| { |
| "best_metric": 0.040521200746297836, |
| "best_model_checkpoint": "./vit-base-beans/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": 0.8627, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.31, |
| "learning_rate": 0.00018461538461538463, |
| "loss": 0.5033, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.46, |
| "learning_rate": 0.00017692307692307693, |
| "loss": 0.444, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.62, |
| "learning_rate": 0.00016923076923076923, |
| "loss": 0.3416, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.77, |
| "learning_rate": 0.00016153846153846155, |
| "loss": 0.1857, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.92, |
| "learning_rate": 0.00015384615384615385, |
| "loss": 0.1631, |
| "step": 60 |
| }, |
| { |
| "epoch": 1.08, |
| "learning_rate": 0.00014615384615384615, |
| "loss": 0.1669, |
| "step": 70 |
| }, |
| { |
| "epoch": 1.23, |
| "learning_rate": 0.00013846153846153847, |
| "loss": 0.0726, |
| "step": 80 |
| }, |
| { |
| "epoch": 1.38, |
| "learning_rate": 0.00013076923076923077, |
| "loss": 0.0391, |
| "step": 90 |
| }, |
| { |
| "epoch": 1.54, |
| "learning_rate": 0.0001230769230769231, |
| "loss": 0.1373, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.54, |
| "eval_accuracy": 0.9924812030075187, |
| "eval_loss": 0.04443267732858658, |
| "eval_runtime": 2.3927, |
| "eval_samples_per_second": 55.586, |
| "eval_steps_per_second": 7.105, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.69, |
| "learning_rate": 0.00011538461538461538, |
| "loss": 0.0624, |
| "step": 110 |
| }, |
| { |
| "epoch": 1.85, |
| "learning_rate": 0.0001076923076923077, |
| "loss": 0.1493, |
| "step": 120 |
| }, |
| { |
| "epoch": 2.0, |
| "learning_rate": 0.0001, |
| "loss": 0.0265, |
| "step": 130 |
| }, |
| { |
| "epoch": 2.15, |
| "learning_rate": 9.230769230769232e-05, |
| "loss": 0.0435, |
| "step": 140 |
| }, |
| { |
| "epoch": 2.31, |
| "learning_rate": 8.461538461538461e-05, |
| "loss": 0.0302, |
| "step": 150 |
| }, |
| { |
| "epoch": 2.46, |
| "learning_rate": 7.692307692307693e-05, |
| "loss": 0.0164, |
| "step": 160 |
| }, |
| { |
| "epoch": 2.62, |
| "learning_rate": 6.923076923076924e-05, |
| "loss": 0.028, |
| "step": 170 |
| }, |
| { |
| "epoch": 2.77, |
| "learning_rate": 6.153846153846155e-05, |
| "loss": 0.0143, |
| "step": 180 |
| }, |
| { |
| "epoch": 2.92, |
| "learning_rate": 5.384615384615385e-05, |
| "loss": 0.013, |
| "step": 190 |
| }, |
| { |
| "epoch": 3.08, |
| "learning_rate": 4.615384615384616e-05, |
| "loss": 0.0369, |
| "step": 200 |
| }, |
| { |
| "epoch": 3.08, |
| "eval_accuracy": 0.9849624060150376, |
| "eval_loss": 0.040521200746297836, |
| "eval_runtime": 2.3381, |
| "eval_samples_per_second": 56.883, |
| "eval_steps_per_second": 7.271, |
| "step": 200 |
| }, |
| { |
| "epoch": 3.23, |
| "learning_rate": 3.846153846153846e-05, |
| "loss": 0.0128, |
| "step": 210 |
| }, |
| { |
| "epoch": 3.38, |
| "learning_rate": 3.0769230769230774e-05, |
| "loss": 0.0119, |
| "step": 220 |
| }, |
| { |
| "epoch": 3.54, |
| "learning_rate": 2.307692307692308e-05, |
| "loss": 0.0125, |
| "step": 230 |
| }, |
| { |
| "epoch": 3.69, |
| "learning_rate": 1.5384615384615387e-05, |
| "loss": 0.0114, |
| "step": 240 |
| }, |
| { |
| "epoch": 3.85, |
| "learning_rate": 7.692307692307694e-06, |
| "loss": 0.0112, |
| "step": 250 |
| }, |
| { |
| "epoch": 4.0, |
| "learning_rate": 0.0, |
| "loss": 0.0112, |
| "step": 260 |
| }, |
| { |
| "epoch": 4.0, |
| "step": 260, |
| "total_flos": 3.205097416476426e+17, |
| "train_loss": 0.13106754241654506, |
| "train_runtime": 152.7249, |
| "train_samples_per_second": 27.081, |
| "train_steps_per_second": 1.702 |
| } |
| ], |
| "max_steps": 260, |
| "num_train_epochs": 4, |
| "total_flos": 3.205097416476426e+17, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
