25] loss: 0.168, train acc: 95.584 test acc: 65.450  28.369 s
[26] loss: 0.167, train acc: 95.576 test acc: 65.450  28.410 s
[27] loss: 0.167, train acc: 95.518 test acc: 65.430  28.366 s
[28] loss: 0.167, train acc: 95.566 test acc: 65.460  28.424 s
[29] loss: 0.167, train acc: 95.584 test acc: 65.300  28.306 s
[30] loss: 0.163, train acc: 95.732 test acc: 65.440  28.288 s
[1] loss: 0.163, train acc: 95.740 test acc: 65.450  28.295 s
[2] loss: 0.160, train acc: 95.808 test acc: 65.650  28.319 s
[3] loss: 0.161, train acc: 95.830 test acc: 65.410  28.342 s
[4] loss: 0.161, train acc: 95.866 test acc: 65.520  28.328 s
[5] loss: 0.160, train acc: 95.922 test acc: 65.410  28.373 s
[6] loss: 0.158, train acc: 96.008 test acc: 65.380  28.353 s
[7] loss: 0.162, train acc: 95.768 test acc: 65.490  28.400 s
[8] loss: 0.159, train acc: 95.872 test acc: 65.500  28.263 s
[9] loss: 0.160, train acc: 95.916 test acc: 65.440  28.378 s
[10] loss: 0.159, train acc: 95.866 test acc: 65.410  28.266 s
[11] loss: 0.161, train acc: 95.764 test acc: 65.400  28.357 s
[12] loss: 0.158, train acc: 95.990 test acc: 65.430  28.365 s
[13] loss: 0.160, train acc: 95.754 test acc: 65.470  28.313 s
[14] loss: 0.160, train acc: 95.864 test acc: 65.360  29.434 s
[15] loss: 0.159, train acc: 95.880 test acc: 65.300  28.343 s
[16] loss: 0.158, train acc: 95.984 test acc: 65.240  28.287 s
[17] loss: 0.159, train acc: 96.034 test acc: 65.290  28.472 s
[18] loss: 0.158, train acc: 95.980 test acc: 65.340  28.310 s
[19] loss: 0.158, train acc: 95.864 test acc: 65.440  28.351 s
[20] loss: 0.158, train acc: 95.916 test acc: 65.390  29.442 s
[21] loss: 0.158, train acc: 95.990 test acc: 65.430  28.330 s
[22] loss: 0.156, train acc: 95.900 test acc: 65.250  28.372 s
[23] loss: 0.156, train acc: 96.012 test acc: 65.450  28.473 s
[24] loss: 0.158, train acc: 95.932 test acc: 65.270  28.501 s
[25] loss: 0.154, train acc: 96.046 test acc: 65.250  28.394 s
[26] loss: 0.158, train acc: 95.918 test acc: 65.220  28.428 s
[27] loss: 0.155, train acc: 96.058 test acc: 65.360  28.399 s
[28] loss: 0.156, train acc: 96.000 test acc: 65.360  28.395 s
[29] loss: 0.155, train acc: 96.002 test acc: 65.240  28.345 s
[30] loss: 0.156, train acc: 95.904 test acc: 65.300  28.425 s
[31] loss: 0.155, train acc: 96.040 test acc: 65.220  28.307 s
[32] loss: 0.154, train acc: 96.118 test acc: 65.350  28.333 s
[33] loss: 0.155, train acc: 96.012 test acc: 65.360  28.264 s
[34] loss: 0.157, train acc: 95.864 test acc: 65.340  28.439 s
[35] loss: 0.158, train acc: 95.930 test acc: 65.290  28.423 s
[36] loss: 0.155, train acc: 95.888 test acc: 65.280  28.437 s
[37] loss: 0.154, train acc: 96.026 test acc: 65.320  28.356 s
[38] loss: 0.154, train acc: 96.058 test acc: 65.330  28.500 s
[39] loss: 0.154, train acc: 96.088 test acc: 65.340  28.476 s
[40] loss: 0.153, train acc: 96.124 test acc: 65.420  28.447 s
[41] loss: 0.154, train acc: 96.076 test acc: 65.340  28.407 s
[42] loss: 0.153, train acc: 96.078 test acc: 65.320  28.328 s
[43] loss: 0.153, train acc: 96.058 test acc: 65.250  28.435 s
[44] loss: 0.153, train acc: 96.060 test acc: 65.200  28.346 s
[45] loss: 0.153, train acc: 96.054 test acc: 65.440  28.386 s
[46] loss: 0.155, train acc: 95.884 test acc: 65.350  28.312 s
[47] loss: 0.153, train acc: 96.076 test acc: 65.290  28.294 s
[48] loss: 0.154, train acc: 96.074 test acc: 65.230  28.312 s
[49] loss: 0.155, train acc: 95.972 test acc: 65.350  28.437 s
[50] loss: 0.152, train acc: 96.080 test acc: 65.400  28.318 s
[1] loss: 2.156, train acc: 47.016 test acc: 55.160  48.729 s
[2] loss: 1.300, train acc: 64.938 test acc: 57.530  49.537 s
[3] loss: 1.169, train acc: 68.278 test acc: 59.230  49.484 s
[4] loss: 1.107, train acc: 69.872 test acc: 59.750  49.527 s
[5] loss: 1.058, train acc: 70.996 test acc: 60.210  49.858 s
[6] loss: 1.025, train acc: 71.816 test acc: 60.640  49.774 s
[7] loss: 0.996, train acc: 72.916 test acc: 61.780  49.952 s
[8] loss: 0.975, train acc: 73.504 test acc: 62.310  49.639 s
[9] loss: 0.970, train acc: 73.346 test acc: 62.330  50.242 s
[10] loss: 0.960, train acc: 73.750 test acc: 62.590  50.355 s
[11] loss: 0.945, train acc: 74.388 test acc: 61.530  50.090 s
[12] loss: 0.939, train acc: 74.392 test acc: 62.160  50.102 s
[13] loss: 0.934, train acc: 74.654 test acc: 62.800  50.238 s
[14] loss: 0.924, train acc: 74.784 test acc: 62.770  50.120 s
[15] loss: 0.922, train acc: 74.908 test acc: 63.100  50.312 s
[16] loss: 0.913, train acc: 75.212 test acc: 62.870  49.743 s
[17] loss: 0.905, train acc: 75.350 test acc: 63.620  49.298 s
[18] loss: 0.904, train acc: 75.548 test acc: 63.280  49.776 s
[19] loss: 0.897, train acc: 75.642 test acc: 63.320  49.591 s
[20] loss: 0.900, train acc: 75.544 test acc: 63.500  50.385 s
[21] loss: 0.901, train acc: 75.690 test acc: 63.090  50.958 s
[22] loss: 0.899, train acc: 75.482 test acc: 63.880  50.181 s
[23] loss: 0.892, train acc: 75.658 test acc: 62.680  50.265 s
[24] loss: 0.888, train acc: 75.846 test acc: 63.490  50.861 s
[25] loss: 0.888, train acc: 75.848 test acc: 63.960  50.855 s
[26] loss: 0.884, train acc: 75.822 test acc: 63.490  50.650 s
[27] loss: 0.884, train acc: 75.984 test acc: 63.470  50.638 s
[28] loss: 0.880, train acc: 76.274 test acc: 63.640  50.459 s
[29] loss: 0.883, train acc: 75.952 test acc: 63.880  49.827 s
[30] loss: 0.874, train acc: 76.454 test acc: 63.200  51.636 s
[31] loss: 0.883, train acc: 76.104 test acc: 64.050  51.383 s
[32] loss: 0.879, train acc: 76.174 test acc: 63.880  51.581 s
[33] loss: 0.877, train acc: 76.446 test acc: 63.700  51.827 s
[34] loss: 0.874, train acc: 76.284 test acc: 64.130  51.843 s
[35] loss: 0.877, train acc: 76.298 test acc: 64.050  51.626 s
[36] loss: 0.879, train acc: 76.172 test acc: 64.030  51.485 s
[37] loss: 0.872, train acc: 76.448 test acc: 64.290  51.860 s
[38] loss: 0.867, train acc: 76.456 test acc: 63.500  51.305 s
[39] loss: 0.868, train acc: 76.622 test acc: 64.320  51.597 s
[40] loss: 0.867, train acc: 76.594 test acc: 63.890  51.163 s
[41] loss: 0.876, train acc: 76.400 test acc: 63.850  50.455 s
[42] loss: 0.875, train acc: 76.442 test acc: 63.850  50.036 s
[43] loss: 0.868, train acc: 76.740 test acc: 63.940  49.634 s
[44] loss: 0.867, train acc: 76.520 test acc: 64.330  51.403 s
[45] loss: 0.869, train acc: 76.596 test acc: 64.180  49.436 s
[46] loss: 0.866, train acc: 76.844 test acc: 64.720  50.585 s
[47] loss: 0.869, train acc: 76.656 test acc: 64.690  50.614 s
[48] loss: 0.870, train acc: 76.794 test acc: 63.460  50.337 s
[49] loss: 0.868, train acc: 76.864 test acc: 63.990  50.397 s
[50] loss: 0.872, train acc: 76.566 test acc: 63.700  50.587 s
[51] loss: 0.872, train acc: 76.562 test acc: 64.590  50.219 s
[52] loss: 0.869, train acc: 76.716 test acc: 64.150  51.118 s
[53] loss: 0.860, train acc: 76.786 test acc: 63.600  50.558 s
[54] loss: 0.871, train acc: 76.464 test acc: 63.880  50.097 s
[55] loss: 0.864, train acc: 76.636 test acc: 64.100  50.269 s
[56] loss: 0.866, train acc: 76.654 test acc: 64.860  52.352 s
[57] loss: 0.858, train acc: 77.070 test acc: 63.770  51.600 s
[58] loss: 0.862, train acc: 76.780 test acc: 64.450  51.254 s
[59] loss: 0.856, train acc: 76.876 test acc: 64.060  51.579 s
[60] loss: 0.861, train acc: 76.832 test acc: 64.810  51.520 s
[61] loss: 0.865, train acc: 76.852 test acc: 64.690  51.499 s
[62] loss: 0.859, train acc: 76.778 test acc: 64.710  52.960 s
[63] loss: 0.860, train acc: 76.826 test acc: 64.970  50.988 s
[64] loss: 0.863, train acc: 76.876 test acc: 64.460  51.456 s
[65] loss: 0.860, train acc: 76.676 test acc: 63.830  51.835 s
[66] loss: 0.857, train acc: 76.870 test acc: 64.740  51.629 s
[67] loss: 0.856, train acc: 76.960 test acc: 65.230  50.696 s
[68] loss: 0.859, train acc: 77.172 test acc: 64.720  49.550 s
[69] loss: 0.858, train acc: 76.850 test acc: 64.520  49.624 s
[70] loss: 0.863, train acc: 76.914 test acc: 64.970  49.460 s
[71] loss: 0.856, train acc: 77.140 test acc: 64.480  49.526 s
[72] loss: 0.864, train acc: 76.972 test acc: 64.300  50.397 s
[73] loss: 0.861, train acc: 77.068 test acc: 64.780  50.665 s
[74] loss: 0.858, train acc: 77.118 test acc: 64.310  50.069 s
[75] loss: 0.860, train acc: 76.822 test acc: 64.080  51.695 s
[76] loss: 0.863, train acc: 77.034 test acc: 64.710  50.364 s
[77] loss: 0.862, train acc: 76.928 test acc: 64.310  50.655 s
[78] loss: 0.860, train acc: 77.036 test acc: 64.620  50.654 s
[79] loss: 0.856, train acc: 77.034 test acc: 64.290  50.319 s
[80] loss: 0.860, train acc: 77.014 test acc: 64.810  50.691 s
[81] loss: 0.856, train acc: 77.182 test acc: 64.100  49.703 s
[82] loss: 0.859, train acc: 76.906 test acc: 64.750  53.251 s
[83] loss: 0.858, train acc: 76.986 test acc: 64.600  52.225 s
[84] loss: 0.855, train acc: 77.028 test acc: 64.580  52.763 s
[85] loss: 0.866, train acc: 76.904 test acc: 64.590  52.442 s
[86] loss: 0.859, train acc: 76.996 test acc: 64.630  52.711 s
[87] loss: 0.858, train acc: 76.970 test acc: 64.620  52.703 s
[88] loss: 0.860, train acc: 77.006 test acc: 64.440  52.544 s
[89] loss: 0.856, train acc: 77.070 test acc: 65.500  51.146 s
[90] loss: 0.859, train acc: 77.254 test acc: 64.630  51.181 s
[91] loss: 0.853, train acc: 77.160 test acc: 64.240  51.536 s
[92] loss: 0.855, train acc: 76.922 test acc: 64.350  51.987 s
[93] loss: 0.852, train acc: 77.246 test acc: 64.660  51.638 s
[94] loss: 0.861, train acc: 77.010 test acc: 64.840  50.845 s
[95] loss: 0.852, train acc: 77.374 test acc: 64.500  49.821 s
[96] loss: 0.850, train acc: 77.336 test acc: 64.210  49.499 s
[97] loss: 0.850, train acc: 77.200 test acc: 64.620  51.200 s
[98] loss: 0.854, train acc: 77.306 test acc: 64.820  49.699 s
[99] loss: 0.850, train acc: 77.324 test acc: 64.760  49.574 s
[100] loss: 0.857, train acc: 77.052 test acc: 64.920  50.131 s
[1] loss: 0.839, train acc: 77.932 test acc: 65.290  50.584 s
[2] loss: 0.851, train acc: 78.018 test acc: 65.070  50.864 s
[3] loss: 0.862, train acc: 77.776 test acc: 65.420  50.869 s
[4] loss: 0.866, train acc: 77.566 test acc: 64.830  51.036 s
[5] loss: 0.867, train acc: 77.872 test acc: 64.980  50.247 s
[6] loss: 0.883, train acc: 77.240 test acc: 65.060  50.428 s
[7] loss: 0.873, train acc: 77.526 test acc: 64.950  50.568 s
[8] loss: 0.874, train acc: 77.516 test acc: 65.320  50.514 s
[9] loss: 0.876, train acc: 77.358 test acc: 65.150  51.861 s
[10] loss: 0.881, train acc: 77.296 test acc: 64.760  51.642 s
[11] loss: 0.877, train acc: 77.394 test acc: 64.790  51.802 s
[12] loss: 0.890, train acc: 76.968 test acc: 65.180  51.477 s
[13] loss: 0.884, train acc: 77.110 test acc: 65.130  51.371 s
[14] loss: 0.888, train acc: 77.212 test acc: 65.150  51.999 s
[15] loss: 0.888, train acc: 77.096 test acc: 65.330  51.968 s
[16] loss: 0.888, train acc: 77.312 test acc: 64.860  52.414 s
[17] loss: 0.893, train acc: 77.130 test acc: 65.140  51.963 s
[18] loss: 0.900, train acc: 76.872 test acc: 65.450  52.311 s
[19] loss: 0.890, train acc: 77.046 test acc: 64.920  51.923 s
[20] loss: 0.897, train acc: 76.914 test acc: 65.300  51.872 s
[21] loss: 0.897, train acc: 76.882 test acc: 64.880  52.140 s
[22] loss: 0.889, train acc: 77.264 test acc: 65.160  51.759 s
[23] loss: 0.900, train acc: 77.090 test acc: 64.820  50.343 s
[24] loss: 0.896, train acc: 76.922 test acc: 65.050  50.373 s
[25] loss: 0.894, train acc: 76.962 test acc: 64.830  52.496 s
[26] loss: 0.901, train acc: 76.952 test acc: 64.720  50.507 s
[27] loss: 0.902, train acc: 76.966 test acc: 64.750  50.839 s
[28] loss: 0.897, train acc: 76.898 test acc: 65.330  50.609 s
[29] loss: 0.903, train acc: 76.632 test acc: 64.770  50.680 s
[30] loss: 0.901, train acc: 76.844 test acc: 64.970  50.769 s
[1] loss: 0.792, train acc: 78.422 test acc: 64.860  40.148 s
[2] loss: 0.802, train acc: 78.424 test acc: 65.580  40.225 s
[3] loss: 0.805, train acc: 78.478 test acc: 65.900  40.109 s
[4] loss: 0.811, train acc: 78.326 test acc: 65.650  39.920 s
[5] loss: 0.816, train acc: 78.242 test acc: 65.100  40.261 s
[6] loss: 0.820, train acc: 78.284 test acc: 65.180  40.040 s
[7] loss: 0.823, train acc: 78.142 test acc: 65.090  39.313 s
[8] loss: 0.829, train acc: 77.914 test acc: 65.090  39.080 s
[9] loss: 0.828, train acc: 77.928 test acc: 65.120  39.428 s
[10] loss: 0.832, train acc: 77.956 test acc: 65.030  39.636 s
[11] loss: 0.835, train acc: 77.738 test acc: 65.300  39.659 s
[12] loss: 0.836, train acc: 77.796 test acc: 65.020  41.073 s
[13] loss: 0.838, train acc: 77.760 test acc: 65.620  39.657 s
[14] loss: 0.841, train acc: 77.830 test acc: 65.220  39.863 s
[15] loss: 0.836, train acc: 77.602 test acc: 65.520  39.850 s
[16] loss: 0.844, train acc: 77.400 test acc: 65.160  39.851 s
[17] loss: 0.841, train acc: 77.656 test acc: 65.000  39.888 s
[18] loss: 0.838, train acc: 77.684 test acc: 65.260  39.942 s
[19] loss: 0.845, train acc: 77.404 test acc: 64.720  40.208 s
[20] loss: 0.846, train acc: 77.436 test acc: 65.190  40.608 s
[21] loss: 0.844, train acc: 77.522 test acc: 65.090  40.029 s
[22] loss: 0.845, train acc: 77.556 test acc: 65.390  39.981 s
[23] loss: 0.845, train acc: 77.510 test acc: 65.440  39.757 s
[24] loss: 0.845, train acc: 77.536 test acc: 65.260  40.245 s
[25] loss: 0.842, train acc: 77.674 test acc: 65.340  39.845 s
[26] loss: 0.854, train acc: 77.462 test acc: 64.790  39.951 s
[27] loss: 0.848, train acc: 77.336 test acc: 65.170  40.246 s
[28] loss: 0.852, train acc: 77.262 test acc: 64.810  40.277 s
[29] loss: 0.850, train acc: 77.408 test acc: 64.830  41.406 s
[30] loss: 0.843, train acc: 77.674 test acc: 65.070  40.268 s
[1] loss: 0.780, train acc: 78.904 test acc: 66.020  35.165 s
[2] loss: 0.787, train acc: 78.836 test acc: 65.480  35.074 s
[3] loss: 0.789, train acc: 78.930 test acc: 65.540  35.085 s
[4] loss: 0.797, train acc: 78.764 test acc: 65.680  35.160 s
[5] loss: 0.799, train acc: 78.746 test acc: 65.350  35.257 s
[6] loss: 0.809, train acc: 78.514 test acc: 65.660  35.210 s
[7] loss: 0.804, train acc: 78.684 test acc: 65.920  35.490 s
[8] loss: 0.804, train acc: 78.722 test acc: 65.660  35.528 s
[9] loss: 0.813, train acc: 78.440 test acc: 65.690  35.510 s
[10] loss: 0.807, train acc: 78.652 test acc: 65.280  35.453 s
[11] loss: 0.818, train acc: 78.318 test acc: 65.470  35.469 s
[12] loss: 0.811, train acc: 78.364 test acc: 65.440  35.583 s
[13] loss: 0.815, train acc: 78.298 test acc: 65.620  35.300 s
[14] loss: 0.816, train acc: 78.394 test acc: 65.220  35.517 s
[15] loss: 0.816, train acc: 78.532 test acc: 65.300  35.576 s
[16] loss: 0.822, train acc: 78.208 test acc: 65.560  35.739 s
[17] loss: 0.818, train acc: 78.360 test acc: 65.540  35.588 s
[18] loss: 0.818, train acc: 78.272 test acc: 65.470  35.570 s
[19] loss: 0.825, train acc: 78.206 test acc: 65.430  35.458 s
[20] loss: 0.822, train acc: 78.314 test acc: 65.050  35.639 s
[21] loss: 0.825, train acc: 78.434 test acc: 65.370  35.697 s
[22] loss: 0.829, train acc: 77.912 test acc: 65.330  36.683 s
[23] loss: 0.828, train acc: 78.200 test acc: 65.440  35.418 s
[24] loss: 0.834, train acc: 77.840 test acc: 65.610  35.515 s
[25] loss: 0.834, train acc: 77.898 test acc: 65.520  35.677 s
[26] loss: 0.828, train acc: 78.186 test acc: 65.460  35.573 s
[27] loss: 0.828, train acc: 78.052 test acc: 65.420  35.679 s
[28] loss: 0.823, train acc: 78.238 test acc: 65.020  35.825 s
[29] loss: 0.831, train acc: 77.982 test acc: 65.420  35.424 s
[30] loss: 0.831, train acc: 78.170 test acc: 65.270  35.645 s
[1] loss: 0.754, train acc: 79.574 test acc: 66.150  32.233 s
[2] loss: 0.743, train acc: 79.626 test acc: 66.320  32.257 s
[3] loss: 0.744, train acc: 79.710 test acc: 66.220  32.426 s
[4] loss: 0.748, train acc: 79.678 test acc: 66.110  32.308 s
[5] loss: 0.750, train acc: 79.698 test acc: 66.090  32.095 s
[6] loss: 0.755, train acc: 79.440 test acc: 65.860  32.188 s
[7] loss: 0.760, train acc: 79.440 test acc: 66.310  32.288 s
[8] loss: 0.760, train acc: 79.436 test acc: 65.860  32.226 s
[9] loss: 0.753, train acc: 79.644 test acc: 65.950  32.925 s
[10] loss: 0.763, train acc: 79.408 test acc: 65.660  32.154 s
[11] loss: 0.759, train acc: 79.520 test acc: 65.700  32.713 s
[12] loss: 0.763, train acc: 79.490 test acc: 66.000  32.120 s
[13] loss: 0.766, train acc: 79.280 test acc: 66.070  32.380 s
[14] loss: 0.765, train acc: 79.332 test acc: 66.010  32.418 s
[15] loss: 0.761, train acc: 79.430 test acc: 65.770  32.347 s
[16] loss: 0.764, train acc: 79.474 test acc: 65.470  32.297 s
[17] loss: 0.772, train acc: 79.088 test acc: 65.680  32.264 s
[18] loss: 0.764, train acc: 79.534 test acc: 66.080  32.229 s
[19] loss: 0.767, train acc: 79.186 test acc: 66.200  32.276 s
[20] loss: 0.765, train acc: 79.376 test acc: 66.080  32.139 s
[21] loss: 0.768, train acc: 79.394 test acc: 66.110  32.186 s
[22] loss: 0.766, train acc: 79.308 test acc: 65.880  32.264 s
[23] loss: 0.767, train acc: 79.478 test acc: 65.840  32.358 s
[24] loss: 0.768, train acc: 79.316 test acc: 65.840  32.280 s
[25] loss: 0.773, train acc: 79.332 test acc: 65.940  32.300 s
[26] loss: 0.766, train acc: 79.310 test acc: 65.870  32.399 s
[27] loss: 0.771, train acc: 79.364 test acc: 65.940  32.203 s
[28] loss: 0.779, train acc: 79.210 test acc: 66.180  32.518 s
[29] loss: 0.778, train acc: 79.106 test acc: 65.540  32.162 s
[30] loss: 0.775, train acc: 79.248 test acc: 65.610  32.509 s
[1] loss: 0.733, train acc: 80.016 test acc: 66.100  30.917 s
[2] loss: 0.732, train acc: 79.684 test acc: 66.090  30.825 s
[3] loss: 0.728, train acc: 79.698 test acc: 65.990  30.875 s
[4] loss: 0.731, train acc: 79.672 test acc: 66.300  30.811 s
[5] loss: 0.729, train acc: 79.858 test acc: 66.060  30.889 s
[6] loss: 0.729, train acc: 79.806 test acc: 66.100  30.695 s
[7] loss: 0.734, train acc: 79.618 test acc: 66.130  30.950 s
[8] loss: 0.733, train acc: 79.768 test acc: 66.260  30.881 s
[9] loss: 0.732, train acc: 79.646 test acc: 65.980  30.803 s
[10] loss: 0.732, train acc: 79.718 test acc: 66.160  30.695 s
[11] loss: 0.733, train acc: 79.686 test acc: 66.060  31.036 s
[12] loss: 0.737, train acc: 79.540 test acc: 65.680  30.954 s
[13] loss: 0.736, train acc: 79.732 test acc: 66.130  30.736 s
[14] loss: 0.740, train acc: 79.572 test acc: 66.380  30.865 s
[15] loss: 0.736, train acc: 79.762 test acc: 66.140  30.722 s
[16] loss: 0.742, train acc: 79.564 test acc: 66.120  30.700 s
[17] loss: 0.740, train acc: 79.848 test acc: 65.840  30.613 s
[18] loss: 0.742, train acc: 79.628 test acc: 66.380  30.650 s
[19] loss: 0.738, train acc: 79.820 test acc: 66.070  30.728 s
[20] loss: 0.742, train acc: 79.568 test acc: 66.100  30.627 s
[21] loss: 0.737, train acc: 79.682 test acc: 66.240  30.681 s
[22] loss: 0.739, train acc: 79.440 test acc: 66.100  30.602 s
[23] loss: 0.742, train acc: 79.516 test acc: 65.980  30.535 s
[24] loss: 0.741, train acc: 79.810 test acc: 66.150  30.661 s
[25] loss: 0.743, train acc: 79.718 test acc: 65.960  30.513 s
[26] loss: 0.747, train acc: 79.568 test acc: 66.170  30.613 s
[27] loss: 0.744, train acc: 79.486 test acc: 66.180  31.912 s
[28] loss: 0.741, train acc: 79.536 test acc: 66.100  30.530 s
[29] loss: 0.747, train acc: 79.442 test acc: 66.090  30.597 s
[30] loss: 0.738, train acc: 79.504 test acc: 65.950  30.758 s
[1] loss: 0.734, train acc: 79.672 test acc: 66.050  30.350 s
[2] loss: 0.727, train acc: 79.668 test acc: 66.320  30.077 s
[3] loss: 0.727, train acc: 79.856 test acc: 66.120  29.967 s
[4] loss: 0.724, train acc: 79.666 test acc: 65.900  30.127 s
[5] loss: 0.724, train acc: 79.758 test acc: 66.020  30.119 s
[6] loss: 0.725, train acc: 79.922 test acc: 66.220  30.508 s
[7] loss: 0.730, train acc: 79.542 test acc: 65.890  30.209 s
[8] loss: 0.730, train acc: 79.550 test acc: 66.230  30.232 s
[9] loss: 0.729, train acc: 79.568 test acc: 66.280  30.226 s
[10] loss: 0.728, train acc: 79.712 test acc: 66.210  30.135 s
[11] loss: 0.728, train acc: 79.866 test acc: 66.080  30.183 s
[12] loss: 0.725, train acc: 79.820 test acc: 66.290  30.045 s
[13] loss: 0.731, train acc: 79.744 test acc: 65.780  30.215 s
[14] loss: 0.724, train acc: 79.716 test acc: 66.180  30.102 s
[15] loss: 0.728, train acc: 79.668 test acc: 65.990  30.320 s
[16] loss: 0.728, train acc: 79.740 test acc: 66.110  30.173 s
[17] loss: 0.727, train acc: 79.818 test acc: 66.100  30.143 s
[18] loss: 0.731, train acc: 79.634 test acc: 66.270  30.248 s
[19] loss: 0.733, train acc: 79.698 test acc: 66.230  30.293 s
[20] loss: 0.731, train acc: 79.614 test acc: 66.110  30.266 s
[21] loss: 0.728, train acc: 79.788 test acc: 66.240  30.304 s
[22] loss: 0.730, train acc: 79.790 test acc: 66.340  30.283 s
[23] loss: 0.732, train acc: 79.710 test acc: 66.220  30.085 s
[24] loss: 0.732, train acc: 79.702 test acc: 66.320  30.323 s
[25] loss: 0.725, train acc: 79.978 test acc: 66.370  30.132 s
[26] loss: 0.728, train acc: 79.832 test acc: 65.880  30.213 s
[27] loss: 0.737, train acc: 79.606 test acc: 65.910  30.210 s
[28] loss: 0.727, train acc: 79.790 test acc: 66.360  30.420 s
[29] loss: 0.733, train acc: 79.830 test acc: 66.100  30.249 s
[30] loss: 0.735, train acc: 79.526 test acc: 66.300  30.141 s