| { |
| "best_metric": 0.05050145834684372, |
| "best_model_checkpoint": "./vit-base-beans-demo-v5/checkpoint-500", |
| "epoch": 8.0, |
| "global_step": 520, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.15, |
| "learning_rate": 0.00019615384615384615, |
| "loss": 0.9379, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.31, |
| "learning_rate": 0.00019230769230769233, |
| "loss": 0.4657, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.46, |
| "learning_rate": 0.00018846153846153847, |
| "loss": 0.1998, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.62, |
| "learning_rate": 0.00018461538461538463, |
| "loss": 0.3618, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.77, |
| "learning_rate": 0.00018076923076923077, |
| "loss": 0.2387, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.92, |
| "learning_rate": 0.00017692307692307693, |
| "loss": 0.1682, |
| "step": 60 |
| }, |
| { |
| "epoch": 1.08, |
| "learning_rate": 0.0001730769230769231, |
| "loss": 0.2817, |
| "step": 70 |
| }, |
| { |
| "epoch": 1.23, |
| "learning_rate": 0.00016923076923076923, |
| "loss": 0.1381, |
| "step": 80 |
| }, |
| { |
| "epoch": 1.38, |
| "learning_rate": 0.0001653846153846154, |
| "loss": 0.1933, |
| "step": 90 |
| }, |
| { |
| "epoch": 1.54, |
| "learning_rate": 0.00016153846153846155, |
| "loss": 0.1166, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.54, |
| "eval_accuracy": 0.9849624060150376, |
| "eval_loss": 0.07644423842430115, |
| "eval_runtime": 1.439, |
| "eval_samples_per_second": 92.425, |
| "eval_steps_per_second": 11.814, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.69, |
| "learning_rate": 0.0001576923076923077, |
| "loss": 0.0955, |
| "step": 110 |
| }, |
| { |
| "epoch": 1.85, |
| "learning_rate": 0.00015384615384615385, |
| "loss": 0.1201, |
| "step": 120 |
| }, |
| { |
| "epoch": 2.0, |
| "learning_rate": 0.00015000000000000001, |
| "loss": 0.1216, |
| "step": 130 |
| }, |
| { |
| "epoch": 2.15, |
| "learning_rate": 0.00014615384615384615, |
| "loss": 0.0981, |
| "step": 140 |
| }, |
| { |
| "epoch": 2.31, |
| "learning_rate": 0.0001423076923076923, |
| "loss": 0.0418, |
| "step": 150 |
| }, |
| { |
| "epoch": 2.46, |
| "learning_rate": 0.00013846153846153847, |
| "loss": 0.0546, |
| "step": 160 |
| }, |
| { |
| "epoch": 2.62, |
| "learning_rate": 0.00013461538461538464, |
| "loss": 0.0333, |
| "step": 170 |
| }, |
| { |
| "epoch": 2.77, |
| "learning_rate": 0.00013076923076923077, |
| "loss": 0.0193, |
| "step": 180 |
| }, |
| { |
| "epoch": 2.92, |
| "learning_rate": 0.00012692307692307693, |
| "loss": 0.0325, |
| "step": 190 |
| }, |
| { |
| "epoch": 3.08, |
| "learning_rate": 0.0001230769230769231, |
| "loss": 0.1607, |
| "step": 200 |
| }, |
| { |
| "epoch": 3.08, |
| "eval_accuracy": 0.9398496240601504, |
| "eval_loss": 0.21135304868221283, |
| "eval_runtime": 1.6144, |
| "eval_samples_per_second": 82.384, |
| "eval_steps_per_second": 10.53, |
| "step": 200 |
| }, |
| { |
| "epoch": 3.23, |
| "learning_rate": 0.00011923076923076923, |
| "loss": 0.0295, |
| "step": 210 |
| }, |
| { |
| "epoch": 3.38, |
| "learning_rate": 0.00011538461538461538, |
| "loss": 0.0193, |
| "step": 220 |
| }, |
| { |
| "epoch": 3.54, |
| "learning_rate": 0.00011153846153846154, |
| "loss": 0.0271, |
| "step": 230 |
| }, |
| { |
| "epoch": 3.69, |
| "learning_rate": 0.0001076923076923077, |
| "loss": 0.0098, |
| "step": 240 |
| }, |
| { |
| "epoch": 3.85, |
| "learning_rate": 0.00010384615384615386, |
| "loss": 0.0104, |
| "step": 250 |
| }, |
| { |
| "epoch": 4.0, |
| "learning_rate": 0.0001, |
| "loss": 0.0079, |
| "step": 260 |
| }, |
| { |
| "epoch": 4.15, |
| "learning_rate": 9.615384615384617e-05, |
| "loss": 0.0077, |
| "step": 270 |
| }, |
| { |
| "epoch": 4.31, |
| "learning_rate": 9.230769230769232e-05, |
| "loss": 0.012, |
| "step": 280 |
| }, |
| { |
| "epoch": 4.46, |
| "learning_rate": 8.846153846153847e-05, |
| "loss": 0.0109, |
| "step": 290 |
| }, |
| { |
| "epoch": 4.62, |
| "learning_rate": 8.461538461538461e-05, |
| "loss": 0.0067, |
| "step": 300 |
| }, |
| { |
| "epoch": 4.62, |
| "eval_accuracy": 0.9774436090225563, |
| "eval_loss": 0.0692465677857399, |
| "eval_runtime": 1.4522, |
| "eval_samples_per_second": 91.585, |
| "eval_steps_per_second": 11.706, |
| "step": 300 |
| }, |
| { |
| "epoch": 4.77, |
| "learning_rate": 8.076923076923078e-05, |
| "loss": 0.0379, |
| "step": 310 |
| }, |
| { |
| "epoch": 4.92, |
| "learning_rate": 7.692307692307693e-05, |
| "loss": 0.0116, |
| "step": 320 |
| }, |
| { |
| "epoch": 5.08, |
| "learning_rate": 7.307692307692307e-05, |
| "loss": 0.0072, |
| "step": 330 |
| }, |
| { |
| "epoch": 5.23, |
| "learning_rate": 6.923076923076924e-05, |
| "loss": 0.0064, |
| "step": 340 |
| }, |
| { |
| "epoch": 5.38, |
| "learning_rate": 6.538461538461539e-05, |
| "loss": 0.0063, |
| "step": 350 |
| }, |
| { |
| "epoch": 5.54, |
| "learning_rate": 6.153846153846155e-05, |
| "loss": 0.0055, |
| "step": 360 |
| }, |
| { |
| "epoch": 5.69, |
| "learning_rate": 5.769230769230769e-05, |
| "loss": 0.0055, |
| "step": 370 |
| }, |
| { |
| "epoch": 5.85, |
| "learning_rate": 5.384615384615385e-05, |
| "loss": 0.0053, |
| "step": 380 |
| }, |
| { |
| "epoch": 6.0, |
| "learning_rate": 5e-05, |
| "loss": 0.0051, |
| "step": 390 |
| }, |
| { |
| "epoch": 6.15, |
| "learning_rate": 4.615384615384616e-05, |
| "loss": 0.005, |
| "step": 400 |
| }, |
| { |
| "epoch": 6.15, |
| "eval_accuracy": 0.9624060150375939, |
| "eval_loss": 0.0944083109498024, |
| "eval_runtime": 1.7559, |
| "eval_samples_per_second": 75.743, |
| "eval_steps_per_second": 9.681, |
| "step": 400 |
| }, |
| { |
| "epoch": 6.31, |
| "learning_rate": 4.230769230769231e-05, |
| "loss": 0.0049, |
| "step": 410 |
| }, |
| { |
| "epoch": 6.46, |
| "learning_rate": 3.846153846153846e-05, |
| "loss": 0.0047, |
| "step": 420 |
| }, |
| { |
| "epoch": 6.62, |
| "learning_rate": 3.461538461538462e-05, |
| "loss": 0.0047, |
| "step": 430 |
| }, |
| { |
| "epoch": 6.77, |
| "learning_rate": 3.0769230769230774e-05, |
| "loss": 0.0045, |
| "step": 440 |
| }, |
| { |
| "epoch": 6.92, |
| "learning_rate": 2.6923076923076923e-05, |
| "loss": 0.0045, |
| "step": 450 |
| }, |
| { |
| "epoch": 7.08, |
| "learning_rate": 2.307692307692308e-05, |
| "loss": 0.0045, |
| "step": 460 |
| }, |
| { |
| "epoch": 7.23, |
| "learning_rate": 1.923076923076923e-05, |
| "loss": 0.0044, |
| "step": 470 |
| }, |
| { |
| "epoch": 7.38, |
| "learning_rate": 1.5384615384615387e-05, |
| "loss": 0.0044, |
| "step": 480 |
| }, |
| { |
| "epoch": 7.54, |
| "learning_rate": 1.153846153846154e-05, |
| "loss": 0.0044, |
| "step": 490 |
| }, |
| { |
| "epoch": 7.69, |
| "learning_rate": 7.692307692307694e-06, |
| "loss": 0.0043, |
| "step": 500 |
| }, |
| { |
| "epoch": 7.69, |
| "eval_accuracy": 0.9849624060150376, |
| "eval_loss": 0.05050145834684372, |
| "eval_runtime": 1.4729, |
| "eval_samples_per_second": 90.301, |
| "eval_steps_per_second": 11.542, |
| "step": 500 |
| }, |
| { |
| "epoch": 7.85, |
| "learning_rate": 3.846153846153847e-06, |
| "loss": 0.0043, |
| "step": 510 |
| }, |
| { |
| "epoch": 8.0, |
| "learning_rate": 0.0, |
| "loss": 0.0043, |
| "step": 520 |
| }, |
| { |
| "epoch": 8.0, |
| "step": 520, |
| "total_flos": 6.410194832952852e+17, |
| "train_loss": 0.08019667567255405, |
| "train_runtime": 186.4742, |
| "train_samples_per_second": 44.36, |
| "train_steps_per_second": 2.789 |
| } |
| ], |
| "max_steps": 520, |
| "num_train_epochs": 8, |
| "total_flos": 6.410194832952852e+17, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
