| { | |
| "best_metric": 0.0222712941467762, | |
| "best_model_checkpoint": "/workspace/training_output/age-gender-vit-large-patch16-224-in21k-v3/checkpoint-9828", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 12285, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 1.9185999185999185e-05, | |
| "loss": 0.3037, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 1.8371998371998375e-05, | |
| "loss": 0.2165, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 1.755799755799756e-05, | |
| "loss": 0.1959, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 1.6743996743996746e-05, | |
| "loss": 0.1868, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9852941176470589, | |
| "eval_loss": 0.05468880385160446, | |
| "eval_runtime": 310.0142, | |
| "eval_samples_per_second": 11.187, | |
| "eval_steps_per_second": 1.4, | |
| "step": 2457 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 1.5929995929995933e-05, | |
| "loss": 0.1677, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "learning_rate": 1.5115995115995116e-05, | |
| "loss": 0.1256, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 1.4301994301994305e-05, | |
| "loss": 0.1379, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "learning_rate": 1.348799348799349e-05, | |
| "loss": 0.1172, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "learning_rate": 1.2673992673992674e-05, | |
| "loss": 0.1209, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9887543252595156, | |
| "eval_loss": 0.040104664862155914, | |
| "eval_runtime": 296.5781, | |
| "eval_samples_per_second": 11.693, | |
| "eval_steps_per_second": 1.463, | |
| "step": 4914 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 1.1859991859991862e-05, | |
| "loss": 0.1226, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "learning_rate": 1.1045991045991047e-05, | |
| "loss": 0.1012, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 1.0231990231990233e-05, | |
| "loss": 0.0843, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 9.417989417989418e-06, | |
| "loss": 0.0938, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 8.603988603988605e-06, | |
| "loss": 0.1027, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9936562860438293, | |
| "eval_loss": 0.026174411177635193, | |
| "eval_runtime": 332.8527, | |
| "eval_samples_per_second": 10.419, | |
| "eval_steps_per_second": 1.304, | |
| "step": 7371 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 7.78998778998779e-06, | |
| "loss": 0.0851, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.26, | |
| "learning_rate": 6.975986975986977e-06, | |
| "loss": 0.0664, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.46, | |
| "learning_rate": 6.161986161986162e-06, | |
| "loss": 0.0826, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 3.66, | |
| "learning_rate": 5.347985347985348e-06, | |
| "loss": 0.0743, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 4.533984533984534e-06, | |
| "loss": 0.0654, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9959630911188004, | |
| "eval_loss": 0.0222712941467762, | |
| "eval_runtime": 299.4708, | |
| "eval_samples_per_second": 11.58, | |
| "eval_steps_per_second": 1.449, | |
| "step": 9828 | |
| }, | |
| { | |
| "epoch": 4.07, | |
| "learning_rate": 3.7199837199837202e-06, | |
| "loss": 0.0632, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.27, | |
| "learning_rate": 2.9059829059829063e-06, | |
| "loss": 0.0424, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 4.48, | |
| "learning_rate": 2.091982091982092e-06, | |
| "loss": 0.0618, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 4.68, | |
| "learning_rate": 1.2779812779812782e-06, | |
| "loss": 0.0461, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 4.88, | |
| "learning_rate": 4.63980463980464e-07, | |
| "loss": 0.0542, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9948096885813149, | |
| "eval_loss": 0.027304569259285927, | |
| "eval_runtime": 302.9795, | |
| "eval_samples_per_second": 11.446, | |
| "eval_steps_per_second": 1.432, | |
| "step": 12285 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 12285, | |
| "total_flos": 2.6917045405252854e+19, | |
| "train_loss": 0.11123836355865317, | |
| "train_runtime": 12631.4535, | |
| "train_samples_per_second": 7.779, | |
| "train_steps_per_second": 0.973 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 12285, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 2.6917045405252854e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |