| { | |
| "best_metric": 0.08533324301242828, | |
| "best_model_checkpoint": "./vit-base-beans-demo/checkpoint-300", | |
| "epoch": 5.0, | |
| "global_step": 325, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.00019384615384615385, | |
| "loss": 0.2116, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0001876923076923077, | |
| "loss": 0.1891, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.00018153846153846155, | |
| "loss": 0.0855, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.0001753846153846154, | |
| "loss": 0.1651, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.00016923076923076923, | |
| "loss": 0.1261, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 0.0001630769230769231, | |
| "loss": 0.2639, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 0.00015692307692307693, | |
| "loss": 0.1101, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 0.00015076923076923077, | |
| "loss": 0.0655, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.0001446153846153846, | |
| "loss": 0.0909, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 0.00013846153846153847, | |
| "loss": 0.0545, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "eval_accuracy": 0.9624060150375939, | |
| "eval_loss": 0.14361602067947388, | |
| "eval_runtime": 4.9212, | |
| "eval_samples_per_second": 27.026, | |
| "eval_steps_per_second": 3.454, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 0.0001323076923076923, | |
| "loss": 0.0318, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 0.00012615384615384615, | |
| "loss": 0.0993, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.00012061538461538462, | |
| "loss": 0.1259, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 0.00011446153846153846, | |
| "loss": 0.098, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 0.00010830769230769231, | |
| "loss": 0.0788, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 0.00010215384615384615, | |
| "loss": 0.0204, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.62, | |
| "learning_rate": 9.6e-05, | |
| "loss": 0.0094, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 8.984615384615384e-05, | |
| "loss": 0.0086, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 8.369230769230769e-05, | |
| "loss": 0.007, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 7.753846153846153e-05, | |
| "loss": 0.006, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "eval_accuracy": 0.9699248120300752, | |
| "eval_loss": 0.10582129657268524, | |
| "eval_runtime": 4.8551, | |
| "eval_samples_per_second": 27.394, | |
| "eval_steps_per_second": 3.501, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.23, | |
| "learning_rate": 7.138461538461538e-05, | |
| "loss": 0.005, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 3.38, | |
| "learning_rate": 6.523076923076923e-05, | |
| "loss": 0.0049, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "learning_rate": 5.907692307692309e-05, | |
| "loss": 0.0046, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "learning_rate": 5.292307692307693e-05, | |
| "loss": 0.0043, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 3.85, | |
| "learning_rate": 4.676923076923077e-05, | |
| "loss": 0.0043, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.0615384615384615e-05, | |
| "loss": 0.0041, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 4.15, | |
| "learning_rate": 3.446153846153846e-05, | |
| "loss": 0.004, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 4.31, | |
| "learning_rate": 2.8307692307692306e-05, | |
| "loss": 0.0039, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 4.46, | |
| "learning_rate": 2.2153846153846154e-05, | |
| "loss": 0.0038, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.0038, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.62, | |
| "eval_accuracy": 0.9774436090225563, | |
| "eval_loss": 0.08533324301242828, | |
| "eval_runtime": 4.9438, | |
| "eval_samples_per_second": 26.903, | |
| "eval_steps_per_second": 3.439, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.77, | |
| "learning_rate": 9.846153846153846e-06, | |
| "loss": 0.0037, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 4.92, | |
| "learning_rate": 3.692307692307693e-06, | |
| "loss": 0.0038, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 325, | |
| "total_flos": 0.0, | |
| "train_loss": 0.05844223602459981, | |
| "train_runtime": 230.414, | |
| "train_samples_per_second": 22.438, | |
| "train_steps_per_second": 1.411 | |
| } | |
| ], | |
| "max_steps": 325, | |
| "num_train_epochs": 5, | |
| "total_flos": 0.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |