{ "best_metric": 1.0, "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-imageclds/checkpoint-1205", "epoch": 5.0, "global_step": 1205, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.04, "learning_rate": 4.132231404958678e-06, "loss": 1.3866, "step": 10 }, { "epoch": 0.08, "learning_rate": 8.264462809917356e-06, "loss": 1.2359, "step": 20 }, { "epoch": 0.12, "learning_rate": 1.2396694214876034e-05, "loss": 0.9595, "step": 30 }, { "epoch": 0.17, "learning_rate": 1.652892561983471e-05, "loss": 0.6781, "step": 40 }, { "epoch": 0.21, "learning_rate": 2.066115702479339e-05, "loss": 0.3948, "step": 50 }, { "epoch": 0.25, "learning_rate": 2.479338842975207e-05, "loss": 0.2208, "step": 60 }, { "epoch": 0.29, "learning_rate": 2.8925619834710744e-05, "loss": 0.1207, "step": 70 }, { "epoch": 0.33, "learning_rate": 3.305785123966942e-05, "loss": 0.0628, "step": 80 }, { "epoch": 0.37, "learning_rate": 3.71900826446281e-05, "loss": 0.0586, "step": 90 }, { "epoch": 0.41, "learning_rate": 4.132231404958678e-05, "loss": 0.044, "step": 100 }, { "epoch": 0.46, "learning_rate": 4.545454545454546e-05, "loss": 0.0497, "step": 110 }, { "epoch": 0.5, "learning_rate": 4.958677685950414e-05, "loss": 0.036, "step": 120 }, { "epoch": 0.54, "learning_rate": 4.9584870848708485e-05, "loss": 0.0364, "step": 130 }, { "epoch": 0.58, "learning_rate": 4.912361623616236e-05, "loss": 0.0447, "step": 140 }, { "epoch": 0.62, "learning_rate": 4.8662361623616234e-05, "loss": 0.0494, "step": 150 }, { "epoch": 0.66, "learning_rate": 4.820110701107011e-05, "loss": 0.0638, "step": 160 }, { "epoch": 0.71, "learning_rate": 4.773985239852399e-05, "loss": 0.0397, "step": 170 }, { "epoch": 0.75, "learning_rate": 4.727859778597786e-05, "loss": 0.0306, "step": 180 }, { "epoch": 0.79, "learning_rate": 4.681734317343174e-05, "loss": 0.0398, "step": 190 }, { "epoch": 0.83, "learning_rate": 4.6356088560885616e-05, "loss": 0.0348, "step": 200 }, { "epoch": 0.87, "learning_rate": 4.589483394833948e-05, "loss": 0.039, "step": 210 }, { "epoch": 0.91, "learning_rate": 4.543357933579336e-05, "loss": 0.0209, "step": 220 }, { "epoch": 0.95, "learning_rate": 4.4972324723247236e-05, "loss": 0.0386, "step": 230 }, { "epoch": 1.0, "learning_rate": 4.451107011070111e-05, "loss": 0.019, "step": 240 }, { "epoch": 1.0, "eval_accuracy": 0.9989621172807472, "eval_loss": 0.003147997660562396, "eval_runtime": 104.8322, "eval_samples_per_second": 73.527, "eval_steps_per_second": 2.299, "step": 241 }, { "epoch": 1.04, "learning_rate": 4.4049815498154985e-05, "loss": 0.0337, "step": 250 }, { "epoch": 1.08, "learning_rate": 4.3588560885608856e-05, "loss": 0.0379, "step": 260 }, { "epoch": 1.12, "learning_rate": 4.3127306273062734e-05, "loss": 0.0249, "step": 270 }, { "epoch": 1.16, "learning_rate": 4.266605166051661e-05, "loss": 0.0276, "step": 280 }, { "epoch": 1.2, "learning_rate": 4.220479704797048e-05, "loss": 0.0532, "step": 290 }, { "epoch": 1.24, "learning_rate": 4.1743542435424353e-05, "loss": 0.0399, "step": 300 }, { "epoch": 1.29, "learning_rate": 4.128228782287823e-05, "loss": 0.0354, "step": 310 }, { "epoch": 1.33, "learning_rate": 4.08210332103321e-05, "loss": 0.0214, "step": 320 }, { "epoch": 1.37, "learning_rate": 4.035977859778598e-05, "loss": 0.0363, "step": 330 }, { "epoch": 1.41, "learning_rate": 3.989852398523986e-05, "loss": 0.0228, "step": 340 }, { "epoch": 1.45, "learning_rate": 3.943726937269373e-05, "loss": 0.0268, "step": 350 }, { "epoch": 1.49, "learning_rate": 3.897601476014761e-05, "loss": 0.0345, "step": 360 }, { "epoch": 1.54, "learning_rate": 3.851476014760148e-05, "loss": 0.0222, "step": 370 }, { "epoch": 1.58, "learning_rate": 3.805350553505535e-05, "loss": 0.0343, "step": 380 }, { "epoch": 1.62, "learning_rate": 3.7592250922509227e-05, "loss": 0.0298, "step": 390 }, { "epoch": 1.66, "learning_rate": 3.71309963099631e-05, "loss": 0.0225, "step": 400 }, { "epoch": 1.7, "learning_rate": 3.6669741697416975e-05, "loss": 0.0339, "step": 410 }, { "epoch": 1.74, "learning_rate": 3.620848708487085e-05, "loss": 0.0245, "step": 420 }, { "epoch": 1.78, "learning_rate": 3.5747232472324724e-05, "loss": 0.0428, "step": 430 }, { "epoch": 1.83, "learning_rate": 3.52859778597786e-05, "loss": 0.0249, "step": 440 }, { "epoch": 1.87, "learning_rate": 3.482472324723247e-05, "loss": 0.0248, "step": 450 }, { "epoch": 1.91, "learning_rate": 3.4363468634686344e-05, "loss": 0.0287, "step": 460 }, { "epoch": 1.95, "learning_rate": 3.390221402214022e-05, "loss": 0.0185, "step": 470 }, { "epoch": 1.99, "learning_rate": 3.34409594095941e-05, "loss": 0.0364, "step": 480 }, { "epoch": 2.0, "eval_accuracy": 0.9994810586403736, "eval_loss": 0.0020151014905422926, "eval_runtime": 103.6986, "eval_samples_per_second": 74.331, "eval_steps_per_second": 2.324, "step": 482 }, { "epoch": 2.03, "learning_rate": 3.297970479704797e-05, "loss": 0.0299, "step": 490 }, { "epoch": 2.07, "learning_rate": 3.251845018450185e-05, "loss": 0.0161, "step": 500 }, { "epoch": 2.12, "learning_rate": 3.205719557195572e-05, "loss": 0.0346, "step": 510 }, { "epoch": 2.16, "learning_rate": 3.15959409594096e-05, "loss": 0.0274, "step": 520 }, { "epoch": 2.2, "learning_rate": 3.113468634686347e-05, "loss": 0.0334, "step": 530 }, { "epoch": 2.24, "learning_rate": 3.067343173431734e-05, "loss": 0.0203, "step": 540 }, { "epoch": 2.28, "learning_rate": 3.0212177121771217e-05, "loss": 0.0367, "step": 550 }, { "epoch": 2.32, "learning_rate": 2.975092250922509e-05, "loss": 0.0379, "step": 560 }, { "epoch": 2.37, "learning_rate": 2.928966789667897e-05, "loss": 0.0297, "step": 570 }, { "epoch": 2.41, "learning_rate": 2.8828413284132844e-05, "loss": 0.0319, "step": 580 }, { "epoch": 2.45, "learning_rate": 2.8367158671586718e-05, "loss": 0.0211, "step": 590 }, { "epoch": 2.49, "learning_rate": 2.7905904059040593e-05, "loss": 0.033, "step": 600 }, { "epoch": 2.53, "learning_rate": 2.7444649446494464e-05, "loss": 0.0245, "step": 610 }, { "epoch": 2.57, "learning_rate": 2.6983394833948338e-05, "loss": 0.0287, "step": 620 }, { "epoch": 2.61, "learning_rate": 2.6522140221402213e-05, "loss": 0.026, "step": 630 }, { "epoch": 2.66, "learning_rate": 2.606088560885609e-05, "loss": 0.023, "step": 640 }, { "epoch": 2.7, "learning_rate": 2.5599630996309965e-05, "loss": 0.0216, "step": 650 }, { "epoch": 2.74, "learning_rate": 2.513837638376384e-05, "loss": 0.0393, "step": 660 }, { "epoch": 2.78, "learning_rate": 2.4677121771217714e-05, "loss": 0.0189, "step": 670 }, { "epoch": 2.82, "learning_rate": 2.4215867158671588e-05, "loss": 0.0207, "step": 680 }, { "epoch": 2.86, "learning_rate": 2.3754612546125462e-05, "loss": 0.0306, "step": 690 }, { "epoch": 2.9, "learning_rate": 2.3293357933579337e-05, "loss": 0.0206, "step": 700 }, { "epoch": 2.95, "learning_rate": 2.283210332103321e-05, "loss": 0.0263, "step": 710 }, { "epoch": 2.99, "learning_rate": 2.2370848708487086e-05, "loss": 0.0151, "step": 720 }, { "epoch": 3.0, "eval_accuracy": 0.9998702646600934, "eval_loss": 0.00030530328513123095, "eval_runtime": 103.6593, "eval_samples_per_second": 74.359, "eval_steps_per_second": 2.325, "step": 723 }, { "epoch": 3.03, "learning_rate": 2.190959409594096e-05, "loss": 0.023, "step": 730 }, { "epoch": 3.07, "learning_rate": 2.1448339483394835e-05, "loss": 0.0255, "step": 740 }, { "epoch": 3.11, "learning_rate": 2.098708487084871e-05, "loss": 0.023, "step": 750 }, { "epoch": 3.15, "learning_rate": 2.0525830258302583e-05, "loss": 0.0283, "step": 760 }, { "epoch": 3.2, "learning_rate": 2.0064575645756458e-05, "loss": 0.0241, "step": 770 }, { "epoch": 3.24, "learning_rate": 1.9603321033210336e-05, "loss": 0.0244, "step": 780 }, { "epoch": 3.28, "learning_rate": 1.9142066420664207e-05, "loss": 0.016, "step": 790 }, { "epoch": 3.32, "learning_rate": 1.868081180811808e-05, "loss": 0.0291, "step": 800 }, { "epoch": 3.36, "learning_rate": 1.8219557195571955e-05, "loss": 0.0204, "step": 810 }, { "epoch": 3.4, "learning_rate": 1.7758302583025833e-05, "loss": 0.0391, "step": 820 }, { "epoch": 3.44, "learning_rate": 1.7297047970479704e-05, "loss": 0.0205, "step": 830 }, { "epoch": 3.49, "learning_rate": 1.683579335793358e-05, "loss": 0.0323, "step": 840 }, { "epoch": 3.53, "learning_rate": 1.6374538745387457e-05, "loss": 0.0172, "step": 850 }, { "epoch": 3.57, "learning_rate": 1.591328413284133e-05, "loss": 0.016, "step": 860 }, { "epoch": 3.61, "learning_rate": 1.5452029520295202e-05, "loss": 0.0223, "step": 870 }, { "epoch": 3.65, "learning_rate": 1.4990774907749078e-05, "loss": 0.0307, "step": 880 }, { "epoch": 3.69, "learning_rate": 1.4529520295202952e-05, "loss": 0.029, "step": 890 }, { "epoch": 3.73, "learning_rate": 1.4068265682656829e-05, "loss": 0.0242, "step": 900 }, { "epoch": 3.78, "learning_rate": 1.3607011070110701e-05, "loss": 0.0245, "step": 910 }, { "epoch": 3.82, "learning_rate": 1.3145756457564576e-05, "loss": 0.0213, "step": 920 }, { "epoch": 3.86, "learning_rate": 1.2684501845018452e-05, "loss": 0.0326, "step": 930 }, { "epoch": 3.9, "learning_rate": 1.2223247232472325e-05, "loss": 0.0172, "step": 940 }, { "epoch": 3.94, "learning_rate": 1.17619926199262e-05, "loss": 0.0191, "step": 950 }, { "epoch": 3.98, "learning_rate": 1.1300738007380073e-05, "loss": 0.025, "step": 960 }, { "epoch": 4.0, "eval_accuracy": 0.9998702646600934, "eval_loss": 0.00045192890684120357, "eval_runtime": 103.7545, "eval_samples_per_second": 74.291, "eval_steps_per_second": 2.323, "step": 964 }, { "epoch": 4.02, "learning_rate": 1.083948339483395e-05, "loss": 0.0171, "step": 970 }, { "epoch": 4.07, "learning_rate": 1.0378228782287822e-05, "loss": 0.0343, "step": 980 }, { "epoch": 4.11, "learning_rate": 9.916974169741698e-06, "loss": 0.0259, "step": 990 }, { "epoch": 4.15, "learning_rate": 9.455719557195573e-06, "loss": 0.0182, "step": 1000 }, { "epoch": 4.19, "learning_rate": 8.994464944649447e-06, "loss": 0.0237, "step": 1010 }, { "epoch": 4.23, "learning_rate": 8.533210332103322e-06, "loss": 0.0273, "step": 1020 }, { "epoch": 4.27, "learning_rate": 8.071955719557196e-06, "loss": 0.0272, "step": 1030 }, { "epoch": 4.32, "learning_rate": 7.6107011070110704e-06, "loss": 0.0242, "step": 1040 }, { "epoch": 4.36, "learning_rate": 7.149446494464946e-06, "loss": 0.016, "step": 1050 }, { "epoch": 4.4, "learning_rate": 6.688191881918819e-06, "loss": 0.0285, "step": 1060 }, { "epoch": 4.44, "learning_rate": 6.2269372693726945e-06, "loss": 0.0206, "step": 1070 }, { "epoch": 4.48, "learning_rate": 5.765682656826569e-06, "loss": 0.0243, "step": 1080 }, { "epoch": 4.52, "learning_rate": 5.304428044280443e-06, "loss": 0.0238, "step": 1090 }, { "epoch": 4.56, "learning_rate": 4.843173431734318e-06, "loss": 0.023, "step": 1100 }, { "epoch": 4.61, "learning_rate": 4.381918819188192e-06, "loss": 0.0181, "step": 1110 }, { "epoch": 4.65, "learning_rate": 3.920664206642067e-06, "loss": 0.0275, "step": 1120 }, { "epoch": 4.69, "learning_rate": 3.4594095940959415e-06, "loss": 0.0237, "step": 1130 }, { "epoch": 4.73, "learning_rate": 2.9981549815498154e-06, "loss": 0.0367, "step": 1140 }, { "epoch": 4.77, "learning_rate": 2.5369003690036903e-06, "loss": 0.0238, "step": 1150 }, { "epoch": 4.81, "learning_rate": 2.0756457564575647e-06, "loss": 0.0353, "step": 1160 }, { "epoch": 4.85, "learning_rate": 1.6143911439114391e-06, "loss": 0.0134, "step": 1170 }, { "epoch": 4.9, "learning_rate": 1.1531365313653137e-06, "loss": 0.0164, "step": 1180 }, { "epoch": 4.94, "learning_rate": 6.918819188191883e-07, "loss": 0.0163, "step": 1190 }, { "epoch": 4.98, "learning_rate": 2.3062730627306274e-07, "loss": 0.0191, "step": 1200 }, { "epoch": 5.0, "eval_accuracy": 1.0, "eval_loss": 0.00034155891626141965, "eval_runtime": 103.1498, "eval_samples_per_second": 74.726, "eval_steps_per_second": 2.336, "step": 1205 }, { "epoch": 5.0, "step": 1205, "total_flos": 3.831764827677696e+18, "train_loss": 0.06846031813700663, "train_runtime": 3280.5201, "train_samples_per_second": 46.989, "train_steps_per_second": 0.367 } ], "max_steps": 1205, "num_train_epochs": 5, "total_flos": 3.831764827677696e+18, "trial_name": null, "trial_params": null }