| { | |
| "best_metric": 0.027647286653518677, | |
| "best_model_checkpoint": "/workspace/training_output/age-gender-vit-large-patch16-224-in21k/checkpoint-12110", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 12110, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 1.9174236168455824e-05, | |
| "loss": 0.2922, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 1.8348472336911643e-05, | |
| "loss": 0.2144, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 1.752270850536747e-05, | |
| "loss": 0.1678, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "learning_rate": 1.6696944673823288e-05, | |
| "loss": 0.1615, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9847953216374269, | |
| "eval_loss": 0.05505400523543358, | |
| "eval_runtime": 258.103, | |
| "eval_samples_per_second": 13.251, | |
| "eval_steps_per_second": 1.658, | |
| "step": 2422 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 1.587118084227911e-05, | |
| "loss": 0.1614, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 1.5045417010734931e-05, | |
| "loss": 0.1273, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 1.4219653179190754e-05, | |
| "loss": 0.1361, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 1.3393889347646574e-05, | |
| "loss": 0.1256, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 1.2568125516102397e-05, | |
| "loss": 0.1032, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9903508771929824, | |
| "eval_loss": 0.040435414761304855, | |
| "eval_runtime": 263.0983, | |
| "eval_samples_per_second": 12.999, | |
| "eval_steps_per_second": 1.627, | |
| "step": 4844 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 1.1742361684558218e-05, | |
| "loss": 0.1154, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 1.091659785301404e-05, | |
| "loss": 0.0984, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "learning_rate": 1.0090834021469859e-05, | |
| "loss": 0.087, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 9.265070189925683e-06, | |
| "loss": 0.0889, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 8.439306358381504e-06, | |
| "loss": 0.095, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9903508771929824, | |
| "eval_loss": 0.04800451546907425, | |
| "eval_runtime": 255.5823, | |
| "eval_samples_per_second": 13.381, | |
| "eval_steps_per_second": 1.675, | |
| "step": 7266 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 7.6135425268373255e-06, | |
| "loss": 0.0604, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "learning_rate": 6.787778695293147e-06, | |
| "loss": 0.0647, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "learning_rate": 5.962014863748969e-06, | |
| "loss": 0.0585, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "learning_rate": 5.13625103220479e-06, | |
| "loss": 0.0746, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "learning_rate": 4.310487200660611e-06, | |
| "loss": 0.0599, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9923976608187135, | |
| "eval_loss": 0.03288862109184265, | |
| "eval_runtime": 256.4303, | |
| "eval_samples_per_second": 13.337, | |
| "eval_steps_per_second": 1.669, | |
| "step": 9688 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "learning_rate": 3.484723369116433e-06, | |
| "loss": 0.0452, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.34, | |
| "learning_rate": 2.658959537572254e-06, | |
| "loss": 0.0528, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 4.54, | |
| "learning_rate": 1.8331957060280761e-06, | |
| "loss": 0.045, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 4.75, | |
| "learning_rate": 1.0074318744838975e-06, | |
| "loss": 0.0504, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 4.95, | |
| "learning_rate": 1.8166804293971927e-07, | |
| "loss": 0.0386, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9944444444444445, | |
| "eval_loss": 0.027647286653518677, | |
| "eval_runtime": 295.0164, | |
| "eval_samples_per_second": 11.593, | |
| "eval_steps_per_second": 1.451, | |
| "step": 12110 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 12110, | |
| "total_flos": 2.653627303487527e+19, | |
| "train_loss": 0.10462298566501659, | |
| "train_runtime": 9666.7071, | |
| "train_samples_per_second": 10.021, | |
| "train_steps_per_second": 1.253 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 12110, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 2.653627303487527e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |