diff --git a/.gitignore b/.gitignore index ee6b6a8..7d9125b 100644 --- a/.gitignore +++ b/.gitignore @@ -1,5 +1,4 @@ -# 创建.gitignore文件排除不需要的文件 -cat > .gitignore << EOF +# Python缓存文件 __pycache__/ *.py[cod] *$py.class @@ -7,17 +6,23 @@ __pycache__/ .env venv/ ENV/ + +# 编辑器配置 .vscode/ .idea/ + +# 日志和模型文件 *.log *.pt *.pth +logs/ + +# 训练数据（避免上传大量图像） data/training/images/* data/training/labels/* !data/training/images/.gitkeep !data/training/labels/.gitkeep -EOF -# 确保创建空目录占位符 -mkdir -p data/training/images data/training/labels -touch data/training/images/.gitkeep data/training/labels/.gitkeep \ No newline at end of file +# 系统文件 +.DS_Store +Thumbs.db \ No newline at end of file diff --git a/config/encryption.key b/config/encryption.key new file mode 100644 index 0000000..11741a1 --- /dev/null +++ b/config/encryption.key @@ -0,0 +1 @@ +W5q7WV5ITm9C3UfY1w8FMHgHk6Mv5PysDo0D2r2hXSE= \ No newline at end of file diff --git a/data/training/data.yaml b/data/training/data.yaml index 5de57bc..843b7d9 100644 --- a/data/training/data.yaml +++ b/data/training/data.yaml @@ -1,7 +1,7 @@ -path: data/training -train: data/training/images/train -val: data/training/images/val -test: data/training/images/test +path: /workspace/dnf-auto-cloud/data/training +train: /workspace/dnf-auto-cloud/data/training/images/train +val: /workspace/dnf-auto-cloud/data/training/images/val +test: /workspace/dnf-auto-cloud/data/training/images/test nc: 10 names: - monster diff --git a/models/weights/F1_curve.png b/models/weights/F1_curve.png new file mode 100644 index 0000000..29c31fb Binary files /dev/null and b/models/weights/F1_curve.png differ diff --git a/models/weights/PR_curve.png b/models/weights/PR_curve.png new file mode 100644 index 0000000..b47fee1 Binary files /dev/null and b/models/weights/PR_curve.png differ diff --git a/models/weights/P_curve.png b/models/weights/P_curve.png new file mode 100644 index 0000000..c5f0df2 Binary files /dev/null and b/models/weights/P_curve.png differ diff --git a/models/weights/R_curve.png b/models/weights/R_curve.png new file mode 100644 index 0000000..01d14b4 Binary files /dev/null and b/models/weights/R_curve.png differ diff --git a/models/weights/confusion_matrix.png b/models/weights/confusion_matrix.png new file mode 100644 index 0000000..362c40c Binary files /dev/null and b/models/weights/confusion_matrix.png differ diff --git a/models/weights/labels.jpg b/models/weights/labels.jpg new file mode 100644 index 0000000..9eebb7b Binary files /dev/null and b/models/weights/labels.jpg differ diff --git a/models/weights/labels_correlogram.jpg b/models/weights/labels_correlogram.jpg new file mode 100644 index 0000000..530fd43 Binary files /dev/null and b/models/weights/labels_correlogram.jpg differ diff --git a/models/weights/results.csv b/models/weights/results.csv new file mode 100644 index 0000000..b9db8d8 --- /dev/null +++ b/models/weights/results.csv @@ -0,0 +1,101 @@ + 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, 0.11062, 0.044713, 0.062774, 0.012864, 0.25694, 0.044005, 0.011607, 0.09363, 0.041389, 0.055328, 0.0883, 0.0013, 0.0013 + 1, 0.084386, 0.052809, 0.045843, 0.012305, 0.58135, 0.037866, 0.01087, 0.081148, 0.041847, 0.040951, 0.075673, 0.0026733, 0.0026733 + 2, 0.080052, 0.046154, 0.039956, 0.55511, 0.13355, 0.073915, 0.019121, 0.075842, 0.027031, 0.031986, 0.063019, 0.0040188, 0.0040188 + 3, 0.078225, 0.032642, 0.03222, 0.89008, 0.1593, 0.2019, 0.093216, 0.081309, 0.020676, 0.028984, 0.050337, 0.0053367, 0.0053367 + 4, 0.075672, 0.030229, 0.030481, 0.38578, 0.42212, 0.20752, 0.055837, 0.088352, 0.018589, 0.026398, 0.037627, 0.0066268, 0.0066268 + 5, 0.081446, 0.027271, 0.0294, 0.73021, 0.30357, 0.3954, 0.14948, 0.085555, 0.016193, 0.023893, 0.024889, 0.0078891, 0.0078891 + 6, 0.075361, 0.023581, 0.027368, 0.45782, 0.62595, 0.29554, 0.1066, 0.090591, 0.017099, 0.022596, 0.012124, 0.0091238, 0.0091238 + 7, 0.076797, 0.022953, 0.026967, 0.46326, 0.63058, 0.37613, 0.10864, 0.080876, 0.019985, 0.021849, 0.009307, 0.009307, 0.009307 + 8, 0.069277, 0.021624, 0.025897, 0.44857, 0.52412, 0.39575, 0.1484, 0.08447, 0.027252, 0.023365, 0.009307, 0.009307, 0.009307 + 9, 0.069657, 0.021723, 0.025119, 0.68102, 0.69233, 0.66763, 0.25607, 0.061653, 0.018906, 0.018222, 0.009208, 0.009208, 0.009208 + 10, 0.057849, 0.020393, 0.025134, 0.58782, 0.6286, 0.62221, 0.38337, 0.053783, 0.017891, 0.017056, 0.009109, 0.009109, 0.009109 + 11, 0.055896, 0.018569, 0.020781, 0.65791, 0.70455, 0.84397, 0.38415, 0.054539, 0.015846, 0.015294, 0.00901, 0.00901, 0.00901 + 12, 0.055146, 0.019553, 0.020457, 0.61275, 0.61576, 0.69215, 0.35269, 0.054875, 0.01627, 0.015222, 0.008911, 0.008911, 0.008911 + 13, 0.051287, 0.01728, 0.019606, 0.61733, 0.67584, 0.77896, 0.44407, 0.054481, 0.015491, 0.012888, 0.008812, 0.008812, 0.008812 + 14, 0.049833, 0.017423, 0.017959, 0.77359, 0.65157, 0.79625, 0.32758, 0.050235, 0.015112, 0.011866, 0.008713, 0.008713, 0.008713 + 15, 0.046434, 0.015612, 0.018171, 0.7663, 0.66557, 0.86331, 0.38905, 0.043334, 0.013628, 0.013504, 0.008614, 0.008614, 0.008614 + 16, 0.046136, 0.016711, 0.01776, 0.83936, 0.64404, 0.90648, 0.43725, 0.043066, 0.013478, 0.012336, 0.008515, 0.008515, 0.008515 + 17, 0.043906, 0.017081, 0.017746, 0.8748, 0.65369, 0.85847, 0.42247, 0.042315, 0.013261, 0.010645, 0.008416, 0.008416, 0.008416 + 18, 0.042014, 0.01525, 0.014397, 0.94607, 0.65739, 0.95437, 0.5543, 0.037817, 0.012558, 0.0096827, 0.008317, 0.008317, 0.008317 + 19, 0.041375, 0.016451, 0.014576, 0.89657, 0.67084, 0.96508, 0.57767, 0.038009, 0.012259, 0.0095654, 0.008218, 0.008218, 0.008218 + 20, 0.041926, 0.016043, 0.013606, 0.66705, 0.71925, 0.84926, 0.53664, 0.035998, 0.011994, 0.0097508, 0.008119, 0.008119, 0.008119 + 21, 0.037189, 0.014944, 0.01277, 0.75159, 0.68896, 0.79352, 0.5062, 0.043335, 0.012637, 0.0087438, 0.00802, 0.00802, 0.00802 + 22, 0.038953, 0.016303, 0.012446, 0.9459, 0.71773, 0.96028, 0.52056, 0.037898, 0.012242, 0.0083613, 0.007921, 0.007921, 0.007921 + 23, 0.036443, 0.015973, 0.011707, 0.88666, 0.70157, 0.96726, 0.66801, 0.035636, 0.012009, 0.0079827, 0.007822, 0.007822, 0.007822 + 24, 0.034551, 0.015153, 0.011661, 0.8874, 0.65796, 0.97283, 0.67398, 0.034467, 0.011672, 0.0087318, 0.007723, 0.007723, 0.007723 + 25, 0.035314, 0.013907, 0.01145, 0.9537, 0.71451, 0.98817, 0.69635, 0.02786, 0.011174, 0.0084147, 0.007624, 0.007624, 0.007624 + 26, 0.034133, 0.013862, 0.011951, 0.72453, 0.89518, 0.98299, 0.60003, 0.034185, 0.011813, 0.0096507, 0.007525, 0.007525, 0.007525 + 27, 0.034647, 0.013215, 0.010499, 0.71244, 0.98283, 0.98149, 0.63632, 0.030054, 0.011369, 0.0081987, 0.007426, 0.007426, 0.007426 + 28, 0.03294, 0.01456, 0.011796, 0.73212, 0.97058, 0.96119, 0.73927, 0.026755, 0.01094, 0.010894, 0.007327, 0.007327, 0.007327 + 29, 0.030987, 0.014169, 0.0097715, 0.67805, 0.97321, 0.96337, 0.67988, 0.02935, 0.011142, 0.0090103, 0.007228, 0.007228, 0.007228 + 30, 0.030368, 0.011704, 0.0094685, 0.72498, 0.98016, 0.98725, 0.68643, 0.027223, 0.010579, 0.0074158, 0.007129, 0.007129, 0.007129 + 31, 0.030643, 0.012915, 0.009158, 0.84399, 0.98909, 0.98542, 0.74619, 0.026265, 0.010507, 0.006691, 0.00703, 0.00703, 0.00703 + 32, 0.029583, 0.012493, 0.0090232, 0.79042, 0.98909, 0.98082, 0.6973, 0.029134, 0.010659, 0.0066166, 0.006931, 0.006931, 0.006931 + 33, 0.02909, 0.014057, 0.0093144, 0.84826, 0.98909, 0.98338, 0.74119, 0.029363, 0.010697, 0.0061455, 0.006832, 0.006832, 0.006832 + 34, 0.030618, 0.013529, 0.0085338, 0.94853, 0.98909, 0.99382, 0.71363, 0.022315, 0.0099039, 0.0059595, 0.006733, 0.006733, 0.006733 + 35, 0.027751, 0.01337, 0.0094638, 0.92628, 0.98902, 0.99414, 0.71161, 0.026125, 0.010315, 0.0057881, 0.006634, 0.006634, 0.006634 + 36, 0.028829, 0.012622, 0.0083021, 0.9495, 0.97817, 0.98928, 0.65483, 0.025652, 0.010463, 0.0067701, 0.006535, 0.006535, 0.006535 + 37, 0.028152, 0.012452, 0.008016, 0.95411, 0.98699, 0.99242, 0.74834, 0.029938, 0.010807, 0.0057073, 0.006436, 0.006436, 0.006436 + 38, 0.028067, 0.0132, 0.0072333, 0.98131, 0.9871, 0.99342, 0.72968, 0.023942, 0.0098715, 0.0056729, 0.006337, 0.006337, 0.006337 + 39, 0.02707, 0.012021, 0.0073622, 0.97159, 0.98606, 0.99106, 0.66866, 0.021726, 0.0096645, 0.0058436, 0.006238, 0.006238, 0.006238 + 40, 0.025963, 0.012655, 0.0087835, 0.97675, 0.98624, 0.99053, 0.61778, 0.026124, 0.010141, 0.0056002, 0.006139, 0.006139, 0.006139 + 41, 0.025687, 0.01183, 0.0075706, 0.88505, 0.98679, 0.99024, 0.74323, 0.025508, 0.0097275, 0.0051115, 0.00604, 0.00604, 0.00604 + 42, 0.026962, 0.013543, 0.0070483, 0.98938, 0.98615, 0.99425, 0.79556, 0.019257, 0.0091076, 0.0049877, 0.005941, 0.005941, 0.005941 + 43, 0.026752, 0.012546, 0.0073945, 0.99243, 0.97926, 0.99414, 0.82536, 0.020015, 0.0091324, 0.0050037, 0.005842, 0.005842, 0.005842 + 44, 0.025512, 0.01291, 0.0065296, 0.99284, 0.98377, 0.99446, 0.8073, 0.022218, 0.0093803, 0.0047843, 0.005743, 0.005743, 0.005743 + 45, 0.025071, 0.012273, 0.0064804, 0.92519, 0.98283, 0.99399, 0.8022, 0.022355, 0.0091429, 0.0051858, 0.005644, 0.005644, 0.005644 + 46, 0.02441, 0.012031, 0.0067756, 0.9232, 0.98173, 0.95315, 0.81619, 0.019415, 0.0091877, 0.0052021, 0.005545, 0.005545, 0.005545 + 47, 0.024003, 0.011556, 0.0061719, 0.9769, 0.98091, 0.99398, 0.79701, 0.018607, 0.0089998, 0.0049653, 0.005446, 0.005446, 0.005446 + 48, 0.023256, 0.013171, 0.0057394, 0.98993, 0.97992, 0.99348, 0.79191, 0.023383, 0.0096035, 0.0049262, 0.005347, 0.005347, 0.005347 + 49, 0.022691, 0.01123, 0.0064885, 0.96746, 0.98329, 0.99362, 0.80438, 0.020979, 0.0090265, 0.0046365, 0.005248, 0.005248, 0.005248 + 50, 0.023009, 0.010828, 0.0061525, 0.97973, 0.9809, 0.99406, 0.78401, 0.021838, 0.009346, 0.0045595, 0.005149, 0.005149, 0.005149 + 51, 0.022762, 0.011628, 0.0055389, 0.98075, 0.97948, 0.99472, 0.79553, 0.020373, 0.0091763, 0.0044677, 0.00505, 0.00505, 0.00505 + 52, 0.022359, 0.011327, 0.0064108, 0.92924, 0.97627, 0.99434, 0.82906, 0.01873, 0.0088215, 0.0047634, 0.004951, 0.004951, 0.004951 + 53, 0.021434, 0.010782, 0.0063598, 0.89751, 0.98395, 0.99438, 0.82292, 0.01758, 0.0084286, 0.0047634, 0.004852, 0.004852, 0.004852 + 54, 0.021491, 0.011245, 0.0060554, 0.93484, 0.9871, 0.99439, 0.76245, 0.019861, 0.0091343, 0.0045495, 0.004753, 0.004753, 0.004753 + 55, 0.02136, 0.010731, 0.005393, 0.98175, 0.97451, 0.99376, 0.81609, 0.017771, 0.0085478, 0.0044975, 0.004654, 0.004654, 0.004654 + 56, 0.020601, 0.011435, 0.0048984, 0.98702, 0.97619, 0.99379, 0.77352, 0.020876, 0.0089264, 0.0044127, 0.004555, 0.004555, 0.004555 + 57, 0.020204, 0.01007, 0.0049571, 0.98484, 0.97553, 0.99377, 0.80822, 0.018683, 0.0087328, 0.0042574, 0.004456, 0.004456, 0.004456 + 58, 0.020187, 0.010405, 0.0048422, 0.99255, 0.97817, 0.99174, 0.77373, 0.018314, 0.0085936, 0.0043662, 0.004357, 0.004357, 0.004357 + 59, 0.019939, 0.011237, 0.0055137, 0.96766, 0.98313, 0.99227, 0.79108, 0.018299, 0.0086794, 0.0046237, 0.004258, 0.004258, 0.004258 + 60, 0.019841, 0.010175, 0.0051564, 0.97467, 0.98512, 0.99305, 0.83424, 0.016635, 0.008214, 0.0044363, 0.004159, 0.004159, 0.004159 + 61, 0.019577, 0.011111, 0.0053072, 0.98888, 0.9871, 0.99366, 0.84728, 0.018191, 0.0086145, 0.0041327, 0.00406, 0.00406, 0.00406 + 62, 0.020016, 0.0097929, 0.0057736, 0.98464, 0.9871, 0.99353, 0.81056, 0.016411, 0.0081959, 0.0042267, 0.003961, 0.003961, 0.003961 + 63, 0.018354, 0.010467, 0.0048361, 0.99336, 0.9851, 0.99354, 0.78057, 0.017278, 0.0084858, 0.0038996, 0.003862, 0.003862, 0.003862 + 64, 0.019343, 0.010997, 0.0050088, 0.99232, 0.98663, 0.99367, 0.83564, 0.016815, 0.0083208, 0.0038519, 0.003763, 0.003763, 0.003763 + 65, 0.018261, 0.0098814, 0.0060664, 0.98882, 0.9871, 0.99358, 0.82697, 0.015722, 0.0081091, 0.0038631, 0.003664, 0.003664, 0.003664 + 66, 0.017361, 0.010034, 0.0045755, 0.9908, 0.9871, 0.99353, 0.86476, 0.014229, 0.0077305, 0.0040259, 0.003565, 0.003565, 0.003565 + 67, 0.018567, 0.010758, 0.0043478, 0.99195, 0.9871, 0.9936, 0.86666, 0.013242, 0.0076628, 0.0036473, 0.003466, 0.003466, 0.003466 + 68, 0.017044, 0.010896, 0.004663, 0.99249, 0.98488, 0.99403, 0.84897, 0.015298, 0.0079508, 0.0035157, 0.003367, 0.003367, 0.003367 + 69, 0.016631, 0.0092014, 0.0043957, 0.98846, 0.98422, 0.99374, 0.81506, 0.015467, 0.0079451, 0.0035839, 0.003268, 0.003268, 0.003268 + 70, 0.01695, 0.010094, 0.0035115, 0.99048, 0.98446, 0.99367, 0.84357, 0.014905, 0.0078063, 0.0035594, 0.003169, 0.003169, 0.003169 + 71, 0.015838, 0.010228, 0.003958, 0.99303, 0.98292, 0.99246, 0.90626, 0.012365, 0.0071888, 0.0033793, 0.00307, 0.00307, 0.00307 + 72, 0.015778, 0.0095427, 0.0062521, 0.99223, 0.98173, 0.99247, 0.85636, 0.014819, 0.0079422, 0.0033822, 0.002971, 0.002971, 0.002971 + 73, 0.015479, 0.0093237, 0.00404, 0.99072, 0.98312, 0.99374, 0.86208, 0.015422, 0.0078058, 0.0033088, 0.002872, 0.002872, 0.002872 + 74, 0.015936, 0.0093511, 0.0045926, 0.98964, 0.98485, 0.99389, 0.87519, 0.012562, 0.0072331, 0.0032785, 0.002773, 0.002773, 0.002773 + 75, 0.014587, 0.0090074, 0.0044003, 0.98713, 0.9871, 0.99287, 0.88673, 0.013244, 0.0074663, 0.0034659, 0.002674, 0.002674, 0.002674 + 76, 0.014562, 0.0085611, 0.003786, 0.98725, 0.9871, 0.99325, 0.86882, 0.012914, 0.007401, 0.0032434, 0.002575, 0.002575, 0.002575 + 77, 0.014627, 0.0093808, 0.0040065, 0.98718, 0.9871, 0.99327, 0.85954, 0.013101, 0.0073853, 0.003063, 0.002476, 0.002476, 0.002476 + 78, 0.014713, 0.0086822, 0.0041137, 0.98856, 0.98458, 0.99338, 0.85969, 0.013281, 0.0073719, 0.0029892, 0.002377, 0.002377, 0.002377 + 79, 0.013479, 0.008951, 0.0032, 0.99063, 0.98275, 0.99299, 0.89536, 0.011977, 0.0070486, 0.0029796, 0.002278, 0.002278, 0.002278 + 80, 0.01334, 0.0088546, 0.0039412, 0.99078, 0.98297, 0.99239, 0.90339, 0.011971, 0.0071206, 0.0030048, 0.002179, 0.002179, 0.002179 + 81, 0.013471, 0.0085181, 0.0038238, 0.99147, 0.98339, 0.99223, 0.9112, 0.011377, 0.0070229, 0.0029151, 0.00208, 0.00208, 0.00208 + 82, 0.013225, 0.0081305, 0.0039529, 0.98983, 0.98394, 0.99189, 0.90095, 0.011049, 0.0069065, 0.0028324, 0.001981, 0.001981, 0.001981 + 83, 0.012782, 0.0087502, 0.0041046, 0.98948, 0.98345, 0.99231, 0.87987, 0.011397, 0.0071707, 0.0028323, 0.001882, 0.001882, 0.001882 + 84, 0.01268, 0.008468, 0.003646, 0.98967, 0.98255, 0.99349, 0.90452, 0.010897, 0.0069289, 0.0028338, 0.001783, 0.001783, 0.001783 + 85, 0.011956, 0.0081656, 0.0033978, 0.99234, 0.98695, 0.99349, 0.90359, 0.011106, 0.0069551, 0.0027155, 0.001684, 0.001684, 0.001684 + 86, 0.011678, 0.0074737, 0.0029689, 0.99059, 0.9871, 0.99331, 0.90539, 0.011038, 0.0070376, 0.0026494, 0.001585, 0.001585, 0.001585 + 87, 0.011452, 0.0082765, 0.0031965, 0.99032, 0.9871, 0.99354, 0.90376, 0.011192, 0.0070019, 0.0026226, 0.001486, 0.001486, 0.001486 + 88, 0.011014, 0.0077558, 0.0029091, 0.98926, 0.9871, 0.99367, 0.93096, 0.0097374, 0.0064735, 0.0025889, 0.001387, 0.001387, 0.001387 + 89, 0.010877, 0.0081713, 0.0035058, 0.98853, 0.9871, 0.99369, 0.89844, 0.010716, 0.0067215, 0.0025122, 0.001288, 0.001288, 0.001288 + 90, 0.01086, 0.0081078, 0.0029808, 0.98859, 0.9871, 0.99369, 0.90179, 0.010313, 0.0066843, 0.002467, 0.001189, 0.001189, 0.001189 + 91, 0.010432, 0.0082593, 0.0032532, 0.98823, 0.98823, 0.99386, 0.90426, 0.00942, 0.0063066, 0.0025923, 0.00109, 0.00109, 0.00109 + 92, 0.0099856, 0.0072484, 0.0027239, 0.98621, 0.98861, 0.99381, 0.91917, 0.010086, 0.0064635, 0.0026085, 0.000991, 0.000991, 0.000991 + 93, 0.0097199, 0.0081254, 0.0029498, 0.98445, 0.98366, 0.99358, 0.92386, 0.0092925, 0.0062866, 0.0023818, 0.000892, 0.000892, 0.000892 + 94, 0.0098492, 0.0087202, 0.0028662, 0.98588, 0.98249, 0.99359, 0.92486, 0.0095873, 0.0064139, 0.0023302, 0.000793, 0.000793, 0.000793 + 95, 0.0095458, 0.0077495, 0.0031524, 0.98675, 0.98195, 0.99366, 0.91152, 0.0091509, 0.0061636, 0.002304, 0.000694, 0.000694, 0.000694 + 96, 0.0093134, 0.0070511, 0.010882, 0.98633, 0.98247, 0.99361, 0.91815, 0.0090792, 0.0062695, 0.0022655, 0.000595, 0.000595, 0.000595 + 97, 0.0092242, 0.0074598, 0.0032303, 0.98604, 0.98365, 0.99366, 0.9308, 0.0082706, 0.005929, 0.0022603, 0.000496, 0.000496, 0.000496 + 98, 0.0088322, 0.007461, 0.0025581, 0.9797, 0.9832, 0.99359, 0.92623, 0.0086755, 0.006093, 0.0023018, 0.000397, 0.000397, 0.000397 + 99, 0.009008, 0.0076601, 0.0027773, 0.97461, 0.98322, 0.99354, 0.93797, 0.0082771, 0.0060029, 0.0023297, 0.000298, 0.000298, 0.000298 diff --git a/models/weights/results.png b/models/weights/results.png new file mode 100644 index 0000000..e3356bb Binary files /dev/null and b/models/weights/results.png differ diff --git a/models/weights/train_batch0.jpg b/models/weights/train_batch0.jpg new file mode 100644 index 0000000..3e1fa75 Binary files /dev/null and b/models/weights/train_batch0.jpg differ diff --git a/models/weights/train_batch1.jpg b/models/weights/train_batch1.jpg new file mode 100644 index 0000000..43a6391 Binary files /dev/null and b/models/weights/train_batch1.jpg differ diff --git a/models/weights/train_batch2.jpg b/models/weights/train_batch2.jpg new file mode 100644 index 0000000..58af309 Binary files /dev/null and b/models/weights/train_batch2.jpg differ diff --git a/models/weights/val_batch0_labels.jpg b/models/weights/val_batch0_labels.jpg new file mode 100644 index 0000000..7d7f242 Binary files /dev/null and b/models/weights/val_batch0_labels.jpg differ diff --git a/models/weights/val_batch0_pred.jpg b/models/weights/val_batch0_pred.jpg new file mode 100644 index 0000000..d2ee8e8 Binary files /dev/null and b/models/weights/val_batch0_pred.jpg differ diff --git a/models/weights/val_batch1_labels.jpg b/models/weights/val_batch1_labels.jpg new file mode 100644 index 0000000..cebc6d5 Binary files /dev/null and b/models/weights/val_batch1_labels.jpg differ diff --git a/models/weights/val_batch1_pred.jpg b/models/weights/val_batch1_pred.jpg new file mode 100644 index 0000000..d83243f Binary files /dev/null and b/models/weights/val_batch1_pred.jpg differ diff --git a/tools/data_collector.py b/tools/data_collector.py index bf2d3d7..c4bbfe1 100644 --- a/tools/data_collector.py +++ b/tools/data_collector.py @@ -9,7 +9,7 @@ import os import sys import time import json -import yaml # 新增导入 +import yaml import argparse import logging import random @@ -215,12 +215,15 @@ class DNFDataCollector: return count def create_dataset_config(self): - """创建YOLO数据集配置文件""" + """创建YOLO数据集配置文件（修复版）""" data_yaml_path = self.output_dir / "data.yaml" - # 数据集配置 + # 使用绝对路径 + abs_data_dir = self.output_dir.absolute() + + # 确保路径正确，YOLOv5期望的格式 data_config = { - "path": str(self.output_dir.absolute()), + "path": str(abs_data_dir), "train": str((self.images_dir / "train").absolute()), "val": str((self.images_dir / "val").absolute()), "test": str((self.images_dir / "test").absolute()), diff --git a/tools/model_trainer.py b/tools/model_trainer.py index 93a1d53..584162d 100644 --- a/tools/model_trainer.py +++ b/tools/model_trainer.py @@ -37,7 +37,7 @@ class YOLOTrainer: self.config = self.load_config() # YOLO仓库路径 - self.yolo_repo = Path(self.config.get("yolo_repo", "yolov5")) + self.yolo_repo = Path(self.config.get("yolo_repo", "tools/yolov5")) # 检查配置 self.validate_config() @@ -82,44 +82,24 @@ class YOLOTrainer: logger.error(f"数据目录不存在: {data_dir}") return False - # 创建YOLO数据配置 - data_yaml_path = data_dir / "data.yaml" + # 验证数据目录结构 + images_dir = data_dir / "images" + train_dir = images_dir / "train" + val_dir = images_dir / "val" + + if not train_dir.exists() or not val_dir.exists(): + logger.warning(f"训练/验证目录不存在: {train_dir} / {val_dir}") + logger.warning("请先运行 data_collector.py 生成训练数据") + return False # 统计图像数量 - images_dir = data_dir / "images" - train_images = list((images_dir / "train").glob("*.jpg")) - val_images = list((images_dir / "val").glob("*.jpg")) + train_images = list(train_dir.glob("*.jpg")) + val_images = list(val_dir.glob("*.jpg")) - if not train_images: - logger.warning(f"没有找到训练图像: {images_dir / 'train'}") - - if not val_images: - logger.warning(f"没有找到验证图像: {images_dir / 'val'}") - - # 获取类别 - classes = self.config.get("classes", [ - "monster", "boss", "door", "item", "npc", "player", - "hp_bar", "mp_bar", "skill_ready", "cooldown" - ]) - - # 创建数据配置文件 - data_config = { - "path": str(data_dir), - "train": str(images_dir / "train"), - "val": str(images_dir / "val"), - "test": str(images_dir / "test"), - "nc": len(classes), - "names": classes - } - - # 保存配置 - with open(data_yaml_path, "w", encoding="utf-8") as f: - yaml.dump(data_config, f, default_flow_style=False, sort_keys=False) - - logger.info(f"已创建数据配置: {data_yaml_path}") + logger.info(f"已创建数据配置: {data_dir / 'data.yaml'}") logger.info(f"训练图像: {len(train_images)}, 验证图像: {len(val_images)}") - return True + return len(train_images) > 0 and len(val_images) > 0 def clone_yolo_repo(self): """克隆YOLOv5仓库"""