| { | |
| "best_metric": 0.26617246866226196, | |
| "best_model_checkpoint": "./apple/mobilevit-small-garbage/checkpoint-500", | |
| "epoch": 5.0, | |
| "global_step": 555, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 0.0001963963963963964, | |
| "loss": 1.7495, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.00019279279279279282, | |
| "loss": 1.6591, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.0001891891891891892, | |
| "loss": 1.5418, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.00018594594594594597, | |
| "loss": 1.3934, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 0.00018234234234234235, | |
| "loss": 1.2802, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.00017873873873873876, | |
| "loss": 1.18, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.00017513513513513516, | |
| "loss": 0.9854, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 0.00017153153153153154, | |
| "loss": 0.9527, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 0.00016792792792792794, | |
| "loss": 0.8574, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "learning_rate": 0.00016432432432432435, | |
| "loss": 0.8589, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "eval_accuracy": 0.8079268292682927, | |
| "eval_loss": 0.661000669002533, | |
| "eval_runtime": 4.4994, | |
| "eval_samples_per_second": 72.898, | |
| "eval_steps_per_second": 9.112, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 0.0001610810810810811, | |
| "loss": 0.7726, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 0.00015747747747747747, | |
| "loss": 0.74, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 0.00015387387387387388, | |
| "loss": 0.6325, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 0.00015063063063063063, | |
| "loss": 0.5792, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 0.00014702702702702703, | |
| "loss": 0.5582, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 0.00014342342342342344, | |
| "loss": 0.5644, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "learning_rate": 0.00013981981981981982, | |
| "loss": 0.4776, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 0.00013621621621621622, | |
| "loss": 0.4537, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "learning_rate": 0.00013261261261261263, | |
| "loss": 0.4813, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 0.000129009009009009, | |
| "loss": 0.4676, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "eval_accuracy": 0.8597560975609756, | |
| "eval_loss": 0.4162694811820984, | |
| "eval_runtime": 4.3992, | |
| "eval_samples_per_second": 74.559, | |
| "eval_steps_per_second": 9.32, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 0.0001254054054054054, | |
| "loss": 0.5374, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 0.0001218018018018018, | |
| "loss": 0.4006, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 0.00011855855855855858, | |
| "loss": 0.4236, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 0.00011495495495495497, | |
| "loss": 0.4494, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 0.00011135135135135135, | |
| "loss": 0.3139, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 0.00010774774774774776, | |
| "loss": 0.3151, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.43, | |
| "learning_rate": 0.00010414414414414416, | |
| "loss": 0.2666, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 0.00010054054054054053, | |
| "loss": 0.404, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 9.693693693693694e-05, | |
| "loss": 0.3745, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "learning_rate": 9.333333333333334e-05, | |
| "loss": 0.3111, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "eval_accuracy": 0.8932926829268293, | |
| "eval_loss": 0.3442399501800537, | |
| "eval_runtime": 4.2748, | |
| "eval_samples_per_second": 76.729, | |
| "eval_steps_per_second": 9.591, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "learning_rate": 8.972972972972973e-05, | |
| "loss": 0.3372, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 8.612612612612613e-05, | |
| "loss": 0.4432, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "learning_rate": 8.252252252252253e-05, | |
| "loss": 0.3007, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 3.06, | |
| "learning_rate": 7.891891891891892e-05, | |
| "loss": 0.2942, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 3.15, | |
| "learning_rate": 7.531531531531531e-05, | |
| "loss": 0.2356, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "learning_rate": 7.171171171171172e-05, | |
| "loss": 0.2971, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 6.810810810810811e-05, | |
| "loss": 0.2004, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "learning_rate": 6.45045045045045e-05, | |
| "loss": 0.3405, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "learning_rate": 6.09009009009009e-05, | |
| "loss": 0.3609, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 5.7297297297297305e-05, | |
| "loss": 0.3709, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "eval_accuracy": 0.9146341463414634, | |
| "eval_loss": 0.3073720932006836, | |
| "eval_runtime": 4.3122, | |
| "eval_samples_per_second": 76.064, | |
| "eval_steps_per_second": 9.508, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "learning_rate": 5.369369369369369e-05, | |
| "loss": 0.3094, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 3.78, | |
| "learning_rate": 5.009009009009009e-05, | |
| "loss": 0.2294, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 4.648648648648649e-05, | |
| "loss": 0.3965, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "learning_rate": 4.2882882882882885e-05, | |
| "loss": 0.2148, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 3.927927927927928e-05, | |
| "loss": 0.3167, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 4.14, | |
| "learning_rate": 3.567567567567568e-05, | |
| "loss": 0.1351, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 4.23, | |
| "learning_rate": 3.207207207207207e-05, | |
| "loss": 0.275, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 4.32, | |
| "learning_rate": 2.8468468468468467e-05, | |
| "loss": 0.1743, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 4.41, | |
| "learning_rate": 2.486486486486487e-05, | |
| "loss": 0.1459, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "learning_rate": 2.126126126126126e-05, | |
| "loss": 0.1942, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "eval_accuracy": 0.9115853658536586, | |
| "eval_loss": 0.26617246866226196, | |
| "eval_runtime": 4.2639, | |
| "eval_samples_per_second": 76.925, | |
| "eval_steps_per_second": 9.616, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "learning_rate": 1.765765765765766e-05, | |
| "loss": 0.1843, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 4.68, | |
| "learning_rate": 1.4054054054054055e-05, | |
| "loss": 0.3644, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.77, | |
| "learning_rate": 1.0450450450450452e-05, | |
| "loss": 0.1587, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 4.86, | |
| "learning_rate": 6.846846846846847e-06, | |
| "loss": 0.1534, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "learning_rate": 3.2432432432432437e-06, | |
| "loss": 0.1954, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 555, | |
| "total_flos": 5.153016901533696e+16, | |
| "train_loss": 0.5183452816696854, | |
| "train_runtime": 159.3173, | |
| "train_samples_per_second": 55.487, | |
| "train_steps_per_second": 3.484 | |
| } | |
| ], | |
| "max_steps": 555, | |
| "num_train_epochs": 5, | |
| "total_flos": 5.153016901533696e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |