|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 3.2467532467532467, |
|
"global_step": 1000, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.32, |
|
"learning_rate": 5.401e-05, |
|
"loss": 0.6612, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.65, |
|
"learning_rate": 4.8020000000000004e-05, |
|
"loss": 0.1793, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.97, |
|
"learning_rate": 4.203e-05, |
|
"loss": 0.1, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy_background": NaN, |
|
"eval_accuracy_left": 0.002532754418037782, |
|
"eval_accuracy_right": 0.0, |
|
"eval_iou_background": 0.0, |
|
"eval_iou_left": 0.0025213406707686975, |
|
"eval_iou_right": 0.0, |
|
"eval_loss": 0.08616962283849716, |
|
"eval_mean_accuracy": 0.001266377209018891, |
|
"eval_mean_iou": 0.0008404468902562325, |
|
"eval_overall_accuracy": 0.0012295585343380525, |
|
"eval_runtime": 611.9523, |
|
"eval_samples_per_second": 1.005, |
|
"eval_steps_per_second": 0.126, |
|
"step": 308 |
|
}, |
|
{ |
|
"epoch": 1.3, |
|
"learning_rate": 3.604e-05, |
|
"loss": 0.0801, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 1.62, |
|
"learning_rate": 3.0050000000000002e-05, |
|
"loss": 0.0679, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 1.95, |
|
"learning_rate": 2.406e-05, |
|
"loss": 0.0596, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy_background": NaN, |
|
"eval_accuracy_left": 0.09400481703955196, |
|
"eval_accuracy_right": 0.13127195102648453, |
|
"eval_iou_background": 0.0, |
|
"eval_iou_left": 0.09073784898499374, |
|
"eval_iou_right": 0.12281592100333956, |
|
"eval_loss": 0.05965268611907959, |
|
"eval_mean_accuracy": 0.11263838403301825, |
|
"eval_mean_iou": 0.0711845899961111, |
|
"eval_overall_accuracy": 0.1131801367085955, |
|
"eval_runtime": 606.9819, |
|
"eval_samples_per_second": 1.013, |
|
"eval_steps_per_second": 0.127, |
|
"step": 616 |
|
}, |
|
{ |
|
"epoch": 2.27, |
|
"learning_rate": 1.807e-05, |
|
"loss": 0.0549, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 2.6, |
|
"learning_rate": 1.2079999999999998e-05, |
|
"loss": 0.0533, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 2.92, |
|
"learning_rate": 6.0899999999999984e-06, |
|
"loss": 0.0506, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy_background": NaN, |
|
"eval_accuracy_left": 0.1805120280897246, |
|
"eval_accuracy_right": 0.05364071543769438, |
|
"eval_iou_background": 0.0, |
|
"eval_iou_left": 0.15389004121309202, |
|
"eval_iou_right": 0.05084237202230107, |
|
"eval_loss": 0.05508435145020485, |
|
"eval_mean_accuracy": 0.1170763717637095, |
|
"eval_mean_iou": 0.06824413774513104, |
|
"eval_overall_accuracy": 0.11523204230420943, |
|
"eval_runtime": 604.1982, |
|
"eval_samples_per_second": 1.018, |
|
"eval_steps_per_second": 0.127, |
|
"step": 924 |
|
}, |
|
{ |
|
"epoch": 3.25, |
|
"learning_rate": 1e-07, |
|
"loss": 0.0494, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 3.25, |
|
"eval_accuracy_background": NaN, |
|
"eval_accuracy_left": 0.18931908534285133, |
|
"eval_accuracy_right": 0.05783050064382548, |
|
"eval_iou_background": 0.0, |
|
"eval_iou_left": 0.1581183213882954, |
|
"eval_iou_right": 0.05440017513134851, |
|
"eval_loss": 0.05480428412556648, |
|
"eval_mean_accuracy": 0.1235747929933384, |
|
"eval_mean_iou": 0.0708394988398813, |
|
"eval_overall_accuracy": 0.12166334220687487, |
|
"eval_runtime": 607.9543, |
|
"eval_samples_per_second": 1.012, |
|
"eval_steps_per_second": 0.127, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 3.25, |
|
"step": 1000, |
|
"total_flos": 1.4002299557904384e+17, |
|
"train_loss": 0.13563978910446167, |
|
"train_runtime": 2724.3965, |
|
"train_samples_per_second": 2.936, |
|
"train_steps_per_second": 0.367 |
|
} |
|
], |
|
"max_steps": 1000, |
|
"num_train_epochs": 4, |
|
"total_flos": 1.4002299557904384e+17, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|