|
{ |
|
"best_metric": 0.7170731707317073, |
|
"best_model_checkpoint": "swinv2-tiny-patch4-window8-256-finetuned-gardner-exp-max/checkpoint-174", |
|
"epoch": 14.482758620689655, |
|
"eval_steps": 500, |
|
"global_step": 210, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.69, |
|
"learning_rate": 2.380952380952381e-05, |
|
"loss": 1.6043, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.97, |
|
"eval_accuracy": 0.5414634146341464, |
|
"eval_loss": 1.5287535190582275, |
|
"eval_runtime": 3.5003, |
|
"eval_samples_per_second": 58.567, |
|
"eval_steps_per_second": 2.0, |
|
"step": 14 |
|
}, |
|
{ |
|
"epoch": 1.38, |
|
"learning_rate": 4.761904761904762e-05, |
|
"loss": 1.4967, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.5414634146341464, |
|
"eval_loss": 1.1718776226043701, |
|
"eval_runtime": 3.523, |
|
"eval_samples_per_second": 58.189, |
|
"eval_steps_per_second": 1.987, |
|
"step": 29 |
|
}, |
|
{ |
|
"epoch": 2.07, |
|
"learning_rate": 4.761904761904762e-05, |
|
"loss": 1.2615, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 2.76, |
|
"learning_rate": 4.4973544973544974e-05, |
|
"loss": 1.1276, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 2.97, |
|
"eval_accuracy": 0.5463414634146342, |
|
"eval_loss": 1.052482008934021, |
|
"eval_runtime": 3.482, |
|
"eval_samples_per_second": 58.874, |
|
"eval_steps_per_second": 2.01, |
|
"step": 43 |
|
}, |
|
{ |
|
"epoch": 3.45, |
|
"learning_rate": 4.232804232804233e-05, |
|
"loss": 1.0796, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.6536585365853659, |
|
"eval_loss": 0.9085918664932251, |
|
"eval_runtime": 3.5324, |
|
"eval_samples_per_second": 58.034, |
|
"eval_steps_per_second": 1.982, |
|
"step": 58 |
|
}, |
|
{ |
|
"epoch": 4.14, |
|
"learning_rate": 3.968253968253968e-05, |
|
"loss": 1.0019, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 4.83, |
|
"learning_rate": 3.7037037037037037e-05, |
|
"loss": 0.9387, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 4.97, |
|
"eval_accuracy": 0.6439024390243903, |
|
"eval_loss": 0.8500024676322937, |
|
"eval_runtime": 3.5195, |
|
"eval_samples_per_second": 58.246, |
|
"eval_steps_per_second": 1.989, |
|
"step": 72 |
|
}, |
|
{ |
|
"epoch": 5.52, |
|
"learning_rate": 3.439153439153439e-05, |
|
"loss": 0.9232, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"eval_accuracy": 0.6731707317073171, |
|
"eval_loss": 0.8189888000488281, |
|
"eval_runtime": 3.532, |
|
"eval_samples_per_second": 58.041, |
|
"eval_steps_per_second": 1.982, |
|
"step": 87 |
|
}, |
|
{ |
|
"epoch": 6.21, |
|
"learning_rate": 3.1746031746031745e-05, |
|
"loss": 0.9018, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 6.9, |
|
"learning_rate": 2.91005291005291e-05, |
|
"loss": 0.8456, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 6.97, |
|
"eval_accuracy": 0.6878048780487804, |
|
"eval_loss": 0.8041830062866211, |
|
"eval_runtime": 3.4986, |
|
"eval_samples_per_second": 58.594, |
|
"eval_steps_per_second": 2.001, |
|
"step": 101 |
|
}, |
|
{ |
|
"epoch": 7.59, |
|
"learning_rate": 2.6455026455026456e-05, |
|
"loss": 0.8348, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"eval_accuracy": 0.6926829268292682, |
|
"eval_loss": 0.7769930362701416, |
|
"eval_runtime": 3.5012, |
|
"eval_samples_per_second": 58.552, |
|
"eval_steps_per_second": 1.999, |
|
"step": 116 |
|
}, |
|
{ |
|
"epoch": 8.28, |
|
"learning_rate": 2.380952380952381e-05, |
|
"loss": 0.8287, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 8.97, |
|
"learning_rate": 2.1164021164021164e-05, |
|
"loss": 0.8057, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 8.97, |
|
"eval_accuracy": 0.7073170731707317, |
|
"eval_loss": 0.7456830143928528, |
|
"eval_runtime": 3.4827, |
|
"eval_samples_per_second": 58.862, |
|
"eval_steps_per_second": 2.01, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 9.66, |
|
"learning_rate": 1.8518518518518518e-05, |
|
"loss": 0.8033, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"eval_accuracy": 0.7024390243902439, |
|
"eval_loss": 0.7352668642997742, |
|
"eval_runtime": 3.5384, |
|
"eval_samples_per_second": 57.935, |
|
"eval_steps_per_second": 1.978, |
|
"step": 145 |
|
}, |
|
{ |
|
"epoch": 10.34, |
|
"learning_rate": 1.5873015873015872e-05, |
|
"loss": 0.7822, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 10.97, |
|
"eval_accuracy": 0.7121951219512195, |
|
"eval_loss": 0.7165755033493042, |
|
"eval_runtime": 3.4957, |
|
"eval_samples_per_second": 58.643, |
|
"eval_steps_per_second": 2.002, |
|
"step": 159 |
|
}, |
|
{ |
|
"epoch": 11.03, |
|
"learning_rate": 1.3227513227513228e-05, |
|
"loss": 0.8041, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 11.72, |
|
"learning_rate": 1.0582010582010582e-05, |
|
"loss": 0.7594, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 12.0, |
|
"eval_accuracy": 0.7170731707317073, |
|
"eval_loss": 0.718829333782196, |
|
"eval_runtime": 3.5638, |
|
"eval_samples_per_second": 57.522, |
|
"eval_steps_per_second": 1.964, |
|
"step": 174 |
|
}, |
|
{ |
|
"epoch": 12.41, |
|
"learning_rate": 7.936507936507936e-06, |
|
"loss": 0.7777, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 12.97, |
|
"eval_accuracy": 0.7170731707317073, |
|
"eval_loss": 0.7086274027824402, |
|
"eval_runtime": 3.4886, |
|
"eval_samples_per_second": 58.762, |
|
"eval_steps_per_second": 2.007, |
|
"step": 188 |
|
}, |
|
{ |
|
"epoch": 13.1, |
|
"learning_rate": 5.291005291005291e-06, |
|
"loss": 0.7863, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 13.79, |
|
"learning_rate": 2.6455026455026455e-06, |
|
"loss": 0.7445, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 14.0, |
|
"eval_accuracy": 0.6878048780487804, |
|
"eval_loss": 0.7139347791671753, |
|
"eval_runtime": 3.4798, |
|
"eval_samples_per_second": 58.912, |
|
"eval_steps_per_second": 2.012, |
|
"step": 203 |
|
}, |
|
{ |
|
"epoch": 14.48, |
|
"learning_rate": 0.0, |
|
"loss": 0.7513, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 14.48, |
|
"eval_accuracy": 0.6878048780487804, |
|
"eval_loss": 0.7138883471488953, |
|
"eval_runtime": 3.48, |
|
"eval_samples_per_second": 58.908, |
|
"eval_steps_per_second": 2.012, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 14.48, |
|
"step": 210, |
|
"total_flos": 8.6685208493713e+17, |
|
"train_loss": 0.9456698463076637, |
|
"train_runtime": 1102.737, |
|
"train_samples_per_second": 25.015, |
|
"train_steps_per_second": 0.19 |
|
} |
|
], |
|
"logging_steps": 10, |
|
"max_steps": 210, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 15, |
|
"save_steps": 500, |
|
"total_flos": 8.6685208493713e+17, |
|
"train_batch_size": 32, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|