{ "best_metric": 0.10869565217391304, "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-DMAE-8e-6\\checkpoint-3", "epoch": 34.285714285714285, "eval_steps": 500, "global_step": 120, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.86, "eval_accuracy": 0.10869565217391304, "eval_loss": 7.94273042678833, "eval_runtime": 0.6742, "eval_samples_per_second": 68.233, "eval_steps_per_second": 4.45, "step": 3 }, { "epoch": 2.0, "eval_accuracy": 0.10869565217391304, "eval_loss": 7.938077926635742, "eval_runtime": 0.7182, "eval_samples_per_second": 64.052, "eval_steps_per_second": 4.177, "step": 7 }, { "epoch": 2.86, "learning_rate": 6.666666666666667e-06, "loss": 7.9636, "step": 10 }, { "epoch": 2.86, "eval_accuracy": 0.10869565217391304, "eval_loss": 7.930100917816162, "eval_runtime": 0.6504, "eval_samples_per_second": 70.725, "eval_steps_per_second": 4.612, "step": 10 }, { "epoch": 4.0, "eval_accuracy": 0.10869565217391304, "eval_loss": 7.908843517303467, "eval_runtime": 0.6571, "eval_samples_per_second": 70.0, "eval_steps_per_second": 4.565, "step": 14 }, { "epoch": 4.86, "eval_accuracy": 0.10869565217391304, "eval_loss": 7.885687828063965, "eval_runtime": 0.6511, "eval_samples_per_second": 70.645, "eval_steps_per_second": 4.607, "step": 17 }, { "epoch": 5.71, "learning_rate": 7.4074074074074075e-06, "loss": 7.8732, "step": 20 }, { "epoch": 6.0, "eval_accuracy": 0.10869565217391304, "eval_loss": 7.8449907302856445, "eval_runtime": 0.6802, "eval_samples_per_second": 67.632, "eval_steps_per_second": 4.411, "step": 21 }, { "epoch": 6.86, "eval_accuracy": 0.10869565217391304, "eval_loss": 7.804883003234863, "eval_runtime": 0.6691, "eval_samples_per_second": 68.744, "eval_steps_per_second": 4.483, "step": 24 }, { "epoch": 8.0, "eval_accuracy": 0.10869565217391304, "eval_loss": 7.737576007843018, "eval_runtime": 0.6657, "eval_samples_per_second": 69.104, "eval_steps_per_second": 4.507, "step": 28 }, { "epoch": 8.57, "learning_rate": 6.666666666666667e-06, "loss": 7.6568, "step": 30 }, { "epoch": 8.86, "eval_accuracy": 0.10869565217391304, "eval_loss": 7.678272247314453, "eval_runtime": 0.6862, "eval_samples_per_second": 67.04, "eval_steps_per_second": 4.372, "step": 31 }, { "epoch": 10.0, "eval_accuracy": 0.10869565217391304, "eval_loss": 7.594333648681641, "eval_runtime": 0.6459, "eval_samples_per_second": 71.215, "eval_steps_per_second": 4.644, "step": 35 }, { "epoch": 10.86, "eval_accuracy": 0.10869565217391304, "eval_loss": 7.528815746307373, "eval_runtime": 0.6537, "eval_samples_per_second": 70.373, "eval_steps_per_second": 4.59, "step": 38 }, { "epoch": 11.43, "learning_rate": 5.925925925925925e-06, "loss": 7.7458, "step": 40 }, { "epoch": 12.0, "eval_accuracy": 0.10869565217391304, "eval_loss": 7.435303211212158, "eval_runtime": 0.6387, "eval_samples_per_second": 72.025, "eval_steps_per_second": 4.697, "step": 42 }, { "epoch": 12.86, "eval_accuracy": 0.10869565217391304, "eval_loss": 7.361030101776123, "eval_runtime": 0.6672, "eval_samples_per_second": 68.947, "eval_steps_per_second": 4.497, "step": 45 }, { "epoch": 14.0, "eval_accuracy": 0.10869565217391304, "eval_loss": 7.261359214782715, "eval_runtime": 0.6457, "eval_samples_per_second": 71.241, "eval_steps_per_second": 4.646, "step": 49 }, { "epoch": 14.29, "learning_rate": 5.185185185185185e-06, "loss": 7.3025, "step": 50 }, { "epoch": 14.86, "eval_accuracy": 0.10869565217391304, "eval_loss": 7.189355373382568, "eval_runtime": 0.6962, "eval_samples_per_second": 66.077, "eval_steps_per_second": 4.309, "step": 52 }, { "epoch": 16.0, "eval_accuracy": 0.10869565217391304, "eval_loss": 7.099257946014404, "eval_runtime": 0.6592, "eval_samples_per_second": 69.786, "eval_steps_per_second": 4.551, "step": 56 }, { "epoch": 16.86, "eval_accuracy": 0.10869565217391304, "eval_loss": 7.034775257110596, "eval_runtime": 0.6762, "eval_samples_per_second": 68.032, "eval_steps_per_second": 4.437, "step": 59 }, { "epoch": 17.14, "learning_rate": 4.444444444444444e-06, "loss": 7.0862, "step": 60 }, { "epoch": 18.0, "eval_accuracy": 0.10869565217391304, "eval_loss": 6.95252799987793, "eval_runtime": 0.6451, "eval_samples_per_second": 71.302, "eval_steps_per_second": 4.65, "step": 63 }, { "epoch": 18.86, "eval_accuracy": 0.10869565217391304, "eval_loss": 6.894467353820801, "eval_runtime": 0.6837, "eval_samples_per_second": 67.285, "eval_steps_per_second": 4.388, "step": 66 }, { "epoch": 20.0, "learning_rate": 3.7037037037037037e-06, "loss": 6.9553, "step": 70 }, { "epoch": 20.0, "eval_accuracy": 0.10869565217391304, "eval_loss": 6.825292110443115, "eval_runtime": 0.6427, "eval_samples_per_second": 71.578, "eval_steps_per_second": 4.668, "step": 70 }, { "epoch": 20.86, "eval_accuracy": 0.10869565217391304, "eval_loss": 6.779515743255615, "eval_runtime": 0.6441, "eval_samples_per_second": 71.413, "eval_steps_per_second": 4.657, "step": 73 }, { "epoch": 22.0, "eval_accuracy": 0.10869565217391304, "eval_loss": 6.7202067375183105, "eval_runtime": 0.6361, "eval_samples_per_second": 72.311, "eval_steps_per_second": 4.716, "step": 77 }, { "epoch": 22.86, "learning_rate": 2.9629629629629625e-06, "loss": 6.8024, "step": 80 }, { "epoch": 22.86, "eval_accuracy": 0.10869565217391304, "eval_loss": 6.675656318664551, "eval_runtime": 0.6491, "eval_samples_per_second": 70.862, "eval_steps_per_second": 4.621, "step": 80 }, { "epoch": 24.0, "eval_accuracy": 0.10869565217391304, "eval_loss": 6.621024131774902, "eval_runtime": 0.6807, "eval_samples_per_second": 67.581, "eval_steps_per_second": 4.407, "step": 84 }, { "epoch": 24.86, "eval_accuracy": 0.10869565217391304, "eval_loss": 6.578530311584473, "eval_runtime": 0.6592, "eval_samples_per_second": 69.785, "eval_steps_per_second": 4.551, "step": 87 }, { "epoch": 25.71, "learning_rate": 2.222222222222222e-06, "loss": 6.6652, "step": 90 }, { "epoch": 26.0, "eval_accuracy": 0.10869565217391304, "eval_loss": 6.5274739265441895, "eval_runtime": 0.6592, "eval_samples_per_second": 69.785, "eval_steps_per_second": 4.551, "step": 91 }, { "epoch": 26.86, "eval_accuracy": 0.10869565217391304, "eval_loss": 6.494857311248779, "eval_runtime": 0.6622, "eval_samples_per_second": 69.469, "eval_steps_per_second": 4.531, "step": 94 }, { "epoch": 28.0, "eval_accuracy": 0.10869565217391304, "eval_loss": 6.458892345428467, "eval_runtime": 0.6467, "eval_samples_per_second": 71.136, "eval_steps_per_second": 4.639, "step": 98 }, { "epoch": 28.57, "learning_rate": 1.4814814814814812e-06, "loss": 6.467, "step": 100 }, { "epoch": 28.86, "eval_accuracy": 0.10869565217391304, "eval_loss": 6.4354329109191895, "eval_runtime": 0.6551, "eval_samples_per_second": 70.213, "eval_steps_per_second": 4.579, "step": 101 }, { "epoch": 30.0, "eval_accuracy": 0.10869565217391304, "eval_loss": 6.409433841705322, "eval_runtime": 0.8987, "eval_samples_per_second": 51.185, "eval_steps_per_second": 3.338, "step": 105 }, { "epoch": 30.86, "eval_accuracy": 0.10869565217391304, "eval_loss": 6.39463996887207, "eval_runtime": 0.6521, "eval_samples_per_second": 70.536, "eval_steps_per_second": 4.6, "step": 108 }, { "epoch": 31.43, "learning_rate": 7.407407407407406e-07, "loss": 6.4984, "step": 110 }, { "epoch": 32.0, "eval_accuracy": 0.10869565217391304, "eval_loss": 6.379604816436768, "eval_runtime": 0.6611, "eval_samples_per_second": 69.576, "eval_steps_per_second": 4.538, "step": 112 }, { "epoch": 32.86, "eval_accuracy": 0.10869565217391304, "eval_loss": 6.371899127960205, "eval_runtime": 0.6278, "eval_samples_per_second": 73.271, "eval_steps_per_second": 4.779, "step": 115 }, { "epoch": 34.0, "eval_accuracy": 0.10869565217391304, "eval_loss": 6.366767406463623, "eval_runtime": 0.6237, "eval_samples_per_second": 73.759, "eval_steps_per_second": 4.81, "step": 119 }, { "epoch": 34.29, "learning_rate": 0.0, "loss": 6.4603, "step": 120 }, { "epoch": 34.29, "eval_accuracy": 0.10869565217391304, "eval_loss": 6.366403102874756, "eval_runtime": 0.7282, "eval_samples_per_second": 63.173, "eval_steps_per_second": 4.12, "step": 120 }, { "epoch": 34.29, "step": 120, "total_flos": 2.3770905934823424e+17, "train_loss": 7.123055585225424, "train_runtime": 193.127, "train_samples_per_second": 44.116, "train_steps_per_second": 0.621 } ], "logging_steps": 10, "max_steps": 120, "num_input_tokens_seen": 0, "num_train_epochs": 40, "save_steps": 500, "total_flos": 2.3770905934823424e+17, "train_batch_size": 16, "trial_name": null, "trial_params": null }