| { | |
| "best_metric": 0.9709876543209877, | |
| "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat\\checkpoint-342", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 342, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 5.281110763549805, | |
| "learning_rate": 1.4285714285714285e-05, | |
| "loss": 2.2704, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 6.057841777801514, | |
| "learning_rate": 2.857142857142857e-05, | |
| "loss": 1.9317, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 17.515657424926758, | |
| "learning_rate": 4.2857142857142856e-05, | |
| "loss": 1.272, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 23.613327026367188, | |
| "learning_rate": 4.9185667752443e-05, | |
| "loss": 0.6814, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 18.322710037231445, | |
| "learning_rate": 4.755700325732899e-05, | |
| "loss": 0.5119, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 25.08259391784668, | |
| "learning_rate": 4.592833876221499e-05, | |
| "loss": 0.4619, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 16.223060607910156, | |
| "learning_rate": 4.429967426710098e-05, | |
| "loss": 0.4132, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 16.03809928894043, | |
| "learning_rate": 4.2671009771986977e-05, | |
| "loss": 0.4062, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 10.12449836730957, | |
| "learning_rate": 4.104234527687297e-05, | |
| "loss": 0.3142, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 9.619889259338379, | |
| "learning_rate": 3.941368078175896e-05, | |
| "loss": 0.3146, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 11.356223106384277, | |
| "learning_rate": 3.778501628664495e-05, | |
| "loss": 0.2841, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9469135802469136, | |
| "eval_loss": 0.17170849442481995, | |
| "eval_runtime": 6.6951, | |
| "eval_samples_per_second": 241.968, | |
| "eval_steps_per_second": 7.618, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 13.724865913391113, | |
| "learning_rate": 3.615635179153095e-05, | |
| "loss": 0.308, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "grad_norm": 18.163928985595703, | |
| "learning_rate": 3.452768729641694e-05, | |
| "loss": 0.2901, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "grad_norm": 10.404041290283203, | |
| "learning_rate": 3.289902280130293e-05, | |
| "loss": 0.2516, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "grad_norm": 24.62061309814453, | |
| "learning_rate": 3.127035830618892e-05, | |
| "loss": 0.2744, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 26.786832809448242, | |
| "learning_rate": 2.9641693811074923e-05, | |
| "loss": 0.24, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "grad_norm": 11.213741302490234, | |
| "learning_rate": 2.8013029315960915e-05, | |
| "loss": 0.2152, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "grad_norm": 14.150710105895996, | |
| "learning_rate": 2.6384364820846906e-05, | |
| "loss": 0.2344, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "grad_norm": 7.0177812576293945, | |
| "learning_rate": 2.47557003257329e-05, | |
| "loss": 0.2265, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 13.405513763427734, | |
| "learning_rate": 2.3127035830618892e-05, | |
| "loss": 0.253, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "grad_norm": 7.144536018371582, | |
| "learning_rate": 2.149837133550489e-05, | |
| "loss": 0.2086, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "grad_norm": 17.144977569580078, | |
| "learning_rate": 1.986970684039088e-05, | |
| "loss": 0.2053, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9561728395061728, | |
| "eval_loss": 0.1471811980009079, | |
| "eval_runtime": 6.8477, | |
| "eval_samples_per_second": 236.576, | |
| "eval_steps_per_second": 7.448, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "grad_norm": 8.713410377502441, | |
| "learning_rate": 1.8241042345276872e-05, | |
| "loss": 0.2034, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "grad_norm": 10.655760765075684, | |
| "learning_rate": 1.6612377850162867e-05, | |
| "loss": 0.1967, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "grad_norm": 22.123258590698242, | |
| "learning_rate": 1.4983713355048862e-05, | |
| "loss": 0.1745, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "grad_norm": 14.105958938598633, | |
| "learning_rate": 1.3355048859934855e-05, | |
| "loss": 0.207, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "grad_norm": 11.178110122680664, | |
| "learning_rate": 1.1726384364820847e-05, | |
| "loss": 0.1848, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "grad_norm": 8.699991226196289, | |
| "learning_rate": 1.009771986970684e-05, | |
| "loss": 0.1661, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "grad_norm": 13.814720153808594, | |
| "learning_rate": 8.469055374592833e-06, | |
| "loss": 0.1564, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "grad_norm": 7.040809631347656, | |
| "learning_rate": 6.840390879478828e-06, | |
| "loss": 0.15, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "grad_norm": 11.934651374816895, | |
| "learning_rate": 5.211726384364822e-06, | |
| "loss": 0.1848, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "grad_norm": 7.909034729003906, | |
| "learning_rate": 3.5830618892508147e-06, | |
| "loss": 0.1724, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "grad_norm": 10.677735328674316, | |
| "learning_rate": 1.9543973941368076e-06, | |
| "loss": 0.1456, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "grad_norm": 7.879863262176514, | |
| "learning_rate": 3.2573289902280136e-07, | |
| "loss": 0.166, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9709876543209877, | |
| "eval_loss": 0.10647899657487869, | |
| "eval_runtime": 6.8337, | |
| "eval_samples_per_second": 237.059, | |
| "eval_steps_per_second": 7.463, | |
| "step": 342 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 342, | |
| "total_flos": 1.0874439903456461e+18, | |
| "train_loss": 0.4010893177916432, | |
| "train_runtime": 447.7729, | |
| "train_samples_per_second": 97.683, | |
| "train_steps_per_second": 0.764 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 342, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 1.0874439903456461e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |