| { | |
| "best_metric": 0.7613636363636364, | |
| "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-dmae-humeda-DAV40/checkpoint-100", | |
| "epoch": 30.0, | |
| "eval_steps": 500, | |
| "global_step": 120, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.3409090909090909, | |
| "eval_loss": 1.5349395275115967, | |
| "eval_runtime": 2.5542, | |
| "eval_samples_per_second": 34.453, | |
| "eval_steps_per_second": 1.175, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.4431818181818182, | |
| "eval_loss": 1.321252703666687, | |
| "eval_runtime": 2.354, | |
| "eval_samples_per_second": 37.383, | |
| "eval_steps_per_second": 1.274, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 8.033442497253418, | |
| "learning_rate": 5e-05, | |
| "loss": 4.7629, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.4431818181818182, | |
| "eval_loss": 1.2540555000305176, | |
| "eval_runtime": 2.642, | |
| "eval_samples_per_second": 33.308, | |
| "eval_steps_per_second": 1.136, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6022727272727273, | |
| "eval_loss": 1.2072031497955322, | |
| "eval_runtime": 1.7756, | |
| "eval_samples_per_second": 49.56, | |
| "eval_steps_per_second": 1.69, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6363636363636364, | |
| "eval_loss": 1.1313027143478394, | |
| "eval_runtime": 1.6972, | |
| "eval_samples_per_second": 51.852, | |
| "eval_steps_per_second": 1.768, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 8.113847732543945, | |
| "learning_rate": 4.849231551964771e-05, | |
| "loss": 3.7987, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.6477272727272727, | |
| "eval_loss": 1.0712097883224487, | |
| "eval_runtime": 1.7097, | |
| "eval_samples_per_second": 51.471, | |
| "eval_steps_per_second": 1.755, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.6590909090909091, | |
| "eval_loss": 0.967653214931488, | |
| "eval_runtime": 1.7586, | |
| "eval_samples_per_second": 50.041, | |
| "eval_steps_per_second": 1.706, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7159090909090909, | |
| "eval_loss": 0.8655213713645935, | |
| "eval_runtime": 1.6895, | |
| "eval_samples_per_second": 52.087, | |
| "eval_steps_per_second": 1.776, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 13.333137512207031, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 3.0437, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.6818181818181818, | |
| "eval_loss": 0.8563566207885742, | |
| "eval_runtime": 1.7069, | |
| "eval_samples_per_second": 51.555, | |
| "eval_steps_per_second": 1.758, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.6818181818181818, | |
| "eval_loss": 0.8003210425376892, | |
| "eval_runtime": 1.7037, | |
| "eval_samples_per_second": 51.653, | |
| "eval_steps_per_second": 1.761, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.7386363636363636, | |
| "eval_loss": 0.7986971139907837, | |
| "eval_runtime": 2.4009, | |
| "eval_samples_per_second": 36.653, | |
| "eval_steps_per_second": 1.25, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 16.13611602783203, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 2.4867, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.7159090909090909, | |
| "eval_loss": 0.7619297504425049, | |
| "eval_runtime": 2.3634, | |
| "eval_samples_per_second": 37.235, | |
| "eval_steps_per_second": 1.269, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.7386363636363636, | |
| "eval_loss": 0.7425692081451416, | |
| "eval_runtime": 1.785, | |
| "eval_samples_per_second": 49.3, | |
| "eval_steps_per_second": 1.681, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.6931818181818182, | |
| "eval_loss": 0.7491652965545654, | |
| "eval_runtime": 1.6916, | |
| "eval_samples_per_second": 52.023, | |
| "eval_steps_per_second": 1.774, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 18.206378936767578, | |
| "learning_rate": 2.9341204441673266e-05, | |
| "loss": 2.147, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.7159090909090909, | |
| "eval_loss": 0.7827097177505493, | |
| "eval_runtime": 1.681, | |
| "eval_samples_per_second": 52.349, | |
| "eval_steps_per_second": 1.785, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.7045454545454546, | |
| "eval_loss": 0.7509434223175049, | |
| "eval_runtime": 1.7041, | |
| "eval_samples_per_second": 51.641, | |
| "eval_steps_per_second": 1.76, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.7386363636363636, | |
| "eval_loss": 0.7364481091499329, | |
| "eval_runtime": 1.6978, | |
| "eval_samples_per_second": 51.831, | |
| "eval_steps_per_second": 1.767, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 16.78498649597168, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 1.8443, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.7159090909090909, | |
| "eval_loss": 0.770459771156311, | |
| "eval_runtime": 1.7059, | |
| "eval_samples_per_second": 51.585, | |
| "eval_steps_per_second": 1.759, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.7272727272727273, | |
| "eval_loss": 0.7515316009521484, | |
| "eval_runtime": 1.7077, | |
| "eval_samples_per_second": 51.532, | |
| "eval_steps_per_second": 1.757, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.7386363636363636, | |
| "eval_loss": 0.747029721736908, | |
| "eval_runtime": 1.8467, | |
| "eval_samples_per_second": 47.652, | |
| "eval_steps_per_second": 1.625, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "grad_norm": 14.945952415466309, | |
| "learning_rate": 1.2500000000000006e-05, | |
| "loss": 1.659, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.7494531273841858, | |
| "eval_runtime": 2.3744, | |
| "eval_samples_per_second": 37.062, | |
| "eval_steps_per_second": 1.263, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.7236538529396057, | |
| "eval_runtime": 2.3525, | |
| "eval_samples_per_second": 37.408, | |
| "eval_steps_per_second": 1.275, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.744009256362915, | |
| "eval_runtime": 2.2107, | |
| "eval_samples_per_second": 39.807, | |
| "eval_steps_per_second": 1.357, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "grad_norm": 21.294570922851562, | |
| "learning_rate": 5.848888922025553e-06, | |
| "loss": 1.5303, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.736692488193512, | |
| "eval_runtime": 1.7151, | |
| "eval_samples_per_second": 51.31, | |
| "eval_steps_per_second": 1.749, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 0.7613636363636364, | |
| "eval_loss": 0.7427918910980225, | |
| "eval_runtime": 1.6602, | |
| "eval_samples_per_second": 53.006, | |
| "eval_steps_per_second": 1.807, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.7406925559043884, | |
| "eval_runtime": 1.671, | |
| "eval_samples_per_second": 52.662, | |
| "eval_steps_per_second": 1.795, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "grad_norm": 18.939655303955078, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 1.4305, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.740644633769989, | |
| "eval_runtime": 1.6889, | |
| "eval_samples_per_second": 52.104, | |
| "eval_steps_per_second": 1.776, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.7422773241996765, | |
| "eval_runtime": 1.6858, | |
| "eval_samples_per_second": 52.2, | |
| "eval_steps_per_second": 1.78, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.7426608800888062, | |
| "eval_runtime": 1.7009, | |
| "eval_samples_per_second": 51.738, | |
| "eval_steps_per_second": 1.764, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "grad_norm": 13.808012962341309, | |
| "learning_rate": 0.0, | |
| "loss": 1.3529, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.7427504062652588, | |
| "eval_runtime": 2.6552, | |
| "eval_samples_per_second": 33.142, | |
| "eval_steps_per_second": 1.13, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "step": 120, | |
| "total_flos": 3.982534914657485e+17, | |
| "train_loss": 2.4056013425191245, | |
| "train_runtime": 469.8187, | |
| "train_samples_per_second": 34.737, | |
| "train_steps_per_second": 0.255 | |
| } | |
| ], | |
| "logging_steps": 12, | |
| "max_steps": 120, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 40, | |
| "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": 3.982534914657485e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |