| { | |
| "best_metric": 0.934375, | |
| "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-finetuned-THFOOD-50/checkpoint-940", | |
| "epoch": 19.894179894179896, | |
| "global_step": 940, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 5.319148936170213e-06, | |
| "loss": 3.9199, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 1.0638297872340426e-05, | |
| "loss": 3.8792, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 1.595744680851064e-05, | |
| "loss": 3.7844, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 2.1276595744680852e-05, | |
| "loss": 3.6558, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_accuracy": 0.28, | |
| "eval_loss": 3.195585012435913, | |
| "eval_runtime": 43.9765, | |
| "eval_samples_per_second": 36.383, | |
| "eval_steps_per_second": 0.568, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 2.6595744680851064e-05, | |
| "loss": 3.4006, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 3.191489361702128e-05, | |
| "loss": 2.9259, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 3.723404255319149e-05, | |
| "loss": 2.4813, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 4.2553191489361704e-05, | |
| "loss": 2.0113, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 4.787234042553192e-05, | |
| "loss": 1.705, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "eval_accuracy": 0.67875, | |
| "eval_loss": 1.1700831651687622, | |
| "eval_runtime": 42.7514, | |
| "eval_samples_per_second": 37.426, | |
| "eval_steps_per_second": 0.585, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 4.964539007092199e-05, | |
| "loss": 1.4876, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 4.905437352245863e-05, | |
| "loss": 1.3025, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 4.846335697399527e-05, | |
| "loss": 1.1542, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 4.787234042553192e-05, | |
| "loss": 1.0516, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 4.728132387706856e-05, | |
| "loss": 0.9805, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "eval_accuracy": 0.8125, | |
| "eval_loss": 0.6491796374320984, | |
| "eval_runtime": 42.6164, | |
| "eval_samples_per_second": 37.544, | |
| "eval_steps_per_second": 0.587, | |
| "step": 141 | |
| }, | |
| { | |
| "epoch": 3.17, | |
| "learning_rate": 4.669030732860521e-05, | |
| "loss": 0.9217, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.39, | |
| "learning_rate": 4.609929078014185e-05, | |
| "loss": 0.8574, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 4.550827423167849e-05, | |
| "loss": 0.7836, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 3.81, | |
| "learning_rate": 4.491725768321513e-05, | |
| "loss": 0.7925, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.864375, | |
| "eval_loss": 0.47244203090667725, | |
| "eval_runtime": 42.4933, | |
| "eval_samples_per_second": 37.653, | |
| "eval_steps_per_second": 0.588, | |
| "step": 189 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 4.432624113475177e-05, | |
| "loss": 0.769, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 4.23, | |
| "learning_rate": 4.373522458628842e-05, | |
| "loss": 0.6721, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.44, | |
| "learning_rate": 4.3144208037825064e-05, | |
| "loss": 0.6727, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 4.66, | |
| "learning_rate": 4.2553191489361704e-05, | |
| "loss": 0.6304, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.87, | |
| "learning_rate": 4.1962174940898345e-05, | |
| "loss": 0.6169, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "eval_accuracy": 0.87375, | |
| "eval_loss": 0.41286996006965637, | |
| "eval_runtime": 42.6335, | |
| "eval_samples_per_second": 37.529, | |
| "eval_steps_per_second": 0.586, | |
| "step": 236 | |
| }, | |
| { | |
| "epoch": 5.08, | |
| "learning_rate": 4.1371158392434986e-05, | |
| "loss": 0.5997, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 5.29, | |
| "learning_rate": 4.078014184397163e-05, | |
| "loss": 0.5769, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 5.5, | |
| "learning_rate": 4.018912529550828e-05, | |
| "loss": 0.5889, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 5.71, | |
| "learning_rate": 3.959810874704492e-05, | |
| "loss": 0.5698, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 5.93, | |
| "learning_rate": 3.900709219858156e-05, | |
| "loss": 0.5343, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 5.99, | |
| "eval_accuracy": 0.8825, | |
| "eval_loss": 0.371685266494751, | |
| "eval_runtime": 42.6664, | |
| "eval_samples_per_second": 37.5, | |
| "eval_steps_per_second": 0.586, | |
| "step": 283 | |
| }, | |
| { | |
| "epoch": 6.14, | |
| "learning_rate": 3.84160756501182e-05, | |
| "loss": 0.5222, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 6.35, | |
| "learning_rate": 3.782505910165485e-05, | |
| "loss": 0.565, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 6.56, | |
| "learning_rate": 3.723404255319149e-05, | |
| "loss": 0.5147, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 6.77, | |
| "learning_rate": 3.664302600472813e-05, | |
| "loss": 0.5166, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 6.98, | |
| "learning_rate": 3.605200945626478e-05, | |
| "loss": 0.5196, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 6.98, | |
| "eval_accuracy": 0.890625, | |
| "eval_loss": 0.3654029965400696, | |
| "eval_runtime": 41.9867, | |
| "eval_samples_per_second": 38.107, | |
| "eval_steps_per_second": 0.595, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "learning_rate": 3.546099290780142e-05, | |
| "loss": 0.4794, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 7.41, | |
| "learning_rate": 3.4869976359338065e-05, | |
| "loss": 0.4747, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 7.62, | |
| "learning_rate": 3.4278959810874706e-05, | |
| "loss": 0.4549, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 7.83, | |
| "learning_rate": 3.3687943262411347e-05, | |
| "loss": 0.5059, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.896875, | |
| "eval_loss": 0.3267361521720886, | |
| "eval_runtime": 42.9793, | |
| "eval_samples_per_second": 37.227, | |
| "eval_steps_per_second": 0.582, | |
| "step": 378 | |
| }, | |
| { | |
| "epoch": 8.04, | |
| "learning_rate": 3.309692671394799e-05, | |
| "loss": 0.4381, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 8.25, | |
| "learning_rate": 3.2505910165484634e-05, | |
| "loss": 0.4277, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 8.47, | |
| "learning_rate": 3.191489361702128e-05, | |
| "loss": 0.4362, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 8.68, | |
| "learning_rate": 3.132387706855792e-05, | |
| "loss": 0.4095, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 8.89, | |
| "learning_rate": 3.073286052009456e-05, | |
| "loss": 0.4432, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 8.99, | |
| "eval_accuracy": 0.908125, | |
| "eval_loss": 0.2996412217617035, | |
| "eval_runtime": 42.09, | |
| "eval_samples_per_second": 38.014, | |
| "eval_steps_per_second": 0.594, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 9.1, | |
| "learning_rate": 3.0141843971631207e-05, | |
| "loss": 0.4369, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 9.31, | |
| "learning_rate": 2.9550827423167847e-05, | |
| "loss": 0.4071, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 9.52, | |
| "learning_rate": 2.895981087470449e-05, | |
| "loss": 0.3956, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 9.74, | |
| "learning_rate": 2.836879432624114e-05, | |
| "loss": 0.3971, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 9.95, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.3819, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 9.99, | |
| "eval_accuracy": 0.90875, | |
| "eval_loss": 0.30557677149772644, | |
| "eval_runtime": 42.6034, | |
| "eval_samples_per_second": 37.556, | |
| "eval_steps_per_second": 0.587, | |
| "step": 472 | |
| }, | |
| { | |
| "epoch": 10.16, | |
| "learning_rate": 2.7186761229314423e-05, | |
| "loss": 0.4018, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 10.37, | |
| "learning_rate": 2.6595744680851064e-05, | |
| "loss": 0.3847, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 10.58, | |
| "learning_rate": 2.6004728132387708e-05, | |
| "loss": 0.3719, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 10.79, | |
| "learning_rate": 2.5413711583924348e-05, | |
| "loss": 0.3627, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 10.98, | |
| "eval_accuracy": 0.92125, | |
| "eval_loss": 0.2795552909374237, | |
| "eval_runtime": 42.2776, | |
| "eval_samples_per_second": 37.845, | |
| "eval_steps_per_second": 0.591, | |
| "step": 519 | |
| }, | |
| { | |
| "epoch": 11.01, | |
| "learning_rate": 2.4822695035460995e-05, | |
| "loss": 0.3823, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 11.22, | |
| "learning_rate": 2.4231678486997636e-05, | |
| "loss": 0.3417, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 11.43, | |
| "learning_rate": 2.364066193853428e-05, | |
| "loss": 0.3663, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 11.64, | |
| "learning_rate": 2.3049645390070924e-05, | |
| "loss": 0.3518, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 11.85, | |
| "learning_rate": 2.2458628841607564e-05, | |
| "loss": 0.3505, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.915, | |
| "eval_loss": 0.2753271758556366, | |
| "eval_runtime": 43.0104, | |
| "eval_samples_per_second": 37.2, | |
| "eval_steps_per_second": 0.581, | |
| "step": 567 | |
| }, | |
| { | |
| "epoch": 12.06, | |
| "learning_rate": 2.186761229314421e-05, | |
| "loss": 0.3182, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 12.28, | |
| "learning_rate": 2.1276595744680852e-05, | |
| "loss": 0.3526, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 12.49, | |
| "learning_rate": 2.0685579196217493e-05, | |
| "loss": 0.3441, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 12.7, | |
| "learning_rate": 2.009456264775414e-05, | |
| "loss": 0.3178, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 12.91, | |
| "learning_rate": 1.950354609929078e-05, | |
| "loss": 0.3224, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 12.99, | |
| "eval_accuracy": 0.920625, | |
| "eval_loss": 0.28300976753234863, | |
| "eval_runtime": 42.2858, | |
| "eval_samples_per_second": 37.838, | |
| "eval_steps_per_second": 0.591, | |
| "step": 614 | |
| }, | |
| { | |
| "epoch": 13.12, | |
| "learning_rate": 1.8912529550827425e-05, | |
| "loss": 0.3412, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 13.33, | |
| "learning_rate": 1.8321513002364065e-05, | |
| "loss": 0.3155, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 13.54, | |
| "learning_rate": 1.773049645390071e-05, | |
| "loss": 0.327, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 13.76, | |
| "learning_rate": 1.7139479905437353e-05, | |
| "loss": 0.3049, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 13.97, | |
| "learning_rate": 1.6548463356973994e-05, | |
| "loss": 0.3206, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 13.99, | |
| "eval_accuracy": 0.923125, | |
| "eval_loss": 0.2796793282032013, | |
| "eval_runtime": 42.1391, | |
| "eval_samples_per_second": 37.97, | |
| "eval_steps_per_second": 0.593, | |
| "step": 661 | |
| }, | |
| { | |
| "epoch": 14.18, | |
| "learning_rate": 1.595744680851064e-05, | |
| "loss": 0.2965, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 14.39, | |
| "learning_rate": 1.536643026004728e-05, | |
| "loss": 0.294, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 14.6, | |
| "learning_rate": 1.4775413711583924e-05, | |
| "loss": 0.3325, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 14.81, | |
| "learning_rate": 1.418439716312057e-05, | |
| "loss": 0.3141, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 14.98, | |
| "eval_accuracy": 0.92875, | |
| "eval_loss": 0.2568666338920593, | |
| "eval_runtime": 42.5767, | |
| "eval_samples_per_second": 37.579, | |
| "eval_steps_per_second": 0.587, | |
| "step": 708 | |
| }, | |
| { | |
| "epoch": 15.03, | |
| "learning_rate": 1.3593380614657212e-05, | |
| "loss": 0.3035, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 15.24, | |
| "learning_rate": 1.3002364066193854e-05, | |
| "loss": 0.2836, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 15.45, | |
| "learning_rate": 1.2411347517730498e-05, | |
| "loss": 0.2892, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 15.66, | |
| "learning_rate": 1.182033096926714e-05, | |
| "loss": 0.3382, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 15.87, | |
| "learning_rate": 1.1229314420803782e-05, | |
| "loss": 0.2946, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.931875, | |
| "eval_loss": 0.25816333293914795, | |
| "eval_runtime": 42.6426, | |
| "eval_samples_per_second": 37.521, | |
| "eval_steps_per_second": 0.586, | |
| "step": 756 | |
| }, | |
| { | |
| "epoch": 16.08, | |
| "learning_rate": 1.0638297872340426e-05, | |
| "loss": 0.279, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 16.3, | |
| "learning_rate": 1.004728132387707e-05, | |
| "loss": 0.2867, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 16.51, | |
| "learning_rate": 9.456264775413712e-06, | |
| "loss": 0.2787, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 16.72, | |
| "learning_rate": 8.865248226950355e-06, | |
| "loss": 0.3116, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 16.93, | |
| "learning_rate": 8.274231678486997e-06, | |
| "loss": 0.3008, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 16.99, | |
| "eval_accuracy": 0.93375, | |
| "eval_loss": 0.2583409249782562, | |
| "eval_runtime": 44.0351, | |
| "eval_samples_per_second": 36.335, | |
| "eval_steps_per_second": 0.568, | |
| "step": 803 | |
| }, | |
| { | |
| "epoch": 17.14, | |
| "learning_rate": 7.68321513002364e-06, | |
| "loss": 0.2553, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 17.35, | |
| "learning_rate": 7.092198581560285e-06, | |
| "loss": 0.2857, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 17.57, | |
| "learning_rate": 6.501182033096927e-06, | |
| "loss": 0.2584, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 17.78, | |
| "learning_rate": 5.91016548463357e-06, | |
| "loss": 0.3162, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 17.99, | |
| "learning_rate": 5.319148936170213e-06, | |
| "loss": 0.2356, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 17.99, | |
| "eval_accuracy": 0.928125, | |
| "eval_loss": 0.2566778063774109, | |
| "eval_runtime": 43.8016, | |
| "eval_samples_per_second": 36.528, | |
| "eval_steps_per_second": 0.571, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 18.2, | |
| "learning_rate": 4.728132387706856e-06, | |
| "loss": 0.2426, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 18.41, | |
| "learning_rate": 4.137115839243498e-06, | |
| "loss": 0.2754, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 18.62, | |
| "learning_rate": 3.5460992907801423e-06, | |
| "loss": 0.2581, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 18.84, | |
| "learning_rate": 2.955082742316785e-06, | |
| "loss": 0.2954, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 18.98, | |
| "eval_accuracy": 0.931875, | |
| "eval_loss": 0.25814077258110046, | |
| "eval_runtime": 42.9843, | |
| "eval_samples_per_second": 37.223, | |
| "eval_steps_per_second": 0.582, | |
| "step": 897 | |
| }, | |
| { | |
| "epoch": 19.05, | |
| "learning_rate": 2.364066193853428e-06, | |
| "loss": 0.2766, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 19.26, | |
| "learning_rate": 1.7730496453900712e-06, | |
| "loss": 0.289, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 19.47, | |
| "learning_rate": 1.182033096926714e-06, | |
| "loss": 0.2598, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 19.68, | |
| "learning_rate": 5.91016548463357e-07, | |
| "loss": 0.2659, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 19.89, | |
| "learning_rate": 0.0, | |
| "loss": 0.2628, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 19.89, | |
| "eval_accuracy": 0.934375, | |
| "eval_loss": 0.25352808833122253, | |
| "eval_runtime": 43.2178, | |
| "eval_samples_per_second": 37.022, | |
| "eval_steps_per_second": 0.578, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 19.89, | |
| "step": 940, | |
| "total_flos": 7.820350673549525e+18, | |
| "train_loss": 0.7086160603989946, | |
| "train_runtime": 9826.2518, | |
| "train_samples_per_second": 24.557, | |
| "train_steps_per_second": 0.096 | |
| }, | |
| { | |
| "epoch": 19.89, | |
| "eval_accuracy": 0.9557397430584335, | |
| "eval_loss": 0.16691383719444275, | |
| "eval_runtime": 327.4304, | |
| "eval_samples_per_second": 36.848, | |
| "eval_steps_per_second": 0.577, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 19.89, | |
| "eval_accuracy": 0.934375, | |
| "eval_loss": 0.25352808833122253, | |
| "eval_runtime": 44.4189, | |
| "eval_samples_per_second": 36.021, | |
| "eval_steps_per_second": 0.563, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 19.89, | |
| "eval_accuracy": 0.9292161520190024, | |
| "eval_loss": 0.2668735980987549, | |
| "eval_runtime": 60.4298, | |
| "eval_samples_per_second": 34.834, | |
| "eval_steps_per_second": 0.546, | |
| "step": 940 | |
| } | |
| ], | |
| "max_steps": 940, | |
| "num_train_epochs": 20, | |
| "total_flos": 7.820350673549525e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |