|
{ |
|
"best_metric": 0.0017420644871890545, |
|
"best_model_checkpoint": "./vit-base-beans-demo-v5/checkpoint-400", |
|
"epoch": 1.0, |
|
"global_step": 441, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.02, |
|
"learning_rate": 0.00019546485260770976, |
|
"loss": 0.273, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.05, |
|
"learning_rate": 0.0001909297052154195, |
|
"loss": 0.0234, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.07, |
|
"learning_rate": 0.00018639455782312926, |
|
"loss": 0.021, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.09, |
|
"learning_rate": 0.000181859410430839, |
|
"loss": 0.0173, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.11, |
|
"learning_rate": 0.00017732426303854876, |
|
"loss": 0.0073, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.14, |
|
"learning_rate": 0.0001727891156462585, |
|
"loss": 0.0081, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 0.16, |
|
"learning_rate": 0.00016825396825396826, |
|
"loss": 0.0219, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 0.18, |
|
"learning_rate": 0.000163718820861678, |
|
"loss": 0.004, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 0.2, |
|
"learning_rate": 0.00015918367346938776, |
|
"loss": 0.0039, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 0.23, |
|
"learning_rate": 0.00015464852607709753, |
|
"loss": 0.0033, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.23, |
|
"eval_f1": 1.0, |
|
"eval_loss": 0.0031821790616959333, |
|
"eval_runtime": 31.2658, |
|
"eval_samples_per_second": 36.781, |
|
"eval_steps_per_second": 4.606, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.25, |
|
"learning_rate": 0.00015011337868480726, |
|
"loss": 0.003, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 0.27, |
|
"learning_rate": 0.000145578231292517, |
|
"loss": 0.0027, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 0.29, |
|
"learning_rate": 0.00014104308390022676, |
|
"loss": 0.0024, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 0.32, |
|
"learning_rate": 0.0001365079365079365, |
|
"loss": 0.0026, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 0.34, |
|
"learning_rate": 0.00013197278911564626, |
|
"loss": 0.0022, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 0.36, |
|
"learning_rate": 0.000127437641723356, |
|
"loss": 0.002, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 0.39, |
|
"learning_rate": 0.00012290249433106578, |
|
"loss": 0.0019, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 0.41, |
|
"learning_rate": 0.00011836734693877552, |
|
"loss": 0.0021, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 0.43, |
|
"learning_rate": 0.00011383219954648527, |
|
"loss": 0.0017, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 0.45, |
|
"learning_rate": 0.000109297052154195, |
|
"loss": 0.0017, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.45, |
|
"eval_f1": 1.0, |
|
"eval_loss": 0.0017858472419902682, |
|
"eval_runtime": 30.1491, |
|
"eval_samples_per_second": 38.144, |
|
"eval_steps_per_second": 4.776, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 0.48, |
|
"learning_rate": 0.00010476190476190477, |
|
"loss": 0.0016, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 0.5, |
|
"learning_rate": 0.0001002267573696145, |
|
"loss": 0.0015, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 0.52, |
|
"learning_rate": 9.569160997732427e-05, |
|
"loss": 0.002, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 0.54, |
|
"learning_rate": 9.115646258503402e-05, |
|
"loss": 0.0014, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 0.57, |
|
"learning_rate": 8.662131519274377e-05, |
|
"loss": 0.0013, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 0.59, |
|
"learning_rate": 8.208616780045352e-05, |
|
"loss": 0.0013, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 0.61, |
|
"learning_rate": 7.755102040816327e-05, |
|
"loss": 0.0013, |
|
"step": 270 |
|
}, |
|
{ |
|
"epoch": 0.63, |
|
"learning_rate": 7.301587301587302e-05, |
|
"loss": 0.0012, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 0.66, |
|
"learning_rate": 6.848072562358277e-05, |
|
"loss": 0.0012, |
|
"step": 290 |
|
}, |
|
{ |
|
"epoch": 0.68, |
|
"learning_rate": 6.394557823129253e-05, |
|
"loss": 0.0012, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.68, |
|
"eval_f1": 0.9990009990009989, |
|
"eval_loss": 0.001951999613083899, |
|
"eval_runtime": 30.3102, |
|
"eval_samples_per_second": 37.941, |
|
"eval_steps_per_second": 4.751, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 0.7, |
|
"learning_rate": 5.9410430839002274e-05, |
|
"loss": 0.0011, |
|
"step": 310 |
|
}, |
|
{ |
|
"epoch": 0.73, |
|
"learning_rate": 5.4875283446712023e-05, |
|
"loss": 0.0011, |
|
"step": 320 |
|
}, |
|
{ |
|
"epoch": 0.75, |
|
"learning_rate": 5.034013605442177e-05, |
|
"loss": 0.0011, |
|
"step": 330 |
|
}, |
|
{ |
|
"epoch": 0.77, |
|
"learning_rate": 4.580498866213152e-05, |
|
"loss": 0.0011, |
|
"step": 340 |
|
}, |
|
{ |
|
"epoch": 0.79, |
|
"learning_rate": 4.126984126984127e-05, |
|
"loss": 0.001, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 0.82, |
|
"learning_rate": 3.673469387755102e-05, |
|
"loss": 0.001, |
|
"step": 360 |
|
}, |
|
{ |
|
"epoch": 0.84, |
|
"learning_rate": 3.219954648526077e-05, |
|
"loss": 0.001, |
|
"step": 370 |
|
}, |
|
{ |
|
"epoch": 0.86, |
|
"learning_rate": 2.7664399092970522e-05, |
|
"loss": 0.001, |
|
"step": 380 |
|
}, |
|
{ |
|
"epoch": 0.88, |
|
"learning_rate": 2.3129251700680275e-05, |
|
"loss": 0.001, |
|
"step": 390 |
|
}, |
|
{ |
|
"epoch": 0.91, |
|
"learning_rate": 1.8594104308390022e-05, |
|
"loss": 0.001, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 0.91, |
|
"eval_f1": 0.9990009990009989, |
|
"eval_loss": 0.0017420644871890545, |
|
"eval_runtime": 30.1934, |
|
"eval_samples_per_second": 38.088, |
|
"eval_steps_per_second": 4.769, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 0.93, |
|
"learning_rate": 1.4058956916099775e-05, |
|
"loss": 0.0009, |
|
"step": 410 |
|
}, |
|
{ |
|
"epoch": 0.95, |
|
"learning_rate": 9.523809523809523e-06, |
|
"loss": 0.0009, |
|
"step": 420 |
|
}, |
|
{ |
|
"epoch": 0.98, |
|
"learning_rate": 4.9886621315192745e-06, |
|
"loss": 0.001, |
|
"step": 430 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"learning_rate": 4.53514739229025e-07, |
|
"loss": 0.001, |
|
"step": 440 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"step": 441, |
|
"total_flos": 5.463185267828736e+17, |
|
"train_loss": 0.009828982074758727, |
|
"train_runtime": 378.8692, |
|
"train_samples_per_second": 18.608, |
|
"train_steps_per_second": 1.164 |
|
} |
|
], |
|
"max_steps": 441, |
|
"num_train_epochs": 1, |
|
"total_flos": 5.463185267828736e+17, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|