| { |
| "best_metric": 0.04421403259038925, |
| "best_model_checkpoint": "./vit-base-beans-tutorial/checkpoint-300", |
| "epoch": 5.0, |
| "eval_steps": 100, |
| "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": 0.7495, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.31, |
| "learning_rate": 0.0001876923076923077, |
| "loss": 0.3076, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.46, |
| "learning_rate": 0.00018153846153846155, |
| "loss": 0.3115, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.62, |
| "learning_rate": 0.0001753846153846154, |
| "loss": 0.2662, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.77, |
| "learning_rate": 0.00016923076923076923, |
| "loss": 0.2006, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.92, |
| "learning_rate": 0.0001630769230769231, |
| "loss": 0.3011, |
| "step": 60 |
| }, |
| { |
| "epoch": 1.08, |
| "learning_rate": 0.00015692307692307693, |
| "loss": 0.2175, |
| "step": 70 |
| }, |
| { |
| "epoch": 1.23, |
| "learning_rate": 0.00015076923076923077, |
| "loss": 0.0878, |
| "step": 80 |
| }, |
| { |
| "epoch": 1.38, |
| "learning_rate": 0.0001446153846153846, |
| "loss": 0.1126, |
| "step": 90 |
| }, |
| { |
| "epoch": 1.54, |
| "learning_rate": 0.00013846153846153847, |
| "loss": 0.0664, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.54, |
| "eval_accuracy": 0.9774436090225563, |
| "eval_loss": 0.06984785199165344, |
| "eval_runtime": 1.8484, |
| "eval_samples_per_second": 71.952, |
| "eval_steps_per_second": 9.197, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.69, |
| "learning_rate": 0.0001323076923076923, |
| "loss": 0.1299, |
| "step": 110 |
| }, |
| { |
| "epoch": 1.85, |
| "learning_rate": 0.00012615384615384615, |
| "loss": 0.1537, |
| "step": 120 |
| }, |
| { |
| "epoch": 2.0, |
| "learning_rate": 0.00012, |
| "loss": 0.0532, |
| "step": 130 |
| }, |
| { |
| "epoch": 2.15, |
| "learning_rate": 0.00011384615384615384, |
| "loss": 0.1094, |
| "step": 140 |
| }, |
| { |
| "epoch": 2.31, |
| "learning_rate": 0.0001076923076923077, |
| "loss": 0.0215, |
| "step": 150 |
| }, |
| { |
| "epoch": 2.46, |
| "learning_rate": 0.00010153846153846153, |
| "loss": 0.056, |
| "step": 160 |
| }, |
| { |
| "epoch": 2.62, |
| "learning_rate": 9.53846153846154e-05, |
| "loss": 0.0197, |
| "step": 170 |
| }, |
| { |
| "epoch": 2.77, |
| "learning_rate": 8.923076923076924e-05, |
| "loss": 0.0202, |
| "step": 180 |
| }, |
| { |
| "epoch": 2.92, |
| "learning_rate": 8.307692307692309e-05, |
| "loss": 0.0175, |
| "step": 190 |
| }, |
| { |
| "epoch": 3.08, |
| "learning_rate": 7.692307692307693e-05, |
| "loss": 0.0115, |
| "step": 200 |
| }, |
| { |
| "epoch": 3.08, |
| "eval_accuracy": 0.9849624060150376, |
| "eval_loss": 0.06000593304634094, |
| "eval_runtime": 1.82, |
| "eval_samples_per_second": 73.075, |
| "eval_steps_per_second": 9.34, |
| "step": 200 |
| }, |
| { |
| "epoch": 3.23, |
| "learning_rate": 7.076923076923078e-05, |
| "loss": 0.0113, |
| "step": 210 |
| }, |
| { |
| "epoch": 3.38, |
| "learning_rate": 6.461538461538462e-05, |
| "loss": 0.0104, |
| "step": 220 |
| }, |
| { |
| "epoch": 3.54, |
| "learning_rate": 5.846153846153847e-05, |
| "loss": 0.0099, |
| "step": 230 |
| }, |
| { |
| "epoch": 3.69, |
| "learning_rate": 5.230769230769231e-05, |
| "loss": 0.0093, |
| "step": 240 |
| }, |
| { |
| "epoch": 3.85, |
| "learning_rate": 4.615384615384616e-05, |
| "loss": 0.0093, |
| "step": 250 |
| }, |
| { |
| "epoch": 4.0, |
| "learning_rate": 4e-05, |
| "loss": 0.009, |
| "step": 260 |
| }, |
| { |
| "epoch": 4.15, |
| "learning_rate": 3.384615384615385e-05, |
| "loss": 0.0086, |
| "step": 270 |
| }, |
| { |
| "epoch": 4.31, |
| "learning_rate": 2.7692307692307694e-05, |
| "loss": 0.0083, |
| "step": 280 |
| }, |
| { |
| "epoch": 4.46, |
| "learning_rate": 2.1538461538461542e-05, |
| "loss": 0.0082, |
| "step": 290 |
| }, |
| { |
| "epoch": 4.62, |
| "learning_rate": 1.5384615384615387e-05, |
| "loss": 0.008, |
| "step": 300 |
| }, |
| { |
| "epoch": 4.62, |
| "eval_accuracy": 0.9849624060150376, |
| "eval_loss": 0.04421403259038925, |
| "eval_runtime": 1.8047, |
| "eval_samples_per_second": 73.696, |
| "eval_steps_per_second": 9.42, |
| "step": 300 |
| }, |
| { |
| "epoch": 4.77, |
| "learning_rate": 9.230769230769232e-06, |
| "loss": 0.008, |
| "step": 310 |
| }, |
| { |
| "epoch": 4.92, |
| "learning_rate": 3.0769230769230774e-06, |
| "loss": 0.0079, |
| "step": 320 |
| }, |
| { |
| "epoch": 5.0, |
| "step": 325, |
| "total_flos": 4.006371770595533e+17, |
| "train_loss": 0.10232065571042208, |
| "train_runtime": 145.332, |
| "train_samples_per_second": 35.574, |
| "train_steps_per_second": 2.236 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 325, |
| "num_train_epochs": 5, |
| "save_steps": 100, |
| "total_flos": 4.006371770595533e+17, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
