| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.120313143798929, | |
| "eval_steps": 500, | |
| "global_step": 10000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04120313143798929, | |
| "grad_norm": 2.6929988861083984, | |
| "learning_rate": 5.9401e-05, | |
| "loss": 1.2381, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.08240626287597858, | |
| "grad_norm": 0.9781515002250671, | |
| "learning_rate": 5.8802000000000004e-05, | |
| "loss": 0.5408, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.12360939431396786, | |
| "grad_norm": 1.4702354669570923, | |
| "learning_rate": 5.8203e-05, | |
| "loss": 0.3679, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.16481252575195715, | |
| "grad_norm": 0.458916038274765, | |
| "learning_rate": 5.7604e-05, | |
| "loss": 0.4243, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.20601565718994644, | |
| "grad_norm": 2.584094524383545, | |
| "learning_rate": 5.7005e-05, | |
| "loss": 0.341, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.24721878862793573, | |
| "grad_norm": 4.845738410949707, | |
| "learning_rate": 5.6406e-05, | |
| "loss": 0.3247, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.288421920065925, | |
| "grad_norm": 0.6360086798667908, | |
| "learning_rate": 5.5806999999999996e-05, | |
| "loss": 0.3736, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.3296250515039143, | |
| "grad_norm": 0.5916720628738403, | |
| "learning_rate": 5.5208000000000004e-05, | |
| "loss": 0.3775, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.37082818294190356, | |
| "grad_norm": 1.5302131175994873, | |
| "learning_rate": 5.4609000000000005e-05, | |
| "loss": 0.3549, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.4120313143798929, | |
| "grad_norm": 1.928609848022461, | |
| "learning_rate": 5.401e-05, | |
| "loss": 0.311, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.45323444581788214, | |
| "grad_norm": 4.012447834014893, | |
| "learning_rate": 5.3411e-05, | |
| "loss": 0.3266, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.49443757725587145, | |
| "grad_norm": 1.2283966541290283, | |
| "learning_rate": 5.2812e-05, | |
| "loss": 0.3581, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.5356407086938607, | |
| "grad_norm": 2.1378538608551025, | |
| "learning_rate": 5.2213e-05, | |
| "loss": 0.3654, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.57684384013185, | |
| "grad_norm": 1.014821171760559, | |
| "learning_rate": 5.1614000000000004e-05, | |
| "loss": 0.3019, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.6180469715698393, | |
| "grad_norm": 0.6981366872787476, | |
| "learning_rate": 5.1015e-05, | |
| "loss": 0.297, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.6592501030078286, | |
| "grad_norm": 0.8797981142997742, | |
| "learning_rate": 5.0416e-05, | |
| "loss": 0.3543, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.7004532344458179, | |
| "grad_norm": 0.34348738193511963, | |
| "learning_rate": 4.9817e-05, | |
| "loss": 0.3102, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.7416563658838071, | |
| "grad_norm": 0.9645235538482666, | |
| "learning_rate": 4.9218e-05, | |
| "loss": 0.2859, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.7828594973217965, | |
| "grad_norm": 1.9135812520980835, | |
| "learning_rate": 4.8619e-05, | |
| "loss": 0.3493, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.8240626287597858, | |
| "grad_norm": 1.7853527069091797, | |
| "learning_rate": 4.8020000000000004e-05, | |
| "loss": 0.3256, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.865265760197775, | |
| "grad_norm": 1.7780035734176636, | |
| "learning_rate": 4.7421000000000006e-05, | |
| "loss": 0.2621, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.9064688916357643, | |
| "grad_norm": 0.8148425221443176, | |
| "learning_rate": 4.6822e-05, | |
| "loss": 0.3273, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.9476720230737536, | |
| "grad_norm": 2.2365009784698486, | |
| "learning_rate": 4.6223e-05, | |
| "loss": 0.2879, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.9888751545117429, | |
| "grad_norm": 1.7118935585021973, | |
| "learning_rate": 4.5624e-05, | |
| "loss": 0.2715, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy_Autre r\u00e9paration": 0.2533258414054248, | |
| "eval_accuracy_D\u00e9coupe": 0.08133862998794246, | |
| "eval_accuracy_Emergence": 0.0, | |
| "eval_accuracy_Gla\u00e7age ou Ressuage": 0.0, | |
| "eval_accuracy_Reflet m\u00e9t\u00e9o": 0.0, | |
| "eval_accuracy_Unlabeled": NaN, | |
| "eval_iou_Autre r\u00e9paration": 0.23621639872040598, | |
| "eval_iou_D\u00e9coupe": 0.07661059644544693, | |
| "eval_iou_Emergence": 0.0, | |
| "eval_iou_Gla\u00e7age ou Ressuage": 0.0, | |
| "eval_iou_Reflet m\u00e9t\u00e9o": 0.0, | |
| "eval_iou_Unlabeled": 0.0, | |
| "eval_loss": 0.2682347893714905, | |
| "eval_mean_accuracy": 0.06693289427867345, | |
| "eval_mean_iou": 0.05213783252764215, | |
| "eval_overall_accuracy": 0.18279080675579093, | |
| "eval_runtime": 150.5227, | |
| "eval_samples_per_second": 12.045, | |
| "eval_steps_per_second": 3.016, | |
| "step": 2427 | |
| }, | |
| { | |
| "epoch": 1.0300782859497322, | |
| "grad_norm": 3.537090539932251, | |
| "learning_rate": 4.5025000000000003e-05, | |
| "loss": 0.2763, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.0712814173877214, | |
| "grad_norm": 2.1730239391326904, | |
| "learning_rate": 4.4426000000000005e-05, | |
| "loss": 0.2981, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 1.1124845488257107, | |
| "grad_norm": 1.0320223569869995, | |
| "learning_rate": 4.3827e-05, | |
| "loss": 0.3227, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 1.1536876802637002, | |
| "grad_norm": 4.7768635749816895, | |
| "learning_rate": 4.3228e-05, | |
| "loss": 0.3398, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.1948908117016894, | |
| "grad_norm": 1.5758723020553589, | |
| "learning_rate": 4.2629e-05, | |
| "loss": 0.334, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 1.2360939431396787, | |
| "grad_norm": 4.915160655975342, | |
| "learning_rate": 4.203e-05, | |
| "loss": 0.2577, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.277297074577668, | |
| "grad_norm": 0.7495476603507996, | |
| "learning_rate": 4.1431e-05, | |
| "loss": 0.2807, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 1.3185002060156572, | |
| "grad_norm": 1.0287623405456543, | |
| "learning_rate": 4.0832e-05, | |
| "loss": 0.3277, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 1.3597033374536465, | |
| "grad_norm": 3.6160237789154053, | |
| "learning_rate": 4.0233e-05, | |
| "loss": 0.3073, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 1.4009064688916357, | |
| "grad_norm": 6.738962173461914, | |
| "learning_rate": 3.9634e-05, | |
| "loss": 0.2744, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 1.442109600329625, | |
| "grad_norm": 0.7060651779174805, | |
| "learning_rate": 3.9035e-05, | |
| "loss": 0.2976, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.4833127317676142, | |
| "grad_norm": 4.404435634613037, | |
| "learning_rate": 3.8436e-05, | |
| "loss": 0.2646, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 1.5245158632056035, | |
| "grad_norm": 1.1246055364608765, | |
| "learning_rate": 3.7837000000000004e-05, | |
| "loss": 0.3497, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 1.5657189946435928, | |
| "grad_norm": 3.132385015487671, | |
| "learning_rate": 3.7238000000000005e-05, | |
| "loss": 0.2437, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 1.6069221260815822, | |
| "grad_norm": 0.3945494592189789, | |
| "learning_rate": 3.6639e-05, | |
| "loss": 0.2616, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 1.6481252575195715, | |
| "grad_norm": 0.8652153015136719, | |
| "learning_rate": 3.604e-05, | |
| "loss": 0.2466, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.6893283889575608, | |
| "grad_norm": 0.44899633526802063, | |
| "learning_rate": 3.544100000000001e-05, | |
| "loss": 0.2562, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 1.73053152039555, | |
| "grad_norm": 3.39601993560791, | |
| "learning_rate": 3.4842e-05, | |
| "loss": 0.2795, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 1.7717346518335395, | |
| "grad_norm": 2.5917625427246094, | |
| "learning_rate": 3.4243000000000004e-05, | |
| "loss": 0.2933, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 1.8129377832715288, | |
| "grad_norm": 1.0517610311508179, | |
| "learning_rate": 3.3644000000000005e-05, | |
| "loss": 0.2632, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 1.854140914709518, | |
| "grad_norm": 1.573089361190796, | |
| "learning_rate": 3.3045000000000006e-05, | |
| "loss": 0.2554, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.8953440461475073, | |
| "grad_norm": 1.3932527303695679, | |
| "learning_rate": 3.2446e-05, | |
| "loss": 0.2676, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 1.9365471775854965, | |
| "grad_norm": 7.98951530456543, | |
| "learning_rate": 3.1847e-05, | |
| "loss": 0.2906, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 1.9777503090234858, | |
| "grad_norm": 0.578360378742218, | |
| "learning_rate": 3.1248e-05, | |
| "loss": 0.2815, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy_Autre r\u00e9paration": 0.19815143518295517, | |
| "eval_accuracy_D\u00e9coupe": 0.11079467411500263, | |
| "eval_accuracy_Emergence": 0.4089615931721195, | |
| "eval_accuracy_Gla\u00e7age ou Ressuage": 0.0, | |
| "eval_accuracy_Reflet m\u00e9t\u00e9o": 0.0, | |
| "eval_accuracy_Unlabeled": NaN, | |
| "eval_iou_Autre r\u00e9paration": 0.19162433877536195, | |
| "eval_iou_D\u00e9coupe": 0.10140688937641373, | |
| "eval_iou_Emergence": 0.40571014840298464, | |
| "eval_iou_Gla\u00e7age ou Ressuage": 0.0, | |
| "eval_iou_Reflet m\u00e9t\u00e9o": 0.0, | |
| "eval_iou_Unlabeled": 0.0, | |
| "eval_loss": 0.26819199323654175, | |
| "eval_mean_accuracy": 0.14358154049401545, | |
| "eval_mean_iou": 0.11645689609246006, | |
| "eval_overall_accuracy": 0.15928522569775833, | |
| "eval_runtime": 140.1786, | |
| "eval_samples_per_second": 12.934, | |
| "eval_steps_per_second": 3.239, | |
| "step": 4854 | |
| }, | |
| { | |
| "epoch": 2.018953440461475, | |
| "grad_norm": 0.5752081871032715, | |
| "learning_rate": 3.0649000000000004e-05, | |
| "loss": 0.2768, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 2.0601565718994643, | |
| "grad_norm": 0.6111757755279541, | |
| "learning_rate": 3.0050000000000002e-05, | |
| "loss": 0.2226, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.1013597033374536, | |
| "grad_norm": 0.48088550567626953, | |
| "learning_rate": 2.9451e-05, | |
| "loss": 0.334, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 2.142562834775443, | |
| "grad_norm": 1.2190054655075073, | |
| "learning_rate": 2.8851999999999998e-05, | |
| "loss": 0.2868, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 2.183765966213432, | |
| "grad_norm": 2.414565324783325, | |
| "learning_rate": 2.8253e-05, | |
| "loss": 0.3291, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 2.2249690976514214, | |
| "grad_norm": 0.2674981653690338, | |
| "learning_rate": 2.7653999999999996e-05, | |
| "loss": 0.2687, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 2.2661722290894106, | |
| "grad_norm": 2.053374767303467, | |
| "learning_rate": 2.7054999999999998e-05, | |
| "loss": 0.2559, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.3073753605274003, | |
| "grad_norm": 3.9835445880889893, | |
| "learning_rate": 2.6455999999999995e-05, | |
| "loss": 0.282, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 2.348578491965389, | |
| "grad_norm": 3.391972303390503, | |
| "learning_rate": 2.5857e-05, | |
| "loss": 0.3191, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 2.389781623403379, | |
| "grad_norm": 0.4526354968547821, | |
| "learning_rate": 2.5258e-05, | |
| "loss": 0.2732, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 2.430984754841368, | |
| "grad_norm": 1.3189719915390015, | |
| "learning_rate": 2.4659e-05, | |
| "loss": 0.242, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 2.4721878862793574, | |
| "grad_norm": 1.6163711547851562, | |
| "learning_rate": 2.406e-05, | |
| "loss": 0.278, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.5133910177173466, | |
| "grad_norm": 1.5330442190170288, | |
| "learning_rate": 2.3460999999999998e-05, | |
| "loss": 0.29, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 2.554594149155336, | |
| "grad_norm": 4.686217784881592, | |
| "learning_rate": 2.2862e-05, | |
| "loss": 0.2586, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 2.595797280593325, | |
| "grad_norm": 3.333735942840576, | |
| "learning_rate": 2.2263e-05, | |
| "loss": 0.2794, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 2.6370004120313144, | |
| "grad_norm": 1.2093195915222168, | |
| "learning_rate": 2.1663999999999998e-05, | |
| "loss": 0.2466, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 2.6782035434693037, | |
| "grad_norm": 1.6071631908416748, | |
| "learning_rate": 2.1065e-05, | |
| "loss": 0.21, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.719406674907293, | |
| "grad_norm": 1.4164949655532837, | |
| "learning_rate": 2.0465999999999997e-05, | |
| "loss": 0.2822, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 2.760609806345282, | |
| "grad_norm": 8.471506118774414, | |
| "learning_rate": 1.9866999999999998e-05, | |
| "loss": 0.2475, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 2.8018129377832715, | |
| "grad_norm": 8.533307075500488, | |
| "learning_rate": 1.9267999999999996e-05, | |
| "loss": 0.2806, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 2.8430160692212607, | |
| "grad_norm": 0.49498608708381653, | |
| "learning_rate": 1.8669e-05, | |
| "loss": 0.2682, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 2.88421920065925, | |
| "grad_norm": 1.339969515800476, | |
| "learning_rate": 1.807e-05, | |
| "loss": 0.2435, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.9254223320972392, | |
| "grad_norm": 1.8642264604568481, | |
| "learning_rate": 1.7471e-05, | |
| "loss": 0.2518, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 2.9666254635352285, | |
| "grad_norm": 2.9471471309661865, | |
| "learning_rate": 1.6872e-05, | |
| "loss": 0.2638, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy_Autre r\u00e9paration": 0.30393303904730357, | |
| "eval_accuracy_D\u00e9coupe": 0.23455367948789083, | |
| "eval_accuracy_Emergence": 0.5085131571199683, | |
| "eval_accuracy_Gla\u00e7age ou Ressuage": 0.003045137463105984, | |
| "eval_accuracy_Reflet m\u00e9t\u00e9o": 0.0, | |
| "eval_accuracy_Unlabeled": NaN, | |
| "eval_iou_Autre r\u00e9paration": 0.2853778307692313, | |
| "eval_iou_D\u00e9coupe": 0.21276477560584842, | |
| "eval_iou_Emergence": 0.49725063677040354, | |
| "eval_iou_Gla\u00e7age ou Ressuage": 0.002998539305038369, | |
| "eval_iou_Reflet m\u00e9t\u00e9o": 0.0, | |
| "eval_iou_Unlabeled": 0.0, | |
| "eval_loss": 0.2419871985912323, | |
| "eval_mean_accuracy": 0.2100090026236537, | |
| "eval_mean_iou": 0.16639863040842026, | |
| "eval_overall_accuracy": 0.2563620151228916, | |
| "eval_runtime": 137.8421, | |
| "eval_samples_per_second": 13.153, | |
| "eval_steps_per_second": 3.294, | |
| "step": 7281 | |
| }, | |
| { | |
| "epoch": 3.0078285949732178, | |
| "grad_norm": 0.4593660533428192, | |
| "learning_rate": 1.6272999999999998e-05, | |
| "loss": 0.2486, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 3.0490317264112075, | |
| "grad_norm": 0.8246074318885803, | |
| "learning_rate": 1.5674e-05, | |
| "loss": 0.2251, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 3.0902348578491967, | |
| "grad_norm": 0.9824215769767761, | |
| "learning_rate": 1.5075000000000002e-05, | |
| "loss": 0.2386, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.131437989287186, | |
| "grad_norm": 6.623724937438965, | |
| "learning_rate": 1.4476e-05, | |
| "loss": 0.2635, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 3.1726411207251752, | |
| "grad_norm": 0.816888689994812, | |
| "learning_rate": 1.3877e-05, | |
| "loss": 0.2821, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 3.2138442521631645, | |
| "grad_norm": 0.45224809646606445, | |
| "learning_rate": 1.3277999999999999e-05, | |
| "loss": 0.2238, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 3.2550473836011538, | |
| "grad_norm": 0.9230859279632568, | |
| "learning_rate": 1.2678999999999998e-05, | |
| "loss": 0.2238, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 3.296250515039143, | |
| "grad_norm": 2.5414812564849854, | |
| "learning_rate": 1.2079999999999998e-05, | |
| "loss": 0.2046, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.3374536464771323, | |
| "grad_norm": 1.6467418670654297, | |
| "learning_rate": 1.1480999999999997e-05, | |
| "loss": 0.2343, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 3.3786567779151215, | |
| "grad_norm": 0.6073494553565979, | |
| "learning_rate": 1.0882000000000004e-05, | |
| "loss": 0.2162, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 3.419859909353111, | |
| "grad_norm": 2.7378017902374268, | |
| "learning_rate": 1.0283000000000003e-05, | |
| "loss": 0.2868, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 3.4610630407911, | |
| "grad_norm": 1.4614454507827759, | |
| "learning_rate": 9.684000000000002e-06, | |
| "loss": 0.2145, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 3.5022661722290893, | |
| "grad_norm": 2.336061954498291, | |
| "learning_rate": 9.085000000000002e-06, | |
| "loss": 0.2918, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 3.5434693036670786, | |
| "grad_norm": 1.7232545614242554, | |
| "learning_rate": 8.486000000000001e-06, | |
| "loss": 0.2854, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 3.584672435105068, | |
| "grad_norm": 0.514677882194519, | |
| "learning_rate": 7.887000000000001e-06, | |
| "loss": 0.2514, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 3.6258755665430575, | |
| "grad_norm": 0.9662112593650818, | |
| "learning_rate": 7.2879999999999995e-06, | |
| "loss": 0.2714, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 3.6670786979810464, | |
| "grad_norm": 10.60983657836914, | |
| "learning_rate": 6.688999999999999e-06, | |
| "loss": 0.2548, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 3.708281829419036, | |
| "grad_norm": 2.669593572616577, | |
| "learning_rate": 6.0899999999999984e-06, | |
| "loss": 0.2591, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.749484960857025, | |
| "grad_norm": 1.071542501449585, | |
| "learning_rate": 5.490999999999998e-06, | |
| "loss": 0.2763, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 3.7906880922950146, | |
| "grad_norm": 2.664677381515503, | |
| "learning_rate": 4.891999999999997e-06, | |
| "loss": 0.2178, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 3.831891223733004, | |
| "grad_norm": 9.70131778717041, | |
| "learning_rate": 4.292999999999997e-06, | |
| "loss": 0.2674, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 3.873094355170993, | |
| "grad_norm": 4.843862056732178, | |
| "learning_rate": 3.694000000000003e-06, | |
| "loss": 0.2581, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 3.9142974866089824, | |
| "grad_norm": 0.8629316091537476, | |
| "learning_rate": 3.0950000000000026e-06, | |
| "loss": 0.2642, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 3.9555006180469716, | |
| "grad_norm": 4.216986179351807, | |
| "learning_rate": 2.496000000000002e-06, | |
| "loss": 0.1965, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 3.996703749484961, | |
| "grad_norm": 2.241065502166748, | |
| "learning_rate": 1.8970000000000013e-06, | |
| "loss": 0.2703, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy_Autre r\u00e9paration": 0.36122042935066995, | |
| "eval_accuracy_D\u00e9coupe": 0.28433059478878114, | |
| "eval_accuracy_Emergence": 0.5473337114203988, | |
| "eval_accuracy_Gla\u00e7age ou Ressuage": 0.0446288018012878, | |
| "eval_accuracy_Reflet m\u00e9t\u00e9o": 0.0, | |
| "eval_accuracy_Unlabeled": NaN, | |
| "eval_iou_Autre r\u00e9paration": 0.33834292206885924, | |
| "eval_iou_D\u00e9coupe": 0.251159370886517, | |
| "eval_iou_Emergence": 0.5319718670461905, | |
| "eval_iou_Gla\u00e7age ou Ressuage": 0.042908421138837, | |
| "eval_iou_Reflet m\u00e9t\u00e9o": 0.0, | |
| "eval_iou_Unlabeled": 0.0, | |
| "eval_loss": 0.2333020269870758, | |
| "eval_mean_accuracy": 0.24750270747222752, | |
| "eval_mean_iou": 0.1940637635234006, | |
| "eval_overall_accuracy": 0.30742270034207847, | |
| "eval_runtime": 138.4117, | |
| "eval_samples_per_second": 13.099, | |
| "eval_steps_per_second": 3.28, | |
| "step": 9708 | |
| }, | |
| { | |
| "epoch": 4.03790688092295, | |
| "grad_norm": 1.052063226699829, | |
| "learning_rate": 1.298000000000001e-06, | |
| "loss": 0.242, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 4.07911001236094, | |
| "grad_norm": 6.82352876663208, | |
| "learning_rate": 6.990000000000005e-07, | |
| "loss": 0.2482, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 4.120313143798929, | |
| "grad_norm": 2.648499011993408, | |
| "learning_rate": 1e-07, | |
| "loss": 0.2197, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.120313143798929, | |
| "eval_accuracy_Autre r\u00e9paration": 0.34369405810457515, | |
| "eval_accuracy_D\u00e9coupe": 0.2864541960267422, | |
| "eval_accuracy_Emergence": 0.5548598133737452, | |
| "eval_accuracy_Gla\u00e7age ou Ressuage": 0.03860482159488221, | |
| "eval_accuracy_Reflet m\u00e9t\u00e9o": 0.0, | |
| "eval_accuracy_Unlabeled": NaN, | |
| "eval_iou_Autre r\u00e9paration": 0.32304877421180617, | |
| "eval_iou_D\u00e9coupe": 0.2515107459482324, | |
| "eval_iou_Emergence": 0.5379450939388203, | |
| "eval_iou_Gla\u00e7age ou Ressuage": 0.03692047935180606, | |
| "eval_iou_Reflet m\u00e9t\u00e9o": 0.0, | |
| "eval_iou_Unlabeled": 0.0, | |
| "eval_loss": 0.23602528870105743, | |
| "eval_mean_accuracy": 0.244722577819989, | |
| "eval_mean_iou": 0.19157084890844414, | |
| "eval_overall_accuracy": 0.29617685609695316, | |
| "eval_runtime": 141.1695, | |
| "eval_samples_per_second": 12.843, | |
| "eval_steps_per_second": 3.216, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.120313143798929, | |
| "step": 10000, | |
| "total_flos": 8.912029734867567e+18, | |
| "train_loss": 0.29374205589294433, | |
| "train_runtime": 4666.1544, | |
| "train_samples_per_second": 8.572, | |
| "train_steps_per_second": 2.143 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 10000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "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": 8.912029734867567e+18, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |