| { | |
| "best_metric": 0.99, | |
| "best_model_checkpoint": "./model-swin-base/checkpoint-550", | |
| "epoch": 30.0, | |
| "eval_steps": 500, | |
| "global_step": 1500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 1.3333333333333334e-06, | |
| "loss": 1.4686, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 2.666666666666667e-06, | |
| "loss": 1.3974, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.3354, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 5.333333333333334e-06, | |
| "loss": 1.2507, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 6.533333333333334e-06, | |
| "loss": 1.1342, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.675, | |
| "eval_loss": 1.0418896675109863, | |
| "eval_runtime": 3.2263, | |
| "eval_samples_per_second": 61.99, | |
| "eval_steps_per_second": 7.749, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 7.866666666666667e-06, | |
| "loss": 0.9423, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 9.200000000000002e-06, | |
| "loss": 0.7573, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 1.04e-05, | |
| "loss": 0.5741, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 1.1733333333333335e-05, | |
| "loss": 0.4699, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.3066666666666668e-05, | |
| "loss": 0.3671, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.93, | |
| "eval_loss": 0.2913907766342163, | |
| "eval_runtime": 3.5132, | |
| "eval_samples_per_second": 56.929, | |
| "eval_steps_per_second": 7.116, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 1.4400000000000001e-05, | |
| "loss": 0.2999, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 1.5733333333333334e-05, | |
| "loss": 0.2729, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 1.6933333333333336e-05, | |
| "loss": 0.2265, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 1.826666666666667e-05, | |
| "loss": 0.2029, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 1.9600000000000002e-05, | |
| "loss": 0.1877, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.96, | |
| "eval_loss": 0.15821456909179688, | |
| "eval_runtime": 3.0285, | |
| "eval_samples_per_second": 66.04, | |
| "eval_steps_per_second": 8.255, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 1.98962962962963e-05, | |
| "loss": 0.1277, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "learning_rate": 1.974814814814815e-05, | |
| "loss": 0.1544, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 1.9600000000000002e-05, | |
| "loss": 0.1202, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "learning_rate": 1.9451851851851852e-05, | |
| "loss": 0.1262, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 1.9303703703703705e-05, | |
| "loss": 0.103, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.96, | |
| "eval_loss": 0.09424378722906113, | |
| "eval_runtime": 2.7746, | |
| "eval_samples_per_second": 72.084, | |
| "eval_steps_per_second": 9.01, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "learning_rate": 1.9155555555555558e-05, | |
| "loss": 0.0513, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "learning_rate": 1.900740740740741e-05, | |
| "loss": 0.0587, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.6, | |
| "learning_rate": 1.885925925925926e-05, | |
| "loss": 0.0652, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 1.8711111111111113e-05, | |
| "loss": 0.0927, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 1.8562962962962966e-05, | |
| "loss": 0.0747, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.98, | |
| "eval_loss": 0.08238686621189117, | |
| "eval_runtime": 3.8058, | |
| "eval_samples_per_second": 52.551, | |
| "eval_steps_per_second": 6.569, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 5.2, | |
| "learning_rate": 1.8414814814814816e-05, | |
| "loss": 0.0371, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 5.4, | |
| "learning_rate": 1.826666666666667e-05, | |
| "loss": 0.038, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "learning_rate": 1.811851851851852e-05, | |
| "loss": 0.0289, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 5.8, | |
| "learning_rate": 1.7970370370370372e-05, | |
| "loss": 0.0359, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 1.782222222222222e-05, | |
| "loss": 0.0415, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.975, | |
| "eval_loss": 0.09845741093158722, | |
| "eval_runtime": 2.7024, | |
| "eval_samples_per_second": 74.009, | |
| "eval_steps_per_second": 9.251, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 6.2, | |
| "learning_rate": 1.7674074074074074e-05, | |
| "loss": 0.0215, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 1.7525925925925927e-05, | |
| "loss": 0.0151, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 6.6, | |
| "learning_rate": 1.737777777777778e-05, | |
| "loss": 0.0148, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 6.8, | |
| "learning_rate": 1.722962962962963e-05, | |
| "loss": 0.0111, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 1.7081481481481483e-05, | |
| "loss": 0.0155, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.98, | |
| "eval_loss": 0.07865051925182343, | |
| "eval_runtime": 3.1716, | |
| "eval_samples_per_second": 63.061, | |
| "eval_steps_per_second": 7.883, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "learning_rate": 1.6933333333333336e-05, | |
| "loss": 0.0171, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 7.4, | |
| "learning_rate": 1.6785185185185186e-05, | |
| "loss": 0.0088, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 7.6, | |
| "learning_rate": 1.663703703703704e-05, | |
| "loss": 0.0157, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 7.8, | |
| "learning_rate": 1.648888888888889e-05, | |
| "loss": 0.0214, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 1.634074074074074e-05, | |
| "loss": 0.0111, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.985, | |
| "eval_loss": 0.05885818600654602, | |
| "eval_runtime": 2.7423, | |
| "eval_samples_per_second": 72.931, | |
| "eval_steps_per_second": 9.116, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 8.2, | |
| "learning_rate": 1.6192592592592594e-05, | |
| "loss": 0.0057, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 8.4, | |
| "learning_rate": 1.6044444444444444e-05, | |
| "loss": 0.0073, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 8.6, | |
| "learning_rate": 1.5896296296296297e-05, | |
| "loss": 0.0026, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 8.8, | |
| "learning_rate": 1.574814814814815e-05, | |
| "loss": 0.0149, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 1.5600000000000003e-05, | |
| "loss": 0.0056, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.985, | |
| "eval_loss": 0.07744328677654266, | |
| "eval_runtime": 3.5391, | |
| "eval_samples_per_second": 56.512, | |
| "eval_steps_per_second": 7.064, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 9.2, | |
| "learning_rate": 1.5451851851851853e-05, | |
| "loss": 0.0012, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 9.4, | |
| "learning_rate": 1.5303703703703706e-05, | |
| "loss": 0.0058, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "learning_rate": 1.5155555555555557e-05, | |
| "loss": 0.0044, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 9.8, | |
| "learning_rate": 1.5007407407407408e-05, | |
| "loss": 0.0044, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 1.485925925925926e-05, | |
| "loss": 0.0018, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.985, | |
| "eval_loss": 0.09671365469694138, | |
| "eval_runtime": 3.037, | |
| "eval_samples_per_second": 65.855, | |
| "eval_steps_per_second": 8.232, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 10.2, | |
| "learning_rate": 1.4711111111111111e-05, | |
| "loss": 0.0097, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 10.4, | |
| "learning_rate": 1.4562962962962962e-05, | |
| "loss": 0.0023, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 10.6, | |
| "learning_rate": 1.4414814814814817e-05, | |
| "loss": 0.0021, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 10.8, | |
| "learning_rate": 1.4266666666666668e-05, | |
| "loss": 0.0047, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 1.411851851851852e-05, | |
| "loss": 0.0043, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.99, | |
| "eval_loss": 0.0876985564827919, | |
| "eval_runtime": 3.0689, | |
| "eval_samples_per_second": 65.169, | |
| "eval_steps_per_second": 8.146, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 11.2, | |
| "learning_rate": 1.3970370370370371e-05, | |
| "loss": 0.0053, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 11.4, | |
| "learning_rate": 1.3822222222222224e-05, | |
| "loss": 0.001, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 11.6, | |
| "learning_rate": 1.3674074074074075e-05, | |
| "loss": 0.001, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 11.8, | |
| "learning_rate": 1.3525925925925927e-05, | |
| "loss": 0.0023, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 1.3377777777777778e-05, | |
| "loss": 0.0104, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.99, | |
| "eval_loss": 0.09939048439264297, | |
| "eval_runtime": 3.4336, | |
| "eval_samples_per_second": 58.248, | |
| "eval_steps_per_second": 7.281, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 12.2, | |
| "learning_rate": 1.322962962962963e-05, | |
| "loss": 0.0097, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 12.4, | |
| "learning_rate": 1.3081481481481482e-05, | |
| "loss": 0.001, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 12.6, | |
| "learning_rate": 1.2933333333333334e-05, | |
| "loss": 0.0011, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 12.8, | |
| "learning_rate": 1.2785185185185187e-05, | |
| "loss": 0.0009, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 1.2637037037037038e-05, | |
| "loss": 0.0002, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.985, | |
| "eval_loss": 0.10399068146944046, | |
| "eval_runtime": 3.0117, | |
| "eval_samples_per_second": 66.407, | |
| "eval_steps_per_second": 8.301, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 13.2, | |
| "learning_rate": 1.2488888888888891e-05, | |
| "loss": 0.0037, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 13.4, | |
| "learning_rate": 1.2340740740740742e-05, | |
| "loss": 0.0026, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 13.6, | |
| "learning_rate": 1.2192592592592594e-05, | |
| "loss": 0.0006, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 13.8, | |
| "learning_rate": 1.2044444444444445e-05, | |
| "loss": 0.0005, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 1.1896296296296296e-05, | |
| "loss": 0.0012, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.99, | |
| "eval_loss": 0.09624689072370529, | |
| "eval_runtime": 2.8208, | |
| "eval_samples_per_second": 70.902, | |
| "eval_steps_per_second": 8.863, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 14.2, | |
| "learning_rate": 1.1748148148148148e-05, | |
| "loss": 0.0001, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 14.4, | |
| "learning_rate": 1.16e-05, | |
| "loss": 0.0034, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 14.6, | |
| "learning_rate": 1.1451851851851852e-05, | |
| "loss": 0.0017, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 14.8, | |
| "learning_rate": 1.1303703703703705e-05, | |
| "loss": 0.0002, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 1.1155555555555556e-05, | |
| "loss": 0.002, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.985, | |
| "eval_loss": 0.11188454926013947, | |
| "eval_runtime": 2.7109, | |
| "eval_samples_per_second": 73.775, | |
| "eval_steps_per_second": 9.222, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 15.2, | |
| "learning_rate": 1.100740740740741e-05, | |
| "loss": 0.0076, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 15.4, | |
| "learning_rate": 1.085925925925926e-05, | |
| "loss": 0.0028, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 15.6, | |
| "learning_rate": 1.0711111111111112e-05, | |
| "loss": 0.0042, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 15.8, | |
| "learning_rate": 1.0562962962962963e-05, | |
| "loss": 0.0025, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 1.0414814814814815e-05, | |
| "loss": 0.0022, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.99, | |
| "eval_loss": 0.10523451119661331, | |
| "eval_runtime": 2.6437, | |
| "eval_samples_per_second": 75.651, | |
| "eval_steps_per_second": 9.456, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 16.2, | |
| "learning_rate": 1.0266666666666668e-05, | |
| "loss": 0.0006, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 16.4, | |
| "learning_rate": 1.0118518518518519e-05, | |
| "loss": 0.0005, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 16.6, | |
| "learning_rate": 9.970370370370372e-06, | |
| "loss": 0.0002, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 16.8, | |
| "learning_rate": 9.822222222222223e-06, | |
| "loss": 0.0012, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 9.674074074074075e-06, | |
| "loss": 0.0005, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.985, | |
| "eval_loss": 0.10460754483938217, | |
| "eval_runtime": 2.8418, | |
| "eval_samples_per_second": 70.378, | |
| "eval_steps_per_second": 8.797, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 17.2, | |
| "learning_rate": 9.525925925925926e-06, | |
| "loss": 0.0003, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 17.4, | |
| "learning_rate": 9.377777777777779e-06, | |
| "loss": 0.0023, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 17.6, | |
| "learning_rate": 9.22962962962963e-06, | |
| "loss": 0.0002, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 17.8, | |
| "learning_rate": 9.081481481481482e-06, | |
| "loss": 0.0081, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 8.933333333333333e-06, | |
| "loss": 0.0006, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.99, | |
| "eval_loss": 0.10591718554496765, | |
| "eval_runtime": 3.1941, | |
| "eval_samples_per_second": 62.615, | |
| "eval_steps_per_second": 7.827, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 18.2, | |
| "learning_rate": 8.785185185185186e-06, | |
| "loss": 0.0002, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 18.4, | |
| "learning_rate": 8.637037037037037e-06, | |
| "loss": 0.001, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 18.6, | |
| "learning_rate": 8.48888888888889e-06, | |
| "loss": 0.0004, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 18.8, | |
| "learning_rate": 8.340740740740742e-06, | |
| "loss": 0.0052, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "learning_rate": 8.192592592592593e-06, | |
| "loss": 0.0002, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.99, | |
| "eval_loss": 0.09751956909894943, | |
| "eval_runtime": 3.4886, | |
| "eval_samples_per_second": 57.33, | |
| "eval_steps_per_second": 7.166, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 19.2, | |
| "learning_rate": 8.044444444444444e-06, | |
| "loss": 0.0176, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 19.4, | |
| "learning_rate": 7.896296296296297e-06, | |
| "loss": 0.0003, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 19.6, | |
| "learning_rate": 7.748148148148149e-06, | |
| "loss": 0.0007, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 19.8, | |
| "learning_rate": 7.600000000000001e-06, | |
| "loss": 0.0066, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 7.451851851851852e-06, | |
| "loss": 0.0013, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.99, | |
| "eval_loss": 0.10124868154525757, | |
| "eval_runtime": 3.6249, | |
| "eval_samples_per_second": 55.174, | |
| "eval_steps_per_second": 6.897, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 20.2, | |
| "learning_rate": 7.303703703703704e-06, | |
| "loss": 0.0001, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 20.4, | |
| "learning_rate": 7.155555555555556e-06, | |
| "loss": 0.0002, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 20.6, | |
| "learning_rate": 7.007407407407408e-06, | |
| "loss": 0.0022, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 20.8, | |
| "learning_rate": 6.85925925925926e-06, | |
| "loss": 0.0019, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "learning_rate": 6.711111111111111e-06, | |
| "loss": 0.0004, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 0.99, | |
| "eval_loss": 0.10131002217531204, | |
| "eval_runtime": 2.7142, | |
| "eval_samples_per_second": 73.688, | |
| "eval_steps_per_second": 9.211, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 21.2, | |
| "learning_rate": 6.562962962962964e-06, | |
| "loss": 0.0001, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 21.4, | |
| "learning_rate": 6.414814814814815e-06, | |
| "loss": 0.0002, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 21.6, | |
| "learning_rate": 6.266666666666668e-06, | |
| "loss": 0.0001, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 21.8, | |
| "learning_rate": 6.118518518518519e-06, | |
| "loss": 0.0, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "learning_rate": 5.970370370370371e-06, | |
| "loss": 0.0, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.99, | |
| "eval_loss": 0.10334394127130508, | |
| "eval_runtime": 2.6878, | |
| "eval_samples_per_second": 74.412, | |
| "eval_steps_per_second": 9.301, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 22.2, | |
| "learning_rate": 5.822222222222223e-06, | |
| "loss": 0.0004, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 22.4, | |
| "learning_rate": 5.674074074074074e-06, | |
| "loss": 0.001, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 22.6, | |
| "learning_rate": 5.525925925925927e-06, | |
| "loss": 0.0005, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 22.8, | |
| "learning_rate": 5.3777777777777784e-06, | |
| "loss": 0.0002, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "learning_rate": 5.22962962962963e-06, | |
| "loss": 0.0003, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 0.99, | |
| "eval_loss": 0.10421876609325409, | |
| "eval_runtime": 3.4412, | |
| "eval_samples_per_second": 58.119, | |
| "eval_steps_per_second": 7.265, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 23.2, | |
| "learning_rate": 5.081481481481482e-06, | |
| "loss": 0.0, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 23.4, | |
| "learning_rate": 4.933333333333334e-06, | |
| "loss": 0.005, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 23.6, | |
| "learning_rate": 4.7851851851851854e-06, | |
| "loss": 0.0013, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 23.8, | |
| "learning_rate": 4.637037037037038e-06, | |
| "loss": 0.0002, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "learning_rate": 4.488888888888889e-06, | |
| "loss": 0.0001, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.99, | |
| "eval_loss": 0.10514242202043533, | |
| "eval_runtime": 3.844, | |
| "eval_samples_per_second": 52.03, | |
| "eval_steps_per_second": 6.504, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 24.2, | |
| "learning_rate": 4.340740740740741e-06, | |
| "loss": 0.0, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 24.4, | |
| "learning_rate": 4.192592592592593e-06, | |
| "loss": 0.0, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 24.6, | |
| "learning_rate": 4.044444444444445e-06, | |
| "loss": 0.0001, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 24.8, | |
| "learning_rate": 3.896296296296297e-06, | |
| "loss": 0.0001, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 3.7481481481481485e-06, | |
| "loss": 0.0008, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 0.985, | |
| "eval_loss": 0.11208448559045792, | |
| "eval_runtime": 3.8866, | |
| "eval_samples_per_second": 51.458, | |
| "eval_steps_per_second": 6.432, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 25.2, | |
| "learning_rate": 3.6000000000000003e-06, | |
| "loss": 0.0002, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 25.4, | |
| "learning_rate": 3.4518518518518524e-06, | |
| "loss": 0.0006, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 25.6, | |
| "learning_rate": 3.3037037037037038e-06, | |
| "loss": 0.0001, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 25.8, | |
| "learning_rate": 3.1555555555555555e-06, | |
| "loss": 0.0001, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "learning_rate": 3.0074074074074077e-06, | |
| "loss": 0.0003, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.985, | |
| "eval_loss": 0.11236526817083359, | |
| "eval_runtime": 2.9457, | |
| "eval_samples_per_second": 67.895, | |
| "eval_steps_per_second": 8.487, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 26.2, | |
| "learning_rate": 2.8592592592592594e-06, | |
| "loss": 0.0002, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 26.4, | |
| "learning_rate": 2.7111111111111116e-06, | |
| "loss": 0.0018, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 26.6, | |
| "learning_rate": 2.5629629629629633e-06, | |
| "loss": 0.0001, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 26.8, | |
| "learning_rate": 2.414814814814815e-06, | |
| "loss": 0.0001, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "learning_rate": 2.266666666666667e-06, | |
| "loss": 0.0004, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy": 0.985, | |
| "eval_loss": 0.1106027290225029, | |
| "eval_runtime": 2.7111, | |
| "eval_samples_per_second": 73.771, | |
| "eval_steps_per_second": 9.221, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 27.2, | |
| "learning_rate": 2.1185185185185186e-06, | |
| "loss": 0.0, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 27.4, | |
| "learning_rate": 1.9703703703703703e-06, | |
| "loss": 0.0003, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 27.6, | |
| "learning_rate": 1.8222222222222225e-06, | |
| "loss": 0.0, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 27.8, | |
| "learning_rate": 1.674074074074074e-06, | |
| "loss": 0.0, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "learning_rate": 1.525925925925926e-06, | |
| "loss": 0.0, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.985, | |
| "eval_loss": 0.10527899116277695, | |
| "eval_runtime": 2.7243, | |
| "eval_samples_per_second": 73.414, | |
| "eval_steps_per_second": 9.177, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 28.2, | |
| "learning_rate": 1.377777777777778e-06, | |
| "loss": 0.0, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 28.4, | |
| "learning_rate": 1.2296296296296297e-06, | |
| "loss": 0.0, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 28.6, | |
| "learning_rate": 1.0814814814814815e-06, | |
| "loss": 0.0, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 28.8, | |
| "learning_rate": 9.333333333333334e-07, | |
| "loss": 0.0001, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "learning_rate": 7.851851851851852e-07, | |
| "loss": 0.0002, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy": 0.985, | |
| "eval_loss": 0.10485915094614029, | |
| "eval_runtime": 2.7609, | |
| "eval_samples_per_second": 72.44, | |
| "eval_steps_per_second": 9.055, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 29.2, | |
| "learning_rate": 6.370370370370371e-07, | |
| "loss": 0.0001, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 29.4, | |
| "learning_rate": 4.88888888888889e-07, | |
| "loss": 0.0161, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 29.6, | |
| "learning_rate": 3.407407407407408e-07, | |
| "loss": 0.0, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 29.8, | |
| "learning_rate": 1.9259259259259263e-07, | |
| "loss": 0.0, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "learning_rate": 4.444444444444445e-08, | |
| "loss": 0.0001, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.99, | |
| "eval_loss": 0.1045214906334877, | |
| "eval_runtime": 2.666, | |
| "eval_samples_per_second": 75.018, | |
| "eval_steps_per_second": 9.377, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "step": 1500, | |
| "total_flos": 8.461519969222656e+18, | |
| "train_loss": 0.08282403266835414, | |
| "train_runtime": 2739.6736, | |
| "train_samples_per_second": 39.421, | |
| "train_steps_per_second": 0.548 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1500, | |
| "num_train_epochs": 30, | |
| "save_steps": 500, | |
| "total_flos": 8.461519969222656e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |