| { | |
| "best_metric": 0.73, | |
| "best_model_checkpoint": "distilhubert-finetuned-gtzan/checkpoint-25", | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 0.0002, | |
| "loss": 0.0001, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 0.0004, | |
| "loss": 0.0427, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 0.0006, | |
| "loss": 0.2245, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0008, | |
| "loss": 0.468, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.001, | |
| "loss": 0.8622, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.73, | |
| "eval_loss": 2.1819140911102295, | |
| "eval_runtime": 144.9991, | |
| "eval_samples_per_second": 1.379, | |
| "eval_steps_per_second": 0.172, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.0009777777777777777, | |
| "loss": 1.6806, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 0.0009555555555555556, | |
| "loss": 2.4373, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.0009333333333333333, | |
| "loss": 2.3049, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 0.0009111111111111111, | |
| "loss": 1.55, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.0008888888888888888, | |
| "loss": 1.2651, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.615, | |
| "eval_loss": 1.3465973138809204, | |
| "eval_runtime": 143.8182, | |
| "eval_samples_per_second": 1.391, | |
| "eval_steps_per_second": 0.174, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 0.0008666666666666667, | |
| "loss": 0.7705, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 0.0008444444444444444, | |
| "loss": 0.6799, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "learning_rate": 0.0008222222222222222, | |
| "loss": 0.7068, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 0.0008, | |
| "loss": 0.934, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 0.0007777777777777778, | |
| "loss": 0.7861, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.605, | |
| "eval_loss": 1.3502554893493652, | |
| "eval_runtime": 149.3628, | |
| "eval_samples_per_second": 1.339, | |
| "eval_steps_per_second": 0.167, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 0.0007555555555555555, | |
| "loss": 0.8192, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "learning_rate": 0.0007333333333333333, | |
| "loss": 0.6076, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "learning_rate": 0.0007111111111111111, | |
| "loss": 0.7342, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "learning_rate": 0.000688888888888889, | |
| "loss": 0.8138, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 0.0006666666666666666, | |
| "loss": 0.7081, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.725, | |
| "eval_loss": 0.7991036176681519, | |
| "eval_runtime": 146.6273, | |
| "eval_samples_per_second": 1.364, | |
| "eval_steps_per_second": 0.171, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 0.0007, | |
| "loss": 0.6077, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 0.0007333333333333333, | |
| "loss": 0.2685, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 0.0007666666666666667, | |
| "loss": 1.048, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 0.0008, | |
| "loss": 1.1636, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "learning_rate": 0.0008333333333333334, | |
| "loss": 1.0052, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "learning_rate": 0.0008666666666666667, | |
| "loss": 2.0122, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 0.0009000000000000001, | |
| "loss": 1.6791, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 0.0009333333333333333, | |
| "loss": 1.3824, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 0.0009666666666666667, | |
| "loss": 1.0703, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 0.001, | |
| "loss": 1.4339, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 0.0009962962962962963, | |
| "loss": 1.5909, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.0009925925925925927, | |
| "loss": 2.139, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 0.000988888888888889, | |
| "loss": 1.8476, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 0.000985185185185185, | |
| "loss": 2.1354, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 0.0009814814814814816, | |
| "loss": 2.0248, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 0.0009777777777777777, | |
| "loss": 2.1732, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 0.0009740740740740741, | |
| "loss": 1.9887, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 0.0009703703703703704, | |
| "loss": 2.0138, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 0.0009666666666666667, | |
| "loss": 1.9447, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.0009629629629629629, | |
| "loss": 1.9785, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.33, | |
| "eval_loss": 2.211935520172119, | |
| "eval_runtime": 145.0026, | |
| "eval_samples_per_second": 1.379, | |
| "eval_steps_per_second": 0.172, | |
| "step": 200 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 1500, | |
| "num_train_epochs": 15, | |
| "save_steps": 500, | |
| "total_flos": 2.725837948512e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |