Invalid JSON: Unexpected token 'N', ..."kground": NaN,
"... is not valid JSON
| { | |
| "best_global_step": 3696, | |
| "best_metric": 0.04337907209992409, | |
| "best_model_checkpoint": "./models/autocrop-bilder/checkpoint-3696", | |
| "epoch": 50.0, | |
| "eval_steps": 500, | |
| "global_step": 5600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08928571428571429, | |
| "grad_norm": 3.270373582839966, | |
| "learning_rate": 9.642857142857142e-07, | |
| "loss": 0.6841774940490722, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.17857142857142858, | |
| "grad_norm": 3.1556570529937744, | |
| "learning_rate": 2.0357142857142854e-06, | |
| "loss": 0.6719733715057373, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.26785714285714285, | |
| "grad_norm": 3.7284576892852783, | |
| "learning_rate": 3.1071428571428574e-06, | |
| "loss": 0.6568694114685059, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.35714285714285715, | |
| "grad_norm": 3.3195960521698, | |
| "learning_rate": 4.178571428571429e-06, | |
| "loss": 0.6387552738189697, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.44642857142857145, | |
| "grad_norm": 4.566788673400879, | |
| "learning_rate": 5.25e-06, | |
| "loss": 0.6023201465606689, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.5357142857142857, | |
| "grad_norm": 3.4431145191192627, | |
| "learning_rate": 6.321428571428571e-06, | |
| "loss": 0.5654563903808594, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.625, | |
| "grad_norm": 4.218961715698242, | |
| "learning_rate": 7.392857142857144e-06, | |
| "loss": 0.5167044162750244, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.7142857142857143, | |
| "grad_norm": 4.118458271026611, | |
| "learning_rate": 8.464285714285714e-06, | |
| "loss": 0.4801907539367676, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.8035714285714286, | |
| "grad_norm": 3.0397183895111084, | |
| "learning_rate": 9.535714285714285e-06, | |
| "loss": 0.4150387763977051, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.8928571428571429, | |
| "grad_norm": 6.0887651443481445, | |
| "learning_rate": 1.0607142857142858e-05, | |
| "loss": 0.36455488204956055, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.9821428571428571, | |
| "grad_norm": 5.4721360206604, | |
| "learning_rate": 1.1678571428571428e-05, | |
| "loss": 0.32769825458526614, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.8947501785580725, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.8947501785580725, | |
| "eval_loss": 0.3303993046283722, | |
| "eval_mean_accuracy": 0.8947501785580725, | |
| "eval_mean_iou": 0.44737508927903624, | |
| "eval_overall_accuracy": 0.8947501785580725, | |
| "eval_runtime": 3.069, | |
| "eval_samples_per_second": 51.483, | |
| "eval_steps_per_second": 6.517, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 1.0714285714285714, | |
| "grad_norm": 5.009084701538086, | |
| "learning_rate": 1.275e-05, | |
| "loss": 0.29956760406494143, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.1607142857142858, | |
| "grad_norm": 4.929739952087402, | |
| "learning_rate": 1.3821428571428571e-05, | |
| "loss": 0.2776294231414795, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 2.2742035388946533, | |
| "learning_rate": 1.4892857142857143e-05, | |
| "loss": 0.25046911239624026, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.3392857142857144, | |
| "grad_norm": 2.3090789318084717, | |
| "learning_rate": 1.5964285714285715e-05, | |
| "loss": 0.262372350692749, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.4285714285714286, | |
| "grad_norm": 1.8420805931091309, | |
| "learning_rate": 1.7035714285714285e-05, | |
| "loss": 0.2225811719894409, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.5178571428571428, | |
| "grad_norm": 2.4408090114593506, | |
| "learning_rate": 1.8107142857142858e-05, | |
| "loss": 0.22525691986083984, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.6071428571428572, | |
| "grad_norm": 1.8533332347869873, | |
| "learning_rate": 1.9178571428571428e-05, | |
| "loss": 0.22061917781829835, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.6964285714285714, | |
| "grad_norm": 2.3673174381256104, | |
| "learning_rate": 2.025e-05, | |
| "loss": 0.1996947169303894, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.7857142857142856, | |
| "grad_norm": 2.2984402179718018, | |
| "learning_rate": 2.1321428571428574e-05, | |
| "loss": 0.19175281524658203, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.875, | |
| "grad_norm": 5.28623628616333, | |
| "learning_rate": 2.2392857142857144e-05, | |
| "loss": 0.19550073146820068, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.9642857142857144, | |
| "grad_norm": 1.4293358325958252, | |
| "learning_rate": 2.3464285714285714e-05, | |
| "loss": 0.18344916105270387, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9449638670840851, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9449638670840851, | |
| "eval_loss": 0.17327815294265747, | |
| "eval_mean_accuracy": 0.9449638670840851, | |
| "eval_mean_iou": 0.47248193354204254, | |
| "eval_overall_accuracy": 0.9449638670840851, | |
| "eval_runtime": 3.0436, | |
| "eval_samples_per_second": 51.912, | |
| "eval_steps_per_second": 6.571, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 2.0535714285714284, | |
| "grad_norm": 1.0139216184616089, | |
| "learning_rate": 2.4535714285714287e-05, | |
| "loss": 0.18806995153427125, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.142857142857143, | |
| "grad_norm": 1.6520318984985352, | |
| "learning_rate": 2.5607142857142857e-05, | |
| "loss": 0.16265342235565186, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.232142857142857, | |
| "grad_norm": 3.5133588314056396, | |
| "learning_rate": 2.6678571428571427e-05, | |
| "loss": 0.1572718858718872, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.3214285714285716, | |
| "grad_norm": 1.2140995264053345, | |
| "learning_rate": 2.7750000000000004e-05, | |
| "loss": 0.15668554306030275, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.4107142857142856, | |
| "grad_norm": 2.7298743724823, | |
| "learning_rate": 2.8821428571428574e-05, | |
| "loss": 0.14472755193710327, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 0.6985360980033875, | |
| "learning_rate": 2.9892857142857143e-05, | |
| "loss": 0.14720585346221923, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.5892857142857144, | |
| "grad_norm": 0.8003563284873962, | |
| "learning_rate": 3.096428571428572e-05, | |
| "loss": 0.14043016433715821, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.678571428571429, | |
| "grad_norm": 1.7715791463851929, | |
| "learning_rate": 3.203571428571428e-05, | |
| "loss": 0.129607629776001, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.767857142857143, | |
| "grad_norm": 0.7600429058074951, | |
| "learning_rate": 3.310714285714286e-05, | |
| "loss": 0.1326908826828003, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 1.411843180656433, | |
| "learning_rate": 3.417857142857143e-05, | |
| "loss": 0.1340739369392395, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.946428571428571, | |
| "grad_norm": 1.2389501333236694, | |
| "learning_rate": 3.525e-05, | |
| "loss": 0.1279363751411438, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9813412891527392, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9813412891527392, | |
| "eval_loss": 0.1177222803235054, | |
| "eval_mean_accuracy": 0.9813412891527392, | |
| "eval_mean_iou": 0.4906706445763696, | |
| "eval_overall_accuracy": 0.9813412891527392, | |
| "eval_runtime": 3.0843, | |
| "eval_samples_per_second": 51.227, | |
| "eval_steps_per_second": 6.484, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 3.0357142857142856, | |
| "grad_norm": 4.182790756225586, | |
| "learning_rate": 3.632142857142857e-05, | |
| "loss": 0.11685030460357666, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 3.125, | |
| "grad_norm": 1.452004313468933, | |
| "learning_rate": 3.739285714285714e-05, | |
| "loss": 0.1146626591682434, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 3.2142857142857144, | |
| "grad_norm": 0.737139105796814, | |
| "learning_rate": 3.8464285714285716e-05, | |
| "loss": 0.1056035041809082, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 3.3035714285714284, | |
| "grad_norm": 0.5500335693359375, | |
| "learning_rate": 3.953571428571429e-05, | |
| "loss": 0.12782689332962036, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 3.392857142857143, | |
| "grad_norm": 0.8951995968818665, | |
| "learning_rate": 4.060714285714286e-05, | |
| "loss": 0.10672543048858643, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.482142857142857, | |
| "grad_norm": 1.5596611499786377, | |
| "learning_rate": 4.167857142857143e-05, | |
| "loss": 0.10761728286743164, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.571428571428571, | |
| "grad_norm": 1.7340526580810547, | |
| "learning_rate": 4.275e-05, | |
| "loss": 0.12569416761398317, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.6607142857142856, | |
| "grad_norm": 0.8778754472732544, | |
| "learning_rate": 4.382142857142857e-05, | |
| "loss": 0.0940355658531189, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "grad_norm": 2.6315760612487793, | |
| "learning_rate": 4.489285714285714e-05, | |
| "loss": 0.10828475952148438, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.8392857142857144, | |
| "grad_norm": 0.5823694467544556, | |
| "learning_rate": 4.5964285714285715e-05, | |
| "loss": 0.09212335348129272, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.928571428571429, | |
| "grad_norm": 1.48265540599823, | |
| "learning_rate": 4.703571428571429e-05, | |
| "loss": 0.08791649341583252, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9857982300059793, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9857982300059793, | |
| "eval_loss": 0.08414151519536972, | |
| "eval_mean_accuracy": 0.9857982300059793, | |
| "eval_mean_iou": 0.49289911500298966, | |
| "eval_overall_accuracy": 0.9857982300059793, | |
| "eval_runtime": 2.8965, | |
| "eval_samples_per_second": 54.549, | |
| "eval_steps_per_second": 6.905, | |
| "step": 448 | |
| }, | |
| { | |
| "epoch": 4.017857142857143, | |
| "grad_norm": 7.7965288162231445, | |
| "learning_rate": 4.810714285714286e-05, | |
| "loss": 0.09109385013580322, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 4.107142857142857, | |
| "grad_norm": 0.60382080078125, | |
| "learning_rate": 4.917857142857143e-05, | |
| "loss": 0.08951844573020935, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 4.196428571428571, | |
| "grad_norm": 28.733182907104492, | |
| "learning_rate": 5.025e-05, | |
| "loss": 0.08306971788406373, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 4.285714285714286, | |
| "grad_norm": 3.4891135692596436, | |
| "learning_rate": 5.132142857142857e-05, | |
| "loss": 0.08310299515724182, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 4.375, | |
| "grad_norm": 0.517422080039978, | |
| "learning_rate": 5.239285714285714e-05, | |
| "loss": 0.08568280935287476, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 4.464285714285714, | |
| "grad_norm": 0.7893465161323547, | |
| "learning_rate": 5.346428571428572e-05, | |
| "loss": 0.09003521800041199, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.553571428571429, | |
| "grad_norm": 4.526234149932861, | |
| "learning_rate": 5.453571428571429e-05, | |
| "loss": 0.09423142075538635, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 4.642857142857143, | |
| "grad_norm": 0.7713887691497803, | |
| "learning_rate": 5.560714285714286e-05, | |
| "loss": 0.07021326422691346, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.732142857142857, | |
| "grad_norm": 0.3963504135608673, | |
| "learning_rate": 5.667857142857143e-05, | |
| "loss": 0.08367589712142945, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 4.821428571428571, | |
| "grad_norm": 1.0999394655227661, | |
| "learning_rate": 5.775e-05, | |
| "loss": 0.08009102344512939, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 4.910714285714286, | |
| "grad_norm": 0.7939938306808472, | |
| "learning_rate": 5.882142857142857e-05, | |
| "loss": 0.09382905960083007, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 2.131338357925415, | |
| "learning_rate": 5.989285714285715e-05, | |
| "loss": 0.0795985221862793, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9741688295995273, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9741688295995273, | |
| "eval_loss": 0.08404607325792313, | |
| "eval_mean_accuracy": 0.9741688295995273, | |
| "eval_mean_iou": 0.48708441479976367, | |
| "eval_overall_accuracy": 0.9741688295995273, | |
| "eval_runtime": 3.1074, | |
| "eval_samples_per_second": 50.846, | |
| "eval_steps_per_second": 6.436, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 5.089285714285714, | |
| "grad_norm": 0.9680992364883423, | |
| "learning_rate": 5.999952792194596e-05, | |
| "loss": 0.07798992395401001, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 5.178571428571429, | |
| "grad_norm": 0.8994519710540771, | |
| "learning_rate": 5.999789606626562e-05, | |
| "loss": 0.06749414801597595, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 5.267857142857143, | |
| "grad_norm": 0.986232340335846, | |
| "learning_rate": 5.99950986682241e-05, | |
| "loss": 0.0746770441532135, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 5.357142857142857, | |
| "grad_norm": 0.6977314352989197, | |
| "learning_rate": 5.999113583651189e-05, | |
| "loss": 0.07597554922103882, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.446428571428571, | |
| "grad_norm": 0.9763554930686951, | |
| "learning_rate": 5.9986007725101386e-05, | |
| "loss": 0.07884644269943238, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 5.535714285714286, | |
| "grad_norm": 0.4131947457790375, | |
| "learning_rate": 5.997971453324095e-05, | |
| "loss": 0.07100933790206909, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 5.625, | |
| "grad_norm": 4.289861679077148, | |
| "learning_rate": 5.99722565054471e-05, | |
| "loss": 0.07310172915458679, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 5.714285714285714, | |
| "grad_norm": 1.0220744609832764, | |
| "learning_rate": 5.996363393149509e-05, | |
| "loss": 0.07721139788627625, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 5.803571428571429, | |
| "grad_norm": 0.3010362684726715, | |
| "learning_rate": 5.995384714640757e-05, | |
| "loss": 0.059419333934783936, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.892857142857143, | |
| "grad_norm": 3.465324640274048, | |
| "learning_rate": 5.994289653044164e-05, | |
| "loss": 0.07689496874809265, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 5.982142857142857, | |
| "grad_norm": 0.6869708895683289, | |
| "learning_rate": 5.993078250907403e-05, | |
| "loss": 0.06406612396240234, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9859818189036227, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9859818189036227, | |
| "eval_loss": 0.07088413834571838, | |
| "eval_mean_accuracy": 0.9859818189036227, | |
| "eval_mean_iou": 0.49299090945181134, | |
| "eval_overall_accuracy": 0.9859818189036227, | |
| "eval_runtime": 3.1076, | |
| "eval_samples_per_second": 50.843, | |
| "eval_steps_per_second": 6.436, | |
| "step": 672 | |
| }, | |
| { | |
| "epoch": 6.071428571428571, | |
| "grad_norm": 1.3824396133422852, | |
| "learning_rate": 5.9917505552984576e-05, | |
| "loss": 0.06562145948410034, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 6.160714285714286, | |
| "grad_norm": 0.8862974047660828, | |
| "learning_rate": 5.990306617803795e-05, | |
| "loss": 0.07552834749221801, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "grad_norm": 1.8890525102615356, | |
| "learning_rate": 5.9887464945263616e-05, | |
| "loss": 0.06953117251396179, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 6.339285714285714, | |
| "grad_norm": 1.3064137697219849, | |
| "learning_rate": 5.9870702460833996e-05, | |
| "loss": 0.05875499844551087, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 6.428571428571429, | |
| "grad_norm": 5.315647125244141, | |
| "learning_rate": 5.985277937604096e-05, | |
| "loss": 0.06155696511268616, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 6.517857142857143, | |
| "grad_norm": 0.5886704921722412, | |
| "learning_rate": 5.98336963872705e-05, | |
| "loss": 0.06854590773582458, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 6.607142857142857, | |
| "grad_norm": 0.36860746145248413, | |
| "learning_rate": 5.981345423597567e-05, | |
| "loss": 0.05901351571083069, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 6.696428571428571, | |
| "grad_norm": 0.5009424090385437, | |
| "learning_rate": 5.979205370864779e-05, | |
| "loss": 0.055243945121765135, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 6.785714285714286, | |
| "grad_norm": 0.5251625180244446, | |
| "learning_rate": 5.976949563678588e-05, | |
| "loss": 0.07381758689880372, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 6.875, | |
| "grad_norm": 0.5076754093170166, | |
| "learning_rate": 5.9745780896864355e-05, | |
| "loss": 0.05926129817962646, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 6.964285714285714, | |
| "grad_norm": 0.5567944645881653, | |
| "learning_rate": 5.972091041029896e-05, | |
| "loss": 0.05233304500579834, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9894160453154839, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9894160453154839, | |
| "eval_loss": 0.06326749175786972, | |
| "eval_mean_accuracy": 0.9894160453154839, | |
| "eval_mean_iou": 0.49470802265774194, | |
| "eval_overall_accuracy": 0.9894160453154839, | |
| "eval_runtime": 3.1323, | |
| "eval_samples_per_second": 50.443, | |
| "eval_steps_per_second": 6.385, | |
| "step": 784 | |
| }, | |
| { | |
| "epoch": 7.053571428571429, | |
| "grad_norm": 0.860552191734314, | |
| "learning_rate": 5.969488514341099e-05, | |
| "loss": 0.0780815064907074, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 7.142857142857143, | |
| "grad_norm": 1.0587300062179565, | |
| "learning_rate": 5.966770610738974e-05, | |
| "loss": 0.056760483980178834, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 7.232142857142857, | |
| "grad_norm": 0.7505276799201965, | |
| "learning_rate": 5.963937435825317e-05, | |
| "loss": 0.059036099910736085, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 7.321428571428571, | |
| "grad_norm": 1.0296212434768677, | |
| "learning_rate": 5.960989099680696e-05, | |
| "loss": 0.05659686326980591, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 7.410714285714286, | |
| "grad_norm": 0.46932515501976013, | |
| "learning_rate": 5.957925716860167e-05, | |
| "loss": 0.05175484418869018, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "grad_norm": 0.718617856502533, | |
| "learning_rate": 5.954747406388824e-05, | |
| "loss": 0.055910295248031615, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 7.589285714285714, | |
| "grad_norm": 0.7998201251029968, | |
| "learning_rate": 5.951454291757181e-05, | |
| "loss": 0.06455008387565613, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 7.678571428571429, | |
| "grad_norm": 0.37768468260765076, | |
| "learning_rate": 5.948046500916361e-05, | |
| "loss": 0.05296936631202698, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 7.767857142857143, | |
| "grad_norm": 1.0487741231918335, | |
| "learning_rate": 5.944524166273137e-05, | |
| "loss": 0.056065672636032106, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 7.857142857142857, | |
| "grad_norm": 0.7201439142227173, | |
| "learning_rate": 5.940887424684782e-05, | |
| "loss": 0.0546242892742157, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 7.946428571428571, | |
| "grad_norm": 0.7990844249725342, | |
| "learning_rate": 5.9371364174537516e-05, | |
| "loss": 0.057676571607589724, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9807289934553552, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9807289934553552, | |
| "eval_loss": 0.06055007502436638, | |
| "eval_mean_accuracy": 0.9807289934553552, | |
| "eval_mean_iou": 0.4903644967276776, | |
| "eval_overall_accuracy": 0.9807289934553552, | |
| "eval_runtime": 2.976, | |
| "eval_samples_per_second": 53.092, | |
| "eval_steps_per_second": 6.72, | |
| "step": 896 | |
| }, | |
| { | |
| "epoch": 8.035714285714286, | |
| "grad_norm": 0.49310821294784546, | |
| "learning_rate": 5.933271290322194e-05, | |
| "loss": 0.051748812198638916, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 8.125, | |
| "grad_norm": 0.9782583117485046, | |
| "learning_rate": 5.929292193466289e-05, | |
| "loss": 0.052764898538589476, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 8.214285714285714, | |
| "grad_norm": 0.4996137022972107, | |
| "learning_rate": 5.925199281490409e-05, | |
| "loss": 0.04635877311229706, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 8.303571428571429, | |
| "grad_norm": 0.7325310111045837, | |
| "learning_rate": 5.920992713421118e-05, | |
| "loss": 0.0518154501914978, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 8.392857142857142, | |
| "grad_norm": 0.6661192178726196, | |
| "learning_rate": 5.916672652700986e-05, | |
| "loss": 0.05230782628059387, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 8.482142857142858, | |
| "grad_norm": 0.6793703436851501, | |
| "learning_rate": 5.9122392671822454e-05, | |
| "loss": 0.061950719356536864, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 8.571428571428571, | |
| "grad_norm": 0.5158465504646301, | |
| "learning_rate": 5.907692729120263e-05, | |
| "loss": 0.052070868015289304, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 8.660714285714286, | |
| "grad_norm": 1.4337759017944336, | |
| "learning_rate": 5.903033215166852e-05, | |
| "loss": 0.05228162407875061, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 8.75, | |
| "grad_norm": 1.1488945484161377, | |
| "learning_rate": 5.898260906363405e-05, | |
| "loss": 0.05639091730117798, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 8.839285714285714, | |
| "grad_norm": 3.3319473266601562, | |
| "learning_rate": 5.8933759881338625e-05, | |
| "loss": 0.06053876280784607, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 8.928571428571429, | |
| "grad_norm": 0.47250106930732727, | |
| "learning_rate": 5.888378650277505e-05, | |
| "loss": 0.05278565287590027, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9903641264589433, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9903641264589433, | |
| "eval_loss": 0.05963556095957756, | |
| "eval_mean_accuracy": 0.9903641264589433, | |
| "eval_mean_iou": 0.49518206322947167, | |
| "eval_overall_accuracy": 0.9903641264589433, | |
| "eval_runtime": 3.041, | |
| "eval_samples_per_second": 51.957, | |
| "eval_steps_per_second": 6.577, | |
| "step": 1008 | |
| }, | |
| { | |
| "epoch": 9.017857142857142, | |
| "grad_norm": 0.806399405002594, | |
| "learning_rate": 5.8832690869615824e-05, | |
| "loss": 0.05292539000511169, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 9.107142857142858, | |
| "grad_norm": 3.02130126953125, | |
| "learning_rate": 5.878047496713768e-05, | |
| "loss": 0.04778439402580261, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 9.196428571428571, | |
| "grad_norm": 2.86051082611084, | |
| "learning_rate": 5.8727140824144435e-05, | |
| "loss": 0.05619266629219055, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 9.285714285714286, | |
| "grad_norm": 0.7149938344955444, | |
| "learning_rate": 5.8672690512888185e-05, | |
| "loss": 0.047925320267677304, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 9.375, | |
| "grad_norm": 0.7016188502311707, | |
| "learning_rate": 5.8617126148988775e-05, | |
| "loss": 0.046968936920166016, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 9.464285714285714, | |
| "grad_norm": 0.4471415579319, | |
| "learning_rate": 5.856044989135162e-05, | |
| "loss": 0.0487202912569046, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 9.553571428571429, | |
| "grad_norm": 0.589865505695343, | |
| "learning_rate": 5.850266394208378e-05, | |
| "loss": 0.05116734504699707, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 9.642857142857142, | |
| "grad_norm": 0.477340966463089, | |
| "learning_rate": 5.844377054640846e-05, | |
| "loss": 0.058660686016082764, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 9.732142857142858, | |
| "grad_norm": 0.38238945603370667, | |
| "learning_rate": 5.83837719925777e-05, | |
| "loss": 0.046440258622169495, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 9.821428571428571, | |
| "grad_norm": 1.1571301221847534, | |
| "learning_rate": 5.8322670611783533e-05, | |
| "loss": 0.05035667419433594, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 9.910714285714286, | |
| "grad_norm": 0.7370873689651489, | |
| "learning_rate": 5.826046877806737e-05, | |
| "loss": 0.05174518227577209, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.7022402882575989, | |
| "learning_rate": 5.819716890822778e-05, | |
| "loss": 0.04487786293029785, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9849750743137399, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9849750743137399, | |
| "eval_loss": 0.056510768830776215, | |
| "eval_mean_accuracy": 0.9849750743137399, | |
| "eval_mean_iou": 0.49248753715686994, | |
| "eval_overall_accuracy": 0.9849750743137399, | |
| "eval_runtime": 2.8666, | |
| "eval_samples_per_second": 55.118, | |
| "eval_steps_per_second": 6.977, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 10.089285714285714, | |
| "grad_norm": 1.81419837474823, | |
| "learning_rate": 5.813277346172658e-05, | |
| "loss": 0.054458075761795045, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 10.178571428571429, | |
| "grad_norm": 0.6672200560569763, | |
| "learning_rate": 5.806728494059325e-05, | |
| "loss": 0.04975816011428833, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 10.267857142857142, | |
| "grad_norm": 0.6396266222000122, | |
| "learning_rate": 5.8000705889327764e-05, | |
| "loss": 0.05723657608032227, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 10.357142857142858, | |
| "grad_norm": 1.1310316324234009, | |
| "learning_rate": 5.79330388948017e-05, | |
| "loss": 0.04499860405921936, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 10.446428571428571, | |
| "grad_norm": 0.5163964629173279, | |
| "learning_rate": 5.7864286586157726e-05, | |
| "loss": 0.04683603048324585, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 10.535714285714286, | |
| "grad_norm": 1.1878842115402222, | |
| "learning_rate": 5.779445163470746e-05, | |
| "loss": 0.053842353820800784, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 10.625, | |
| "grad_norm": 0.9575569033622742, | |
| "learning_rate": 5.772353675382766e-05, | |
| "loss": 0.050610685348510744, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 10.714285714285714, | |
| "grad_norm": 0.46333664655685425, | |
| "learning_rate": 5.765154469885482e-05, | |
| "loss": 0.04571016728878021, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 10.803571428571429, | |
| "grad_norm": 0.5651761293411255, | |
| "learning_rate": 5.75784782669781e-05, | |
| "loss": 0.05275800824165344, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 10.892857142857142, | |
| "grad_norm": 1.1750099658966064, | |
| "learning_rate": 5.7504340297130654e-05, | |
| "loss": 0.04366275668144226, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 10.982142857142858, | |
| "grad_norm": 1.4780343770980835, | |
| "learning_rate": 5.742913366987931e-05, | |
| "loss": 0.04656153619289398, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9852730923489182, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9852730923489182, | |
| "eval_loss": 0.05328800156712532, | |
| "eval_mean_accuracy": 0.9852730923489182, | |
| "eval_mean_iou": 0.4926365461744591, | |
| "eval_overall_accuracy": 0.9852730923489182, | |
| "eval_runtime": 3.133, | |
| "eval_samples_per_second": 50.431, | |
| "eval_steps_per_second": 6.384, | |
| "step": 1232 | |
| }, | |
| { | |
| "epoch": 11.071428571428571, | |
| "grad_norm": 0.8470721244812012, | |
| "learning_rate": 5.735286130731266e-05, | |
| "loss": 0.049132627248764035, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 11.160714285714286, | |
| "grad_norm": 0.8392991423606873, | |
| "learning_rate": 5.7275526172927524e-05, | |
| "loss": 0.045812690258026124, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 11.25, | |
| "grad_norm": 0.7748396396636963, | |
| "learning_rate": 5.71971312715138e-05, | |
| "loss": 0.05167960524559021, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 11.339285714285714, | |
| "grad_norm": 1.7050870656967163, | |
| "learning_rate": 5.711767964903773e-05, | |
| "loss": 0.05110126733779907, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 11.428571428571429, | |
| "grad_norm": 0.4576941430568695, | |
| "learning_rate": 5.7037174392523523e-05, | |
| "loss": 0.04657388925552368, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 11.517857142857142, | |
| "grad_norm": 7.09236478805542, | |
| "learning_rate": 5.695561862993344e-05, | |
| "loss": 0.055950915813446044, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 11.607142857142858, | |
| "grad_norm": 0.43980512022972107, | |
| "learning_rate": 5.687301553004625e-05, | |
| "loss": 0.04865770637989044, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 11.696428571428571, | |
| "grad_norm": 0.4581732153892517, | |
| "learning_rate": 5.6789368302334114e-05, | |
| "loss": 0.04953811764717102, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 11.785714285714286, | |
| "grad_norm": 0.4522385001182556, | |
| "learning_rate": 5.670468019683786e-05, | |
| "loss": 0.04517539143562317, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 11.875, | |
| "grad_norm": 5.039539813995361, | |
| "learning_rate": 5.661895450404073e-05, | |
| "loss": 0.04825035333633423, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 11.964285714285714, | |
| "grad_norm": 0.7777697443962097, | |
| "learning_rate": 5.653219455474054e-05, | |
| "loss": 0.04639602601528168, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9873568681936357, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9873568681936357, | |
| "eval_loss": 0.05003076419234276, | |
| "eval_mean_accuracy": 0.9873568681936357, | |
| "eval_mean_iou": 0.49367843409681783, | |
| "eval_overall_accuracy": 0.9873568681936357, | |
| "eval_runtime": 3.1124, | |
| "eval_samples_per_second": 50.764, | |
| "eval_steps_per_second": 6.426, | |
| "step": 1344 | |
| }, | |
| { | |
| "epoch": 12.053571428571429, | |
| "grad_norm": 0.48987045884132385, | |
| "learning_rate": 5.6444403719920244e-05, | |
| "loss": 0.0484184205532074, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 12.142857142857142, | |
| "grad_norm": 0.774320662021637, | |
| "learning_rate": 5.6355585410616946e-05, | |
| "loss": 0.04041492640972137, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 12.232142857142858, | |
| "grad_norm": 1.0305246114730835, | |
| "learning_rate": 5.62657430777894e-05, | |
| "loss": 0.04853481948375702, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 12.321428571428571, | |
| "grad_norm": 1.1221197843551636, | |
| "learning_rate": 5.617488021218392e-05, | |
| "loss": 0.04230141043663025, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 12.410714285714286, | |
| "grad_norm": 0.3958747088909149, | |
| "learning_rate": 5.608300034419871e-05, | |
| "loss": 0.04070273339748383, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "grad_norm": 0.6378204822540283, | |
| "learning_rate": 5.599010704374675e-05, | |
| "loss": 0.040793830156326295, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 12.589285714285714, | |
| "grad_norm": 0.2917404770851135, | |
| "learning_rate": 5.589620392011705e-05, | |
| "loss": 0.04370662868022919, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 12.678571428571429, | |
| "grad_norm": 0.3751641511917114, | |
| "learning_rate": 5.580129462183444e-05, | |
| "loss": 0.04144810438156128, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 12.767857142857142, | |
| "grad_norm": 0.8122061491012573, | |
| "learning_rate": 5.570538283651778e-05, | |
| "loss": 0.043512603640556334, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 12.857142857142858, | |
| "grad_norm": 0.7271958589553833, | |
| "learning_rate": 5.560847229073673e-05, | |
| "loss": 0.04332651495933533, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 12.946428571428571, | |
| "grad_norm": 0.5666247010231018, | |
| "learning_rate": 5.551056674986689e-05, | |
| "loss": 0.04556152820587158, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9913747025252939, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9913747025252939, | |
| "eval_loss": 0.050282228738069534, | |
| "eval_mean_accuracy": 0.9913747025252939, | |
| "eval_mean_iou": 0.49568735126264696, | |
| "eval_overall_accuracy": 0.9913747025252939, | |
| "eval_runtime": 3.1533, | |
| "eval_samples_per_second": 50.106, | |
| "eval_steps_per_second": 6.343, | |
| "step": 1456 | |
| }, | |
| { | |
| "epoch": 13.035714285714286, | |
| "grad_norm": 1.259560227394104, | |
| "learning_rate": 5.541167001794355e-05, | |
| "loss": 0.044156748056411746, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 13.125, | |
| "grad_norm": 0.8817604184150696, | |
| "learning_rate": 5.5311785937513874e-05, | |
| "loss": 0.05123854279518127, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 13.214285714285714, | |
| "grad_norm": 0.5233781337738037, | |
| "learning_rate": 5.52109183894876e-05, | |
| "loss": 0.04295460879802704, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 13.303571428571429, | |
| "grad_norm": 0.4079657793045044, | |
| "learning_rate": 5.5109071292986255e-05, | |
| "loss": 0.0446503072977066, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 13.392857142857142, | |
| "grad_norm": 0.3609340190887451, | |
| "learning_rate": 5.5006248605190865e-05, | |
| "loss": 0.04136324226856232, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 13.482142857142858, | |
| "grad_norm": 0.6312282681465149, | |
| "learning_rate": 5.4902454321188225e-05, | |
| "loss": 0.04388892650604248, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 13.571428571428571, | |
| "grad_norm": 0.9294518232345581, | |
| "learning_rate": 5.479769247381566e-05, | |
| "loss": 0.04297168850898743, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 13.660714285714286, | |
| "grad_norm": 0.7301751375198364, | |
| "learning_rate": 5.4691967133504326e-05, | |
| "loss": 0.041097778081893924, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 13.75, | |
| "grad_norm": 0.588044285774231, | |
| "learning_rate": 5.458528240812107e-05, | |
| "loss": 0.036137142777442934, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 13.839285714285714, | |
| "grad_norm": 0.49514254927635193, | |
| "learning_rate": 5.447764244280884e-05, | |
| "loss": 0.04436479806900025, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 13.928571428571429, | |
| "grad_norm": 0.8887337446212769, | |
| "learning_rate": 5.436905141982555e-05, | |
| "loss": 0.039365378022193906, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9875765824408328, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9875765824408328, | |
| "eval_loss": 0.04912761226296425, | |
| "eval_mean_accuracy": 0.9875765824408328, | |
| "eval_mean_iou": 0.4937882912204164, | |
| "eval_overall_accuracy": 0.9875765824408328, | |
| "eval_runtime": 3.1041, | |
| "eval_samples_per_second": 50.9, | |
| "eval_steps_per_second": 6.443, | |
| "step": 1568 | |
| }, | |
| { | |
| "epoch": 14.017857142857142, | |
| "grad_norm": 0.6825718879699707, | |
| "learning_rate": 5.425951355838171e-05, | |
| "loss": 0.04728492200374603, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 14.107142857142858, | |
| "grad_norm": 0.6529393196105957, | |
| "learning_rate": 5.4149033114476364e-05, | |
| "loss": 0.04193180501461029, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 14.196428571428571, | |
| "grad_norm": 1.142166256904602, | |
| "learning_rate": 5.403761438073182e-05, | |
| "loss": 0.05213127732276916, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 14.285714285714286, | |
| "grad_norm": 0.2536723017692566, | |
| "learning_rate": 5.3925261686226815e-05, | |
| "loss": 0.040300771594047546, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 14.375, | |
| "grad_norm": 0.5792322754859924, | |
| "learning_rate": 5.381197939632832e-05, | |
| "loss": 0.04092682600021362, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 14.464285714285714, | |
| "grad_norm": 0.41399243474006653, | |
| "learning_rate": 5.369777191252192e-05, | |
| "loss": 0.037803399562835696, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 14.553571428571429, | |
| "grad_norm": 0.633029043674469, | |
| "learning_rate": 5.358264367224087e-05, | |
| "loss": 0.03901064395904541, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 14.642857142857142, | |
| "grad_norm": 0.44275617599487305, | |
| "learning_rate": 5.3466599148693555e-05, | |
| "loss": 0.04518336057662964, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 14.732142857142858, | |
| "grad_norm": 0.5145997405052185, | |
| "learning_rate": 5.334964285068978e-05, | |
| "loss": 0.04450685381889343, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 14.821428571428571, | |
| "grad_norm": 0.7258153557777405, | |
| "learning_rate": 5.323177932246557e-05, | |
| "loss": 0.03933931291103363, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 14.910714285714286, | |
| "grad_norm": 1.4203119277954102, | |
| "learning_rate": 5.3113013143506585e-05, | |
| "loss": 0.043070229887962344, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.701845109462738, | |
| "learning_rate": 5.2993348928370205e-05, | |
| "loss": 0.04021662771701813, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9920743816439693, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9920743816439693, | |
| "eval_loss": 0.05143989622592926, | |
| "eval_mean_accuracy": 0.9920743816439693, | |
| "eval_mean_iou": 0.49603719082198466, | |
| "eval_overall_accuracy": 0.9920743816439693, | |
| "eval_runtime": 2.9833, | |
| "eval_samples_per_second": 52.962, | |
| "eval_steps_per_second": 6.704, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 15.089285714285714, | |
| "grad_norm": 1.7581926584243774, | |
| "learning_rate": 5.287279132650623e-05, | |
| "loss": 0.04302600026130676, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 15.178571428571429, | |
| "grad_norm": 0.6842783689498901, | |
| "learning_rate": 5.2751345022076234e-05, | |
| "loss": 0.045688962936401366, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 15.267857142857142, | |
| "grad_norm": 5.589493274688721, | |
| "learning_rate": 5.2629014733771556e-05, | |
| "loss": 0.040783023834228514, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 15.357142857142858, | |
| "grad_norm": 0.9302808046340942, | |
| "learning_rate": 5.2505805214629965e-05, | |
| "loss": 0.04296206533908844, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 15.446428571428571, | |
| "grad_norm": 0.5288134813308716, | |
| "learning_rate": 5.238172125185099e-05, | |
| "loss": 0.03719891309738159, | |
| "step": 1730 | |
| }, | |
| { | |
| "epoch": 15.535714285714286, | |
| "grad_norm": 0.38548046350479126, | |
| "learning_rate": 5.225676766660991e-05, | |
| "loss": 0.0392879843711853, | |
| "step": 1740 | |
| }, | |
| { | |
| "epoch": 15.625, | |
| "grad_norm": 0.40487754344940186, | |
| "learning_rate": 5.213094931387043e-05, | |
| "loss": 0.042755302786827085, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 15.714285714285714, | |
| "grad_norm": 0.6676396727561951, | |
| "learning_rate": 5.2004271082196064e-05, | |
| "loss": 0.04512317776679993, | |
| "step": 1760 | |
| }, | |
| { | |
| "epoch": 15.803571428571429, | |
| "grad_norm": 0.6934312582015991, | |
| "learning_rate": 5.187673789356017e-05, | |
| "loss": 0.04026437401771545, | |
| "step": 1770 | |
| }, | |
| { | |
| "epoch": 15.892857142857142, | |
| "grad_norm": 1.3655362129211426, | |
| "learning_rate": 5.1748354703154706e-05, | |
| "loss": 0.045592480897903444, | |
| "step": 1780 | |
| }, | |
| { | |
| "epoch": 15.982142857142858, | |
| "grad_norm": 2.6899712085723877, | |
| "learning_rate": 5.1619126499197756e-05, | |
| "loss": 0.04209406077861786, | |
| "step": 1790 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9909822820615134, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9909822820615134, | |
| "eval_loss": 0.048930149525403976, | |
| "eval_mean_accuracy": 0.9909822820615134, | |
| "eval_mean_iou": 0.4954911410307567, | |
| "eval_overall_accuracy": 0.9909822820615134, | |
| "eval_runtime": 3.199, | |
| "eval_samples_per_second": 49.39, | |
| "eval_steps_per_second": 6.252, | |
| "step": 1792 | |
| }, | |
| { | |
| "epoch": 16.071428571428573, | |
| "grad_norm": 3.0663135051727295, | |
| "learning_rate": 5.148905830273964e-05, | |
| "loss": 0.044573307037353516, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 16.160714285714285, | |
| "grad_norm": 0.4636627435684204, | |
| "learning_rate": 5.1358155167467856e-05, | |
| "loss": 0.03752498030662536, | |
| "step": 1810 | |
| }, | |
| { | |
| "epoch": 16.25, | |
| "grad_norm": 0.47981229424476624, | |
| "learning_rate": 5.1226422179510766e-05, | |
| "loss": 0.03836653232574463, | |
| "step": 1820 | |
| }, | |
| { | |
| "epoch": 16.339285714285715, | |
| "grad_norm": 0.5740169286727905, | |
| "learning_rate": 5.1093864457239915e-05, | |
| "loss": 0.04181548357009888, | |
| "step": 1830 | |
| }, | |
| { | |
| "epoch": 16.428571428571427, | |
| "grad_norm": 0.5714458227157593, | |
| "learning_rate": 5.09604871510712e-05, | |
| "loss": 0.04018245935440064, | |
| "step": 1840 | |
| }, | |
| { | |
| "epoch": 16.517857142857142, | |
| "grad_norm": 0.8310384750366211, | |
| "learning_rate": 5.082629544326473e-05, | |
| "loss": 0.038782325387001035, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 16.607142857142858, | |
| "grad_norm": 1.200546383857727, | |
| "learning_rate": 5.06912945477235e-05, | |
| "loss": 0.044720190763473514, | |
| "step": 1860 | |
| }, | |
| { | |
| "epoch": 16.696428571428573, | |
| "grad_norm": 0.4382542371749878, | |
| "learning_rate": 5.055548970979077e-05, | |
| "loss": 0.041479668021202086, | |
| "step": 1870 | |
| }, | |
| { | |
| "epoch": 16.785714285714285, | |
| "grad_norm": 0.41013088822364807, | |
| "learning_rate": 5.041888620604634e-05, | |
| "loss": 0.035242030024528505, | |
| "step": 1880 | |
| }, | |
| { | |
| "epoch": 16.875, | |
| "grad_norm": 2.843520164489746, | |
| "learning_rate": 5.0281489344101436e-05, | |
| "loss": 0.0387931764125824, | |
| "step": 1890 | |
| }, | |
| { | |
| "epoch": 16.964285714285715, | |
| "grad_norm": 0.6174224615097046, | |
| "learning_rate": 5.014330446239254e-05, | |
| "loss": 0.04532441198825836, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9893786642719616, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9893786642719616, | |
| "eval_loss": 0.046091217547655106, | |
| "eval_mean_accuracy": 0.9893786642719616, | |
| "eval_mean_iou": 0.4946893321359808, | |
| "eval_overall_accuracy": 0.9893786642719616, | |
| "eval_runtime": 3.0874, | |
| "eval_samples_per_second": 51.176, | |
| "eval_steps_per_second": 6.478, | |
| "step": 1904 | |
| }, | |
| { | |
| "epoch": 17.053571428571427, | |
| "grad_norm": 0.4036897122859955, | |
| "learning_rate": 5.0004336929973956e-05, | |
| "loss": 0.03707077205181122, | |
| "step": 1910 | |
| }, | |
| { | |
| "epoch": 17.142857142857142, | |
| "grad_norm": 0.5742233991622925, | |
| "learning_rate": 4.9864592146309236e-05, | |
| "loss": 0.03820139169692993, | |
| "step": 1920 | |
| }, | |
| { | |
| "epoch": 17.232142857142858, | |
| "grad_norm": 0.7360264658927917, | |
| "learning_rate": 4.972407554106134e-05, | |
| "loss": 0.0436991274356842, | |
| "step": 1930 | |
| }, | |
| { | |
| "epoch": 17.321428571428573, | |
| "grad_norm": 0.5104629397392273, | |
| "learning_rate": 4.9582792573881675e-05, | |
| "loss": 0.033945232629776, | |
| "step": 1940 | |
| }, | |
| { | |
| "epoch": 17.410714285714285, | |
| "grad_norm": 1.295369267463684, | |
| "learning_rate": 4.9440748734197995e-05, | |
| "loss": 0.04245510697364807, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 17.5, | |
| "grad_norm": 0.3614010214805603, | |
| "learning_rate": 4.929794954100111e-05, | |
| "loss": 0.039192336797714236, | |
| "step": 1960 | |
| }, | |
| { | |
| "epoch": 17.589285714285715, | |
| "grad_norm": 0.3201848268508911, | |
| "learning_rate": 4.9154400542630405e-05, | |
| "loss": 0.0368089109659195, | |
| "step": 1970 | |
| }, | |
| { | |
| "epoch": 17.678571428571427, | |
| "grad_norm": 0.3553301692008972, | |
| "learning_rate": 4.901010731655833e-05, | |
| "loss": 0.04043708443641662, | |
| "step": 1980 | |
| }, | |
| { | |
| "epoch": 17.767857142857142, | |
| "grad_norm": 0.4751736521720886, | |
| "learning_rate": 4.886507546917364e-05, | |
| "loss": 0.033682060241699216, | |
| "step": 1990 | |
| }, | |
| { | |
| "epoch": 17.857142857142858, | |
| "grad_norm": 0.19512499868869781, | |
| "learning_rate": 4.871931063556361e-05, | |
| "loss": 0.03278649747371674, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 17.946428571428573, | |
| "grad_norm": 0.5196254849433899, | |
| "learning_rate": 4.857281847929503e-05, | |
| "loss": 0.04488261938095093, | |
| "step": 2010 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9857635535340814, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9857635535340814, | |
| "eval_loss": 0.04853605851531029, | |
| "eval_mean_accuracy": 0.9857635535340814, | |
| "eval_mean_iou": 0.4928817767670407, | |
| "eval_overall_accuracy": 0.9857635535340814, | |
| "eval_runtime": 3.1354, | |
| "eval_samples_per_second": 50.393, | |
| "eval_steps_per_second": 6.379, | |
| "step": 2016 | |
| }, | |
| { | |
| "epoch": 18.035714285714285, | |
| "grad_norm": 0.30406859517097473, | |
| "learning_rate": 4.842560469219418e-05, | |
| "loss": 0.03401051759719849, | |
| "step": 2020 | |
| }, | |
| { | |
| "epoch": 18.125, | |
| "grad_norm": 0.5000826716423035, | |
| "learning_rate": 4.827767499412573e-05, | |
| "loss": 0.034174400568008426, | |
| "step": 2030 | |
| }, | |
| { | |
| "epoch": 18.214285714285715, | |
| "grad_norm": 0.4069335460662842, | |
| "learning_rate": 4.8129035132770424e-05, | |
| "loss": 0.03663274347782135, | |
| "step": 2040 | |
| }, | |
| { | |
| "epoch": 18.303571428571427, | |
| "grad_norm": 0.28684502840042114, | |
| "learning_rate": 4.797969088340177e-05, | |
| "loss": 0.0345532089471817, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 18.392857142857142, | |
| "grad_norm": 0.4621877074241638, | |
| "learning_rate": 4.782964804866168e-05, | |
| "loss": 0.034203451871871945, | |
| "step": 2060 | |
| }, | |
| { | |
| "epoch": 18.482142857142858, | |
| "grad_norm": 0.274502694606781, | |
| "learning_rate": 4.7678912458335025e-05, | |
| "loss": 0.033478057384490965, | |
| "step": 2070 | |
| }, | |
| { | |
| "epoch": 18.571428571428573, | |
| "grad_norm": 2.0716018676757812, | |
| "learning_rate": 4.7527489969123035e-05, | |
| "loss": 0.03838878571987152, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 18.660714285714285, | |
| "grad_norm": 0.5928405523300171, | |
| "learning_rate": 4.737538646441587e-05, | |
| "loss": 0.04016884863376617, | |
| "step": 2090 | |
| }, | |
| { | |
| "epoch": 18.75, | |
| "grad_norm": 0.3344573974609375, | |
| "learning_rate": 4.722260785406391e-05, | |
| "loss": 0.034842535853385925, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 18.839285714285715, | |
| "grad_norm": 0.31821951270103455, | |
| "learning_rate": 4.706916007414821e-05, | |
| "loss": 0.03510299324989319, | |
| "step": 2110 | |
| }, | |
| { | |
| "epoch": 18.928571428571427, | |
| "grad_norm": 1.2290804386138916, | |
| "learning_rate": 4.691504908674982e-05, | |
| "loss": 0.034851866960525515, | |
| "step": 2120 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9924935580484623, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9924935580484623, | |
| "eval_loss": 0.04678338021039963, | |
| "eval_mean_accuracy": 0.9924935580484623, | |
| "eval_mean_iou": 0.49624677902423114, | |
| "eval_overall_accuracy": 0.9924935580484623, | |
| "eval_runtime": 3.1887, | |
| "eval_samples_per_second": 49.55, | |
| "eval_steps_per_second": 6.272, | |
| "step": 2128 | |
| }, | |
| { | |
| "epoch": 19.017857142857142, | |
| "grad_norm": 0.8945170640945435, | |
| "learning_rate": 4.6760280879718145e-05, | |
| "loss": 0.03648801743984222, | |
| "step": 2130 | |
| }, | |
| { | |
| "epoch": 19.107142857142858, | |
| "grad_norm": 0.7037824988365173, | |
| "learning_rate": 4.660486146643829e-05, | |
| "loss": 0.036881595849990845, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 19.196428571428573, | |
| "grad_norm": 0.6412319540977478, | |
| "learning_rate": 4.644879688559742e-05, | |
| "loss": 0.03466321229934692, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 19.285714285714285, | |
| "grad_norm": 0.5471871495246887, | |
| "learning_rate": 4.629209320095013e-05, | |
| "loss": 0.031432312726974485, | |
| "step": 2160 | |
| }, | |
| { | |
| "epoch": 19.375, | |
| "grad_norm": 0.4182012379169464, | |
| "learning_rate": 4.613475650108284e-05, | |
| "loss": 0.03879809677600861, | |
| "step": 2170 | |
| }, | |
| { | |
| "epoch": 19.464285714285715, | |
| "grad_norm": 0.7332907319068909, | |
| "learning_rate": 4.5976792899177225e-05, | |
| "loss": 0.03283085525035858, | |
| "step": 2180 | |
| }, | |
| { | |
| "epoch": 19.553571428571427, | |
| "grad_norm": 0.7331687808036804, | |
| "learning_rate": 4.58182085327727e-05, | |
| "loss": 0.03797485828399658, | |
| "step": 2190 | |
| }, | |
| { | |
| "epoch": 19.642857142857142, | |
| "grad_norm": 1.1568264961242676, | |
| "learning_rate": 4.5659009563527966e-05, | |
| "loss": 0.040525627136230466, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 19.732142857142858, | |
| "grad_norm": 1.1485722064971924, | |
| "learning_rate": 4.549920217698156e-05, | |
| "loss": 0.03370160162448883, | |
| "step": 2210 | |
| }, | |
| { | |
| "epoch": 19.821428571428573, | |
| "grad_norm": 1.1431678533554077, | |
| "learning_rate": 4.533879258231156e-05, | |
| "loss": 0.03664481043815613, | |
| "step": 2220 | |
| }, | |
| { | |
| "epoch": 19.910714285714285, | |
| "grad_norm": 0.6225838661193848, | |
| "learning_rate": 4.5177787012094334e-05, | |
| "loss": 0.035591110587120056, | |
| "step": 2230 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 1.3148231506347656, | |
| "learning_rate": 4.5016191722062336e-05, | |
| "loss": 0.03507326543331146, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9924366332552259, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9924366332552259, | |
| "eval_loss": 0.04701998829841614, | |
| "eval_mean_accuracy": 0.9924366332552259, | |
| "eval_mean_iou": 0.49621831662761295, | |
| "eval_overall_accuracy": 0.9924366332552259, | |
| "eval_runtime": 3.013, | |
| "eval_samples_per_second": 52.44, | |
| "eval_steps_per_second": 6.638, | |
| "step": 2240 | |
| }, | |
| { | |
| "epoch": 20.089285714285715, | |
| "grad_norm": 0.6287689208984375, | |
| "learning_rate": 4.48540129908611e-05, | |
| "loss": 0.03255096971988678, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 20.178571428571427, | |
| "grad_norm": 0.7971874475479126, | |
| "learning_rate": 4.469125711980525e-05, | |
| "loss": 0.0385926216840744, | |
| "step": 2260 | |
| }, | |
| { | |
| "epoch": 20.267857142857142, | |
| "grad_norm": 0.4999794065952301, | |
| "learning_rate": 4.452793043263369e-05, | |
| "loss": 0.0330315887928009, | |
| "step": 2270 | |
| }, | |
| { | |
| "epoch": 20.357142857142858, | |
| "grad_norm": 1.0392268896102905, | |
| "learning_rate": 4.4364039275263886e-05, | |
| "loss": 0.03553598821163177, | |
| "step": 2280 | |
| }, | |
| { | |
| "epoch": 20.446428571428573, | |
| "grad_norm": 0.748245120048523, | |
| "learning_rate": 4.4199590015545306e-05, | |
| "loss": 0.03817463517189026, | |
| "step": 2290 | |
| }, | |
| { | |
| "epoch": 20.535714285714285, | |
| "grad_norm": 3.1416192054748535, | |
| "learning_rate": 4.4034589043012e-05, | |
| "loss": 0.03555123507976532, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 20.625, | |
| "grad_norm": 1.6041500568389893, | |
| "learning_rate": 4.386904276863435e-05, | |
| "loss": 0.04191716015338898, | |
| "step": 2310 | |
| }, | |
| { | |
| "epoch": 20.714285714285715, | |
| "grad_norm": 0.3397618234157562, | |
| "learning_rate": 4.370295762456997e-05, | |
| "loss": 0.03920762240886688, | |
| "step": 2320 | |
| }, | |
| { | |
| "epoch": 20.803571428571427, | |
| "grad_norm": 0.47967231273651123, | |
| "learning_rate": 4.3536340063913794e-05, | |
| "loss": 0.03428833186626434, | |
| "step": 2330 | |
| }, | |
| { | |
| "epoch": 20.892857142857142, | |
| "grad_norm": 0.504014253616333, | |
| "learning_rate": 4.336919656044731e-05, | |
| "loss": 0.03250396251678467, | |
| "step": 2340 | |
| }, | |
| { | |
| "epoch": 20.982142857142858, | |
| "grad_norm": 0.3230708837509155, | |
| "learning_rate": 4.320153360838713e-05, | |
| "loss": 0.032353276014328004, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9897066901732577, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9897066901732577, | |
| "eval_loss": 0.04517462104558945, | |
| "eval_mean_accuracy": 0.9897066901732577, | |
| "eval_mean_iou": 0.49485334508662887, | |
| "eval_overall_accuracy": 0.9897066901732577, | |
| "eval_runtime": 3.124, | |
| "eval_samples_per_second": 50.576, | |
| "eval_steps_per_second": 6.402, | |
| "step": 2352 | |
| }, | |
| { | |
| "epoch": 21.071428571428573, | |
| "grad_norm": 1.0374712944030762, | |
| "learning_rate": 4.3033357722132514e-05, | |
| "loss": 0.03456159830093384, | |
| "step": 2360 | |
| }, | |
| { | |
| "epoch": 21.160714285714285, | |
| "grad_norm": 0.5611382126808167, | |
| "learning_rate": 4.28646754360124e-05, | |
| "loss": 0.03961513936519623, | |
| "step": 2370 | |
| }, | |
| { | |
| "epoch": 21.25, | |
| "grad_norm": 1.1374410390853882, | |
| "learning_rate": 4.26954933040314e-05, | |
| "loss": 0.03642260730266571, | |
| "step": 2380 | |
| }, | |
| { | |
| "epoch": 21.339285714285715, | |
| "grad_norm": 1.065623164176941, | |
| "learning_rate": 4.252581789961527e-05, | |
| "loss": 0.0321618914604187, | |
| "step": 2390 | |
| }, | |
| { | |
| "epoch": 21.428571428571427, | |
| "grad_norm": 0.344766229391098, | |
| "learning_rate": 4.235565581535539e-05, | |
| "loss": 0.03374730050563812, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 21.517857142857142, | |
| "grad_norm": 0.6901429295539856, | |
| "learning_rate": 4.218501366275267e-05, | |
| "loss": 0.037382254004478456, | |
| "step": 2410 | |
| }, | |
| { | |
| "epoch": 21.607142857142858, | |
| "grad_norm": 0.6230500340461731, | |
| "learning_rate": 4.2013898071960676e-05, | |
| "loss": 0.03607453107833862, | |
| "step": 2420 | |
| }, | |
| { | |
| "epoch": 21.696428571428573, | |
| "grad_norm": 0.33789336681365967, | |
| "learning_rate": 4.184231569152802e-05, | |
| "loss": 0.03156797885894776, | |
| "step": 2430 | |
| }, | |
| { | |
| "epoch": 21.785714285714285, | |
| "grad_norm": 0.6263316869735718, | |
| "learning_rate": 4.167027318813999e-05, | |
| "loss": 0.02936406135559082, | |
| "step": 2440 | |
| }, | |
| { | |
| "epoch": 21.875, | |
| "grad_norm": 0.3787066638469696, | |
| "learning_rate": 4.149777724635957e-05, | |
| "loss": 0.03306094706058502, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 21.964285714285715, | |
| "grad_norm": 1.0717002153396606, | |
| "learning_rate": 4.132483456836771e-05, | |
| "loss": 0.03666483759880066, | |
| "step": 2460 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9897153834391932, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9897153834391932, | |
| "eval_loss": 0.04611481726169586, | |
| "eval_mean_accuracy": 0.9897153834391932, | |
| "eval_mean_iou": 0.4948576917195966, | |
| "eval_overall_accuracy": 0.9897153834391932, | |
| "eval_runtime": 3.1604, | |
| "eval_samples_per_second": 49.994, | |
| "eval_steps_per_second": 6.328, | |
| "step": 2464 | |
| }, | |
| { | |
| "epoch": 22.053571428571427, | |
| "grad_norm": 0.26346009969711304, | |
| "learning_rate": 4.1151451873702884e-05, | |
| "loss": 0.036768025159835814, | |
| "step": 2470 | |
| }, | |
| { | |
| "epoch": 22.142857142857142, | |
| "grad_norm": 0.3978211581707001, | |
| "learning_rate": 4.0977635899000066e-05, | |
| "loss": 0.037500059604644774, | |
| "step": 2480 | |
| }, | |
| { | |
| "epoch": 22.232142857142858, | |
| "grad_norm": 0.6863635778427124, | |
| "learning_rate": 4.080339339772893e-05, | |
| "loss": 0.03716857135295868, | |
| "step": 2490 | |
| }, | |
| { | |
| "epoch": 22.321428571428573, | |
| "grad_norm": 0.8167490363121033, | |
| "learning_rate": 4.0628731139931474e-05, | |
| "loss": 0.031910479068756104, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 22.410714285714285, | |
| "grad_norm": 0.6655036807060242, | |
| "learning_rate": 4.045365591195899e-05, | |
| "loss": 0.03525538444519043, | |
| "step": 2510 | |
| }, | |
| { | |
| "epoch": 22.5, | |
| "grad_norm": 0.6085823774337769, | |
| "learning_rate": 4.027817451620835e-05, | |
| "loss": 0.03340524137020111, | |
| "step": 2520 | |
| }, | |
| { | |
| "epoch": 22.589285714285715, | |
| "grad_norm": 0.8327236771583557, | |
| "learning_rate": 4.0102293770857734e-05, | |
| "loss": 0.03699168860912323, | |
| "step": 2530 | |
| }, | |
| { | |
| "epoch": 22.678571428571427, | |
| "grad_norm": 0.6012502312660217, | |
| "learning_rate": 3.99260205096017e-05, | |
| "loss": 0.0321900874376297, | |
| "step": 2540 | |
| }, | |
| { | |
| "epoch": 22.767857142857142, | |
| "grad_norm": 0.9624703526496887, | |
| "learning_rate": 3.974936158138568e-05, | |
| "loss": 0.032026082277297974, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 22.857142857142858, | |
| "grad_norm": 0.4007648229598999, | |
| "learning_rate": 3.957232385013986e-05, | |
| "loss": 0.03329742252826691, | |
| "step": 2560 | |
| }, | |
| { | |
| "epoch": 22.946428571428573, | |
| "grad_norm": 0.48390379548072815, | |
| "learning_rate": 3.9394914194512486e-05, | |
| "loss": 0.03496589064598084, | |
| "step": 2570 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9903151865914551, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9903151865914551, | |
| "eval_loss": 0.04514744132757187, | |
| "eval_mean_accuracy": 0.9903151865914551, | |
| "eval_mean_iou": 0.49515759329572756, | |
| "eval_overall_accuracy": 0.9903151865914551, | |
| "eval_runtime": 3.1477, | |
| "eval_samples_per_second": 50.196, | |
| "eval_steps_per_second": 6.354, | |
| "step": 2576 | |
| }, | |
| { | |
| "epoch": 23.035714285714285, | |
| "grad_norm": 0.40391215682029724, | |
| "learning_rate": 3.9217139507602614e-05, | |
| "loss": 0.03333309590816498, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 23.125, | |
| "grad_norm": 0.3738088309764862, | |
| "learning_rate": 3.9039006696692276e-05, | |
| "loss": 0.03630165755748749, | |
| "step": 2590 | |
| }, | |
| { | |
| "epoch": 23.214285714285715, | |
| "grad_norm": 0.8210233449935913, | |
| "learning_rate": 3.886052268297809e-05, | |
| "loss": 0.033744516968727115, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 23.303571428571427, | |
| "grad_norm": 0.362053245306015, | |
| "learning_rate": 3.868169440130236e-05, | |
| "loss": 0.03140257894992828, | |
| "step": 2610 | |
| }, | |
| { | |
| "epoch": 23.392857142857142, | |
| "grad_norm": 0.5485966205596924, | |
| "learning_rate": 3.850252879988364e-05, | |
| "loss": 0.0314506471157074, | |
| "step": 2620 | |
| }, | |
| { | |
| "epoch": 23.482142857142858, | |
| "grad_norm": 0.5351651906967163, | |
| "learning_rate": 3.8323032840046734e-05, | |
| "loss": 0.030066084861755372, | |
| "step": 2630 | |
| }, | |
| { | |
| "epoch": 23.571428571428573, | |
| "grad_norm": 0.49327391386032104, | |
| "learning_rate": 3.8143213495952224e-05, | |
| "loss": 0.031948789954185486, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 23.660714285714285, | |
| "grad_norm": 0.554222822189331, | |
| "learning_rate": 3.7963077754325563e-05, | |
| "loss": 0.0361316055059433, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 23.75, | |
| "grad_norm": 0.4138738811016083, | |
| "learning_rate": 3.7782632614185495e-05, | |
| "loss": 0.03466886579990387, | |
| "step": 2660 | |
| }, | |
| { | |
| "epoch": 23.839285714285715, | |
| "grad_norm": 0.34648677706718445, | |
| "learning_rate": 3.760188508657221e-05, | |
| "loss": 0.028808239102363586, | |
| "step": 2670 | |
| }, | |
| { | |
| "epoch": 23.928571428571427, | |
| "grad_norm": 1.9430650472640991, | |
| "learning_rate": 3.742084219427489e-05, | |
| "loss": 0.03538582324981689, | |
| "step": 2680 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9913732214503568, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9913732214503568, | |
| "eval_loss": 0.04687141999602318, | |
| "eval_mean_accuracy": 0.9913732214503568, | |
| "eval_mean_iou": 0.4956866107251784, | |
| "eval_overall_accuracy": 0.9913732214503568, | |
| "eval_runtime": 2.9475, | |
| "eval_samples_per_second": 53.605, | |
| "eval_steps_per_second": 6.785, | |
| "step": 2688 | |
| }, | |
| { | |
| "epoch": 24.017857142857142, | |
| "grad_norm": 0.3821694850921631, | |
| "learning_rate": 3.7239510971558875e-05, | |
| "loss": 0.029324138164520265, | |
| "step": 2690 | |
| }, | |
| { | |
| "epoch": 24.107142857142858, | |
| "grad_norm": 0.4301556348800659, | |
| "learning_rate": 3.705789846389231e-05, | |
| "loss": 0.03461351692676544, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 24.196428571428573, | |
| "grad_norm": 0.4085342586040497, | |
| "learning_rate": 3.687601172767243e-05, | |
| "loss": 0.033036014437675475, | |
| "step": 2710 | |
| }, | |
| { | |
| "epoch": 24.285714285714285, | |
| "grad_norm": 0.4198435842990875, | |
| "learning_rate": 3.6693857829951396e-05, | |
| "loss": 0.028960457444190978, | |
| "step": 2720 | |
| }, | |
| { | |
| "epoch": 24.375, | |
| "grad_norm": 0.3448272943496704, | |
| "learning_rate": 3.6511443848161685e-05, | |
| "loss": 0.033338889479637146, | |
| "step": 2730 | |
| }, | |
| { | |
| "epoch": 24.464285714285715, | |
| "grad_norm": 0.700137734413147, | |
| "learning_rate": 3.632877686984114e-05, | |
| "loss": 0.03215117454528808, | |
| "step": 2740 | |
| }, | |
| { | |
| "epoch": 24.553571428571427, | |
| "grad_norm": 0.5219758152961731, | |
| "learning_rate": 3.6145863992357524e-05, | |
| "loss": 0.030151611566543578, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 24.642857142857142, | |
| "grad_norm": 1.1105842590332031, | |
| "learning_rate": 3.5962712322632844e-05, | |
| "loss": 0.032877811789512636, | |
| "step": 2760 | |
| }, | |
| { | |
| "epoch": 24.732142857142858, | |
| "grad_norm": 0.5468029379844666, | |
| "learning_rate": 3.577932897686714e-05, | |
| "loss": 0.03059898316860199, | |
| "step": 2770 | |
| }, | |
| { | |
| "epoch": 24.821428571428573, | |
| "grad_norm": 0.43046024441719055, | |
| "learning_rate": 3.559572108026204e-05, | |
| "loss": 0.02730157971382141, | |
| "step": 2780 | |
| }, | |
| { | |
| "epoch": 24.910714285714285, | |
| "grad_norm": 0.7536216378211975, | |
| "learning_rate": 3.541189576674391e-05, | |
| "loss": 0.03176028728485107, | |
| "step": 2790 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "grad_norm": 1.3762458562850952, | |
| "learning_rate": 3.522786017868663e-05, | |
| "loss": 0.0352538138628006, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9889600030239687, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9889600030239687, | |
| "eval_loss": 0.045154131948947906, | |
| "eval_mean_accuracy": 0.9889600030239687, | |
| "eval_mean_iou": 0.49448000151198435, | |
| "eval_overall_accuracy": 0.9889600030239687, | |
| "eval_runtime": 2.8182, | |
| "eval_samples_per_second": 56.065, | |
| "eval_steps_per_second": 7.097, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 25.089285714285715, | |
| "grad_norm": 0.3534994125366211, | |
| "learning_rate": 3.504362146663416e-05, | |
| "loss": 0.03126908540725708, | |
| "step": 2810 | |
| }, | |
| { | |
| "epoch": 25.178571428571427, | |
| "grad_norm": 0.5898330807685852, | |
| "learning_rate": 3.485918678902263e-05, | |
| "loss": 0.03366010189056397, | |
| "step": 2820 | |
| }, | |
| { | |
| "epoch": 25.267857142857142, | |
| "grad_norm": 0.7023333311080933, | |
| "learning_rate": 3.4674563311902265e-05, | |
| "loss": 0.030235046148300172, | |
| "step": 2830 | |
| }, | |
| { | |
| "epoch": 25.357142857142858, | |
| "grad_norm": 0.40369901061058044, | |
| "learning_rate": 3.448975820865892e-05, | |
| "loss": 0.02650986909866333, | |
| "step": 2840 | |
| }, | |
| { | |
| "epoch": 25.446428571428573, | |
| "grad_norm": 0.47843503952026367, | |
| "learning_rate": 3.430477865973538e-05, | |
| "loss": 0.0322486013174057, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 25.535714285714285, | |
| "grad_norm": 1.224301815032959, | |
| "learning_rate": 3.411963185235238e-05, | |
| "loss": 0.03207373321056366, | |
| "step": 2860 | |
| }, | |
| { | |
| "epoch": 25.625, | |
| "grad_norm": 0.9510029554367065, | |
| "learning_rate": 3.39343249802293e-05, | |
| "loss": 0.031137165427207947, | |
| "step": 2870 | |
| }, | |
| { | |
| "epoch": 25.714285714285715, | |
| "grad_norm": 0.4742358326911926, | |
| "learning_rate": 3.374886524330474e-05, | |
| "loss": 0.032748347520828246, | |
| "step": 2880 | |
| }, | |
| { | |
| "epoch": 25.803571428571427, | |
| "grad_norm": 0.48148679733276367, | |
| "learning_rate": 3.3563259847456706e-05, | |
| "loss": 0.02805485725402832, | |
| "step": 2890 | |
| }, | |
| { | |
| "epoch": 25.892857142857142, | |
| "grad_norm": 0.4079000949859619, | |
| "learning_rate": 3.337751600422267e-05, | |
| "loss": 0.03277623355388641, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 25.982142857142858, | |
| "grad_norm": 0.502567708492279, | |
| "learning_rate": 3.319164093051934e-05, | |
| "loss": 0.03339660167694092, | |
| "step": 2910 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9923893676463623, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9923893676463623, | |
| "eval_loss": 0.04475132003426552, | |
| "eval_mean_accuracy": 0.9923893676463623, | |
| "eval_mean_iou": 0.49619468382318116, | |
| "eval_overall_accuracy": 0.9923893676463623, | |
| "eval_runtime": 3.0957, | |
| "eval_samples_per_second": 51.038, | |
| "eval_steps_per_second": 6.461, | |
| "step": 2912 | |
| }, | |
| { | |
| "epoch": 26.071428571428573, | |
| "grad_norm": 0.7227765917778015, | |
| "learning_rate": 3.300564184836229e-05, | |
| "loss": 0.029735544323921205, | |
| "step": 2920 | |
| }, | |
| { | |
| "epoch": 26.160714285714285, | |
| "grad_norm": 1.2855943441390991, | |
| "learning_rate": 3.281952598458532e-05, | |
| "loss": 0.0299568772315979, | |
| "step": 2930 | |
| }, | |
| { | |
| "epoch": 26.25, | |
| "grad_norm": 0.6255515813827515, | |
| "learning_rate": 3.263330057055968e-05, | |
| "loss": 0.030430236458778383, | |
| "step": 2940 | |
| }, | |
| { | |
| "epoch": 26.339285714285715, | |
| "grad_norm": 1.8101065158843994, | |
| "learning_rate": 3.244697284191312e-05, | |
| "loss": 0.03257070183753967, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 26.428571428571427, | |
| "grad_norm": 0.7128406167030334, | |
| "learning_rate": 3.2260550038248705e-05, | |
| "loss": 0.029836279153823853, | |
| "step": 2960 | |
| }, | |
| { | |
| "epoch": 26.517857142857142, | |
| "grad_norm": 0.7769372463226318, | |
| "learning_rate": 3.207403940286358e-05, | |
| "loss": 0.037280154228210446, | |
| "step": 2970 | |
| }, | |
| { | |
| "epoch": 26.607142857142858, | |
| "grad_norm": 0.7605101466178894, | |
| "learning_rate": 3.188744818246751e-05, | |
| "loss": 0.03187412917613983, | |
| "step": 2980 | |
| }, | |
| { | |
| "epoch": 26.696428571428573, | |
| "grad_norm": 1.993949294090271, | |
| "learning_rate": 3.170078362690133e-05, | |
| "loss": 0.030369496345520018, | |
| "step": 2990 | |
| }, | |
| { | |
| "epoch": 26.785714285714285, | |
| "grad_norm": 0.6963241696357727, | |
| "learning_rate": 3.1514052988855223e-05, | |
| "loss": 0.030344370007514953, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 26.875, | |
| "grad_norm": 0.4319656193256378, | |
| "learning_rate": 3.1327263523586985e-05, | |
| "loss": 0.029322752356529237, | |
| "step": 3010 | |
| }, | |
| { | |
| "epoch": 26.964285714285715, | |
| "grad_norm": 0.48957666754722595, | |
| "learning_rate": 3.114042248864007e-05, | |
| "loss": 0.026906794309616087, | |
| "step": 3020 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9915391984351606, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9915391984351606, | |
| "eval_loss": 0.044823113828897476, | |
| "eval_mean_accuracy": 0.9915391984351606, | |
| "eval_mean_iou": 0.4957695992175803, | |
| "eval_overall_accuracy": 0.9915391984351606, | |
| "eval_runtime": 3.147, | |
| "eval_samples_per_second": 50.207, | |
| "eval_steps_per_second": 6.355, | |
| "step": 3024 | |
| }, | |
| { | |
| "epoch": 27.053571428571427, | |
| "grad_norm": 0.8572945594787598, | |
| "learning_rate": 3.0953537143561624e-05, | |
| "loss": 0.0366124838590622, | |
| "step": 3030 | |
| }, | |
| { | |
| "epoch": 27.142857142857142, | |
| "grad_norm": 0.6149380207061768, | |
| "learning_rate": 3.076661474962045e-05, | |
| "loss": 0.027939742803573607, | |
| "step": 3040 | |
| }, | |
| { | |
| "epoch": 27.232142857142858, | |
| "grad_norm": 0.5692083239555359, | |
| "learning_rate": 3.057966256952481e-05, | |
| "loss": 0.03376020193099975, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 27.321428571428573, | |
| "grad_norm": 0.32568567991256714, | |
| "learning_rate": 3.0392687867140333e-05, | |
| "loss": 0.031811395287513734, | |
| "step": 3060 | |
| }, | |
| { | |
| "epoch": 27.410714285714285, | |
| "grad_norm": 0.8710930943489075, | |
| "learning_rate": 3.0205697907207702e-05, | |
| "loss": 0.032558149099349974, | |
| "step": 3070 | |
| }, | |
| { | |
| "epoch": 27.5, | |
| "grad_norm": 0.41898801922798157, | |
| "learning_rate": 3.0018699955060417e-05, | |
| "loss": 0.02802801728248596, | |
| "step": 3080 | |
| }, | |
| { | |
| "epoch": 27.589285714285715, | |
| "grad_norm": 0.3447811007499695, | |
| "learning_rate": 2.9831701276342514e-05, | |
| "loss": 0.02836693823337555, | |
| "step": 3090 | |
| }, | |
| { | |
| "epoch": 27.678571428571427, | |
| "grad_norm": 0.3702087104320526, | |
| "learning_rate": 2.9644709136726268e-05, | |
| "loss": 0.03142592906951904, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 27.767857142857142, | |
| "grad_norm": 0.4320816099643707, | |
| "learning_rate": 2.9457730801629877e-05, | |
| "loss": 0.02944057583808899, | |
| "step": 3110 | |
| }, | |
| { | |
| "epoch": 27.857142857142858, | |
| "grad_norm": 0.6954593658447266, | |
| "learning_rate": 2.927077353593516e-05, | |
| "loss": 0.030368226766586303, | |
| "step": 3120 | |
| }, | |
| { | |
| "epoch": 27.946428571428573, | |
| "grad_norm": 0.6075822710990906, | |
| "learning_rate": 2.9083844603705326e-05, | |
| "loss": 0.03189859688282013, | |
| "step": 3130 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9898321951755401, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9898321951755401, | |
| "eval_loss": 0.04428162798285484, | |
| "eval_mean_accuracy": 0.9898321951755401, | |
| "eval_mean_iou": 0.49491609758777005, | |
| "eval_overall_accuracy": 0.9898321951755401, | |
| "eval_runtime": 3.1745, | |
| "eval_samples_per_second": 49.772, | |
| "eval_steps_per_second": 6.3, | |
| "step": 3136 | |
| }, | |
| { | |
| "epoch": 28.035714285714285, | |
| "grad_norm": 0.972698450088501, | |
| "learning_rate": 2.8896951267902705e-05, | |
| "loss": 0.030377256870269775, | |
| "step": 3140 | |
| }, | |
| { | |
| "epoch": 28.125, | |
| "grad_norm": 0.5971301198005676, | |
| "learning_rate": 2.8710100790106555e-05, | |
| "loss": 0.029175907373428345, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 28.214285714285715, | |
| "grad_norm": 0.3869675397872925, | |
| "learning_rate": 2.8523300430230912e-05, | |
| "loss": 0.030972689390182495, | |
| "step": 3160 | |
| }, | |
| { | |
| "epoch": 28.303571428571427, | |
| "grad_norm": 0.4772997796535492, | |
| "learning_rate": 2.833655744624256e-05, | |
| "loss": 0.02870742082595825, | |
| "step": 3170 | |
| }, | |
| { | |
| "epoch": 28.392857142857142, | |
| "grad_norm": 0.49730241298675537, | |
| "learning_rate": 2.8149879093878954e-05, | |
| "loss": 0.0323731005191803, | |
| "step": 3180 | |
| }, | |
| { | |
| "epoch": 28.482142857142858, | |
| "grad_norm": 0.48639583587646484, | |
| "learning_rate": 2.7963272626366367e-05, | |
| "loss": 0.02942431569099426, | |
| "step": 3190 | |
| }, | |
| { | |
| "epoch": 28.571428571428573, | |
| "grad_norm": 0.38169947266578674, | |
| "learning_rate": 2.7776745294138023e-05, | |
| "loss": 0.03691979944705963, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 28.660714285714285, | |
| "grad_norm": 0.4951934516429901, | |
| "learning_rate": 2.759030434455244e-05, | |
| "loss": 0.031536877155303955, | |
| "step": 3210 | |
| }, | |
| { | |
| "epoch": 28.75, | |
| "grad_norm": 0.39789584279060364, | |
| "learning_rate": 2.7403957021611794e-05, | |
| "loss": 0.029964196681976318, | |
| "step": 3220 | |
| }, | |
| { | |
| "epoch": 28.839285714285715, | |
| "grad_norm": 0.6697349548339844, | |
| "learning_rate": 2.7217710565680494e-05, | |
| "loss": 0.02921389937400818, | |
| "step": 3230 | |
| }, | |
| { | |
| "epoch": 28.928571428571427, | |
| "grad_norm": 0.6118156909942627, | |
| "learning_rate": 2.7031572213203838e-05, | |
| "loss": 0.029344850778579713, | |
| "step": 3240 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9923891744626748, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9923891744626748, | |
| "eval_loss": 0.045042846351861954, | |
| "eval_mean_accuracy": 0.9923891744626748, | |
| "eval_mean_iou": 0.4961945872313374, | |
| "eval_overall_accuracy": 0.9923891744626748, | |
| "eval_runtime": 3.0338, | |
| "eval_samples_per_second": 52.08, | |
| "eval_steps_per_second": 6.592, | |
| "step": 3248 | |
| }, | |
| { | |
| "epoch": 29.017857142857142, | |
| "grad_norm": 0.3650244474411011, | |
| "learning_rate": 2.6845549196426856e-05, | |
| "loss": 0.025525736808776855, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 29.107142857142858, | |
| "grad_norm": 0.46227961778640747, | |
| "learning_rate": 2.665964874311331e-05, | |
| "loss": 0.03523986339569092, | |
| "step": 3260 | |
| }, | |
| { | |
| "epoch": 29.196428571428573, | |
| "grad_norm": 0.449188232421875, | |
| "learning_rate": 2.6473878076264875e-05, | |
| "loss": 0.030411550402641298, | |
| "step": 3270 | |
| }, | |
| { | |
| "epoch": 29.285714285714285, | |
| "grad_norm": 0.4849155843257904, | |
| "learning_rate": 2.6288244413840478e-05, | |
| "loss": 0.02755580246448517, | |
| "step": 3280 | |
| }, | |
| { | |
| "epoch": 29.375, | |
| "grad_norm": 0.5743690729141235, | |
| "learning_rate": 2.6102754968475846e-05, | |
| "loss": 0.029297468066215516, | |
| "step": 3290 | |
| }, | |
| { | |
| "epoch": 29.464285714285715, | |
| "grad_norm": 0.7058063745498657, | |
| "learning_rate": 2.5917416947203298e-05, | |
| "loss": 0.032971051335334775, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 29.553571428571427, | |
| "grad_norm": 0.6479628086090088, | |
| "learning_rate": 2.5732237551171683e-05, | |
| "loss": 0.029575762152671815, | |
| "step": 3310 | |
| }, | |
| { | |
| "epoch": 29.642857142857142, | |
| "grad_norm": 0.9846095442771912, | |
| "learning_rate": 2.5547223975366605e-05, | |
| "loss": 0.030085331201553343, | |
| "step": 3320 | |
| }, | |
| { | |
| "epoch": 29.732142857142858, | |
| "grad_norm": 0.4668130576610565, | |
| "learning_rate": 2.5362383408330884e-05, | |
| "loss": 0.031483760476112364, | |
| "step": 3330 | |
| }, | |
| { | |
| "epoch": 29.821428571428573, | |
| "grad_norm": 1.0946239233016968, | |
| "learning_rate": 2.5177723031885214e-05, | |
| "loss": 0.026578304171562196, | |
| "step": 3340 | |
| }, | |
| { | |
| "epoch": 29.910714285714285, | |
| "grad_norm": 0.601983368396759, | |
| "learning_rate": 2.499325002084915e-05, | |
| "loss": 0.02976991832256317, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 0.40559685230255127, | |
| "learning_rate": 2.480897154276232e-05, | |
| "loss": 0.03062085509300232, | |
| "step": 3360 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9923291265331581, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9923291265331581, | |
| "eval_loss": 0.04375786706805229, | |
| "eval_mean_accuracy": 0.9923291265331581, | |
| "eval_mean_iou": 0.49616456326657904, | |
| "eval_overall_accuracy": 0.9923291265331581, | |
| "eval_runtime": 3.0864, | |
| "eval_samples_per_second": 51.193, | |
| "eval_steps_per_second": 6.48, | |
| "step": 3360 | |
| }, | |
| { | |
| "epoch": 30.089285714285715, | |
| "grad_norm": 2.2518370151519775, | |
| "learning_rate": 2.4624894757605942e-05, | |
| "loss": 0.028901228308677675, | |
| "step": 3370 | |
| }, | |
| { | |
| "epoch": 30.178571428571427, | |
| "grad_norm": 0.48904410004615784, | |
| "learning_rate": 2.4441026817524623e-05, | |
| "loss": 0.03043428063392639, | |
| "step": 3380 | |
| }, | |
| { | |
| "epoch": 30.267857142857142, | |
| "grad_norm": 0.4779645502567291, | |
| "learning_rate": 2.4257374866548496e-05, | |
| "loss": 0.031688454747200015, | |
| "step": 3390 | |
| }, | |
| { | |
| "epoch": 30.357142857142858, | |
| "grad_norm": 0.5410107374191284, | |
| "learning_rate": 2.40739460403156e-05, | |
| "loss": 0.029933920502662657, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 30.446428571428573, | |
| "grad_norm": 0.5752859711647034, | |
| "learning_rate": 2.389074746579468e-05, | |
| "loss": 0.029010221362113953, | |
| "step": 3410 | |
| }, | |
| { | |
| "epoch": 30.535714285714285, | |
| "grad_norm": 0.7090629935264587, | |
| "learning_rate": 2.3707786261008217e-05, | |
| "loss": 0.029284483194351195, | |
| "step": 3420 | |
| }, | |
| { | |
| "epoch": 30.625, | |
| "grad_norm": 0.9575799107551575, | |
| "learning_rate": 2.3525069534755923e-05, | |
| "loss": 0.028822487592697142, | |
| "step": 3430 | |
| }, | |
| { | |
| "epoch": 30.714285714285715, | |
| "grad_norm": 0.27941980957984924, | |
| "learning_rate": 2.3342604386338497e-05, | |
| "loss": 0.029538267850875856, | |
| "step": 3440 | |
| }, | |
| { | |
| "epoch": 30.803571428571427, | |
| "grad_norm": 0.7441650032997131, | |
| "learning_rate": 2.31603979052818e-05, | |
| "loss": 0.025286585092544556, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 30.892857142857142, | |
| "grad_norm": 0.5763952136039734, | |
| "learning_rate": 2.2978457171061393e-05, | |
| "loss": 0.031220585107803345, | |
| "step": 3460 | |
| }, | |
| { | |
| "epoch": 30.982142857142858, | |
| "grad_norm": 0.5611822009086609, | |
| "learning_rate": 2.2796789252827475e-05, | |
| "loss": 0.02782772183418274, | |
| "step": 3470 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.992043246873008, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.992043246873008, | |
| "eval_loss": 0.04470207169651985, | |
| "eval_mean_accuracy": 0.992043246873008, | |
| "eval_mean_iou": 0.496021623436504, | |
| "eval_overall_accuracy": 0.992043246873008, | |
| "eval_runtime": 3.1674, | |
| "eval_samples_per_second": 49.883, | |
| "eval_steps_per_second": 6.314, | |
| "step": 3472 | |
| }, | |
| { | |
| "epoch": 31.071428571428573, | |
| "grad_norm": 0.9794915914535522, | |
| "learning_rate": 2.261540120913021e-05, | |
| "loss": 0.031087765097618104, | |
| "step": 3480 | |
| }, | |
| { | |
| "epoch": 31.160714285714285, | |
| "grad_norm": 0.625676691532135, | |
| "learning_rate": 2.2434300087645497e-05, | |
| "loss": 0.027998083829879762, | |
| "step": 3490 | |
| }, | |
| { | |
| "epoch": 31.25, | |
| "grad_norm": 0.3041325509548187, | |
| "learning_rate": 2.22534929249011e-05, | |
| "loss": 0.025890350341796875, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 31.339285714285715, | |
| "grad_norm": 0.8442551493644714, | |
| "learning_rate": 2.207298674600328e-05, | |
| "loss": 0.0253944993019104, | |
| "step": 3510 | |
| }, | |
| { | |
| "epoch": 31.428571428571427, | |
| "grad_norm": 1.6644678115844727, | |
| "learning_rate": 2.1892788564363828e-05, | |
| "loss": 0.03048495650291443, | |
| "step": 3520 | |
| }, | |
| { | |
| "epoch": 31.517857142857142, | |
| "grad_norm": 0.4767031967639923, | |
| "learning_rate": 2.171290538142757e-05, | |
| "loss": 0.02629549503326416, | |
| "step": 3530 | |
| }, | |
| { | |
| "epoch": 31.607142857142858, | |
| "grad_norm": 0.5214105248451233, | |
| "learning_rate": 2.1533344186400313e-05, | |
| "loss": 0.027956664562225342, | |
| "step": 3540 | |
| }, | |
| { | |
| "epoch": 31.696428571428573, | |
| "grad_norm": 0.43287837505340576, | |
| "learning_rate": 2.135411195597732e-05, | |
| "loss": 0.02858385443687439, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 31.785714285714285, | |
| "grad_norm": 0.6804491877555847, | |
| "learning_rate": 2.117521565407222e-05, | |
| "loss": 0.024375714361667633, | |
| "step": 3560 | |
| }, | |
| { | |
| "epoch": 31.875, | |
| "grad_norm": Infinity, | |
| "learning_rate": 2.0996662231546415e-05, | |
| "loss": 0.03603389859199524, | |
| "step": 3570 | |
| }, | |
| { | |
| "epoch": 31.964285714285715, | |
| "grad_norm": 0.4699368476867676, | |
| "learning_rate": 2.0818458625939044e-05, | |
| "loss": 0.026799708604812622, | |
| "step": 3580 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9924100061036386, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9924100061036386, | |
| "eval_loss": 0.045893941074609756, | |
| "eval_mean_accuracy": 0.9924100061036386, | |
| "eval_mean_iou": 0.4962050030518193, | |
| "eval_overall_accuracy": 0.9924100061036386, | |
| "eval_runtime": 3.1947, | |
| "eval_samples_per_second": 49.457, | |
| "eval_steps_per_second": 6.26, | |
| "step": 3584 | |
| }, | |
| { | |
| "epoch": 32.05357142857143, | |
| "grad_norm": 0.5421997308731079, | |
| "learning_rate": 2.0640611761197383e-05, | |
| "loss": 0.03136940598487854, | |
| "step": 3590 | |
| }, | |
| { | |
| "epoch": 32.142857142857146, | |
| "grad_norm": 0.5340573191642761, | |
| "learning_rate": 2.046312854740787e-05, | |
| "loss": 0.024993625283241273, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 32.232142857142854, | |
| "grad_norm": 0.854223370552063, | |
| "learning_rate": 2.0286015880527604e-05, | |
| "loss": 0.032940977811813356, | |
| "step": 3610 | |
| }, | |
| { | |
| "epoch": 32.32142857142857, | |
| "grad_norm": 0.4844415485858917, | |
| "learning_rate": 2.010928064211639e-05, | |
| "loss": 0.027138784527778625, | |
| "step": 3620 | |
| }, | |
| { | |
| "epoch": 32.410714285714285, | |
| "grad_norm": 0.8010973930358887, | |
| "learning_rate": 1.993292969906938e-05, | |
| "loss": 0.029732125997543334, | |
| "step": 3630 | |
| }, | |
| { | |
| "epoch": 32.5, | |
| "grad_norm": 0.5439932942390442, | |
| "learning_rate": 1.9756969903350243e-05, | |
| "loss": 0.02823580801486969, | |
| "step": 3640 | |
| }, | |
| { | |
| "epoch": 32.589285714285715, | |
| "grad_norm": 0.5155593156814575, | |
| "learning_rate": 1.9581408091725015e-05, | |
| "loss": 0.02901386618614197, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 32.67857142857143, | |
| "grad_norm": 0.7744992971420288, | |
| "learning_rate": 1.9406251085496335e-05, | |
| "loss": 0.028212952613830566, | |
| "step": 3660 | |
| }, | |
| { | |
| "epoch": 32.767857142857146, | |
| "grad_norm": 0.5233684182167053, | |
| "learning_rate": 1.9231505690238513e-05, | |
| "loss": 0.029126328229904175, | |
| "step": 3670 | |
| }, | |
| { | |
| "epoch": 32.857142857142854, | |
| "grad_norm": 0.5012129545211792, | |
| "learning_rate": 1.9057178695533066e-05, | |
| "loss": 0.02480347752571106, | |
| "step": 3680 | |
| }, | |
| { | |
| "epoch": 32.94642857142857, | |
| "grad_norm": 0.3605862259864807, | |
| "learning_rate": 1.888327687470491e-05, | |
| "loss": 0.026929333806037903, | |
| "step": 3690 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9899491678982931, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9899491678982931, | |
| "eval_loss": 0.04337907209992409, | |
| "eval_mean_accuracy": 0.9899491678982931, | |
| "eval_mean_iou": 0.49497458394914656, | |
| "eval_overall_accuracy": 0.9899491678982931, | |
| "eval_runtime": 3.1897, | |
| "eval_samples_per_second": 49.535, | |
| "eval_steps_per_second": 6.27, | |
| "step": 3696 | |
| }, | |
| { | |
| "epoch": 33.035714285714285, | |
| "grad_norm": 0.7721555233001709, | |
| "learning_rate": 1.8709806984559214e-05, | |
| "loss": 0.027222782373428345, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 33.125, | |
| "grad_norm": 0.400393009185791, | |
| "learning_rate": 1.8536775765118827e-05, | |
| "loss": 0.02593788206577301, | |
| "step": 3710 | |
| }, | |
| { | |
| "epoch": 33.214285714285715, | |
| "grad_norm": 0.5267050266265869, | |
| "learning_rate": 1.8364189939362448e-05, | |
| "loss": 0.031792303919792174, | |
| "step": 3720 | |
| }, | |
| { | |
| "epoch": 33.30357142857143, | |
| "grad_norm": 0.4609811007976532, | |
| "learning_rate": 1.819205621296338e-05, | |
| "loss": 0.025541388988494874, | |
| "step": 3730 | |
| }, | |
| { | |
| "epoch": 33.392857142857146, | |
| "grad_norm": 0.47067102789878845, | |
| "learning_rate": 1.8020381274029008e-05, | |
| "loss": 0.025906682014465332, | |
| "step": 3740 | |
| }, | |
| { | |
| "epoch": 33.482142857142854, | |
| "grad_norm": 0.5264711976051331, | |
| "learning_rate": 1.78491717928409e-05, | |
| "loss": 0.02825929820537567, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 33.57142857142857, | |
| "grad_norm": 0.7650744915008545, | |
| "learning_rate": 1.7678434421595687e-05, | |
| "loss": 0.02745148241519928, | |
| "step": 3760 | |
| }, | |
| { | |
| "epoch": 33.660714285714285, | |
| "grad_norm": 0.7770243883132935, | |
| "learning_rate": 1.7508175794146576e-05, | |
| "loss": 0.03023013174533844, | |
| "step": 3770 | |
| }, | |
| { | |
| "epoch": 33.75, | |
| "grad_norm": 0.5987495183944702, | |
| "learning_rate": 1.7338402525745577e-05, | |
| "loss": 0.02483441084623337, | |
| "step": 3780 | |
| }, | |
| { | |
| "epoch": 33.839285714285715, | |
| "grad_norm": 0.35336533188819885, | |
| "learning_rate": 1.716912121278651e-05, | |
| "loss": 0.02692849636077881, | |
| "step": 3790 | |
| }, | |
| { | |
| "epoch": 33.92857142857143, | |
| "grad_norm": 0.39276322722435, | |
| "learning_rate": 1.7000338432548702e-05, | |
| "loss": 0.026848804950714112, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9906004867005427, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9906004867005427, | |
| "eval_loss": 0.04445888474583626, | |
| "eval_mean_accuracy": 0.9906004867005427, | |
| "eval_mean_iou": 0.49530024335027134, | |
| "eval_overall_accuracy": 0.9906004867005427, | |
| "eval_runtime": 3.0691, | |
| "eval_samples_per_second": 51.48, | |
| "eval_steps_per_second": 6.516, | |
| "step": 3808 | |
| }, | |
| { | |
| "epoch": 34.017857142857146, | |
| "grad_norm": 0.4166980981826782, | |
| "learning_rate": 1.683206074294139e-05, | |
| "loss": 0.025022292137145997, | |
| "step": 3810 | |
| }, | |
| { | |
| "epoch": 34.107142857142854, | |
| "grad_norm": 0.5631642937660217, | |
| "learning_rate": 1.666429468224898e-05, | |
| "loss": 0.025662198662757874, | |
| "step": 3820 | |
| }, | |
| { | |
| "epoch": 34.19642857142857, | |
| "grad_norm": 0.9187061786651611, | |
| "learning_rate": 1.6497046768876947e-05, | |
| "loss": 0.02872176170349121, | |
| "step": 3830 | |
| }, | |
| { | |
| "epoch": 34.285714285714285, | |
| "grad_norm": 0.5928124785423279, | |
| "learning_rate": 1.633032350109864e-05, | |
| "loss": 0.028657811880111694, | |
| "step": 3840 | |
| }, | |
| { | |
| "epoch": 34.375, | |
| "grad_norm": 0.41895392537117004, | |
| "learning_rate": 1.616413135680272e-05, | |
| "loss": 0.02543511092662811, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 34.464285714285715, | |
| "grad_norm": 0.5905092358589172, | |
| "learning_rate": 1.599847679324151e-05, | |
| "loss": 0.026353082060813902, | |
| "step": 3860 | |
| }, | |
| { | |
| "epoch": 34.55357142857143, | |
| "grad_norm": 0.5018086433410645, | |
| "learning_rate": 1.5833366246780132e-05, | |
| "loss": 0.029970470070838928, | |
| "step": 3870 | |
| }, | |
| { | |
| "epoch": 34.642857142857146, | |
| "grad_norm": 0.7949761152267456, | |
| "learning_rate": 1.566880613264633e-05, | |
| "loss": 0.025467506051063536, | |
| "step": 3880 | |
| }, | |
| { | |
| "epoch": 34.732142857142854, | |
| "grad_norm": 0.5539437532424927, | |
| "learning_rate": 1.550480284468134e-05, | |
| "loss": 0.027051180601119995, | |
| "step": 3890 | |
| }, | |
| { | |
| "epoch": 34.82142857142857, | |
| "grad_norm": 0.740074098110199, | |
| "learning_rate": 1.534136275509136e-05, | |
| "loss": 0.028169989585876465, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 34.910714285714285, | |
| "grad_norm": 0.5646836161613464, | |
| "learning_rate": 1.5178492214199992e-05, | |
| "loss": 0.028034254908561707, | |
| "step": 3910 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "grad_norm": 0.6476469039916992, | |
| "learning_rate": 1.5016197550201563e-05, | |
| "loss": 0.03015502095222473, | |
| "step": 3920 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9891122961642446, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9891122961642446, | |
| "eval_loss": 0.04427924379706383, | |
| "eval_mean_accuracy": 0.9891122961642446, | |
| "eval_mean_iou": 0.4945561480821223, | |
| "eval_overall_accuracy": 0.9891122961642446, | |
| "eval_runtime": 2.9945, | |
| "eval_samples_per_second": 52.763, | |
| "eval_steps_per_second": 6.679, | |
| "step": 3920 | |
| }, | |
| { | |
| "epoch": 35.089285714285715, | |
| "grad_norm": 1.9664608240127563, | |
| "learning_rate": 1.4854485068915138e-05, | |
| "loss": 0.0286253958940506, | |
| "step": 3930 | |
| }, | |
| { | |
| "epoch": 35.17857142857143, | |
| "grad_norm": 0.35685494542121887, | |
| "learning_rate": 1.469336105353964e-05, | |
| "loss": 0.02511138916015625, | |
| "step": 3940 | |
| }, | |
| { | |
| "epoch": 35.267857142857146, | |
| "grad_norm": 0.6063141822814941, | |
| "learning_rate": 1.4532831764409593e-05, | |
| "loss": 0.027212321758270264, | |
| "step": 3950 | |
| }, | |
| { | |
| "epoch": 35.357142857142854, | |
| "grad_norm": 0.6104390621185303, | |
| "learning_rate": 1.4372903438752002e-05, | |
| "loss": 0.025636497139930724, | |
| "step": 3960 | |
| }, | |
| { | |
| "epoch": 35.44642857142857, | |
| "grad_norm": 0.616294264793396, | |
| "learning_rate": 1.4213582290443913e-05, | |
| "loss": 0.02637772560119629, | |
| "step": 3970 | |
| }, | |
| { | |
| "epoch": 35.535714285714285, | |
| "grad_norm": 0.7684455513954163, | |
| "learning_rate": 1.405487450977105e-05, | |
| "loss": 0.02701924443244934, | |
| "step": 3980 | |
| }, | |
| { | |
| "epoch": 35.625, | |
| "grad_norm": 0.3559069037437439, | |
| "learning_rate": 1.389678626318724e-05, | |
| "loss": 0.030377697944641114, | |
| "step": 3990 | |
| }, | |
| { | |
| "epoch": 35.714285714285715, | |
| "grad_norm": 0.4215583801269531, | |
| "learning_rate": 1.373932369307488e-05, | |
| "loss": 0.022440439462661742, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 35.80357142857143, | |
| "grad_norm": 0.3485909104347229, | |
| "learning_rate": 1.3582492917506218e-05, | |
| "loss": 0.024866998195648193, | |
| "step": 4010 | |
| }, | |
| { | |
| "epoch": 35.892857142857146, | |
| "grad_norm": 0.7256842851638794, | |
| "learning_rate": 1.3426300030005673e-05, | |
| "loss": 0.024919840693473815, | |
| "step": 4020 | |
| }, | |
| { | |
| "epoch": 35.982142857142854, | |
| "grad_norm": 0.3357905447483063, | |
| "learning_rate": 1.3270751099313106e-05, | |
| "loss": 0.02386535108089447, | |
| "step": 4030 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.991872182717768, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.991872182717768, | |
| "eval_loss": 0.04389407858252525, | |
| "eval_mean_accuracy": 0.991872182717768, | |
| "eval_mean_iou": 0.495936091358884, | |
| "eval_overall_accuracy": 0.991872182717768, | |
| "eval_runtime": 3.1411, | |
| "eval_samples_per_second": 50.301, | |
| "eval_steps_per_second": 6.367, | |
| "step": 4032 | |
| }, | |
| { | |
| "epoch": 36.07142857142857, | |
| "grad_norm": 1.1024103164672852, | |
| "learning_rate": 1.3115852169147942e-05, | |
| "loss": 0.02843126356601715, | |
| "step": 4040 | |
| }, | |
| { | |
| "epoch": 36.160714285714285, | |
| "grad_norm": 0.5255261659622192, | |
| "learning_rate": 1.2961609257974446e-05, | |
| "loss": 0.028734481334686278, | |
| "step": 4050 | |
| }, | |
| { | |
| "epoch": 36.25, | |
| "grad_norm": 1.559330940246582, | |
| "learning_rate": 1.28080283587678e-05, | |
| "loss": 0.02784898579120636, | |
| "step": 4060 | |
| }, | |
| { | |
| "epoch": 36.339285714285715, | |
| "grad_norm": 1.8099925518035889, | |
| "learning_rate": 1.2655115438781276e-05, | |
| "loss": 0.02688506543636322, | |
| "step": 4070 | |
| }, | |
| { | |
| "epoch": 36.42857142857143, | |
| "grad_norm": 1.2787431478500366, | |
| "learning_rate": 1.2502876439314429e-05, | |
| "loss": 0.027660322189331055, | |
| "step": 4080 | |
| }, | |
| { | |
| "epoch": 36.517857142857146, | |
| "grad_norm": 0.6661645770072937, | |
| "learning_rate": 1.2351317275482168e-05, | |
| "loss": 0.028162682056427003, | |
| "step": 4090 | |
| }, | |
| { | |
| "epoch": 36.607142857142854, | |
| "grad_norm": 0.659756600856781, | |
| "learning_rate": 1.2200443835985017e-05, | |
| "loss": 0.02568315863609314, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 36.69642857142857, | |
| "grad_norm": 0.32298824191093445, | |
| "learning_rate": 1.2050261982880229e-05, | |
| "loss": 0.025612333416938783, | |
| "step": 4110 | |
| }, | |
| { | |
| "epoch": 36.785714285714285, | |
| "grad_norm": 0.5366314053535461, | |
| "learning_rate": 1.1900777551354101e-05, | |
| "loss": 0.02462407052516937, | |
| "step": 4120 | |
| }, | |
| { | |
| "epoch": 36.875, | |
| "grad_norm": 0.3352350890636444, | |
| "learning_rate": 1.1751996349495216e-05, | |
| "loss": 0.023358260095119477, | |
| "step": 4130 | |
| }, | |
| { | |
| "epoch": 36.964285714285715, | |
| "grad_norm": 0.45832622051239014, | |
| "learning_rate": 1.1603924158068747e-05, | |
| "loss": 0.026781818270683287, | |
| "step": 4140 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.991538457897692, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.991538457897692, | |
| "eval_loss": 0.044210035353899, | |
| "eval_mean_accuracy": 0.991538457897692, | |
| "eval_mean_iou": 0.495769228948846, | |
| "eval_overall_accuracy": 0.991538457897692, | |
| "eval_runtime": 3.158, | |
| "eval_samples_per_second": 50.031, | |
| "eval_steps_per_second": 6.333, | |
| "step": 4144 | |
| }, | |
| { | |
| "epoch": 37.05357142857143, | |
| "grad_norm": 0.33479002118110657, | |
| "learning_rate": 1.1456566730291887e-05, | |
| "loss": 0.024782487750053407, | |
| "step": 4150 | |
| }, | |
| { | |
| "epoch": 37.142857142857146, | |
| "grad_norm": 0.4473731815814972, | |
| "learning_rate": 1.1309929791610332e-05, | |
| "loss": 0.02461591362953186, | |
| "step": 4160 | |
| }, | |
| { | |
| "epoch": 37.232142857142854, | |
| "grad_norm": 0.5323212146759033, | |
| "learning_rate": 1.1164019039475756e-05, | |
| "loss": 0.029087895154953004, | |
| "step": 4170 | |
| }, | |
| { | |
| "epoch": 37.32142857142857, | |
| "grad_norm": 0.34507501125335693, | |
| "learning_rate": 1.1018840143124519e-05, | |
| "loss": 0.023842757940292357, | |
| "step": 4180 | |
| }, | |
| { | |
| "epoch": 37.410714285714285, | |
| "grad_norm": 0.6882377862930298, | |
| "learning_rate": 1.087439874335733e-05, | |
| "loss": 0.026730936765670777, | |
| "step": 4190 | |
| }, | |
| { | |
| "epoch": 37.5, | |
| "grad_norm": 0.7014771699905396, | |
| "learning_rate": 1.0730700452320104e-05, | |
| "loss": 0.02618339955806732, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 37.589285714285715, | |
| "grad_norm": 0.5191277861595154, | |
| "learning_rate": 1.0587750853285933e-05, | |
| "loss": 0.02311095893383026, | |
| "step": 4210 | |
| }, | |
| { | |
| "epoch": 37.67857142857143, | |
| "grad_norm": 0.32207751274108887, | |
| "learning_rate": 1.0445555500438094e-05, | |
| "loss": 0.024721568822860716, | |
| "step": 4220 | |
| }, | |
| { | |
| "epoch": 37.767857142857146, | |
| "grad_norm": 0.4454548954963684, | |
| "learning_rate": 1.0304119918654315e-05, | |
| "loss": 0.026396998763084413, | |
| "step": 4230 | |
| }, | |
| { | |
| "epoch": 37.857142857142854, | |
| "grad_norm": 0.4234572947025299, | |
| "learning_rate": 1.0163449603292046e-05, | |
| "loss": 0.025258558988571166, | |
| "step": 4240 | |
| }, | |
| { | |
| "epoch": 37.94642857142857, | |
| "grad_norm": 0.36918315291404724, | |
| "learning_rate": 1.0023550019974975e-05, | |
| "loss": 0.031771954894065854, | |
| "step": 4250 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9915934830513351, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9915934830513351, | |
| "eval_loss": 0.04508010670542717, | |
| "eval_mean_accuracy": 0.9915934830513351, | |
| "eval_mean_iou": 0.49579674152566755, | |
| "eval_overall_accuracy": 0.9915934830513351, | |
| "eval_runtime": 3.1481, | |
| "eval_samples_per_second": 50.188, | |
| "eval_steps_per_second": 6.353, | |
| "step": 4256 | |
| }, | |
| { | |
| "epoch": 38.035714285714285, | |
| "grad_norm": 0.3639099597930908, | |
| "learning_rate": 9.884426604380682e-06, | |
| "loss": 0.03158144354820251, | |
| "step": 4260 | |
| }, | |
| { | |
| "epoch": 38.125, | |
| "grad_norm": 0.3746374249458313, | |
| "learning_rate": 9.746084762029392e-06, | |
| "loss": 0.027127474546432495, | |
| "step": 4270 | |
| }, | |
| { | |
| "epoch": 38.214285714285715, | |
| "grad_norm": 0.7463239431381226, | |
| "learning_rate": 9.608529868073992e-06, | |
| "loss": 0.02929815649986267, | |
| "step": 4280 | |
| }, | |
| { | |
| "epoch": 38.30357142857143, | |
| "grad_norm": 0.2743181586265564, | |
| "learning_rate": 9.471767267091185e-06, | |
| "loss": 0.024283508956432342, | |
| "step": 4290 | |
| }, | |
| { | |
| "epoch": 38.392857142857146, | |
| "grad_norm": 0.4589490294456482, | |
| "learning_rate": 9.335802272873778e-06, | |
| "loss": 0.0359749972820282, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 38.482142857142854, | |
| "grad_norm": 0.6548665761947632, | |
| "learning_rate": 9.200640168224273e-06, | |
| "loss": 0.02526823282241821, | |
| "step": 4310 | |
| }, | |
| { | |
| "epoch": 38.57142857142857, | |
| "grad_norm": 0.8196024894714355, | |
| "learning_rate": 9.066286204749602e-06, | |
| "loss": 0.029026347398757934, | |
| "step": 4320 | |
| }, | |
| { | |
| "epoch": 38.660714285714285, | |
| "grad_norm": 0.954537034034729, | |
| "learning_rate": 8.932745602657037e-06, | |
| "loss": 0.028312593698501587, | |
| "step": 4330 | |
| }, | |
| { | |
| "epoch": 38.75, | |
| "grad_norm": 0.23943273723125458, | |
| "learning_rate": 8.800023550551443e-06, | |
| "loss": 0.021906113624572753, | |
| "step": 4340 | |
| }, | |
| { | |
| "epoch": 38.839285714285715, | |
| "grad_norm": 0.4769308567047119, | |
| "learning_rate": 8.668125205233592e-06, | |
| "loss": 0.024248187243938447, | |
| "step": 4350 | |
| }, | |
| { | |
| "epoch": 38.92857142857143, | |
| "grad_norm": 0.8277206420898438, | |
| "learning_rate": 8.537055691499845e-06, | |
| "loss": 0.027632582187652587, | |
| "step": 4360 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9912071800709905, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9912071800709905, | |
| "eval_loss": 0.044417619705200195, | |
| "eval_mean_accuracy": 0.9912071800709905, | |
| "eval_mean_iou": 0.49560359003549526, | |
| "eval_overall_accuracy": 0.9912071800709905, | |
| "eval_runtime": 3.0475, | |
| "eval_samples_per_second": 51.846, | |
| "eval_steps_per_second": 6.563, | |
| "step": 4368 | |
| }, | |
| { | |
| "epoch": 39.017857142857146, | |
| "grad_norm": 0.5799295902252197, | |
| "learning_rate": 8.406820101943053e-06, | |
| "loss": 0.029939955472946166, | |
| "step": 4370 | |
| }, | |
| { | |
| "epoch": 39.107142857142854, | |
| "grad_norm": 0.4670115113258362, | |
| "learning_rate": 8.277423496754626e-06, | |
| "loss": 0.02502877414226532, | |
| "step": 4380 | |
| }, | |
| { | |
| "epoch": 39.19642857142857, | |
| "grad_norm": 0.3573046922683716, | |
| "learning_rate": 8.148870903527995e-06, | |
| "loss": 0.027808144688606262, | |
| "step": 4390 | |
| }, | |
| { | |
| "epoch": 39.285714285714285, | |
| "grad_norm": 0.48127973079681396, | |
| "learning_rate": 8.021167317063218e-06, | |
| "loss": 0.024090613424777984, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 39.375, | |
| "grad_norm": 0.5244694352149963, | |
| "learning_rate": 7.894317699172912e-06, | |
| "loss": 0.027278715372085573, | |
| "step": 4410 | |
| }, | |
| { | |
| "epoch": 39.464285714285715, | |
| "grad_norm": 0.5811761021614075, | |
| "learning_rate": 7.768326978489516e-06, | |
| "loss": 0.027140852808952332, | |
| "step": 4420 | |
| }, | |
| { | |
| "epoch": 39.55357142857143, | |
| "grad_norm": 0.7176029086112976, | |
| "learning_rate": 7.643200050273741e-06, | |
| "loss": 0.026023095846176146, | |
| "step": 4430 | |
| }, | |
| { | |
| "epoch": 39.642857142857146, | |
| "grad_norm": 0.28565794229507446, | |
| "learning_rate": 7.518941776224381e-06, | |
| "loss": 0.024949049949645995, | |
| "step": 4440 | |
| }, | |
| { | |
| "epoch": 39.732142857142854, | |
| "grad_norm": 0.42266854643821716, | |
| "learning_rate": 7.395556984289442e-06, | |
| "loss": 0.02561112344264984, | |
| "step": 4450 | |
| }, | |
| { | |
| "epoch": 39.82142857142857, | |
| "grad_norm": 0.9398337602615356, | |
| "learning_rate": 7.273050468478509e-06, | |
| "loss": 0.024365237355232237, | |
| "step": 4460 | |
| }, | |
| { | |
| "epoch": 39.910714285714285, | |
| "grad_norm": 0.370419979095459, | |
| "learning_rate": 7.151426988676538e-06, | |
| "loss": 0.025940805673599243, | |
| "step": 4470 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 0.46006032824516296, | |
| "learning_rate": 7.0306912704588566e-06, | |
| "loss": 0.02482418715953827, | |
| "step": 4480 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9920959538224017, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9920959538224017, | |
| "eval_loss": 0.045616794377565384, | |
| "eval_mean_accuracy": 0.9920959538224017, | |
| "eval_mean_iou": 0.49604797691120084, | |
| "eval_overall_accuracy": 0.9920959538224017, | |
| "eval_runtime": 2.8701, | |
| "eval_samples_per_second": 55.05, | |
| "eval_steps_per_second": 6.968, | |
| "step": 4480 | |
| }, | |
| { | |
| "epoch": 40.089285714285715, | |
| "grad_norm": 0.3945085406303406, | |
| "learning_rate": 6.910848004907583e-06, | |
| "loss": 0.026185140013694763, | |
| "step": 4490 | |
| }, | |
| { | |
| "epoch": 40.17857142857143, | |
| "grad_norm": 0.6535700559616089, | |
| "learning_rate": 6.791901848429387e-06, | |
| "loss": 0.02537115216255188, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 40.267857142857146, | |
| "grad_norm": 0.5408845543861389, | |
| "learning_rate": 6.673857422574499e-06, | |
| "loss": 0.025524777173995972, | |
| "step": 4510 | |
| }, | |
| { | |
| "epoch": 40.357142857142854, | |
| "grad_norm": 0.6645027995109558, | |
| "learning_rate": 6.556719313857219e-06, | |
| "loss": 0.0288840651512146, | |
| "step": 4520 | |
| }, | |
| { | |
| "epoch": 40.44642857142857, | |
| "grad_norm": 0.2951991856098175, | |
| "learning_rate": 6.440492073577659e-06, | |
| "loss": 0.02452797442674637, | |
| "step": 4530 | |
| }, | |
| { | |
| "epoch": 40.535714285714285, | |
| "grad_norm": 0.4799725413322449, | |
| "learning_rate": 6.325180217644914e-06, | |
| "loss": 0.024873518943786622, | |
| "step": 4540 | |
| }, | |
| { | |
| "epoch": 40.625, | |
| "grad_norm": 0.3771775960922241, | |
| "learning_rate": 6.2107882264016446e-06, | |
| "loss": 0.025704348087310792, | |
| "step": 4550 | |
| }, | |
| { | |
| "epoch": 40.714285714285715, | |
| "grad_norm": 0.36452680826187134, | |
| "learning_rate": 6.097320544449921e-06, | |
| "loss": 0.02450278103351593, | |
| "step": 4560 | |
| }, | |
| { | |
| "epoch": 40.80357142857143, | |
| "grad_norm": 0.7826708555221558, | |
| "learning_rate": 5.984781580478611e-06, | |
| "loss": 0.02239405959844589, | |
| "step": 4570 | |
| }, | |
| { | |
| "epoch": 40.892857142857146, | |
| "grad_norm": 0.4126140773296356, | |
| "learning_rate": 5.873175707092019e-06, | |
| "loss": 0.029208868741989136, | |
| "step": 4580 | |
| }, | |
| { | |
| "epoch": 40.982142857142854, | |
| "grad_norm": 0.4714967906475067, | |
| "learning_rate": 5.7625072606400375e-06, | |
| "loss": 0.02443849891424179, | |
| "step": 4590 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9904924004274125, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9904924004274125, | |
| "eval_loss": 0.04487628862261772, | |
| "eval_mean_accuracy": 0.9904924004274125, | |
| "eval_mean_iou": 0.49524620021370624, | |
| "eval_overall_accuracy": 0.9904924004274125, | |
| "eval_runtime": 3.1369, | |
| "eval_samples_per_second": 50.367, | |
| "eval_steps_per_second": 6.376, | |
| "step": 4592 | |
| }, | |
| { | |
| "epoch": 41.07142857142857, | |
| "grad_norm": 0.8939480781555176, | |
| "learning_rate": 5.652780541049631e-06, | |
| "loss": 0.024046406149864197, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 41.160714285714285, | |
| "grad_norm": 0.9223622679710388, | |
| "learning_rate": 5.543999811657797e-06, | |
| "loss": 0.026213526725769043, | |
| "step": 4610 | |
| }, | |
| { | |
| "epoch": 41.25, | |
| "grad_norm": 0.2867562770843506, | |
| "learning_rate": 5.436169299045876e-06, | |
| "loss": 0.028421574831008913, | |
| "step": 4620 | |
| }, | |
| { | |
| "epoch": 41.339285714285715, | |
| "grad_norm": 0.5164304375648499, | |
| "learning_rate": 5.329293192875387e-06, | |
| "loss": 0.02325395345687866, | |
| "step": 4630 | |
| }, | |
| { | |
| "epoch": 41.42857142857143, | |
| "grad_norm": 0.34760358929634094, | |
| "learning_rate": 5.223375645725183e-06, | |
| "loss": 0.02278565615415573, | |
| "step": 4640 | |
| }, | |
| { | |
| "epoch": 41.517857142857146, | |
| "grad_norm": 0.3044535219669342, | |
| "learning_rate": 5.118420772930138e-06, | |
| "loss": 0.025249192118644716, | |
| "step": 4650 | |
| }, | |
| { | |
| "epoch": 41.607142857142854, | |
| "grad_norm": 0.7161318063735962, | |
| "learning_rate": 5.01443265242126e-06, | |
| "loss": 0.025399017333984374, | |
| "step": 4660 | |
| }, | |
| { | |
| "epoch": 41.69642857142857, | |
| "grad_norm": 0.48156532645225525, | |
| "learning_rate": 4.911415324567213e-06, | |
| "loss": 0.02383261322975159, | |
| "step": 4670 | |
| }, | |
| { | |
| "epoch": 41.785714285714285, | |
| "grad_norm": 0.5352237224578857, | |
| "learning_rate": 4.809372792017372e-06, | |
| "loss": 0.02105867862701416, | |
| "step": 4680 | |
| }, | |
| { | |
| "epoch": 41.875, | |
| "grad_norm": 0.5803747773170471, | |
| "learning_rate": 4.708309019546263e-06, | |
| "loss": 0.02634103000164032, | |
| "step": 4690 | |
| }, | |
| { | |
| "epoch": 41.964285714285715, | |
| "grad_norm": 0.47478434443473816, | |
| "learning_rate": 4.608227933899527e-06, | |
| "loss": 0.023492810130119324, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9922144076200919, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9922144076200919, | |
| "eval_loss": 0.044545359909534454, | |
| "eval_mean_accuracy": 0.9922144076200919, | |
| "eval_mean_iou": 0.49610720381004597, | |
| "eval_overall_accuracy": 0.9922144076200919, | |
| "eval_runtime": 3.0802, | |
| "eval_samples_per_second": 51.295, | |
| "eval_steps_per_second": 6.493, | |
| "step": 4704 | |
| }, | |
| { | |
| "epoch": 42.05357142857143, | |
| "grad_norm": 0.4991433918476105, | |
| "learning_rate": 4.509133423641388e-06, | |
| "loss": 0.028053835034370422, | |
| "step": 4710 | |
| }, | |
| { | |
| "epoch": 42.142857142857146, | |
| "grad_norm": 0.5921125411987305, | |
| "learning_rate": 4.4110293390035e-06, | |
| "loss": 0.024588567018508912, | |
| "step": 4720 | |
| }, | |
| { | |
| "epoch": 42.232142857142854, | |
| "grad_norm": 0.4289485216140747, | |
| "learning_rate": 4.313919491735413e-06, | |
| "loss": 0.026377671957015993, | |
| "step": 4730 | |
| }, | |
| { | |
| "epoch": 42.32142857142857, | |
| "grad_norm": 0.2543030381202698, | |
| "learning_rate": 4.217807654956419e-06, | |
| "loss": 0.02101795971393585, | |
| "step": 4740 | |
| }, | |
| { | |
| "epoch": 42.410714285714285, | |
| "grad_norm": 0.46004971861839294, | |
| "learning_rate": 4.122697563008991e-06, | |
| "loss": 0.023616319894790648, | |
| "step": 4750 | |
| }, | |
| { | |
| "epoch": 42.5, | |
| "grad_norm": 0.4452957808971405, | |
| "learning_rate": 4.028592911313669e-06, | |
| "loss": 0.02816382348537445, | |
| "step": 4760 | |
| }, | |
| { | |
| "epoch": 42.589285714285715, | |
| "grad_norm": 1.8105047941207886, | |
| "learning_rate": 3.935497356225467e-06, | |
| "loss": 0.03450767695903778, | |
| "step": 4770 | |
| }, | |
| { | |
| "epoch": 42.67857142857143, | |
| "grad_norm": 0.3400633633136749, | |
| "learning_rate": 3.843414514891828e-06, | |
| "loss": 0.02481631338596344, | |
| "step": 4780 | |
| }, | |
| { | |
| "epoch": 42.767857142857146, | |
| "grad_norm": 0.3335254490375519, | |
| "learning_rate": 3.752347965112084e-06, | |
| "loss": 0.02071080356836319, | |
| "step": 4790 | |
| }, | |
| { | |
| "epoch": 42.857142857142854, | |
| "grad_norm": 0.36557644605636597, | |
| "learning_rate": 3.6623012451984194e-06, | |
| "loss": 0.024737077951431274, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 42.94642857142857, | |
| "grad_norm": 0.5764870047569275, | |
| "learning_rate": 3.5732778538384336e-06, | |
| "loss": 0.024080076813697816, | |
| "step": 4810 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9919902823453646, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9919902823453646, | |
| "eval_loss": 0.04448164626955986, | |
| "eval_mean_accuracy": 0.9919902823453646, | |
| "eval_mean_iou": 0.4959951411726823, | |
| "eval_overall_accuracy": 0.9919902823453646, | |
| "eval_runtime": 3.171, | |
| "eval_samples_per_second": 49.827, | |
| "eval_steps_per_second": 6.307, | |
| "step": 4816 | |
| }, | |
| { | |
| "epoch": 43.035714285714285, | |
| "grad_norm": 0.4781723916530609, | |
| "learning_rate": 3.485281249959158e-06, | |
| "loss": 0.026803573966026305, | |
| "step": 4820 | |
| }, | |
| { | |
| "epoch": 43.125, | |
| "grad_norm": 0.3812042772769928, | |
| "learning_rate": 3.3983148525926798e-06, | |
| "loss": 0.024715760350227357, | |
| "step": 4830 | |
| }, | |
| { | |
| "epoch": 43.214285714285715, | |
| "grad_norm": 0.4324609041213989, | |
| "learning_rate": 3.312382040743319e-06, | |
| "loss": 0.027126455307006837, | |
| "step": 4840 | |
| }, | |
| { | |
| "epoch": 43.30357142857143, | |
| "grad_norm": 0.8323965668678284, | |
| "learning_rate": 3.2274861532563093e-06, | |
| "loss": 0.02302350103855133, | |
| "step": 4850 | |
| }, | |
| { | |
| "epoch": 43.392857142857146, | |
| "grad_norm": 0.40330275893211365, | |
| "learning_rate": 3.143630488688095e-06, | |
| "loss": 0.024910798668861388, | |
| "step": 4860 | |
| }, | |
| { | |
| "epoch": 43.482142857142854, | |
| "grad_norm": 0.7080916166305542, | |
| "learning_rate": 3.060818305178145e-06, | |
| "loss": 0.024807298183441163, | |
| "step": 4870 | |
| }, | |
| { | |
| "epoch": 43.57142857142857, | |
| "grad_norm": 2.711069107055664, | |
| "learning_rate": 2.979052820322364e-06, | |
| "loss": 0.024544814229011537, | |
| "step": 4880 | |
| }, | |
| { | |
| "epoch": 43.660714285714285, | |
| "grad_norm": 7.1795220375061035, | |
| "learning_rate": 2.898337211048119e-06, | |
| "loss": 0.0320773184299469, | |
| "step": 4890 | |
| }, | |
| { | |
| "epoch": 43.75, | |
| "grad_norm": 0.5481147766113281, | |
| "learning_rate": 2.8186746134907225e-06, | |
| "loss": 0.02236861139535904, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 43.839285714285715, | |
| "grad_norm": 0.5208039879798889, | |
| "learning_rate": 2.7400681228716583e-06, | |
| "loss": 0.025225913524627684, | |
| "step": 4910 | |
| }, | |
| { | |
| "epoch": 43.92857142857143, | |
| "grad_norm": 0.8581423163414001, | |
| "learning_rate": 2.6625207933782827e-06, | |
| "loss": 0.029476696252822877, | |
| "step": 4920 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9918761107860795, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9918761107860795, | |
| "eval_loss": 0.044531892985105515, | |
| "eval_mean_accuracy": 0.9918761107860795, | |
| "eval_mean_iou": 0.49593805539303976, | |
| "eval_overall_accuracy": 0.9918761107860795, | |
| "eval_runtime": 2.9802, | |
| "eval_samples_per_second": 53.017, | |
| "eval_steps_per_second": 6.711, | |
| "step": 4928 | |
| }, | |
| { | |
| "epoch": 44.017857142857146, | |
| "grad_norm": 0.33332762122154236, | |
| "learning_rate": 2.5860356380451466e-06, | |
| "loss": 0.021295398473739624, | |
| "step": 4930 | |
| }, | |
| { | |
| "epoch": 44.107142857142854, | |
| "grad_norm": 0.9368950128555298, | |
| "learning_rate": 2.5106156286369418e-06, | |
| "loss": 0.024914734065532684, | |
| "step": 4940 | |
| }, | |
| { | |
| "epoch": 44.19642857142857, | |
| "grad_norm": 0.4496859610080719, | |
| "learning_rate": 2.4362636955330543e-06, | |
| "loss": 0.02527157962322235, | |
| "step": 4950 | |
| }, | |
| { | |
| "epoch": 44.285714285714285, | |
| "grad_norm": 0.6395953893661499, | |
| "learning_rate": 2.362982727613658e-06, | |
| "loss": 0.023759710788726806, | |
| "step": 4960 | |
| }, | |
| { | |
| "epoch": 44.375, | |
| "grad_norm": 0.7984117269515991, | |
| "learning_rate": 2.2907755721475286e-06, | |
| "loss": 0.022670991718769073, | |
| "step": 4970 | |
| }, | |
| { | |
| "epoch": 44.464285714285715, | |
| "grad_norm": 0.4340791404247284, | |
| "learning_rate": 2.219645034681367e-06, | |
| "loss": 0.024120573699474335, | |
| "step": 4980 | |
| }, | |
| { | |
| "epoch": 44.55357142857143, | |
| "grad_norm": 0.5286053419113159, | |
| "learning_rate": 2.1495938789308055e-06, | |
| "loss": 0.02684595584869385, | |
| "step": 4990 | |
| }, | |
| { | |
| "epoch": 44.642857142857146, | |
| "grad_norm": 0.5050071477890015, | |
| "learning_rate": 2.0806248266730554e-06, | |
| "loss": 0.026467105746269225, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 44.732142857142854, | |
| "grad_norm": 0.2880910038948059, | |
| "learning_rate": 2.012740557641105e-06, | |
| "loss": 0.021798455715179445, | |
| "step": 5010 | |
| }, | |
| { | |
| "epoch": 44.82142857142857, | |
| "grad_norm": 0.3844790458679199, | |
| "learning_rate": 1.945943709419652e-06, | |
| "loss": 0.02199341207742691, | |
| "step": 5020 | |
| }, | |
| { | |
| "epoch": 44.910714285714285, | |
| "grad_norm": 0.5376768112182617, | |
| "learning_rate": 1.8802368773425739e-06, | |
| "loss": 0.025217828154563905, | |
| "step": 5030 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "grad_norm": 0.8637105822563171, | |
| "learning_rate": 1.8156226143921162e-06, | |
| "loss": 0.025202780961990356, | |
| "step": 5040 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9919142645643516, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9919142645643516, | |
| "eval_loss": 0.04428554326295853, | |
| "eval_mean_accuracy": 0.9919142645643516, | |
| "eval_mean_iou": 0.4959571322821758, | |
| "eval_overall_accuracy": 0.9919142645643516, | |
| "eval_runtime": 3.2068, | |
| "eval_samples_per_second": 49.271, | |
| "eval_steps_per_second": 6.237, | |
| "step": 5040 | |
| }, | |
| { | |
| "epoch": 45.089285714285715, | |
| "grad_norm": 0.36045849323272705, | |
| "learning_rate": 1.7521034310997053e-06, | |
| "loss": 0.023151762783527374, | |
| "step": 5050 | |
| }, | |
| { | |
| "epoch": 45.17857142857143, | |
| "grad_norm": 0.5583776831626892, | |
| "learning_rate": 1.689681795448389e-06, | |
| "loss": 0.027086129784584044, | |
| "step": 5060 | |
| }, | |
| { | |
| "epoch": 45.267857142857146, | |
| "grad_norm": 0.28848934173583984, | |
| "learning_rate": 1.6283601327769371e-06, | |
| "loss": 0.024879464507102968, | |
| "step": 5070 | |
| }, | |
| { | |
| "epoch": 45.357142857142854, | |
| "grad_norm": 0.5597347021102905, | |
| "learning_rate": 1.56814082568564e-06, | |
| "loss": 0.025781130790710448, | |
| "step": 5080 | |
| }, | |
| { | |
| "epoch": 45.44642857142857, | |
| "grad_norm": 0.42186370491981506, | |
| "learning_rate": 1.5090262139436896e-06, | |
| "loss": 0.026576164364814758, | |
| "step": 5090 | |
| }, | |
| { | |
| "epoch": 45.535714285714285, | |
| "grad_norm": 0.6118794083595276, | |
| "learning_rate": 1.4510185943983112e-06, | |
| "loss": 0.026358428597450256, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 45.625, | |
| "grad_norm": 0.46963897347450256, | |
| "learning_rate": 1.3941202208854965e-06, | |
| "loss": 0.025634509325027467, | |
| "step": 5110 | |
| }, | |
| { | |
| "epoch": 45.714285714285715, | |
| "grad_norm": 0.8034435510635376, | |
| "learning_rate": 1.338333304142435e-06, | |
| "loss": 0.02395763248205185, | |
| "step": 5120 | |
| }, | |
| { | |
| "epoch": 45.80357142857143, | |
| "grad_norm": 0.6469313502311707, | |
| "learning_rate": 1.2836600117216346e-06, | |
| "loss": 0.024726195633411406, | |
| "step": 5130 | |
| }, | |
| { | |
| "epoch": 45.892857142857146, | |
| "grad_norm": 0.4325697124004364, | |
| "learning_rate": 1.2301024679066808e-06, | |
| "loss": 0.02584630250930786, | |
| "step": 5140 | |
| }, | |
| { | |
| "epoch": 45.982142857142854, | |
| "grad_norm": 0.39687296748161316, | |
| "learning_rate": 1.177662753629717e-06, | |
| "loss": 0.021341803669929504, | |
| "step": 5150 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9921635681130104, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9921635681130104, | |
| "eval_loss": 0.044276703149080276, | |
| "eval_mean_accuracy": 0.9921635681130104, | |
| "eval_mean_iou": 0.4960817840565052, | |
| "eval_overall_accuracy": 0.9921635681130104, | |
| "eval_runtime": 3.1403, | |
| "eval_samples_per_second": 50.313, | |
| "eval_steps_per_second": 6.369, | |
| "step": 5152 | |
| }, | |
| { | |
| "epoch": 46.07142857142857, | |
| "grad_norm": 0.5137065649032593, | |
| "learning_rate": 1.126342906390585e-06, | |
| "loss": 0.02478640079498291, | |
| "step": 5160 | |
| }, | |
| { | |
| "epoch": 46.160714285714285, | |
| "grad_norm": 0.5068938136100769, | |
| "learning_rate": 1.0761449201776464e-06, | |
| "loss": 0.022511668503284454, | |
| "step": 5170 | |
| }, | |
| { | |
| "epoch": 46.25, | |
| "grad_norm": 0.49176064133644104, | |
| "learning_rate": 1.0270707453903428e-06, | |
| "loss": 0.025940099358558656, | |
| "step": 5180 | |
| }, | |
| { | |
| "epoch": 46.339285714285715, | |
| "grad_norm": 0.4621039927005768, | |
| "learning_rate": 9.791222887633711e-07, | |
| "loss": 0.02201744318008423, | |
| "step": 5190 | |
| }, | |
| { | |
| "epoch": 46.42857142857143, | |
| "grad_norm": 3.4700331687927246, | |
| "learning_rate": 9.323014132926355e-07, | |
| "loss": 0.03096678853034973, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 46.517857142857146, | |
| "grad_norm": 0.49781036376953125, | |
| "learning_rate": 8.86609938162839e-07, | |
| "loss": 0.02691877484321594, | |
| "step": 5210 | |
| }, | |
| { | |
| "epoch": 46.607142857142854, | |
| "grad_norm": 0.5014402866363525, | |
| "learning_rate": 8.420496386768195e-07, | |
| "loss": 0.02660224735736847, | |
| "step": 5220 | |
| }, | |
| { | |
| "epoch": 46.69642857142857, | |
| "grad_norm": 0.5916736125946045, | |
| "learning_rate": 7.986222461865455e-07, | |
| "loss": 0.021325075626373292, | |
| "step": 5230 | |
| }, | |
| { | |
| "epoch": 46.785714285714285, | |
| "grad_norm": 0.4295494258403778, | |
| "learning_rate": 7.563294480258798e-07, | |
| "loss": 0.019958339631557465, | |
| "step": 5240 | |
| }, | |
| { | |
| "epoch": 46.875, | |
| "grad_norm": 0.30950236320495605, | |
| "learning_rate": 7.151728874449848e-07, | |
| "loss": 0.026697978377342224, | |
| "step": 5250 | |
| }, | |
| { | |
| "epoch": 46.964285714285715, | |
| "grad_norm": 0.481393426656723, | |
| "learning_rate": 6.751541635465075e-07, | |
| "loss": 0.023805753886699678, | |
| "step": 5260 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9916805123025328, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9916805123025328, | |
| "eval_loss": 0.04457485303282738, | |
| "eval_mean_accuracy": 0.9916805123025328, | |
| "eval_mean_iou": 0.4958402561512664, | |
| "eval_overall_accuracy": 0.9916805123025328, | |
| "eval_runtime": 3.1185, | |
| "eval_samples_per_second": 50.666, | |
| "eval_steps_per_second": 6.413, | |
| "step": 5264 | |
| }, | |
| { | |
| "epoch": 47.05357142857143, | |
| "grad_norm": 0.4641105830669403, | |
| "learning_rate": 6.362748312234223e-07, | |
| "loss": 0.02568671405315399, | |
| "step": 5270 | |
| }, | |
| { | |
| "epoch": 47.142857142857146, | |
| "grad_norm": 0.5612961649894714, | |
| "learning_rate": 5.985364010986261e-07, | |
| "loss": 0.02504618465900421, | |
| "step": 5280 | |
| }, | |
| { | |
| "epoch": 47.232142857142854, | |
| "grad_norm": 0.7722740173339844, | |
| "learning_rate": 5.619403394662614e-07, | |
| "loss": 0.028320273756980895, | |
| "step": 5290 | |
| }, | |
| { | |
| "epoch": 47.32142857142857, | |
| "grad_norm": 0.5712404251098633, | |
| "learning_rate": 5.264880682347095e-07, | |
| "loss": 0.025172898173332216, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 47.410714285714285, | |
| "grad_norm": 0.448178231716156, | |
| "learning_rate": 4.921809648713737e-07, | |
| "loss": 0.025293144583702087, | |
| "step": 5310 | |
| }, | |
| { | |
| "epoch": 47.5, | |
| "grad_norm": 0.8815834522247314, | |
| "learning_rate": 4.590203623491429e-07, | |
| "loss": 0.025543418526649476, | |
| "step": 5320 | |
| }, | |
| { | |
| "epoch": 47.589285714285715, | |
| "grad_norm": 0.6897329688072205, | |
| "learning_rate": 4.270075490946124e-07, | |
| "loss": 0.023450803756713868, | |
| "step": 5330 | |
| }, | |
| { | |
| "epoch": 47.67857142857143, | |
| "grad_norm": 0.2958068549633026, | |
| "learning_rate": 3.9614376893800763e-07, | |
| "loss": 0.021414244174957277, | |
| "step": 5340 | |
| }, | |
| { | |
| "epoch": 47.767857142857146, | |
| "grad_norm": 0.438213586807251, | |
| "learning_rate": 3.664302210648729e-07, | |
| "loss": 0.02141619473695755, | |
| "step": 5350 | |
| }, | |
| { | |
| "epoch": 47.857142857142854, | |
| "grad_norm": 0.6407127976417542, | |
| "learning_rate": 3.378680599694617e-07, | |
| "loss": 0.02390940934419632, | |
| "step": 5360 | |
| }, | |
| { | |
| "epoch": 47.94642857142857, | |
| "grad_norm": 0.6588020324707031, | |
| "learning_rate": 3.1045839540989273e-07, | |
| "loss": 0.023431585729122163, | |
| "step": 5370 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9918207636596241, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9918207636596241, | |
| "eval_loss": 0.04451696202158928, | |
| "eval_mean_accuracy": 0.9918207636596241, | |
| "eval_mean_iou": 0.49591038182981206, | |
| "eval_overall_accuracy": 0.9918207636596241, | |
| "eval_runtime": 3.1745, | |
| "eval_samples_per_second": 49.771, | |
| "eval_steps_per_second": 6.3, | |
| "step": 5376 | |
| }, | |
| { | |
| "epoch": 48.035714285714285, | |
| "grad_norm": 0.23221944272518158, | |
| "learning_rate": 2.842022923650278e-07, | |
| "loss": 0.026059791445732117, | |
| "step": 5380 | |
| }, | |
| { | |
| "epoch": 48.125, | |
| "grad_norm": 0.4213135540485382, | |
| "learning_rate": 2.5910077099309147e-07, | |
| "loss": 0.023841065168380738, | |
| "step": 5390 | |
| }, | |
| { | |
| "epoch": 48.214285714285715, | |
| "grad_norm": 0.47122108936309814, | |
| "learning_rate": 2.3515480659202616e-07, | |
| "loss": 0.025115084648132325, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 48.30357142857143, | |
| "grad_norm": 0.7628607749938965, | |
| "learning_rate": 2.123653295616157e-07, | |
| "loss": 0.02465710639953613, | |
| "step": 5410 | |
| }, | |
| { | |
| "epoch": 48.392857142857146, | |
| "grad_norm": 0.4654390513896942, | |
| "learning_rate": 1.9073322536732107e-07, | |
| "loss": 0.023554743826389314, | |
| "step": 5420 | |
| }, | |
| { | |
| "epoch": 48.482142857142854, | |
| "grad_norm": 0.433553546667099, | |
| "learning_rate": 1.7025933450587784e-07, | |
| "loss": 0.025577667355537414, | |
| "step": 5430 | |
| }, | |
| { | |
| "epoch": 48.57142857142857, | |
| "grad_norm": 0.8381280303001404, | |
| "learning_rate": 1.509444524726489e-07, | |
| "loss": 0.023599877953529358, | |
| "step": 5440 | |
| }, | |
| { | |
| "epoch": 48.660714285714285, | |
| "grad_norm": 0.5404291749000549, | |
| "learning_rate": 1.3278932973070257e-07, | |
| "loss": 0.02920941114425659, | |
| "step": 5450 | |
| }, | |
| { | |
| "epoch": 48.75, | |
| "grad_norm": 0.5444062352180481, | |
| "learning_rate": 1.1579467168166269e-07, | |
| "loss": 0.025456267595291137, | |
| "step": 5460 | |
| }, | |
| { | |
| "epoch": 48.839285714285715, | |
| "grad_norm": 0.489740252494812, | |
| "learning_rate": 9.996113863830369e-08, | |
| "loss": 0.023142220079898836, | |
| "step": 5470 | |
| }, | |
| { | |
| "epoch": 48.92857142857143, | |
| "grad_norm": 0.40527722239494324, | |
| "learning_rate": 8.528934579888459e-08, | |
| "loss": 0.02229561060667038, | |
| "step": 5480 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.9914031971191933, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.9914031971191933, | |
| "eval_loss": 0.04447474330663681, | |
| "eval_mean_accuracy": 0.9914031971191933, | |
| "eval_mean_iou": 0.49570159855959667, | |
| "eval_overall_accuracy": 0.9914031971191933, | |
| "eval_runtime": 3.066, | |
| "eval_samples_per_second": 51.533, | |
| "eval_steps_per_second": 6.523, | |
| "step": 5488 | |
| }, | |
| { | |
| "epoch": 49.017857142857146, | |
| "grad_norm": 1.488918662071228, | |
| "learning_rate": 7.177986322325137e-08, | |
| "loss": 0.026862525939941408, | |
| "step": 5490 | |
| }, | |
| { | |
| "epoch": 49.107142857142854, | |
| "grad_norm": 0.9656788110733032, | |
| "learning_rate": 5.943321581069139e-08, | |
| "loss": 0.027856603264808655, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 49.19642857142857, | |
| "grad_norm": 1.4815126657485962, | |
| "learning_rate": 4.824988327953639e-08, | |
| "loss": 0.027299723029136656, | |
| "step": 5510 | |
| }, | |
| { | |
| "epoch": 49.285714285714285, | |
| "grad_norm": 0.2918512225151062, | |
| "learning_rate": 3.823030014851403e-08, | |
| "loss": 0.021877503395080565, | |
| "step": 5520 | |
| }, | |
| { | |
| "epoch": 49.375, | |
| "grad_norm": 0.4252746105194092, | |
| "learning_rate": 2.9374855719884742e-08, | |
| "loss": 0.026218003034591673, | |
| "step": 5530 | |
| }, | |
| { | |
| "epoch": 49.464285714285715, | |
| "grad_norm": 0.2831713557243347, | |
| "learning_rate": 2.168389406429716e-08, | |
| "loss": 0.025749343633651733, | |
| "step": 5540 | |
| }, | |
| { | |
| "epoch": 49.55357142857143, | |
| "grad_norm": 0.4766254127025604, | |
| "learning_rate": 1.5157714007425495e-08, | |
| "loss": 0.023514069616794586, | |
| "step": 5550 | |
| }, | |
| { | |
| "epoch": 49.642857142857146, | |
| "grad_norm": 0.9032825827598572, | |
| "learning_rate": 9.796569118368793e-09, | |
| "loss": 0.027806323766708375, | |
| "step": 5560 | |
| }, | |
| { | |
| "epoch": 49.732142857142854, | |
| "grad_norm": 0.3083033263683319, | |
| "learning_rate": 5.60066769977885e-09, | |
| "loss": 0.022344766557216643, | |
| "step": 5570 | |
| }, | |
| { | |
| "epoch": 49.82142857142857, | |
| "grad_norm": 0.4118826985359192, | |
| "learning_rate": 2.570172779789992e-09, | |
| "loss": 0.02296028733253479, | |
| "step": 5580 | |
| }, | |
| { | |
| "epoch": 49.910714285714285, | |
| "grad_norm": 0.44650617241859436, | |
| "learning_rate": 7.052021056608293e-10, | |
| "loss": 0.02406269907951355, | |
| "step": 5590 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "grad_norm": 0.5170707106590271, | |
| "learning_rate": 5.8281392212311545e-12, | |
| "loss": 0.02451837360858917, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_accuracy_background": NaN, | |
| "eval_accuracy_crop": 0.991746323545392, | |
| "eval_iou_background": 0.0, | |
| "eval_iou_crop": 0.991746323545392, | |
| "eval_loss": 0.04466014727950096, | |
| "eval_mean_accuracy": 0.991746323545392, | |
| "eval_mean_iou": 0.495873161772696, | |
| "eval_overall_accuracy": 0.991746323545392, | |
| "eval_runtime": 2.9949, | |
| "eval_samples_per_second": 52.756, | |
| "eval_steps_per_second": 6.678, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "step": 5600, | |
| "total_flos": 7.834996118126592e+17, | |
| "train_loss": 0.05321301036647388, | |
| "train_runtime": 744.0706, | |
| "train_samples_per_second": 60.075, | |
| "train_steps_per_second": 7.526 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 5600, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.834996118126592e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |