| { | |
| "best_metric": 0.03212887421250343, | |
| "best_model_checkpoint": "./my_bean_VIT/checkpoint-200", | |
| "epoch": 4.0, | |
| "global_step": 260, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.00019230769230769233, | |
| "loss": 0.8792, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.00018461538461538463, | |
| "loss": 0.4134, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.00017692307692307693, | |
| "loss": 0.3521, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.00016923076923076923, | |
| "loss": 0.3909, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.00016153846153846155, | |
| "loss": 0.2881, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 0.00015384615384615385, | |
| "loss": 0.2193, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 0.00014615384615384615, | |
| "loss": 0.1367, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 0.00013846153846153847, | |
| "loss": 0.0824, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.00013076923076923077, | |
| "loss": 0.084, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 0.0001230769230769231, | |
| "loss": 0.2698, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "eval_accuracy": 0.9548872180451128, | |
| "eval_loss": 0.1349506825208664, | |
| "eval_runtime": 1.7784, | |
| "eval_samples_per_second": 74.788, | |
| "eval_steps_per_second": 9.559, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 0.00011538461538461538, | |
| "loss": 0.1188, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 0.0001076923076923077, | |
| "loss": 0.0767, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.0001, | |
| "loss": 0.059, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 9.230769230769232e-05, | |
| "loss": 0.0507, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 8.461538461538461e-05, | |
| "loss": 0.023, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 7.692307692307693e-05, | |
| "loss": 0.1037, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 7e-05, | |
| "loss": 0.0724, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 6.23076923076923e-05, | |
| "loss": 0.0279, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 5.461538461538461e-05, | |
| "loss": 0.0411, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 4.692307692307693e-05, | |
| "loss": 0.0147, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "eval_accuracy": 0.9924812030075187, | |
| "eval_loss": 0.03212887421250343, | |
| "eval_runtime": 1.8163, | |
| "eval_samples_per_second": 73.225, | |
| "eval_steps_per_second": 9.36, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 3.923076923076923e-05, | |
| "loss": 0.0137, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "learning_rate": 3.153846153846154e-05, | |
| "loss": 0.0134, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "learning_rate": 2.384615384615385e-05, | |
| "loss": 0.0145, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "learning_rate": 1.6153846153846154e-05, | |
| "loss": 0.0167, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "learning_rate": 8.461538461538462e-06, | |
| "loss": 0.0142, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 7.692307692307694e-07, | |
| "loss": 0.0212, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "step": 260, | |
| "total_flos": 3.205097416476426e+17, | |
| "train_loss": 0.14605647795475446, | |
| "train_runtime": 150.6081, | |
| "train_samples_per_second": 27.462, | |
| "train_steps_per_second": 1.726 | |
| } | |
| ], | |
| "max_steps": 260, | |
| "num_train_epochs": 4, | |
| "total_flos": 3.205097416476426e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |