{ "best_metric": 0.2074955552816391, "best_model_checkpoint": "./vit-accident-image/checkpoint-200", "epoch": 10.0, "global_step": 500, "is_hyper_param_search": false, "is_local_process_zero": true, "is_world_process_zero": true, "log_history": [ { "epoch": 0.2, "learning_rate": 0.000196, "loss": 0.6842, "step": 10 }, { "epoch": 0.4, "learning_rate": 0.000192, "loss": 0.6515, "step": 20 }, { "epoch": 0.6, "learning_rate": 0.000188, "loss": 0.5938, "step": 30 }, { "epoch": 0.8, "learning_rate": 0.00018400000000000003, "loss": 0.5845, "step": 40 }, { "epoch": 1.0, "learning_rate": 0.00018, "loss": 0.5319, "step": 50 }, { "epoch": 1.2, "learning_rate": 0.00017600000000000002, "loss": 0.2824, "step": 60 }, { "epoch": 1.4, "learning_rate": 0.000172, "loss": 0.2948, "step": 70 }, { "epoch": 1.6, "learning_rate": 0.000168, "loss": 0.6173, "step": 80 }, { "epoch": 1.8, "learning_rate": 0.000164, "loss": 0.2863, "step": 90 }, { "epoch": 2.0, "learning_rate": 0.00016, "loss": 0.3546, "step": 100 }, { "epoch": 2.0, "eval_accuracy": 0.9183673469387755, "eval_f1": 0.918435459290099, "eval_loss": 0.23273873329162598, "eval_runtime": 2.7024, "eval_samples_per_second": 36.264, "eval_steps_per_second": 4.81, "step": 100 }, { "epoch": 2.2, "learning_rate": 0.00015600000000000002, "loss": 0.1067, "step": 110 }, { "epoch": 2.4, "learning_rate": 0.000152, "loss": 0.2548, "step": 120 }, { "epoch": 2.6, "learning_rate": 0.000148, "loss": 0.1812, "step": 130 }, { "epoch": 2.8, "learning_rate": 0.000144, "loss": 0.2835, "step": 140 }, { "epoch": 3.0, "learning_rate": 0.00014, "loss": 0.2162, "step": 150 }, { "epoch": 3.2, "learning_rate": 0.00013600000000000003, "loss": 0.1976, "step": 160 }, { "epoch": 3.4, "learning_rate": 0.000132, "loss": 0.1068, "step": 170 }, { "epoch": 3.6, "learning_rate": 0.00012800000000000002, "loss": 0.0842, "step": 180 }, { "epoch": 3.8, "learning_rate": 0.000124, "loss": 0.168, "step": 190 }, { "epoch": 4.0, "learning_rate": 0.00012, "loss": 0.1654, "step": 200 }, { "epoch": 4.0, "eval_accuracy": 0.9387755102040817, "eval_f1": 0.9387755102040817, "eval_loss": 0.2074955552816391, "eval_runtime": 2.2462, "eval_samples_per_second": 43.629, "eval_steps_per_second": 5.787, "step": 200 }, { "epoch": 4.2, "learning_rate": 0.000116, "loss": 0.0843, "step": 210 }, { "epoch": 4.4, "learning_rate": 0.00011200000000000001, "loss": 0.032, "step": 220 }, { "epoch": 4.6, "learning_rate": 0.00010800000000000001, "loss": 0.1275, "step": 230 }, { "epoch": 4.8, "learning_rate": 0.00010400000000000001, "loss": 0.073, "step": 240 }, { "epoch": 5.0, "learning_rate": 0.0001, "loss": 0.1203, "step": 250 }, { "epoch": 5.2, "learning_rate": 9.6e-05, "loss": 0.0723, "step": 260 }, { "epoch": 5.4, "learning_rate": 9.200000000000001e-05, "loss": 0.0184, "step": 270 }, { "epoch": 5.6, "learning_rate": 8.800000000000001e-05, "loss": 0.0633, "step": 280 }, { "epoch": 5.8, "learning_rate": 8.4e-05, "loss": 0.0989, "step": 290 }, { "epoch": 6.0, "learning_rate": 8e-05, "loss": 0.0146, "step": 300 }, { "epoch": 6.0, "eval_accuracy": 0.9387755102040817, "eval_f1": 0.9386728276216146, "eval_loss": 0.24968579411506653, "eval_runtime": 2.3881, "eval_samples_per_second": 41.036, "eval_steps_per_second": 5.444, "step": 300 }, { "epoch": 6.2, "learning_rate": 7.6e-05, "loss": 0.0295, "step": 310 }, { "epoch": 6.4, "learning_rate": 7.2e-05, "loss": 0.0698, "step": 320 }, { "epoch": 6.6, "learning_rate": 6.800000000000001e-05, "loss": 0.0289, "step": 330 }, { "epoch": 6.8, "learning_rate": 6.400000000000001e-05, "loss": 0.0336, "step": 340 }, { "epoch": 7.0, "learning_rate": 6e-05, "loss": 0.0384, "step": 350 }, { "epoch": 7.2, "learning_rate": 5.6000000000000006e-05, "loss": 0.0344, "step": 360 }, { "epoch": 7.4, "learning_rate": 5.2000000000000004e-05, "loss": 0.0085, "step": 370 }, { "epoch": 7.6, "learning_rate": 4.8e-05, "loss": 0.0145, "step": 380 }, { "epoch": 7.8, "learning_rate": 4.4000000000000006e-05, "loss": 0.0064, "step": 390 }, { "epoch": 8.0, "learning_rate": 4e-05, "loss": 0.0317, "step": 400 }, { "epoch": 8.0, "eval_accuracy": 0.9285714285714286, "eval_f1": 0.9285190999476715, "eval_loss": 0.21786272525787354, "eval_runtime": 3.0699, "eval_samples_per_second": 31.922, "eval_steps_per_second": 4.235, "step": 400 }, { "epoch": 8.2, "learning_rate": 3.6e-05, "loss": 0.0081, "step": 410 }, { "epoch": 8.4, "learning_rate": 3.2000000000000005e-05, "loss": 0.0057, "step": 420 }, { "epoch": 8.6, "learning_rate": 2.8000000000000003e-05, "loss": 0.0364, "step": 430 }, { "epoch": 8.8, "learning_rate": 2.4e-05, "loss": 0.0062, "step": 440 }, { "epoch": 9.0, "learning_rate": 2e-05, "loss": 0.025, "step": 450 }, { "epoch": 9.2, "learning_rate": 1.6000000000000003e-05, "loss": 0.0087, "step": 460 }, { "epoch": 9.4, "learning_rate": 1.2e-05, "loss": 0.0136, "step": 470 }, { "epoch": 9.6, "learning_rate": 8.000000000000001e-06, "loss": 0.0228, "step": 480 }, { "epoch": 9.8, "learning_rate": 4.000000000000001e-06, "loss": 0.005, "step": 490 }, { "epoch": 10.0, "learning_rate": 0.0, "loss": 0.0192, "step": 500 }, { "epoch": 10.0, "eval_accuracy": 0.9285714285714286, "eval_f1": 0.9286087125109987, "eval_loss": 0.2254808396100998, "eval_runtime": 2.2207, "eval_samples_per_second": 44.131, "eval_steps_per_second": 5.854, "step": 500 }, { "epoch": 10.0, "step": 500, "total_flos": 6.129616378514227e+17, "train_loss": 0.15593594065308572, "train_runtime": 520.5132, "train_samples_per_second": 15.197, "train_steps_per_second": 0.961 } ], "max_steps": 500, "num_train_epochs": 10, "total_flos": 6.129616378514227e+17, "trial_name": null, "trial_params": null }