repvgg-v5s
v5lite-g-sppf-conv6-chunk 36.7map
v5lite-g-sppf-conv6 0.3645
v5-g2 0.3559
yolov5s-LC3-chunk2-rep-512 0.36192
LCB4 0.35085,
chunk2 0.35664
shuffle 0.35926
bs512
| map50-95 | v5s(0.357) | v5s-C3-shuffle | v5s-LC3chunk2 | v5s-LCB5( 0.3661) |
|---|---|---|---|---|
| epoch10 | 0.14827 | 0.14992 | 0.15495 | |
| epoch20 | 0.21336 | 0.21745 | 0.22601 | |
| epoch30 | 0.25346 | 0.25368 | 0.26217 | |
| epoch50 | 0.28785 | 0.288 | 0.29716 | |
| epoch80 | 0.30128 | ---- | 0.31201 | 0.3137 |
| epoch150 | 0.30411 | ---- | 0.31427 | 0.314 |
| map50-95 | v5s | v5s-LCBlock3 | v5s-LCBlock4 |
|---|---|---|---|
| epoch10 | 0.14827 | 0.15102 | 0.14323 |
| epoch20 | 0.21336 | 0.22605 | 0.21101 |
| epoch30 | 0.25346 | 0.26368 | 0.25108 |
| epoch50 | 0.28785 | 0.29997 | 0.28791 |
| epoch80 | 0.30128 | 0.31368 | 0.30026 |
| epoch150 | 0.30411 | 0.31626 | 0.29736 |
| epoch200 | xxxxxxxxxxx | 0.32948 | 0.31126 |
map50-95 | v5s | v5s-LCB3-chunk2|chunk2
-------- | ----- |
epoch10 | 0.14827 | 0.15495 |
epoch20 | 0.21336 | 0.22601
epoch30 | 0.25346 | 0.26217
epoch50 | 0.28785 | 0.29716
epoch80 | 0.30128 | 0.31201
epoch150 | 0.30411 | 0.31427 | 0.301
epoch200|xxxxxxxxxxx |
bs128
| map50-95 | v5s | v5s-g | v5s-lite-g |
|---|---|---|---|
| epoch10 | 0.20609 | 0.21338 | 0.21598 |
| epoch20 | 0.24671 | 0.25771 | 0.25994 |
| epoch30 | 0.2564 | 0.26827 | 0.27078 |
| epoch50 | 0.27553 | 0.28786 | 0.29147 |
| epoch80 | 0.30649 | 0.31702 | 0.32149 |
| epoch150 | 0.34457 0.34509(sppf) |
| map50-95 | v5s | v5s-g-sppf | v5s-repconv |
|---|---|---|---|
| epoch10 | 0.20609 | 0.21671 | 0.20472 |
| epoch20 | 0.24671 | 0.26189 | 0.24597 |
| epoch30 | 0.2564 | 0.27277 | 0.25588 |
| epoch50 | 0.27553 | 0.2919 | 0.27353 |
| epoch80 | 0.30649 | 0.31806 | 0.30048 |
| map50-95 | v5s | v5g-Focus->6 | v5s-lite-g |
|---|---|---|---|
| epoch10 | 0.20609 | 0.21466 | 0.17235 |
| epoch20 | 0.24671 | 0.25808 | 0.21379 |
| epoch30 | 0.2564 | 0.26928 | 0.21894 |
| epoch50 | 0.27553 | 0.28971 | 0.23417 |
| epoch80 | 0.30649 | 0.31824 | ---- |
| map50-95 | v5s | v5-LCBlock | v5s-LCBlock2 | v5s-LCB3-chunk2 |
|---|---|---|---|---|
| epoch10 | 0.20609 | 0.20474 | 0.20699 | 0.21211 |
| epoch20 | 0.24671 | 0.24611 | 0.24932 | 0.25482 |
| epoch30 | 0.2564 | 0.25629 | 0.2584 | 0.26505 |
| epoch50 | 0.27553 | 0.27647 | 0.27931 | 0.28598, |
| epoch80 | 0.30649 | 0.30654 | 0.30737 | |
| epoch150 | 0.33338 | 0.33523 | ---- |
v5s-lite-g
epoch, train/box_loss, train/obj_loss, train/cls_loss, metrics/precision, metrics/recall, metrics/mAP_0.5,metrics/mAP_0.5:0.95, val/box_loss, val/obj_loss, val/cls_loss, x/lr0, x/lr1, x/lr2
0, 0.091764, 0.083605, 0.083667, 0.016046, 0.056931, 0.0041556, 0.0014131, 0.07822, 0.0657, 0.0787, 0.0033297, 0.0033297, 0.070032
1, 0.073686, 0.083846, 0.075502, 0.32996, 0.057096, 0.020941, 0.0088895, 0.069293, 0.063517, 0.068443, 0.0066629, 0.0066629, 0.040032
2, 0.06799, 0.082278, 0.063985, 0.3211, 0.13373, 0.072504, 0.03121, 0.063444, 0.062623, 0.054571, 0.0099954, 0.0099954, 0.010031
3, 0.063771, 0.080539, 0.053533, 0.33994, 0.21106, 0.14836, 0.070977, 0.058648, 0.060657, 0.044437, 0.0099978, 0.0099978, 0.0099978
4, 0.061017, 0.077907, 0.048235, 0.34807, 0.244, 0.19708, 0.10435, 0.056184, 0.058426, 0.039842, 0.0099978, 0.0099978, 0.0099978
5, 0.059494, 0.077005, 0.044746, 0.39311, 0.27296, 0.24094, 0.13153, 0.054184, 0.05767, 0.036521, 0.0099961, 0.0099961, 0.0099961
6, 0.058339, 0.076628, 0.042245, 0.42702, 0.29912, 0.278, 0.15608, 0.052886, 0.056604, 0.034162, 0.0099938, 0.0099938, 0.0099938
7, 0.05765, 0.076134, 0.040563, 0.43897, 0.32715, 0.30912, 0.17754, 0.051631, 0.055954, 0.032001, 0.0099911, 0.0099911, 0.0099911
8, 0.056604, 0.07506, 0.039379, 0.46377, 0.34762, 0.33084, 0.19195, 0.050889, 0.055378, 0.030631, 0.0099879, 0.0099879, 0.0099879
9, 0.056356, 0.075845, 0.03818, 0.51648, 0.34182, 0.34928, 0.2052, 0.050079, 0.054832, 0.029381, 0.0099842, 0.0099842, 0.0099842
10, 0.055955, 0.075936, 0.037408, 0.49384, 0.37324, 0.36567, 0.21598, 0.049515, 0.054541, 0.028365, 0.00998, 0.00998, 0.00998
11, 0.055443, 0.07421, 0.036685, 0.50411, 0.38051, 0.37694, 0.22426, 0.049056, 0.054128, 0.027581, 0.0099753, 0.0099753, 0.0099753
12, 0.055402, 0.074553, 0.035941, 0.53431, 0.37807, 0.38842, 0.23277, 0.048627, 0.053846, 0.027027, 0.0099702, 0.0099702, 0.0099702
13, 0.054766, 0.074203, 0.035406, 0.52697, 0.39371, 0.39624, 0.2389, 0.048288, 0.05367, 0.026508, 0.0099645, 0.0099645, 0.0099645
14, 0.054625, 0.073913, 0.034909, 0.55272, 0.38366, 0.40125, 0.24303, 0.048002, 0.053492, 0.026123, 0.0099584, 0.0099584, 0.0099584
15, 0.054317, 0.073803, 0.034637, 0.54119, 0.39693, 0.40664, 0.24744, 0.047798, 0.053324, 0.025797, 0.0099517, 0.0099517, 0.0099517
16, 0.054204, 0.073825, 0.034358, 0.54239, 0.40287, 0.41212, 0.25056, 0.047653, 0.053182, 0.025574, 0.0099446, 0.0099446, 0.0099446
17, 0.053936, 0.072659, 0.03417, 0.53188, 0.41191, 0.416, 0.25368, 0.047536, 0.053067, 0.025362, 0.009937, 0.009937, 0.009937
18, 0.0539, 0.073468, 0.033832, 0.57219, 0.3927, 0.41887, 0.25594, 0.047424, 0.052961, 0.025217, 0.0099289, 0.0099289, 0.0099289
19, 0.053708, 0.072787, 0.033427, 0.57544, 0.39629, 0.42192, 0.25798, 0.04732, 0.052878, 0.025078, 0.0099203, 0.0099203, 0.0099203
20, 0.05354, 0.072535, 0.033043, 0.57687, 0.39726, 0.42415, 0.25994, 0.047237, 0.052805, 0.024953, 0.0099112, 0.0099112, 0.0099112
21, 0.053139, 0.072018, 0.032922, 0.56312, 0.40651, 0.42616, 0.26174, 0.047163, 0.052731, 0.024833, 0.0099017, 0.0099017, 0.0099017
22, 0.05324, 0.072314, 0.032852, 0.56698, 0.40809, 0.42802, 0.26296, 0.047094, 0.052672, 0.024743, 0.0098916, 0.0098916, 0.0098916
23, 0.052984, 0.071691, 0.032502, 0.5688, 0.40835, 0.42939, 0.26433, 0.047033, 0.052621, 0.024654, 0.0098811, 0.0098811, 0.0098811
24, 0.053075, 0.072245, 0.032542, 0.55917, 0.41549, 0.43074, 0.26544, 0.046973, 0.052579, 0.024576, 0.0098701, 0.0098701, 0.0098701
25, 0.052899, 0.073219, 0.032326, 0.57717, 0.40883, 0.43238, 0.26645, 0.046922, 0.052545, 0.024498, 0.0098586, 0.0098586, 0.0098586
26, 0.052927, 0.072695, 0.032286, 0.56001, 0.41737, 0.43341, 0.26761, 0.046877, 0.052522, 0.024433, 0.0098467, 0.0098467, 0.0098467
27, 0.052769, 0.072309, 0.032052, 0.55659, 0.42052, 0.43427, 0.2685, 0.046839, 0.052514, 0.024381, 0.0098342, 0.0098342, 0.0098342
28, 0.052567, 0.071681, 0.031901, 0.58324, 0.40771, 0.43537, 0.26933, 0.046809, 0.052515, 0.024332, 0.0098213, 0.0098213, 0.0098213
29, 0.052512, 0.071409, 0.031706, 0.55935, 0.4201, 0.4366, 0.27006, 0.04678, 0.052536, 0.024281, 0.0098079, 0.0098079, 0.0098079
30, 0.05257, 0.072451, 0.031773, 0.60038, 0.40082, 0.4375, 0.27078, 0.046756, 0.052566, 0.024233, 0.0097941, 0.0097941, 0.0097941
31, 0.052397, 0.072427, 0.031523, 0.59812, 0.40353, 0.43831, 0.27164, 0.046734, 0.052607, 0.024196, 0.0097798, 0.0097798, 0.0097798
32, 0.052493, 0.072668, 0.03139, 0.5401, 0.43229, 0.43888, 0.27255, 0.046715, 0.052658, 0.024157, 0.009765, 0.009765, 0.009765
33, 0.05221, 0.071673, 0.031336, 0.60056, 0.40494, 0.44013, 0.27315, 0.046691, 0.052701, 0.024123, 0.0097497, 0.0097497, 0.0097497
34, 0.052178, 0.071612, 0.031145, 0.56314, 0.42246, 0.44092, 0.27383, 0.046666, 0.052762, 0.024081, 0.009734, 0.009734, 0.009734
v5s-g
epoch, train/box_loss, train/obj_loss, train/cls_loss, metrics/precision, metrics/recall, metrics/mAP_0.5,metrics/mAP_0.5:0.95, val/box_loss, val/obj_loss, val/cls_loss, x/lr0, x/lr1, x/lr2
0, 0.090561, 0.083676, 0.083902, 0.0058319, 0.047981, 0.0049234, 0.0016317, 0.07946, 0.065755, 0.078896, 0.0033297, 0.0033297, 0.070032
1, 0.07336, 0.08333, 0.075127, 0.39751, 0.056787, 0.020281, 0.0077558, 0.070273, 0.063855, 0.067316, 0.0066629, 0.0066629, 0.040032
2, 0.06794, 0.082012, 0.063558, 0.33755, 0.13032, 0.071825, 0.031181, 0.063522, 0.062164, 0.053643, 0.0099954, 0.0099954, 0.010031
3, 0.063912, 0.079575, 0.0537, 0.35576, 0.18729, 0.14409, 0.070619, 0.059045, 0.059896, 0.044766, 0.0099978, 0.0099978, 0.0099978
4, 0.061241, 0.077947, 0.048031, 0.39008, 0.22238, 0.19577, 0.099902, 0.056331, 0.058283, 0.039873, 0.0099978, 0.0099978, 0.0099978
5, 0.059561, 0.07713, 0.044272, 0.39467, 0.27314, 0.24088, 0.13006, 0.054377, 0.057369, 0.036662, 0.0099961, 0.0099961, 0.0099961
6, 0.058417, 0.076726, 0.042186, 0.44821, 0.29642, 0.28312, 0.15765, 0.052918, 0.056584, 0.03407, 0.0099938, 0.0099938, 0.0099938
7, 0.057537, 0.075668, 0.04065, 0.44709, 0.32061, 0.3053, 0.17384, 0.051819, 0.05594, 0.032175, 0.0099911, 0.0099911, 0.0099911
8, 0.05707, 0.075075, 0.039091, 0.48689, 0.32388, 0.32769, 0.18949, 0.0509, 0.05535, 0.030464, 0.0099879, 0.0099879, 0.0099879
9, 0.056446, 0.075107, 0.038153, 0.46891, 0.3638, 0.3453, 0.20173, 0.050224, 0.054859, 0.029229, 0.0099842, 0.0099842, 0.0099842
10, 0.05599, 0.074498, 0.037155, 0.52217, 0.35487, 0.36348, 0.21338, 0.049652, 0.054643, 0.0284, 0.00998, 0.00998, 0.00998
11, 0.055954, 0.074994, 0.036722, 0.49573, 0.37614, 0.3719, 0.22167, 0.049202, 0.054211, 0.027769, 0.0099753, 0.0099753, 0.0099753
12, 0.055413, 0.074568, 0.035754, 0.49957, 0.38414, 0.38323, 0.22898, 0.048783, 0.053969, 0.027122, 0.0099702, 0.0099702, 0.0099702
13, 0.054981, 0.07344, 0.035519, 0.52262, 0.38747, 0.39345, 0.2357, 0.048482, 0.053742, 0.026598, 0.0099645, 0.0099645, 0.0099645
14, 0.054688, 0.072956, 0.035128, 0.52054, 0.39699, 0.39944, 0.24077, 0.048217, 0.05351, 0.026196, 0.0099584, 0.0099584, 0.0099584
15, 0.054571, 0.073867, 0.034747, 0.56586, 0.38297, 0.40535, 0.24487, 0.047954, 0.05336, 0.025864, 0.0099517, 0.0099517, 0.0099517
16, 0.054332, 0.073118, 0.034194, 0.58583, 0.38081, 0.41132, 0.24901, 0.047765, 0.053222, 0.025573, 0.0099446, 0.0099446, 0.0099446
17, 0.054319, 0.073307, 0.034254, 0.5522, 0.40069, 0.41508, 0.2516, 0.047618, 0.05312, 0.02536, 0.009937, 0.009937, 0.009937
18, 0.053818, 0.073281, 0.033555, 0.55216, 0.40321, 0.41761, 0.25416, 0.047509, 0.053005, 0.025187, 0.0099289, 0.0099289, 0.0099289
19, 0.053636, 0.072146, 0.033614, 0.58506, 0.39111, 0.42018, 0.25602, 0.047416, 0.052914, 0.025055, 0.0099203, 0.0099203, 0.0099203
20, 0.053749, 0.072678, 0.033477, 0.58027, 0.39386, 0.42221, 0.25771, 0.047338, 0.052827, 0.024943, 0.0099112, 0.0099112, 0.0099112
21, 0.053523, 0.07243, 0.03298, 0.56371, 0.40463, 0.42434, 0.25931, 0.047264, 0.052753, 0.024834, 0.0099017, 0.0099017, 0.0099017
22, 0.053351, 0.071941, 0.032575, 0.56743, 0.40402, 0.42539, 0.26088, 0.047201, 0.052685, 0.024746, 0.0098916, 0.0098916, 0.0098916
23, 0.053263, 0.072275, 0.032751, 0.56791, 0.40374, 0.42676, 0.26203, 0.047139, 0.052629, 0.024666, 0.0098811, 0.0098811, 0.0098811
24, 0.052973, 0.07208, 0.032323, 0.55382, 0.41207, 0.42797, 0.26304, 0.047087, 0.052588, 0.024598, 0.0098701, 0.0098701, 0.0098701
25, 0.052886, 0.071604, 0.03219, 0.55647, 0.41262, 0.42945, 0.26401, 0.047041, 0.052566, 0.024535, 0.0098586, 0.0098586, 0.0098586
26, 0.053032, 0.072313, 0.032136, 0.55879, 0.41282, 0.43074, 0.265, 0.047002, 0.052565, 0.024472, 0.0098467, 0.0098467, 0.0098467
27, 0.052872, 0.072334, 0.032071, 0.55561, 0.41583, 0.43179, 0.26574, 0.046965, 0.052578, 0.024414, 0.0098342, 0.0098342, 0.0098342
28, 0.052599, 0.071546, 0.031794, 0.58263, 0.40555, 0.43298, 0.26681, 0.046929, 0.0526, 0.024354, 0.0098213, 0.0098213, 0.0098213
29, 0.052618, 0.071915, 0.031572, 0.55361, 0.42016, 0.43356, 0.26752, 0.046899, 0.052632, 0.024299, 0.0098079, 0.0098079, 0.0098079
30, 0.05259, 0.071716, 0.031395, 0.55701, 0.42011, 0.43425, 0.26827, 0.046869, 0.052684, 0.024257, 0.0097941, 0.0097941, 0.0097941
31, 0.052528, 0.071275, 0.031778, 0.55873, 0.42122, 0.43573, 0.26875, 0.046841, 0.052745, 0.024209, 0.0097798, 0.0097798, 0.0097798
32, 0.052462, 0.072346, 0.031427, 0.56271, 0.4202, 0.43623, 0.269, 0.046816, 0.052822, 0.02416, 0.009765, 0.009765, 0.009765
v5m-bs128
epoch, train/box_loss, train/obj_loss, train/cls_loss, metrics/precision, metrics/recall, metrics/mAP_0.5,metrics/mAP_0.5:0.95, val/box_loss, val/obj_loss, val/cls_loss, x/lr0, x/lr1, x/lr2
0, 0.088815, 0.083758, 0.083028, 0.10967, 0.060776, 0.0084766, 0.0028998, 0.07658, 0.065096, 0.075192, 0.0033297, 0.0033297, 0.070032
1, 0.07153, 0.082675, 0.069671, 0.25921, 0.12556, 0.050926, 0.021388, 0.067456, 0.063615, 0.058973, 0.0066629, 0.0066629, 0.040032
2, 0.065678, 0.079942, 0.056423, 0.28735, 0.19372, 0.13283, 0.059436, 0.060568, 0.061682, 0.046271, 0.0099954, 0.0099954, 0.010031
3, 0.061338, 0.078504, 0.04723, 0.40063, 0.24958, 0.22877, 0.11608, 0.055906, 0.058371, 0.037091, 0.0099978, 0.0099978, 0.0099978
4, 0.058455, 0.076397, 0.042069, 0.42689, 0.30947, 0.28819, 0.15312, 0.053267, 0.057035, 0.032819, 0.0099978, 0.0099978, 0.0099978
5, 0.056665, 0.075084, 0.038737, 0.48373, 0.33178, 0.32958, 0.18127, 0.051449, 0.05598, 0.03003, 0.0099961, 0.0099961, 0.0099961
6, 0.055591, 0.075298, 0.036506, 0.49513, 0.36799, 0.36491, 0.2067, 0.049934, 0.054871, 0.027925, 0.0099938, 0.0099938, 0.0099938
7, 0.054502, 0.073199, 0.035426, 0.49643, 0.39808, 0.38942, 0.22583, 0.048767, 0.054358, 0.026238, 0.0099911, 0.0099911, 0.0099911
8, 0.053876, 0.07238, 0.033299, 0.52135, 0.41381, 0.41562, 0.24501, 0.04781, 0.053518, 0.024843, 0.0099879, 0.0099879, 0.0099879
9, 0.053236, 0.072389, 0.032481, 0.54612, 0.41269, 0.43074, 0.25796, 0.046989, 0.053086, 0.023894, 0.0099842, 0.0099842, 0.0099842
10, 0.052832, 0.07241, 0.031733, 0.55228, 0.43234, 0.44671, 0.26953, 0.046382, 0.052618, 0.023047, 0.00998, 0.00998, 0.00998
11, 0.05282, 0.074112, 0.031097, 0.5533, 0.44669, 0.46024, 0.28065, 0.045857, 0.052293, 0.022248, 0.0099753, 0.0099753, 0.0099753
12, 0.052155, 0.071912, 0.030904, 0.55584, 0.45631, 0.46781, 0.28701, 0.045427, 0.051986, 0.021778, 0.0099702, 0.0099702, 0.0099702
13, 0.051747, 0.072218, 0.03014, 0.60159, 0.44424, 0.47684, 0.29523, 0.045071, 0.051834, 0.021251, 0.0099645, 0.0099645, 0.0099645
14, 0.051199, 0.071139, 0.029753, 0.60888, 0.44745, 0.4834, 0.30013, 0.044758, 0.051615, 0.020943, 0.0099584, 0.0099584, 0.0099584
15, 0.051116, 0.071044, 0.029682, 0.61669, 0.45095, 0.48788, 0.30411, 0.044482, 0.051461, 0.020708, 0.0099517, 0.0099517, 0.0099517
16, 0.051165, 0.071407, 0.029131, 0.58867, 0.46949, 0.49298, 0.30788, 0.044266, 0.051328, 0.020452, 0.0099446, 0.0099446, 0.0099446
17, 0.050888, 0.070982, 0.029018, 0.60247, 0.46829, 0.49698, 0.3109, 0.044097, 0.051225, 0.020252, 0.009937, 0.009937, 0.009937
18, 0.050531, 0.070728, 0.028839, 0.62038, 0.46144, 0.49886, 0.31304, 0.04396, 0.051144, 0.020106, 0.0099289, 0.0099289, 0.0099289
19, 0.050503, 0.070549, 0.028389, 0.61684, 0.4655, 0.50101, 0.31475, 0.043849, 0.051041, 0.019979, 0.0099203, 0.0099203, 0.0099203
20, 0.050645, 0.070725, 0.028245, 0.60357, 0.47478, 0.5033, 0.31661, 0.043755, 0.050947, 0.019874, 0.0099112, 0.0099112, 0.0099112
21, 0.050031, 0.069934, 0.027804, 0.60496, 0.47581, 0.50527, 0.31852, 0.043667, 0.050866, 0.01977, 0.0099017, 0.0099017, 0.0099017
22, 0.049924, 0.069962, 0.027742, 0.61439, 0.47212, 0.50718, 0.32007, 0.043592, 0.050781, 0.019677, 0.0098916, 0.0098916, 0.0098916
23, 0.049774, 0.070259, 0.027356, 0.62347, 0.4707, 0.50993, 0.32225, 0.043526, 0.050693, 0.019594, 0.0098811, 0.0098811, 0.0098811
24, 0.049478, 0.06908, 0.0275, 0.59173, 0.49002, 0.51115, 0.32364, 0.043463, 0.050619, 0.019518, 0.0098701, 0.0098701, 0.0098701
25, 0.049521, 0.070211, 0.027425, 0.59696, 0.4898, 0.51278, 0.32486, 0.043407, 0.050557, 0.019451, 0.0098586, 0.0098586, 0.0098586
26, 0.049435, 0.07008, 0.027264, 0.61438, 0.48112, 0.51381, 0.32583, 0.043351, 0.050498, 0.019382, 0.0098467, 0.0098467, 0.0098467
27, 0.049354, 0.070545, 0.026766, 0.6064, 0.48604, 0.51499, 0.32726, 0.043301, 0.050441, 0.019321, 0.0098342, 0.0098342, 0.0098342
28, 0.049133, 0.069629, 0.02687, 0.63298, 0.47289, 0.51647, 0.32861, 0.043251, 0.050395, 0.019262, 0.0098213, 0.0098213, 0.0098213
29, 0.049244, 0.070239, 0.026565, 0.61425, 0.48472, 0.51749, 0.32981, 0.043202, 0.050362, 0.0192, 0.0098079, 0.0098079, 0.0098079
30, 0.049247, 0.069914, 0.026879, 0.62765, 0.47932, 0.51836, 0.33032, 0.043154, 0.050333, 0.019143, 0.0097941, 0.0097941, 0.0097941
31, 0.048997, 0.070243, 0.026534, 0.64588, 0.47083, 0.51933, 0.33128, 0.04311, 0.05031, 0.019085, 0.0097798, 0.0097798, 0.0097798
32, 0.048746, 0.06989, 0.026604, 0.62604, 0.48261, 0.52032, 0.3324, 0.043065, 0.050294, 0.019024, 0.009765, 0.009765, 0.009765
33, 0.048844, 0.069789, 0.026362, 0.60369, 0.49778, 0.52143, 0.3331, 0.04302, 0.050284, 0.018967, 0.0097497, 0.0097497, 0.0097497
34, 0.048838, 0.06909, 0.026239, 0.6214, 0.48986, 0.52276, 0.33407, 0.042976, 0.050265, 0.018912, 0.009734, 0.009734, 0.009734
35, 0.048356, 0.069284, 0.025822, 0.61233, 0.49574, 0.52344, 0.33489, 0.042929, 0.050244, 0.018861, 0.0097178, 0.0097178, 0.0097178
36, 0.048366, 0.068073, 0.025951, 0.61853, 0.49364, 0.5246, 0.3363, 0.042877, 0.050227, 0.018803, 0.0097011, 0.0097011, 0.0097011
37, 0.048487, 0.069332, 0.025808, 0.63126, 0.48776, 0.52577, 0.33738, 0.042821, 0.050208, 0.01875, 0.009684, 0.009684, 0.009684
38, 0.048189, 0.069316, 0.025894, 0.64172, 0.48407, 0.52724, 0.33821, 0.042764, 0.050177, 0.018681, 0.0096664, 0.0096664, 0.0096664
39, 0.048339, 0.069556, 0.025417, 0.65185, 0.48204, 0.52911, 0.33938, 0.042711, 0.050147, 0.018621, 0.0096484, 0.0096484, 0.0096484
40, 0.048066, 0.068588, 0.025469, 0.65607, 0.4814, 0.53022, 0.34057, 0.042654, 0.050112, 0.018563, 0.0096299, 0.0096299, 0.0096299
41, 0.048106, 0.069846, 0.025577, 0.66145, 0.48068, 0.53182, 0.34192, 0.042593, 0.050081, 0.018496, 0.009611, 0.009611, 0.009611
42, 0.04817, 0.069178, 0.025513, 0.65954, 0.48355, 0.53306, 0.34304, 0.042531, 0.050039, 0.018439, 0.0095916, 0.0095916, 0.0095916
43, 0.047838, 0.068633, 0.0254, 0.63862, 0.49544, 0.5345, 0.34428, 0.042466, 0.049994, 0.01837, 0.0095717, 0.0095717, 0.0095717
44, 0.047805, 0.068237, 0.025414, 0.62767, 0.50361, 0.53601, 0.34541, 0.0424, 0.049956, 0.018304, 0.0095514, 0.0095514, 0.0095514
45, 0.047678, 0.068207, 0.025349, 0.63458, 0.5013, 0.53745, 0.34681, 0.042332, 0.04991, 0.018235, 0.0095307, 0.0095307, 0.0095307
46, 0.047662, 0.068054, 0.02531, 0.62932, 0.50579, 0.53886, 0.34872, 0.042266, 0.049872, 0.018165, 0.0095095, 0.0095095, 0.0095095
47, 0.048001, 0.069291, 0.024793, 0.63726, 0.50436, 0.54061, 0.35014, 0.0422, 0.049825, 0.0181, 0.0094879, 0.0094879, 0.0094879
48, 0.047774, 0.068069, 0.025169, 0.65425, 0.4975, 0.54264, 0.3516, 0.042133, 0.049773, 0.018031, 0.0094659, 0.0094659, 0.0094659
49, 0.047674, 0.06858, 0.024896, 0.64272, 0.50575, 0.54415, 0.35326, 0.04207, 0.049724, 0.017963, 0.0094434, 0.0094434, 0.0094434
50, 0.047703, 0.068074, 0.025099, 0.64057, 0.50878, 0.54578, 0.35462, 0.042001, 0.04967, 0.017888, 0.0094205, 0.0094205, 0.0094205
51, 0.047624, 0.068676, 0.025119, 0.65135, 0.50244, 0.54779, 0.35586, 0.041937, 0.049611, 0.017818, 0.0093971, 0.0093971, 0.0093971
52, 0.047448, 0.068091, 0.024714, 0.62977, 0.51681, 0.54856, 0.35716, 0.041875, 0.049551, 0.017752, 0.0093733, 0.0093733, 0.0093733
53, 0.047625, 0.068465, 0.024857, 0.64038, 0.51344, 0.55018, 0.3584, 0.041815, 0.049497, 0.017687, 0.0093491, 0.0093491, 0.0093491
54, 0.047647, 0.067882, 0.02476, 0.64501, 0.51334, 0.55105, 0.35947, 0.041749, 0.049447, 0.017623, 0.0093245, 0.0093245, 0.0093245
55, 0.047092, 0.068041, 0.024622, 0.64294, 0.51586, 0.55202, 0.36056, 0.041692, 0.049388, 0.017557, 0.0092995, 0.0092995, 0.0092995
56, 0.047486, 0.069097, 0.024603, 0.64305, 0.51754, 0.55371, 0.36145, 0.041635, 0.049338, 0.017489, 0.009274, 0.009274, 0.009274
57, 0.047281, 0.067822, 0.024274, 0.65651, 0.51091, 0.55552, 0.36304, 0.041588, 0.049283, 0.017427, 0.0092481, 0.0092481, 0.0092481
58, 0.047146, 0.068453, 0.024524, 0.64516, 0.51803, 0.55695, 0.3643, 0.041535, 0.04923, 0.017362, 0.0092219, 0.0092219, 0.0092219
59, 0.047279, 0.068181, 0.024352, 0.6465, 0.51983, 0.55847, 0.36542, 0.041479, 0.049175, 0.017288, 0.0091952, 0.0091952, 0.0091952
60, 0.047158, 0.067661, 0.024416, 0.66316, 0.51224, 0.55995, 0.3668, 0.041424, 0.049119, 0.017216, 0.0091681, 0.0091681, 0.0091681
61, 0.047229, 0.067677, 0.024336, 0.64764, 0.52375, 0.56066, 0.36761, 0.041372, 0.049065, 0.017156, 0.0091406, 0.0091406, 0.0091406
62, 0.047253, 0.068938, 0.024497, 0.66099, 0.51725, 0.56189, 0.36853, 0.041325, 0.049013, 0.017093, 0.0091127, 0.0091127, 0.0091127
63, 0.04662, 0.067084, 0.024585, 0.67994, 0.50749, 0.56263, 0.36913, 0.041279, 0.048957, 0.01703, 0.0090844, 0.0090844, 0.0090844
64, 0.047323, 0.06862, 0.024134, 0.67348, 0.51133, 0.56362, 0.37034, 0.041234, 0.048909, 0.01698, 0.0090557, 0.0090557, 0.0090557
65, 0.046989, 0.068175, 0.024199, 0.65902, 0.52133, 0.56497, 0.37145, 0.041186, 0.04886, 0.016933, 0.0090266, 0.0090266, 0.0090266
66, 0.04687, 0.068229, 0.02443, 0.65823, 0.52366, 0.56575, 0.3723, 0.041136, 0.048812, 0.016881, 0.0089972, 0.0089972, 0.0089972
67, 0.046718, 0.067423, 0.023916, 0.66329, 0.52201, 0.56709, 0.37339, 0.041089, 0.048763, 0.016834, 0.0089673, 0.0089673, 0.0089673
68, 0.046628, 0.067355, 0.024175, 0.66367, 0.52275, 0.56778, 0.37432, 0.041048, 0.04872, 0.016788, 0.0089371, 0.0089371, 0.0089371
69, 0.0465, 0.067259, 0.024017, 0.6681, 0.52258, 0.56856, 0.37542, 0.041009, 0.048675, 0.016744, 0.0089065, 0.0089065, 0.0089065
70, 0.046715, 0.067115, 0.023709, 0.67535, 0.51958, 0.56963, 0.37629, 0.040966, 0.04863, 0.016697, 0.0088755, 0.0088755, 0.0088755
71, 0.046674, 0.067407, 0.023887, 0.68412, 0.51669, 0.57097, 0.37746, 0.040926, 0.048595, 0.016648, 0.0088442, 0.0088442, 0.0088442
72, 0.04641, 0.066941, 0.023573, 0.68191, 0.51886, 0.5722, 0.37853, 0.040879, 0.048562, 0.0166, 0.0088124, 0.0088124, 0.0088124
73, 0.046671, 0.067873, 0.023917, 0.6693, 0.52578, 0.57309, 0.37919, 0.040841, 0.048532, 0.016555, 0.0087804, 0.0087804, 0.0087804
74, 0.046719, 0.068519, 0.023678, 0.67838, 0.52318, 0.57385, 0.37985, 0.040802, 0.048502, 0.01651, 0.0087479, 0.0087479, 0.0087479
75, 0.046399, 0.067485, 0.02343, 0.6341, 0.55018, 0.57521, 0.3805, 0.040763, 0.048473, 0.016464, 0.0087151, 0.0087151, 0.0087151
76, 0.046487, 0.067537, 0.023685, 0.6393, 0.54889, 0.57615, 0.38104, 0.040725, 0.048438, 0.016416, 0.008682, 0.008682, 0.008682
77, 0.046465, 0.067602, 0.023841, 0.63906, 0.55104, 0.57683, 0.38147, 0.040689, 0.048402, 0.016367, 0.0086485, 0.0086485, 0.0086485
78, 0.046769, 0.068515, 0.023673, 0.65271, 0.54478, 0.578, 0.38235, 0.040655, 0.048372, 0.016318, 0.0086146, 0.0086146, 0.0086146
79, 0.046306, 0.067563, 0.0236, 0.6569, 0.54514, 0.57891, 0.38297, 0.040619, 0.048347, 0.01627, 0.0085805, 0.0085805, 0.0085805
80, 0.046484, 0.067084, 0.023284, 0.66444, 0.54124, 0.57938, 0.38361, 0.040583, 0.04832, 0.016229, 0.0085459, 0.0085459, 0.0085459
81, 0.046424, 0.067326, 0.023579, 0.66784, 0.53932, 0.58006, 0.38428, 0.040549, 0.048297, 0.016193, 0.0085111, 0.0085111, 0.0085111
82, 0.046469, 0.066749, 0.02311, 0.67043, 0.53995, 0.58087, 0.38432, 0.040513, 0.048274, 0.01616, 0.0084759, 0.0084759, 0.0084759
83, 0.046417, 0.068277, 0.023495, 0.66224, 0.54308, 0.58145, 0.38531, 0.040482, 0.048251, 0.016126, 0.0084404, 0.0084404, 0.0084404
yolov5s
epoch, box_loss obj_loss, cls_loss, precision,recall, mAP_0.5, 0.5:0.95, box_loss,obj_loss, cls_loss, x/lr0, x/lr1, x/lr2
0/299 0.514G 0.08983 0.08378 0.0827 0.2563 35 640 0.1194 0.04001 0.004798 0.001399 0.07812 0.06498 0.077
1/299 12.1G 0.0736 0.08291 0.07077 0.2273 43 640 0.3335 0.09528 0.04203 0.01542 0.07008 0.06207 0.06096
2/299 13.9G 0.06825 0.08104 0.05821 0.2075 25 640 0.3315 0.1591 0.09668 0.03747 0.06493 0.06124 0.04935
3/299 13.9G 0.06433 0.07907 0.04916 0.1926 41 640 0.3415 0.2254 0.1707 0.07781 0.05992 0.05944 0.04079
4/299 13.9G 0.06164 0.07765 0.04405 0.1833 17 640 0.4311 0.249 0.2297 0.1112 0.05726 0.05856 0.03643
5/299 13.9G 0.06012 0.07706 0.04113 0.1783 36 640 0.4069 0.2922 0.2669 0.1348 0.05562 0.05759 0.03344
6/299 13.9G 0.05902 0.07624 0.03907 0.1743 71 640 0.4815 0.3003 0.3001 0.1566 0.05415 0.05684 0.03096
7/299 13.9G 0.05829 0.07524 0.03744 0.171 74 640 0.4619 0.3395 0.3239 0.1722 0.05314 0.05624 0.02939
8/299 13.9G 0.05778 0.07568 0.03638 0.1698 80 640 0.5126 0.3358 0.3444 0.1866 0.05225 0.05577 0.02818
9/299 13.9G 0.05721 0.07519 0.0356 0.168 46 640 0.521 0.3554 0.3634 0.1998 0.05158 0.05533 0.027
10/299 13.9G 0.05662 0.07448 0.03487 0.166 32 640 0.5044 0.3743 0.3757 0.2083 0.05103 0.05493 0.02623
11/299 13.9G 0.05629 0.07418 0.03408 0.1646 75 640 0.517 0.3836 0.3881 0.2169 0.05059 0.05473 0.02556
12/299 13.9G 0.05601 0.07415 0.03357 0.1637 65 640 0.5281 0.3862 0.3939 0.2217 0.05021 0.05443 0.02503
13/299 13.9G 0.05573 0.0742 0.03318 0.1631 63 640 0.5521 0.3839 0.401 0.2279 0.04981 0.05421 0.02453
14/299 13.9G 0.05541 0.0737 0.03265 0.1618 37 640 0.5591 0.391 0.4085 0.2338 0.04954 0.05401 0.0241
15/299 13.9G 0.05519 0.07384 0.03234 0.1614 28 640 0.5382 0.4074 0.413 0.2376 0.04932 0.05388 0.02382
16/299 13.9G 0.05492 0.07396 0.03209 0.161 42 640 0.5667 0.3987 0.4175 0.2409 0.04913 0.05374 0.0236
17/299 13.9G 0.05482 0.07353 0.0319 0.1602 43 640 0.5424 0.4097 0.4211 0.2431 0.04898 0.05364 0.02344
18/299 13.9G 0.05465 0.07341 0.0314 0.1595 38 640 0.5381 0.4128 0.4235 0.2455 0.04884 0.05354 0.02328
19/299 13.9G 0.05451 0.07297 0.03113 0.1586 81 640 0.5598 0.4054 0.4264 0.2473 0.04874 0.05344 0.02314
20/299 13.9G 0.05443 0.07332 0.03108 0.1588 95 640 0.5749 0.3995 0.4289 0.2495 0.04866 0.05337 0.02303
21/299 13.9G 0.05414 0.0727 0.03092 0.1578 26 640 0.5735 0.4018 0.4306 0.2506 0.04859 0.0533 0.02295
22/299 13.9G 0.05417 0.0729 0.0308 0.1579 76 640 0.5559 0.4127 0.4322 0.2518 0.04853 0.05324 0.02287
23/299 13.9G 0.05401 0.0726 0.03051 0.1571 97 640 0.5797 0.4016 0.433 0.253 0.04847 0.05319 0.02282
24/299 13.9G 0.05376 0.07268 0.03013 0.1566 77 640 0.5843 0.4011 0.4338 0.2538 0.04843 0.05315 0.02277
25/299 13.9G 0.05367 0.07225 0.03033 0.1563 33 640 0.5896 0.4006 0.4349 0.2547 0.04838 0.05311 0.02272
26/299 13.9G 0.05358 0.07266 0.02993 0.1562 62 640 0.5885 0.4028 0.4363 0.2554 0.04834 0.05309 0.02267
27/299 13.9G 0.05354 0.07231 0.03007 0.1559 80 640 0.5864 0.4041 0.4371 0.2563 0.04831 0.05307 0.02263
28/299 13.9G 0.05331 0.07255 0.02963 0.1555 62 640 0.5884 0.4063 0.4386 0.2571 0.04827 0.05306 0.02259
29/299 13.9G 0.05326 0.07198 0.02965 0.1549 66 640 0.5837 0.4108 0.4395 0.2579 0.04824 0.05306 0.02255
30/299 13.9G 0.05322 0.07234 0.02951 0.1551 54 640 0.5867 0.4096 0.4399 0.2587 0.04821 0.05306 0.02252
31/299 13.9G 0.05315 0.07232 0.02951 0.155 46 640 0.5665 0.4219 0.4407 0.259 0.04818 0.05307 0.02248
32/299 13.9G 0.05315 0.07247 0.02926 0.1549 29 640 0.5637 0.4239 0.4418 0.2596 0.04815 0.05309 0.02244
33/299 13.9G 0.05295 0.07147 0.02943 0.1538 26 640 0.5673 0.4228 0.443 0.2605 0.04811 0.0531 0.02241
34/299 13.9G 0.05298 0.07178 0.02915 0.1539 37 640 0.5634 0.426 0.4439 0.2611 0.04808 0.05311 0.02237
35/299 13.9G 0.05283 0.07181 0.02887 0.1535 34 640 0.5717 0.4236 0.4453 0.2618 0.04804 0.05311 0.02233
36/299 13.9G 0.05278 0.07196 0.02874 0.1535 73 640 0.5685 0.426 0.4459 0.2625 0.048 0.05312 0.02228
37/299 13.9G 0.05267 0.07162 0.0289 0.1532 103 640 0.5722 0.425 0.4473 0.2634 0.04796 0.05313 0.02222
38/299 13.9G 0.05264 0.07175 0.02878 0.1532 18 640 0.5657 0.4295 0.4478 0.264 0.04792 0.05312 0.02217
39/299 13.9G 0.05255 0.07183 0.0286 0.153 58 640 0.5646 0.4302 0.4492 0.2649 0.04788 0.05312 0.02212
40/299 13.9G 0.05259 0.07194 0.02856 0.1531 86 640 0.5832 0.4236 0.4505 0.2658 0.04783 0.0531 0.02205
41/299 13.9G 0.05239 0.07185 0.02851 0.1527 36 640 0.5801 0.4264 0.4518 0.2667 0.04778 0.05308 0.02199
42/299 13.9G 0.05235 0.07162 0.02835 0.1523 73 640 0.5863 0.4254 0.4527 0.2675 0.04773 0.05306 0.02192
43/299 13.9G 0.05235 0.07114 0.02857 0.1521 30 640 0.5815 0.4291 0.4545 0.2686 0.04767 0.05302 0.02185
44/299 13.9G 0.0523 0.07158 0.02833 0.1522 23 640 0.5914 0.4252 0.4554 0.2698 0.04761 0.05299 0.02178
45/299 13.9G 0.05239 0.07187 0.02841 0.1527 89 640 0.587 0.4306 0.4572 0.2711 0.04755 0.05295 0.0217
46/299 13.9G 0.05219 0.07143 0.02829 0.1519 67 640 0.5747 0.4384 0.4589 0.2721 0.04749 0.05291 0.02163
47/299 13.9G 0.05213 0.0717 0.02804 0.1519 54 640 0.5909 0.4336 0.4604 0.2732 0.04743 0.05287 0.02156
48/299 13.9G 0.05219 0.07187 0.02816 0.1522 47 640 0.5879 0.4372 0.4621 0.2746 0.04736 0.05282 0.02148
49/299 13.9G 0.05191 0.0712 0.02806 0.1512 86 640 0.5875 0.4391 0.4636 0.2761 0.04729 0.05276 0.02141
50/299 13.9G 0.05195 0.07123 0.02797 0.1512 55 640 0.5889 0.4404 0.4648 0.2773 0.04723 0.05271 0.02133
51/299 13.9G 0.05188 0.07089 0.02786 0.1506 104 640 0.6019 0.4355 0.4664 0.2783 0.04716 0.05265 0.02126
52/299 13.9G 0.0519 0.07134 0.02792 0.1512 48 640 0.5904 0.4421 0.4674 0.2795 0.04709 0.05261 0.02119
53/299 13.9G 0.05184 0.0709 0.02775 0.1505 45 640 0.6038 0.4361 0.4684 0.2806 0.04703 0.05255 0.02111
54/299 13.9G 0.05182 0.07142 0.02777 0.151 61 640 0.5985 0.4413 0.4701 0.2815 0.04697 0.0525 0.02104
55/299 13.9G 0.05179 0.07121 0.02773 0.1507 56 640 0.5998 0.443 0.4713 0.2828 0.04691 0.05243 0.02096
56/299 13.9G 0.05177 0.07102 0.02769 0.1505 51 640 0.6019 0.4418 0.4721 0.2842 0.04684 0.05238 0.0209
57/299 13.9G 0.05174 0.0713 0.02759 0.1506 60 640 0.62 0.4344 0.4737 0.285 0.04678 0.05232 0.02083
58/299 13.9G 0.05167 0.07133 0.02746 0.1505 31 640 0.6075 0.4405 0.4753 0.2858 0.04673 0.05227 0.02077
59/299 13.9G 0.05163 0.07157 0.02744 0.1506 84 640 0.5966 0.4474 0.4763 0.287 0.04667 0.05222 0.0207
60/299 13.9G 0.05152 0.07094 0.0274 0.1499 86 640 0.5997 0.4486 0.4776 0.2879 0.04662 0.05217 0.02064
61/299 13.9G 0.05151 0.0712 0.02744 0.1501 72 640 0.6265 0.4381 0.4787 0.2891 0.04657 0.05212 0.02058
62/299 13.9G 0.05158 0.07181 0.02733 0.1507 35 640 0.6248 0.4403 0.4802 0.29 0.04651 0.05207 0.02052
63/299 13.9G 0.05141 0.07083 0.02729 0.1495 17 640 0.6215 0.4429 0.4811 0.2912 0.04646 0.05201 0.02047
64/299 13.9G 0.05142 0.07106 0.02736 0.1498 61 640 0.6253 0.4408 0.482 0.2921 0.0464 0.05196 0.0204
65/299 13.9G 0.05138 0.07059 0.02712 0.1491 35 640 0.5948 0.4569 0.4831 0.2932 0.04635 0.05191 0.02035
66/299 13.9G 0.05139 0.07051 0.02736 0.1493 73 640 0.6007 0.4564 0.4844 0.2939 0.0463 0.05186 0.0203
67/299 13.9G 0.05125 0.07068 0.02727 0.1492 19 640 0.6065 0.4557 0.4855 0.2951 0.04624 0.05182 0.02024
68/299 13.9G 0.05118 0.07086 0.02705 0.1491 27 640 0.6141 0.4528 0.4864 0.296 0.0462 0.05178 0.02018
69/299 13.9G 0.05119 0.07093 0.02711 0.1492 38 640 0.6169 0.4514 0.4878 0.2965 0.04615 0.05174 0.02014
70/299 13.9G 0.05123 0.07064 0.02716 0.149 42 640 0.6163 0.4524 0.4887 0.297 0.0461 0.05169 0.02008
71/299 13.9G 0.0512 0.07054 0.02688 0.1486 26 640 0.6059 0.4585 0.4896 0.2979 0.04606 0.05164 0.02002
72/299 13.9G 0.05109 0.07064 0.02706 0.1488 43 640 0.6122 0.4571 0.4907 0.2987 0.04601 0.05159 0.01997
73/299 13.9G 0.05119 0.07036 0.02696 0.1485 45 640 0.6171 0.4542 0.4911 0.2997 0.04596 0.05155 0.01992
74/299 13.9G 0.05109 0.07071 0.02697 0.1488 87 640 0.6161 0.4564 0.4923 0.3004 0.04592 0.05151 0.01987
75/299 13.9G 0.05106 0.07082 0.0268 0.1487 93 640 0.6187 0.4561 0.493 0.3008 0.04588 0.05147 0.01982
76/299 13.9G 0.051 0.07073 0.02673 0.1485 54 640 0.613 0.459 0.4938 0.3015 0.04584 0.05142 0.01977
77/299 13.9G 0.05089 0.07088 0.02671 0.1485 26 640 0.6137 0.4607 0.4949 0.3022 0.04579 0.05138 0.01972
78/299 13.9G 0.05101 0.07076 0.02676 0.1485 44 640 0.5912 0.4766 0.496 0.3031 0.04575 0.05135 0.01968
79/299 13.9G 0.05083 0.07027 0.02652 0.1476 65 640 0.5986 0.4736 0.4969 0.3036 0.04571 0.05131 0.01965
80/299 13.9G 0.05089 0.07033 0.02673 0.148 28 640 0.6124 0.4689 0.498 0.3045 0.04568 0.05129 0.0196
81/299 13.9G 0.05093 0.0707 0.02661 0.1482 40 640 0.6135 0.4679 0.4988 0.3051 0.04565 0.05125 0.01957
82/299 13.9G 0.05075 0.07079 0.02667 0.1482 34 640 0.6106 0.47 0.499 0.3056 0.04561 0.05122 0.01953
83/299 13.9G 0.05073 0.07052 0.02658 0.1478 25 640 0.5994 0.4765 0.5001 0.3062 0.04558 0.05119 0.01949
84/299 13.9G 0.05066 0.07025 0.02652 0.1474 43 640 0.6155 0.4676 0.4997 0.3064 0.04555 0.05116 0.01945
85/299 13.9G 0.05064 0.07047 0.0265 0.1476 53 640 0.6222 0.4661 0.501 0.307 0.04552 0.05113 0.01941
86/299 13.9G 0.05073 0.07078 0.02638 0.1479 44 640 0.6253 0.4657 0.5019 0.3077 0.04549 0.05111 0.01937
87/299 13.9G 0.05085 0.07053 0.02638 0.1478 48 640 0.6244 0.4669 0.5038 0.3089 0.04546 0.05108 0.01933
88/299 13.9G 0.05058 0.07019 0.02623 0.147 26 640 0.6253 0.467 0.5047 0.3095 0.04543 0.05106 0.01929
89/299 13.9G 0.05057 0.07035 0.02644 0.1474 36 640 0.6291 0.4643 0.5052 0.3098 0.0454 0.05104 0.01925
90/299 13.9G 0.05063 0.07044 0.02618 0.1473 45 640 0.6227 0.4677 0.5058 0.3105 0.04538 0.05102 0.01921
91/299 13.9G 0.05046 0.07021 0.02621 0.1469 37 640 0.6151 0.4731 0.5067 0.3111 0.04535 0.05099 0.01917
92/299 13.9G 0.05044 0.07041 0.0262 0.1471 81 640 0.6155 0.4746 0.508 0.312 0.04532 0.05097 0.01913
93/299 13.9G 0.05041 0.06993 0.02607 0.1464 62 640 0.6212 0.4727 0.5085 0.3125 0.0453 0.05095 0.01909
94/299 13.9G 0.05054 0.07044 0.02621 0.1472 44 640 0.6101 0.4793 0.5092 0.3129 0.04528 0.05093 0.01906
95/299 13.9G 0.05033 0.0698 0.02606 0.1462 39 640 0.6045 0.4828 0.5093 0.3131 0.04525 0.05091 0.01903
96/299 13.9G 0.05044 0.07008 0.02611 0.1466 81 640 0.6158 0.478 0.51 0.3134 0.04523 0.05089 0.019
97/299 13.9G 0.05042 0.07099 0.02594 0.1473 56 640 0.612 0.4809 0.51 0.3133 0.0452 0.05087 0.01896
98/299 13.9G 0.05024 0.06972 0.02603 0.146 48 640 0.63 0.4728 0.5108 0.3141 0.04518 0.05086 0.01893
99/299 13.9G 0.0503 0.07 0.0261 0.1464 60 640 0.6109 0.4824 0.5115 0.3143 0.04515 0.05084 0.01888
100/299 13.9G 0.05016 0.06984 0.02592 0.1459 50 640 0.6407 0.4697 0.5124 0.315 0.04513 0.05082 0.01885
101/299 13.9G 0.0503 0.07008 0.02591 0.1463 63 640 0.6279 0.4771 0.5128 0.3154 0.0451 0.05081 0.01882
102/299 13.9G 0.05026 0.06997 0.02589 0.1461 59 640 0.629 0.4772 0.5137 0.3158 0.04508 0.05079 0.01879
103/299 13.9G 0.05017 0.07046 0.02587 0.1465 30 640 0.638 0.4745 0.5142 0.316 0.04506 0.05077 0.01877
104/299 13.9G 0.05025 0.07051 0.02572 0.1465 53 640 0.6317 0.4763 0.5143 0.3166 0.04503 0.05075 0.01874
105/299 13.9G 0.05015 0.06961 0.02557 0.1453 61 640 0.6316 0.4775 0.5152 0.3175 0.045 0.05073 0.01871
106/299 13.9G 0.0501 0.07019 0.02576 0.146 30 640 0.6228 0.4834 0.516 0.3177 0.04498 0.05071 0.01869
107/299 13.9G 0.0502 0.06981 0.02575 0.1458 17 640 0.6276 0.4798 0.5162 0.3182 0.04496 0.0507 0.01868
108/299 13.9G 0.05004 0.07002 0.02569 0.1457 68 640 0.6091 0.4907 0.5164 0.3184 0.04493 0.05069 0.01867
109/299 13.9G 0.04992 0.06985 0.0255 0.1453 56 640 0.6029 0.4943 0.5166 0.3192 0.04491 0.05067 0.01864
110/299 13.9G 0.05001 0.06975 0.02565 0.1454 35 640 0.6214 0.4847 0.5173 0.3196 0.04489 0.05066 0.01862
111/299 13.9G 0.05002 0.0702 0.02554 0.1458 41 640 0.6133 0.4906 0.5177 0.3197 0.04487 0.05065 0.0186
112/299 13.9G 0.04995 0.06921 0.02565 0.1448 41 640 0.6152 0.4897 0.5187 0.3206 0.04485 0.05064 0.01859
113/299 13.9G 0.0499 0.06982 0.02538 0.1451 71 640 0.6226 0.487 0.5191 0.3205 0.04482 0.05063 0.01856
114/299 13.9G 0.04982 0.06954 0.02548 0.1448 81 640 0.6252 0.4857 0.5195 0.3208 0.0448 0.05062 0.01853
115/299 13.9G 0.04997 0.06978 0.02545 0.1452 76 640 0.6257 0.486 0.5197 0.321 0.04478 0.05061 0.01851
116/299 13.9G 0.04984 0.06965 0.02557 0.1451 42 640 0.6258 0.487 0.5206 0.3213 0.04476 0.05061 0.01849
117/299 13.9G 0.04983 0.07003 0.02536 0.1452 31 640 0.6384 0.4796 0.5208 0.322 0.04475 0.05059 0.01848
118/299 13.9G 0.04978 0.0695 0.02539 0.1447 38 640 0.6274 0.4866 0.5213 0.3226 0.04473 0.05058 0.01846
119/299 13.9G 0.04963 0.06953 0.02525 0.1444 30 640 0.6304 0.4853 0.5215 0.323 0.04471 0.05058 0.01843
120/299 13.9G 0.04971 0.06967 0.02537 0.1448 29 640 0.6465 0.4776 0.5217 0.3234 0.0447 0.05057 0.01841
121/299 13.9G 0.04979 0.06958 0.02512 0.1445 37 640 0.6441 0.4775 0.5217 0.3235 0.04469 0.05056 0.0184
122/299 13.9G 0.04993 0.06982 0.02532 0.1451 57 640 0.6404 0.479 0.5218 0.3239 0.04467 0.05055 0.01838
123/299 13.9G 0.04958 0.06963 0.02497 0.1442 64 640 0.6192 0.4901 0.5222 0.324 0.04465 0.05054 0.01836
124/299 13.9G 0.04956 0.06996 0.02511 0.1446 70 640 0.619 0.4912 0.5229 0.3244 0.04464 0.05053 0.01834
125/299 13.9G 0.04962 0.06923 0.02494 0.1438 38 640 0.6195 0.4927 0.5235 0.3247 0.04462 0.05052 0.01833
126/299 13.9G 0.04964 0.06938 0.02518 0.1442 36 640 0.6149 0.4959 0.5243 0.3252 0.0446 0.05051 0.01831
127/299 13.9G 0.04964 0.06994 0.02506 0.1446 19 640 0.614 0.4964 0.5246 0.3258 0.04459 0.0505 0.01829
128/299 13.9G 0.04955 0.06917 0.02492 0.1436 19 640 0.6218 0.4921 0.5255 0.3263 0.04457 0.0505 0.01825
129/299 13.9G 0.04957 0.06944 0.02487 0.1439 79 640 0.6164 0.4975 0.5264 0.3263 0.04455 0.05049 0.01823
130/299 13.9G 0.04945 0.06973 0.02481 0.144 62 640 0.6147 0.4984 0.5266 0.3269 0.04453 0.05048 0.01821
131/299 13.9G 0.04948 0.0693 0.02493 0.1437 34 640 0.6257 0.4916 0.5269 0.3269 0.04452 0.05047 0.01819
132/299 13.9G 0.04933 0.06903 0.0247 0.1431 94 640 0.6251 0.4918 0.5273 0.3272 0.0445 0.05046 0.01818
133/299 13.9G 0.04933 0.06906 0.0247 0.1431 73 640 0.6171 0.4963 0.5274 0.3281 0.04447 0.05045 0.01815
134/299 13.9G 0.04941 0.06976 0.02479 0.144 70 640 0.6161 0.4967 0.5276 0.3285 0.04445 0.05043 0.01813
135/299 13.9G 0.04931 0.06943 0.02469 0.1434 49 640 0.627 0.4911 0.5279 0.3288 0.04443 0.05042 0.01811
136/299 13.9G 0.04944 0.0695 0.02473 0.1437 72 640 0.6436 0.4825 0.5282 0.3291 0.04441 0.0504 0.01809
137/299 13.9G 0.04926 0.06901 0.02465 0.1429 51 640 0.645 0.4823 0.5287 0.3296 0.0444 0.05038 0.01808
138/299 13.9G 0.04932 0.06938 0.02476 0.1435 60 640 0.6409 0.4852 0.5289 0.3298 0.04439 0.05037 0.01806
139/299 13.9G 0.04929 0.0694 0.02463 0.1433 60 640 0.6259 0.4928 0.5297 0.3298 0.04437 0.05034 0.01804
140/299 13.9G 0.04934 0.06922 0.02455 0.1431 120 640 0.6344 0.4888 0.5301 0.3303 0.04435 0.05033 0.01802
141/299 13.9G 0.04924 0.06931 0.02464 0.1432 35 640 0.6347 0.4891 0.5306 0.3306 0.04433 0.05032 0.01801
142/299 13.9G 0.04906 0.06904 0.02447 0.1426 58 640 0.6469 0.4829 0.5308 0.3312 0.04432 0.05031 0.01799
143/299 13.9G 0.04917 0.06905 0.02435 0.1426 81 640 0.625 0.4941 0.531 0.3315 0.04431 0.05029 0.01798
144/299 13.9G 0.04919 0.06887 0.02427 0.1423 45 640 0.6417 0.4864 0.5316 0.332 0.04429 0.05029 0.01796
145/299 13.9G 0.04917 0.06891 0.0244 0.1425 28 640 0.646 0.4843 0.5314 0.332 0.04428 0.05027 0.01794
146/299 13.9G 0.04912 0.06936 0.02445 0.1429 51 640 0.6232 0.4977 0.5316 0.3322 0.04427 0.05026 0.01794
147/299 13.9G 0.04904 0.06917 0.02426 0.1425 22 640 0.6438 0.4869 0.5318 0.3323 0.04426 0.05025 0.01793
148/299 13.9G 0.04906 0.06916 0.0244 0.1426 39 640 0.6556 0.4817 0.5322 0.3323 0.04425 0.05024 0.01792
149/299 13.9G 0.04902 0.06903 0.02424 0.1423 38 640 0.6462 0.4871 0.5324 0.3323 0.04423 0.05023 0.01791
150/299 13.9G 0.04899 0.06909 0.02422 0.1423 46 640 0.6337 0.494 0.5325 0.3324 0.04422 0.05022 0.01789
151/299 13.9G 0.04903 0.06961 0.02415 0.1428 41 640 0.6375 0.4921 0.5327 0.3325 0.0442 0.05022 0.01788
152/299 13.9G 0.04908 0.06921 0.02427 0.1426 47 640 0.6348 0.4936 0.5329 0.3327 0.0442 0.05021 0.01787
153/299 13.9G 0.04887 0.06909 0.02405 0.142 106 640 0.6533 0.483 0.5331 0.3333 0.04419 0.0502 0.01786
154/299 13.9G 0.04895 0.06891 0.02404 0.1419 65 640 0.6428 0.4896 0.5338 0.3338 0.04417 0.05019 0.01784
155/299 13.9G 0.04889 0.06886 0.02401 0.1418 51 640 0.6619 0.4801 0.5342 0.3343 0.04415 0.05017 0.01783
156/299 13.9G 0.04885 0.06849 0.02394 0.1413 60 640 0.6539 0.4861 0.535 0.3348 0.04414 0.05016 0.01782
157/299 13.9G 0.04871 0.06865 0.02396 0.1413 52 640 0.6521 0.4879 0.5351 0.3352 0.04413 0.05016 0.0178
158/299 13.9G 0.04867 0.06844 0.02382 0.1409 29 640 0.6486 0.4892 0.5349 0.3354 0.04412 0.05015 0.01779
159/299 13.9G 0.04879 0.06946 0.02388 0.1421 50 640 0.6634 0.4824 0.5354 0.3355 0.04411 0.05014 0.01778
160/299 13.9G 0.04865 0.06895 0.02383 0.1414 34 640 0.6586 0.4849 0.5354 0.3357 0.0441 0.05013 0.01777
161/299 13.9G 0.04873 0.06927 0.02371 0.1417 52 640 0.6594 0.485 0.5357 0.3357 0.04409 0.05013 0.01775
162/299 13.9G 0.04861 0.06876 0.02369 0.141 33 640 0.6413 0.4939 0.5354 0.3362 0.04408 0.05013 0.01773
163/299 13.9G 0.04857 0.06862 0.02371 0.1409 13 640 0.6578 0.4855 0.5356 0.3363 0.04406 0.05012 0.01771
164/299 13.9G 0.04859 0.06874 0.02356 0.1409 48 640 0.6417 0.4947 0.5362 0.3366 0.04405 0.05012 0.01769
165/299 13.9G 0.04851 0.06865 0.02355 0.1407 21 640 0.6484 0.4914 0.5369 0.3366 0.04404 0.05011 0.01767
166/299 13.9G 0.04848 0.06861 0.02344 0.1405 46 640 0.6631 0.4844 0.5374 0.3368 0.04403 0.0501 0.01766
167/299 13.9G 0.04848 0.06861 0.02365 0.1407 33 640 0.6623 0.4858 0.5378 0.337 0.04401 0.05009 0.01765
168/299 13.9G 0.04852 0.0687 0.0236 0.1408 29 640 0.6438 0.4965 0.5381 0.3369 0.044 0.05008 0.01764
169/299 13.9G 0.0485 0.06875 0.02348 0.1407 48 640 0.6591 0.4891 0.5382 0.337 0.04399 0.05007 0.01763
170/299 13.9G 0.0484 0.06853 0.02337 0.1403 37 640 0.6628 0.4871 0.5381 0.3373 0.04398 0.05006 0.01762
171/299 13.9G 0.04844 0.06891 0.02338 0.1407 92 640 0.6126 0.5132 0.5384 0.3373 0.04397 0.05005 0.0176
172/299 13.9G 0.04844 0.0687 0.02338 0.1405 24 640 0.6146 0.5128 0.538 0.3373 0.04397 0.05005 0.01758
repvgg-v5s1
这个部分我将只讲C3-Bottleneck的cv1-cv2进行了替换，其他部分没有替换。效果提升的不是很明显。
-
#batchsize=128 v5s 结构重参数化bottleneck #95M 参数量 FLOPs 22.4
epoch, train/box_loss, train/obj_loss, train/cls_loss, metrics/precision, metrics/recall, metrics/mAP_0.5,metrics/mAP_0.5:0.95, val/box_loss, val/obj_loss, val/cls_loss, x/lr0, x/lr1, x/lr2
0, 0.091325, 0.084251, 0.083019, 0.045864, 0.040693, 0.0064491, 0.0018495, 0.082129, 0.062835, 0.077701, 0.0033297, 0.0033297, 0.070032
1, 0.074439, 0.083221, 0.071887, 0.31976, 0.085781, 0.036077, 0.013726, 0.070137, 0.061606, 0.063021, 0.0066629, 0.0066629, 0.040032
2, 0.068349, 0.080858, 0.058965, 0.31089, 0.15308, 0.099369, 0.042348, 0.064392, 0.059422, 0.049325, 0.0099954, 0.0099954, 0.010031
3, 0.063965, 0.078798, 0.04917, 0.32458, 0.22688, 0.18459, 0.088395, 0.059321, 0.057765, 0.040668, 0.0099978, 0.0099978, 0.0099978
4, 0.061199, 0.077382, 0.043838, 0.38715, 0.25856, 0.23561, 0.11888, 0.056412, 0.056409, 0.035993, 0.0099978, 0.0099978, 0.0099978
5, 0.059543, 0.076267, 0.040438, 0.41801, 0.31225, 0.28473, 0.15017, 0.05461, 0.055364, 0.032676, 0.0099961, 0.0099961, 0.0099961
6, 0.058327, 0.07549, 0.038161, 0.46336, 0.31775, 0.31134, 0.16864, 0.053299, 0.0548, 0.030894, 0.0099938, 0.0099938, 0.0099938
7, 0.057373, 0.075205, 0.036591, 0.49252, 0.33666, 0.34447, 0.19211, 0.052136, 0.053898, 0.028695, 0.0099911, 0.0099911, 0.0099911
8, 0.056636, 0.074615, 0.035219, 0.52364, 0.34815, 0.36206, 0.20422, 0.051213, 0.053404, 0.027527, 0.0099879, 0.0099879, 0.0099879
9, 0.05602, 0.073934, 0.034359, 0.50992, 0.38009, 0.38225, 0.21813, 0.05043, 0.052847, 0.026286, 0.0099842, 0.0099842, 0.0099842
10, 0.055668, 0.073962, 0.033563, 0.51525, 0.38478, 0.3973, 0.22897, 0.049851, 0.052506, 0.025292, 0.00998, 0.00998, 0.00998
11, 0.055276, 0.073709, 0.032925, 0.54625, 0.38502, 0.40782, 0.23643, 0.049367, 0.05212, 0.02475, 0.0099753, 0.0099753, 0.0099753
12, 0.054877, 0.07297, 0.032321, 0.55703, 0.39541, 0.41659, 0.24355, 0.048975, 0.051898, 0.024132, 0.0099702, 0.0099702, 0.0099702
13, 0.054596, 0.073085, 0.031765, 0.54956, 0.40851, 0.42707, 0.25115, 0.048565, 0.051643, 0.023694, 0.0099645, 0.0099645, 0.0099645
14, 0.05431, 0.073009, 0.03143, 0.56696, 0.41039, 0.4336, 0.2563, 0.048272, 0.051451, 0.023302, 0.0099584, 0.0099584, 0.0099584
15, 0.05406, 0.072771, 0.030999, 0.56164, 0.41919, 0.43919, 0.26051, 0.048036, 0.051295, 0.023009, 0.0099517, 0.0099517, 0.0099517
16, 0.053834, 0.072276, 0.030678, 0.57196, 0.42093, 0.44328, 0.26414, 0.04783, 0.051172, 0.022729, 0.0099446, 0.0099446, 0.0099446
17, 0.053577, 0.072335, 0.030367, 0.58215, 0.419, 0.44688, 0.2668, 0.047664, 0.051044, 0.022535, 0.009937, 0.009937, 0.009937
18, 0.053448, 0.072006, 0.030096, 0.575, 0.42522, 0.44958, 0.26966, 0.047528, 0.050944, 0.022371, 0.0099289, 0.0099289, 0.0099289
19, 0.053259, 0.071957, 0.029836, 0.5959, 0.41816, 0.45258, 0.27204, 0.047402, 0.050845, 0.022239, 0.0099203, 0.0099203, 0.0099203
20, 0.053077, 0.071917, 0.029526, 0.60879, 0.41392, 0.45441, 0.27368, 0.047301, 0.050761, 0.02214, 0.0099112, 0.0099112, 0.0099112
21, 0.052961, 0.071599, 0.029387, 0.56503, 0.43677, 0.45631, 0.27456, 0.04722, 0.05069, 0.022046, 0.0099017, 0.0099017, 0.0099017
22, 0.052864, 0.071617, 0.029193, 0.58445, 0.42799, 0.45746, 0.2759, 0.047137, 0.050628, 0.021959, 0.0098916, 0.0098916, 0.0098916
23, 0.052629, 0.071329, 0.028944, 0.59464, 0.42518, 0.45866, 0.277, 0.047065, 0.050569, 0.021889, 0.0098811, 0.0098811, 0.0098811
24, 0.052497, 0.071371, 0.028764, 0.57254, 0.43689, 0.45942, 0.278, 0.047001, 0.050518, 0.021832, 0.0098701, 0.0098701, 0.0098701
25, 0.052393, 0.071428, 0.028541, 0.56765, 0.44037, 0.46032, 0.2787, 0.046941, 0.050474, 0.021784, 0.0098586, 0.0098586, 0.0098586
26, 0.052246, 0.071214, 0.028439, 0.5729, 0.43879, 0.46132, 0.27977, 0.046887, 0.050435, 0.02172, 0.0098467, 0.0098467, 0.0098467
27, 0.052175, 0.07089, 0.028303, 0.59449, 0.42917, 0.46239, 0.28053, 0.046835, 0.050403, 0.021667, 0.0098342, 0.0098342, 0.0098342
28, 0.052075, 0.070701, 0.028244, 0.60295, 0.42625, 0.46385, 0.28191, 0.046789, 0.050377, 0.021617, 0.0098213, 0.0098213, 0.0098213
29, 0.051997, 0.071067, 0.028108, 0.58521, 0.43614, 0.46475, 0.28292, 0.046745, 0.050361, 0.02157, 0.0098079, 0.0098079, 0.0098079
30, 0.051922, 0.071, 0.027901, 0.61236, 0.42383, 0.46511, 0.28362, 0.046701, 0.050348, 0.021524, 0.0097941, 0.0097941, 0.0097941
31, 0.051872, 0.070737, 0.027774, 0.62071, 0.42165, 0.46607, 0.28453, 0.046654, 0.05034, 0.021475, 0.0097798, 0.0097798, 0.0097798
32, 0.05168, 0.070582, 0.027777, 0.58882, 0.43789, 0.46703, 0.28551, 0.046607, 0.050339, 0.02142, 0.009765, 0.009765, 0.009765
33, 0.051612, 0.070762, 0.027565, 0.59094, 0.43816, 0.46795, 0.28637, 0.046563, 0.050342, 0.02138, 0.0097497, 0.0097497, 0.0097497
34, 0.051544, 0.070445, 0.027422, 0.58, 0.44565, 0.46896, 0.28706, 0.046517, 0.050339, 0.021334, 0.009734, 0.009734, 0.009734
35, 0.051499, 0.070687, 0.027361, 0.60279, 0.43692, 0.46993, 0.28788, 0.046476, 0.050338, 0.02129, 0.0097178, 0.0097178, 0.0097178
36, 0.051421, 0.070385, 0.027363, 0.57595, 0.45075, 0.47137, 0.28923, 0.046433, 0.050337, 0.021241, 0.0097011, 0.0097011, 0.0097011
37, 0.051288, 0.070366, 0.027199, 0.57659, 0.4512, 0.47223, 0.28996, 0.046385, 0.050332, 0.021189, 0.009684, 0.009684, 0.009684
38, 0.051249, 0.070556, 0.027186, 0.58312, 0.44984, 0.47376, 0.29134, 0.046333, 0.050325, 0.021127, 0.0096664, 0.0096664, 0.0096664
39, 0.051224, 0.070369, 0.026999, 0.56718, 0.46055, 0.47543, 0.29283, 0.046283, 0.050315, 0.02107, 0.0096484, 0.0096484, 0.0096484
40, 0.051065, 0.07025, 0.026815, 0.57623, 0.45931, 0.47704, 0.29384, 0.04624, 0.050295, 0.021016, 0.0096299, 0.0096299, 0.0096299
repvgg-v5s2-not-fuse
这个部分是只将卷积大小为3x3，并且stride=1的进行替换，对于其他大小的卷积或者下采样的Conv没有替换。
- Model Summary: 529 layers, 18323901 parameters, 18323901 gradients, 22.1 GFLOPs bs=256
epoch, train/box_loss, train/obj_loss, train/cls_loss, metrics/precision, metrics/recall, metrics/mAP_0.5,metrics/mAP_0.5:0.95, val/box_loss, val/obj_loss, val/cls_loss, x/lr0, x/lr1, x/lr2
0, 0.092004, 0.084015, 0.082964, 0.029228, 0.040308, 0.0036937, 0.0011781, 0.08216, 0.064867, 0.077113, 0.0033297, 0.0033297, 0.070032
1, 0.074385, 0.083075, 0.071424, 0.36081, 0.083457, 0.032672, 0.010698, 0.070906, 0.063654, 0.062175, 0.0066629, 0.0066629, 0.040032
2, 0.068213, 0.080396, 0.058372, 0.33255, 0.16075, 0.099448, 0.042357, 0.064713, 0.060401, 0.051163, 0.0099954, 0.0099954, 0.010031
3, 0.063901, 0.079088, 0.048938, 0.37051, 0.21196, 0.17536, 0.082413, 0.059104, 0.057996, 0.04138, 0.0099978, 0.0099978, 0.0099978
4, 0.06108, 0.077382, 0.043561, 0.40702, 0.26596, 0.23681, 0.12011, 0.056339, 0.056411, 0.035908, 0.0099978, 0.0099978, 0.0099978
5, 0.059405, 0.076233, 0.040388, 0.41548, 0.3136, 0.28489, 0.15177, 0.054522, 0.055313, 0.032694, 0.0099961, 0.0099961, 0.0099961
6, 0.058182, 0.075428, 0.038158, 0.4367, 0.33047, 0.3094, 0.1678, 0.05316, 0.054771, 0.030924, 0.0099938, 0.0099938, 0.0099938
7, 0.05735, 0.075031, 0.036529, 0.51722, 0.33262, 0.3434, 0.18961, 0.052, 0.053987, 0.028712, 0.0099911, 0.0099911, 0.0099911
8, 0.056641, 0.07458, 0.035336, 0.53165, 0.34665, 0.36402, 0.20534, 0.051109, 0.05336, 0.027521, 0.0099879, 0.0099879, 0.0099879
9, 0.05608, 0.074169, 0.034309, 0.51649, 0.3766, 0.38203, 0.21764, 0.050346, 0.052864, 0.026372, 0.0099842, 0.0099842, 0.0099842
10, 0.055596, 0.073769, 0.033533, 0.52767, 0.39115, 0.39924, 0.22918, 0.049731, 0.052498, 0.02528, 0.00998, 0.00998, 0.00998
11, 0.055214, 0.073445, 0.032875, 0.51122, 0.40731, 0.40802, 0.23782, 0.049252, 0.052116, 0.024745, 0.0099753, 0.0099753, 0.0099753
12, 0.054828, 0.073024, 0.032236, 0.545, 0.3989, 0.41737, 0.24483, 0.048801, 0.051849, 0.024273, 0.0099702, 0.0099702, 0.0099702
13, 0.054548, 0.073003, 0.031797, 0.53235, 0.41676, 0.42498, 0.25128, 0.048508, 0.051585, 0.023823, 0.0099645, 0.0099645, 0.0099645
14, 0.054265, 0.072806, 0.031344, 0.54557, 0.41953, 0.43157, 0.25618, 0.048195, 0.051413, 0.023407, 0.0099584, 0.0099584, 0.0099584
15, 0.054041, 0.07252, 0.030969, 0.56267, 0.4155, 0.43702, 0.26028, 0.047964, 0.051246, 0.023063, 0.0099517, 0.0099517, 0.0099517
repvgg-v5s2-fuse
在上次 repvgg-v5s2修改中，只修改了kernel_size=3，stride=1的卷积，对于其他大小的卷积或者stride!=1的卷积，没有进行fuse_conv_bn操作，而在这次实验中则完善了这一操作！这个BottleNeck里面的cv1,cv2的kernel_size都是3，但是yolov5s里面的是k1=1与k2=3，所以这个需要变一下
#experimental.py
# detect.py加载pt文件时，调用了yolo.py中的fuse操作（下面），对conv与bn进行了fuse
def attempt_load(weights, map_location=None, inplace=True, fuse=True):
from models.yolo import Detect, Model
# Loads an ensemble of models weights=[a,b,c] or a single model weights=[a] or weights=a
model = Ensemble()
for w in weights if isinstance(weights, list) else [weights]:
ckpt = torch.load(attempt_download(w), map_location=map_location) # load
if fuse:
model.append(ckpt['ema' if ckpt.get('ema') else 'model'].float().fuse().eval()) # FP32 model
else:
model.append(ckpt['ema' if ckpt.get('ema') else 'model'].float().eval()) # without layer fuse
#yolo.py
def fuse(self): # fuse model Conv2d() + BatchNorm2d() layers
LOGGER.info('Fusing layers... ')
for m in self.model.modules():
if isinstance(m, (Conv, DWConv)) and hasattr(m, 'bn'):
m.brb_rep = fuse_conv_and_bn(m.brb_rep, m.bn) # update conv
delattr(m, 'bn') # remove batchnorm
m.forward = m.forward_fuse # update forward
self.info()
return self
#torch_utils.py
#这个可以根据自己的需要进行修改，比如conv.in_channels修改成conv.conv.in_channels，具体要看自己的网络模块是什么样的。
def fuse_conv_and_bn(conv, bn):
# Fuse convolution and batchnorm layers https://tehnokv.com/posts/fusing-batchnorm-and-conv/
fusedconv = nn.Conv2d(conv.in_channels,
conv.out_channels,
kernel_size=conv.kernel_size,
stride=conv.stride,
padding=conv.padding,
groups=conv.groups,
bias=True).requires_grad_(False).to(conv.weight.device)
# prepare filters
w_conv = conv.weight.clone().view(conv.out_channels, -1)
w_bn = torch.diag(bn.weight.div(torch.sqrt(bn.eps + bn.running_var)))
fusedconv.weight.copy_(torch.mm(w_bn, w_conv).view(fusedconv.weight.shape))
# prepare spatial bias
b_conv = torch.zeros(conv.weight.size(0), device=conv.weight.device) if conv.bias is None else conv.bias
b_bn = bn.bias - bn.weight.mul(bn.running_mean).div(torch.sqrt(bn.running_var + bn.eps))
fusedconv.bias.copy_(torch.mm(w_bn, b_conv.reshape(-1, 1)).reshape(-1) + b_bn)
return fusedconv
class Conv(nn.Module):
def __init__(self, c1, c2, k=3, s=1, p=None, g=1, act=True,deploy=False): # ch_in, ch_out, kernel, stride, padding, groups
super().__init__()
......
def forward(self, inputs):
#如果不进行_switch_to_deploy,下面两个return就足够了
if (self.deploy):#Conv._switch_to_deploy 会改变deploy
if hasattr(self,'bn'):#说明不是结构重参数化
return self.act(self.bn(self.brb_rep(inputs)))
else:#说明是结构重参数化，没有bn操作
return self.act(self.brb_rep(inputs))
#下面是为了不进行_switch_to_deploy，进行的forward操作
if (self.brb_identity == None):
identity_out = 0
else:
identity_out = self.brb_identity(inputs)
return self.act(self.brb_1x1(inputs) + self.brb_3x3(inputs) + identity_out)
def forward_fuse(self, inputs):
if (self.deploy):#Conv._switch_to_deploy 会改变deploy
# print(self.brb_rep(torch.randn(1,3,4,4)).shape)
# print('self.brb_rep.kernel_size',self.brb_rep.kernel_size)
return self.act(self.brb_rep(inputs))
if (self.brb_identity == None):
identity_out = 0
else:
identity_out = self.brb_identity(inputs)
return self.act(self.brb_1x1(inputs) + self.brb_3x3(inputs) + identity_out)
记录
epoch, train/box_loss, train/obj_loss, train/cls_loss, metrics/precision, metrics/recall, metrics/mAP_0.5,metrics/mAP_0.5:0.95, val/box_loss, val/obj_loss, val/cls_loss, x/lr0, x/lr1, x/lr2
0, 0.091684, 0.083873, 0.083003, 0.11889, 0.036322, 0.0047963, 0.0013873, 0.080382, 0.062765, 0.077113, 0.0033297, 0.0033297, 0.070032
1, 0.074254, 0.08325, 0.071897, 0.28534, 0.1069, 0.038758, 0.014097, 0.069871, 0.060908, 0.063017, 0.0066629, 0.0066629, 0.040032
2, 0.068055, 0.080653, 0.058831, 0.30226, 0.15678, 0.097203, 0.041169, 0.064179, 0.060451, 0.049972, 0.0099954, 0.0099954, 0.010031
3, 0.063864, 0.078561, 0.049357, 0.37208, 0.20302, 0.17114, 0.081694, 0.059796, 0.057443, 0.041513, 0.0099978, 0.0099978, 0.0099978
4, 0.061138, 0.077359, 0.04392, 0.34799, 0.27356, 0.23034, 0.11579, 0.056402, 0.056551, 0.036675, 0.0099978, 0.0099978, 0.0099978
5, 0.059469, 0.076227, 0.040648, 0.45373, 0.2866, 0.27911, 0.14818, 0.054577, 0.055503, 0.033043, 0.0099961, 0.0099961, 0.0099961
6, 0.058296, 0.075598, 0.038351, 0.45885, 0.32383, 0.31248, 0.17017, 0.053094, 0.054646, 0.030785, 0.0099938, 0.0099938, 0.0099938
7, 0.057482, 0.075143, 0.036739, 0.48984, 0.34399, 0.34029, 0.18835, 0.052042, 0.053939, 0.028925, 0.0099911, 0.0099911, 0.0099911
8, 0.056705, 0.074649, 0.035441, 0.4915, 0.3654, 0.36197, 0.20541, 0.051114, 0.05332, 0.027579, 0.0099879, 0.0099879, 0.0099879
9, 0.056144, 0.074082, 0.034403, 0.52628, 0.37434, 0.38172, 0.21816, 0.050361, 0.052842, 0.026346, 0.0099842, 0.0099842, 0.0099842
10, 0.055666, 0.073836, 0.033553, 0.51832, 0.39206, 0.39602, 0.2287, 0.049761, 0.052426, 0.025471, 0.00998, 0.00998, 0.00998
11, 0.055231, 0.073437, 0.032997, 0.56388, 0.37978, 0.40593, 0.23718, 0.04923, 0.0521, 0.024748, 0.0099753, 0.0099753, 0.0099753
12, 0.054889, 0.073394, 0.032308, 0.57626, 0.38826, 0.41714, 0.24481, 0.048803, 0.051869, 0.024245, 0.0099702, 0.0099702, 0.0099702
13, 0.054554, 0.072821, 0.031894, 0.53876, 0.41098, 0.42532, 0.25248, 0.048478, 0.051596, 0.023739, 0.0099645, 0.0099645, 0.0099645
14, 0.054293, 0.072961, 0.031428, 0.55517, 0.41513, 0.43209, 0.25723, 0.048174, 0.051402, 0.02335, 0.0099584, 0.0099584, 0.0099584
15, 0.054051, 0.072379, 0.03106, 0.5711, 0.4148, 0.43923, 0.26223, 0.047915, 0.051225, 0.023025, 0.0099517, 0.0099517, 0.0099517
16, 0.053788, 0.072349, 0.030638, 0.56422, 0.42605, 0.44355, 0.26615, 0.047721, 0.051081, 0.022759, 0.0099446, 0.0099446, 0.0099446
17, 0.053498, 0.072054, 0.030305, 0.58218, 0.4219, 0.44725, 0.26785, 0.047562, 0.050984, 0.022552, 0.009937, 0.009937, 0.009937
18, 0.053371, 0.071944, 0.030003, 0.57388, 0.42997, 0.45016, 0.27031, 0.04744, 0.050874, 0.022405, 0.0099289, 0.0099289, 0.0099289
19, 0.053229, 0.071684, 0.029826, 0.58115, 0.42835, 0.4526, 0.27211, 0.047332, 0.050791, 0.02227, 0.0099203, 0.0099203, 0.0099203
20, 0.053079, 0.071805, 0.029529, 0.58502, 0.42681, 0.45375, 0.27357, 0.047245, 0.050707, 0.022169, 0.0099112, 0.0099112, 0.0099112
21, 0.052902, 0.071675, 0.029337, 0.58125, 0.43068, 0.45569, 0.27526, 0.047169, 0.050641, 0.022076, 0.0099017, 0.0099017, 0.0099017
22, 0.052779, 0.071798, 0.029114, 0.57623, 0.43864, 0.45683, 0.27636, 0.047096, 0.050588, 0.021995, 0.0098916, 0.0098916, 0.0098916
23, 0.052688, 0.07148, 0.028993, 0.5592, 0.44757, 0.45836, 0.27782, 0.047032, 0.050533, 0.021918, 0.0098811, 0.0098811, 0.0098811
24, 0.052492, 0.071561, 0.028844, 0.56807, 0.4431, 0.45924, 0.27891, 0.046975, 0.050497, 0.021851, 0.0098701, 0.0098701, 0.0098701
25, 0.052359, 0.071399, 0.028671, 0.5645, 0.44534, 0.46058, 0.28024, 0.046923, 0.05046, 0.021784, 0.0098586, 0.0098586, 0.0098586
26, 0.052267, 0.071078, 0.028537, 0.56427, 0.44617, 0.46203, 0.28136, 0.046877, 0.05043, 0.021728, 0.0098467, 0.0098467, 0.0098467
27, 0.052167, 0.071205, 0.02834, 0.56993, 0.44432, 0.463, 0.28212, 0.046831, 0.050411, 0.021681, 0.0098342, 0.0098342, 0.0098342
28, 0.052017, 0.070836, 0.028275, 0.58049, 0.44164, 0.46401, 0.28307, 0.046781, 0.050408, 0.021631, 0.0098213, 0.0098213, 0.0098213
29, 0.051979, 0.070958, 0.028054, 0.57612, 0.44456, 0.46563, 0.28385, 0.046736, 0.050405, 0.021588, 0.0098079, 0.0098079, 0.0098079
30, 0.051819, 0.070996, 0.028032, 0.58498, 0.4407, 0.46687, 0.28457, 0.046697, 0.050407, 0.021544, 0.0097941, 0.0097941, 0.0097941
31, 0.051814, 0.070874, 0.027923, 0.59036, 0.43892, 0.46802, 0.28548, 0.046656, 0.050411, 0.021493, 0.0097798, 0.0097798, 0.0097798
32, 0.051686, 0.070831, 0.027725, 0.57675, 0.44805, 0.46862, 0.28645, 0.046615, 0.050426, 0.021445, 0.009765, 0.009765, 0.009765
33, 0.051667, 0.070634, 0.0277, 0.58037, 0.44705, 0.46963, 0.28724, 0.046574, 0.050447, 0.021389, 0.0097497, 0.0097497, 0.0097497
34, 0.051521, 0.070458, 0.027497, 0.58719, 0.44374, 0.47044, 0.28792, 0.046535, 0.050461, 0.021332, 0.009734, 0.009734, 0.009734
35, 0.051506, 0.070662, 0.027441, 0.5885, 0.44439, 0.47186, 0.28942, 0.046491, 0.050471, 0.021269, 0.0097178, 0.0097178, 0.0097178
36, 0.051418, 0.07035, 0.027413, 0.59529, 0.44237, 0.47313, 0.29034, 0.04645, 0.050473, 0.021213, 0.0097011, 0.0097011, 0.0097011
37, 0.051342, 0.07035, 0.027347, 0.59307, 0.4456, 0.47436, 0.2915, 0.046409, 0.050474, 0.021146, 0.009684, 0.009684, 0.009684
38, 0.051249, 0.070294, 0.027215, 0.60617, 0.44072, 0.47558, 0.29213, 0.046363, 0.050472, 0.02108, 0.0096664, 0.0096664, 0.0096664
bs128-v5m-only3-repvgg-失败
epoch, train/box_loss, train/obj_loss, train/cls_loss, metrics/precision, metrics/recall, metrics/mAP_0.5,metrics/mAP_0.5:0.95, val/box_loss, val/obj_loss, val/cls_loss, x/lr0, x/lr1, x/lr2
0, 0.091429, 0.083921, 0.083333, 0.18123, 0.037406, 0.0054993, 0.0017863, 0.080575, 0.065993, 0.077819, 0.0033297, 0.0033297, 0.070032
1, 0.07315, 0.083372, 0.072354, 0.34694, 0.080206, 0.043477, 0.016405, 0.068264, 0.064052, 0.060966, 0.0066629, 0.0066629, 0.040032
2, 0.066702, 0.081285, 0.059143, 0.35, 0.15084, 0.10672, 0.04587, 0.062273, 0.061312, 0.048442, 0.0099954, 0.0099954, 0.010031
3, 0.062169, 0.079481, 0.049128, 0.30166, 0.24924, 0.1853, 0.091003, 0.057319, 0.059455, 0.039904, 0.0099978, 0.0099978, 0.0099978
4, 0.059529, 0.076886, 0.043603, 0.40681, 0.28908, 0.25994, 0.13575, 0.054467, 0.057602, 0.035494, 0.0099978, 0.0099978, 0.0099978
5, 0.057298, 0.076357, 0.040525, 0.48307, 0.30996, 0.30937, 0.16804, 0.052252, 0.056752, 0.031491, 0.0099961, 0.0099961, 0.0099961
6, 0.056366, 0.075171, 0.038025, 0.47274, 0.35327, 0.34361, 0.19272, 0.050701, 0.055754, 0.02923, 0.0099938, 0.0099938, 0.0099938
7, 0.055512, 0.075627, 0.036506, 0.51957, 0.36741, 0.37435, 0.21306, 0.049501, 0.054915, 0.027218, 0.0099911, 0.0099911, 0.0099911
8, 0.054382, 0.072708, 0.034843, 0.54941, 0.37464, 0.39539, 0.22905, 0.048637, 0.054252, 0.025949, 0.0099879, 0.0099879, 0.0099879
9, 0.054007, 0.073536, 0.033998, 0.53968, 0.39345, 0.41214, 0.24421, 0.047677, 0.05369, 0.024703, 0.0099842, 0.0099842, 0.0099842
10, 0.053224, 0.072996, 0.03322, 0.53395, 0.41462, 0.4247, 0.2539, 0.047103, 0.053297, 0.023964, 0.00998, 0.00998, 0.00998
11, 0.053371, 0.07458, 0.032661, 0.54502, 0.42801, 0.43888, 0.26576, 0.046506, 0.052943, 0.023176, 0.0099753, 0.0099753, 0.0099753
12, 0.052681, 0.072668, 0.032079, 0.57409, 0.42684, 0.44958, 0.27447, 0.046053, 0.052622, 0.022701, 0.0099702, 0.0099702, 0.0099702
13, 0.052236, 0.071905, 0.031348, 0.56773, 0.43854, 0.45804, 0.28024, 0.045605, 0.052407, 0.022196, 0.0099645, 0.0099645, 0.0099645
14, 0.052264, 0.072108, 0.030936, 0.57746, 0.4412, 0.46565, 0.28636, 0.045278, 0.052168, 0.02181, 0.0099584, 0.0099584, 0.0099584
15, 0.051566, 0.071671, 0.0305, 0.57607, 0.45039, 0.47213, 0.29103, 0.045031, 0.05199, 0.02146, 0.0099517, 0.0099517, 0.0099517
16, 0.051606, 0.071175, 0.030377, 0.58814, 0.44854, 0.47788, 0.29613, 0.044854, 0.051837, 0.021206, 0.0099446, 0.0099446, 0.0099446
17, 0.051317, 0.071775, 0.029695, 0.59891, 0.44795, 0.48104, 0.29932, 0.044679, 0.051704, 0.021012, 0.009937, 0.009937, 0.009937
v5m
epoch, train/box_loss, train/obj_loss, train/cls_loss, metrics/precision, metrics/recall, metrics/mAP_0.5,metrics/mAP_0.5:0.95, val/box_loss, val/obj_loss, val/cls_loss, x/lr0, x/lr1, x/lr2
0, 0.088815, 0.083758, 0.083028, 0.10967, 0.060776, 0.0084766, 0.0028998, 0.07658, 0.065096, 0.075192, 0.0033297, 0.0033297, 0.070032
1, 0.07153, 0.082675, 0.069671, 0.25921, 0.12556, 0.050926, 0.021388, 0.067456, 0.063615, 0.058973, 0.0066629, 0.0066629, 0.040032
2, 0.065678, 0.079942, 0.056423, 0.28735, 0.19372, 0.13283, 0.059436, 0.060568, 0.061682, 0.046271, 0.0099954, 0.0099954, 0.010031
3, 0.061338, 0.078504, 0.04723, 0.40063, 0.24958, 0.22877, 0.11608, 0.055906, 0.058371, 0.037091, 0.0099978, 0.0099978, 0.0099978
4, 0.058455, 0.076397, 0.042069, 0.42689, 0.30947, 0.28819, 0.15312, 0.053267, 0.057035, 0.032819, 0.0099978, 0.0099978, 0.0099978
5, 0.056665, 0.075084, 0.038737, 0.48373, 0.33178, 0.32958, 0.18127, 0.051449, 0.05598, 0.03003, 0.0099961, 0.0099961, 0.0099961
6, 0.055591, 0.075298, 0.036506, 0.49513, 0.36799, 0.36491, 0.2067, 0.049934, 0.054871, 0.027925, 0.0099938, 0.0099938, 0.0099938
7, 0.054502, 0.073199, 0.035426, 0.49643, 0.39808, 0.38942, 0.22583, 0.048767, 0.054358, 0.026238, 0.0099911, 0.0099911, 0.0099911
8, 0.053876, 0.07238, 0.033299, 0.52135, 0.41381, 0.41562, 0.24501, 0.04781, 0.053518, 0.024843, 0.0099879, 0.0099879, 0.0099879
9, 0.053236, 0.072389, 0.032481, 0.54612, 0.41269, 0.43074, 0.25796, 0.046989, 0.053086, 0.023894, 0.0099842, 0.0099842, 0.0099842
10, 0.052832, 0.07241, 0.031733, 0.55228, 0.43234, 0.44671, 0.26953, 0.046382, 0.052618, 0.023047, 0.00998, 0.00998, 0.00998
11, 0.05282, 0.074112, 0.031097, 0.5533, 0.44669, 0.46024, 0.28065, 0.045857, 0.052293, 0.022248, 0.0099753, 0.0099753, 0.0099753
12, 0.052155, 0.071912, 0.030904, 0.55584, 0.45631, 0.46781, 0.28701, 0.045427, 0.051986, 0.021778, 0.0099702, 0.0099702, 0.0099702
13, 0.051747, 0.072218, 0.03014, 0.60159, 0.44424, 0.47684, 0.29523, 0.045071, 0.051834, 0.021251, 0.0099645, 0.0099645, 0.0099645
14, 0.051199, 0.071139, 0.029753, 0.60888, 0.44745, 0.4834, 0.30013, 0.044758, 0.051615, 0.020943, 0.0099584, 0.0099584, 0.0099584
15, 0.051116, 0.071044, 0.029682, 0.61669, 0.45095, 0.48788, 0.30411, 0.044482, 0.051461, 0.020708, 0.0099517, 0.0099517, 0.0099517
16, 0.051165, 0.071407, 0.029131, 0.58867, 0.46949, 0.49298, 0.30788, 0.044266, 0.051328, 0.020452, 0.0099446, 0.0099446, 0.0099446
17, 0.050888, 0.070982, 0.029018, 0.60247, 0.46829, 0.49698, 0.3109, 0.044097, 0.051225, 0.020252, 0.009937, 0.009937, 0.009937
18, 0.050531, 0.070728, 0.028839, 0.62038, 0.46144, 0.49886, 0.31304, 0.04396, 0.051144, 0.020106, 0.0099289, 0.0099289, 0.0099289
19, 0.050503, 0.070549, 0.028389, 0.61684, 0.4655, 0.50101, 0.31475, 0.043849, 0.051041, 0.019979, 0.0099203, 0.0099203, 0.0099203
20, 0.050645, 0.070725, 0.028245, 0.60357, 0.47478, 0.5033, 0.31661, 0.043755, 0.050947, 0.019874, 0.0099112, 0.0099112, 0.0099112
21, 0.050031, 0.069934, 0.027804, 0.60496, 0.47581, 0.50527, 0.31852, 0.043667, 0.050866, 0.01977, 0.0099017, 0.0099017, 0.0099017
22, 0.049924, 0.069962, 0.027742, 0.61439, 0.47212, 0.50718, 0.32007, 0.043592, 0.050781, 0.019677, 0.0098916, 0.0098916, 0.0098916
23, 0.049774, 0.070259, 0.027356, 0.62347, 0.4707, 0.50993, 0.32225, 0.043526, 0.050693, 0.019594, 0.0098811, 0.0098811, 0.0098811
24, 0.049478, 0.06908, 0.0275, 0.59173, 0.49002, 0.51115, 0.32364, 0.043463, 0.050619, 0.019518, 0.0098701, 0.0098701, 0.0098701
25, 0.049521, 0.070211, 0.027425, 0.59696, 0.4898, 0.51278, 0.32486, 0.043407, 0.050557, 0.019451, 0.0098586, 0.0098586, 0.0098586
26, 0.049435, 0.07008, 0.027264, 0.61438, 0.48112, 0.51381, 0.32583, 0.043351, 0.050498, 0.019382, 0.0098467, 0.0098467, 0.0098467
27, 0.049354, 0.070545, 0.026766, 0.6064, 0.48604, 0.51499, 0.32726, 0.043301, 0.050441, 0.019321, 0.0098342, 0.0098342, 0.0098342
28, 0.049133, 0.069629, 0.02687, 0.63298, 0.47289, 0.51647, 0.32861, 0.043251, 0.050395, 0.019262, 0.0098213, 0.0098213, 0.0098213
29, 0.049244, 0.070239, 0.026565, 0.61425, 0.48472, 0.51749, 0.32981, 0.043202, 0.050362, 0.0192, 0.0098079, 0.0098079, 0.0098079
30, 0.049247, 0.069914, 0.026879, 0.62765, 0.47932, 0.51836, 0.33032, 0.043154, 0.050333, 0.019143, 0.0097941, 0.0097941, 0.0097941
31, 0.048997, 0.070243, 0.026534, 0.64588, 0.47083, 0.51933, 0.33128, 0.04311, 0.05031, 0.019085, 0.0097798, 0.0097798, 0.0097798
32, 0.048746, 0.06989, 0.026604, 0.62604, 0.48261, 0.52032, 0.3324, 0.043065, 0.050294, 0.019024, 0.009765, 0.009765, 0.009765
33, 0.048844, 0.069789, 0.026362, 0.60369, 0.49778, 0.52143, 0.3331, 0.04302, 0.050284, 0.018967, 0.0097497, 0.0097497, 0.0097497
34, 0.048838, 0.06909, 0.026239, 0.6214, 0.48986, 0.52276, 0.33407, 0.042976, 0.050265, 0.018912, 0.009734, 0.009734, 0.009734
35, 0.048356, 0.069284, 0.025822, 0.61233, 0.49574, 0.52344, 0.33489, 0.042929, 0.050244, 0.018861, 0.0097178, 0.0097178, 0.0097178
36, 0.048366, 0.068073, 0.025951, 0.61853, 0.49364, 0.5246, 0.3363, 0.042877, 0.050227, 0.018803, 0.0097011, 0.0097011, 0.0097011
37, 0.048487, 0.069332, 0.025808, 0.63126, 0.48776, 0.52577, 0.33738, 0.042821, 0.050208, 0.01875, 0.009684, 0.009684, 0.009684
38, 0.048189, 0.069316, 0.025894, 0.64172, 0.48407, 0.52724, 0.33821, 0.042764, 0.050177, 0.018681, 0.0096664, 0.0096664, 0.0096664
39, 0.048339, 0.069556, 0.025417, 0.65185, 0.48204, 0.52911, 0.33938, 0.042711, 0.050147, 0.018621, 0.0096484, 0.0096484, 0.0096484
40, 0.048066, 0.068588, 0.025469, 0.65607, 0.4814, 0.53022, 0.34057, 0.042654, 0.050112, 0.018563, 0.0096299, 0.0096299, 0.0096299
41, 0.048106, 0.069846, 0.025577, 0.66145, 0.48068, 0.53182, 0.34192, 0.042593, 0.050081, 0.018496, 0.009611, 0.009611, 0.009611
42, 0.04817, 0.069178, 0.025513, 0.65954, 0.48355, 0.53306, 0.34304, 0.042531, 0.050039, 0.018439, 0.0095916, 0.0095916, 0.0095916
43, 0.047838, 0.068633, 0.0254, 0.63862, 0.49544, 0.5345, 0.34428, 0.042466, 0.049994, 0.01837, 0.0095717, 0.0095717, 0.0095717
44, 0.047805, 0.068237, 0.025414, 0.62767, 0.50361, 0.53601, 0.34541, 0.0424, 0.049956, 0.018304, 0.0095514, 0.0095514, 0.0095514
45, 0.047678, 0.068207, 0.025349, 0.63458, 0.5013, 0.53745, 0.34681, 0.042332, 0.04991, 0.018235, 0.0095307, 0.0095307, 0.0095307
46, 0.047662, 0.068054, 0.02531, 0.62932, 0.50579, 0.53886, 0.34872, 0.042266, 0.049872, 0.018165, 0.0095095, 0.0095095, 0.0095095
47, 0.048001, 0.069291, 0.024793, 0.63726, 0.50436, 0.54061, 0.35014, 0.0422, 0.049825, 0.0181, 0.0094879, 0.0094879, 0.0094879
48, 0.047774, 0.068069, 0.025169, 0.65425, 0.4975, 0.54264, 0.3516, 0.042133, 0.049773, 0.018031, 0.0094659, 0.0094659, 0.0094659
49, 0.047674, 0.06858, 0.024896, 0.64272, 0.50575, 0.54415, 0.35326, 0.04207, 0.049724, 0.017963, 0.0094434, 0.0094434, 0.0094434
50, 0.047703, 0.068074, 0.025099, 0.64057, 0.50878, 0.54578, 0.35462, 0.042001, 0.04967, 0.017888, 0.0094205, 0.0094205, 0.0094205
51, 0.047624, 0.068676, 0.025119, 0.65135, 0.50244, 0.54779, 0.35586, 0.041937, 0.049611, 0.017818, 0.0093971, 0.0093971, 0.0093971
52, 0.047448, 0.068091, 0.024714, 0.62977, 0.51681, 0.54856, 0.35716, 0.041875, 0.049551, 0.017752, 0.0093733, 0.0093733, 0.0093733
53, 0.047625, 0.068465, 0.024857, 0.64038, 0.51344, 0.55018, 0.3584, 0.041815, 0.049497, 0.017687, 0.0093491, 0.0093491, 0.0093491
54, 0.047647, 0.067882, 0.02476, 0.64501, 0.51334, 0.55105, 0.35947, 0.041749, 0.049447, 0.017623, 0.0093245, 0.0093245, 0.0093245
55, 0.047092, 0.068041, 0.024622, 0.64294, 0.51586, 0.55202, 0.36056, 0.041692, 0.049388, 0.017557, 0.0092995, 0.0092995, 0.0092995
56, 0.047486, 0.069097, 0.024603, 0.64305, 0.51754, 0.55371, 0.36145, 0.041635, 0.049338, 0.017489, 0.009274, 0.009274, 0.009274
57, 0.047281, 0.067822, 0.024274, 0.65651, 0.51091, 0.55552, 0.36304, 0.041588, 0.049283, 0.017427, 0.0092481, 0.0092481, 0.0092481
58, 0.047146, 0.068453, 0.024524, 0.64516, 0.51803, 0.55695, 0.3643, 0.041535, 0.04923, 0.017362, 0.0092219, 0.0092219, 0.0092219
1万+
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
