|
{ |
|
"best_metric": 0.042032137513160706, |
|
"best_model_checkpoint": "./beit-base-ches-demo-v0/checkpoint-300", |
|
"epoch": 4.0, |
|
"global_step": 348, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.11, |
|
"learning_rate": 0.00019540229885057472, |
|
"loss": 0.9259, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.23, |
|
"learning_rate": 0.00019022988505747128, |
|
"loss": 0.384, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.34, |
|
"learning_rate": 0.00018448275862068968, |
|
"loss": 0.1864, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.46, |
|
"learning_rate": 0.00017873563218390805, |
|
"loss": 0.1273, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.57, |
|
"learning_rate": 0.00017298850574712645, |
|
"loss": 0.1345, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.69, |
|
"learning_rate": 0.00016724137931034482, |
|
"loss": 0.1201, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 0.8, |
|
"learning_rate": 0.00016149425287356324, |
|
"loss": 0.1063, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 0.92, |
|
"learning_rate": 0.00015574712643678162, |
|
"loss": 0.1019, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 1.03, |
|
"learning_rate": 0.00015000000000000001, |
|
"loss": 0.0904, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 1.15, |
|
"learning_rate": 0.00014425287356321839, |
|
"loss": 0.0526, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 1.26, |
|
"learning_rate": 0.00013850574712643678, |
|
"loss": 0.0761, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 1.38, |
|
"learning_rate": 0.00013275862068965518, |
|
"loss": 0.0569, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 1.49, |
|
"learning_rate": 0.00012701149425287358, |
|
"loss": 0.064, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 1.61, |
|
"learning_rate": 0.00012126436781609197, |
|
"loss": 0.0463, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 1.72, |
|
"learning_rate": 0.00011551724137931035, |
|
"loss": 0.0657, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 1.84, |
|
"learning_rate": 0.00010977011494252874, |
|
"loss": 0.0571, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 1.95, |
|
"learning_rate": 0.00010402298850574712, |
|
"loss": 0.0481, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 2.07, |
|
"learning_rate": 9.827586206896552e-05, |
|
"loss": 0.0474, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 2.18, |
|
"learning_rate": 9.252873563218392e-05, |
|
"loss": 0.0173, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 2.3, |
|
"learning_rate": 8.67816091954023e-05, |
|
"loss": 0.0284, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 2.41, |
|
"learning_rate": 8.103448275862069e-05, |
|
"loss": 0.0322, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 2.53, |
|
"learning_rate": 7.528735632183909e-05, |
|
"loss": 0.0288, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 2.64, |
|
"learning_rate": 6.954022988505747e-05, |
|
"loss": 0.0165, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 2.76, |
|
"learning_rate": 6.379310344827587e-05, |
|
"loss": 0.0251, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 2.87, |
|
"learning_rate": 5.8045977011494254e-05, |
|
"loss": 0.0256, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 2.99, |
|
"learning_rate": 5.2298850574712646e-05, |
|
"loss": 0.0271, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 3.1, |
|
"learning_rate": 4.655172413793104e-05, |
|
"loss": 0.0115, |
|
"step": 270 |
|
}, |
|
{ |
|
"epoch": 3.22, |
|
"learning_rate": 4.080459770114943e-05, |
|
"loss": 0.0084, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 3.33, |
|
"learning_rate": 3.505747126436782e-05, |
|
"loss": 0.0049, |
|
"step": 290 |
|
}, |
|
{ |
|
"epoch": 3.45, |
|
"learning_rate": 2.9310344827586206e-05, |
|
"loss": 0.0183, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 3.45, |
|
"eval_accuracy": 0.9870689655172413, |
|
"eval_loss": 0.042032137513160706, |
|
"eval_runtime": 12.8713, |
|
"eval_samples_per_second": 162.221, |
|
"eval_steps_per_second": 20.278, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 3.56, |
|
"learning_rate": 2.3563218390804597e-05, |
|
"loss": 0.0096, |
|
"step": 310 |
|
}, |
|
{ |
|
"epoch": 3.68, |
|
"learning_rate": 1.781609195402299e-05, |
|
"loss": 0.0138, |
|
"step": 320 |
|
}, |
|
{ |
|
"epoch": 3.79, |
|
"learning_rate": 1.206896551724138e-05, |
|
"loss": 0.003, |
|
"step": 330 |
|
}, |
|
{ |
|
"epoch": 3.91, |
|
"learning_rate": 6.321839080459771e-06, |
|
"loss": 0.0037, |
|
"step": 340 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"step": 348, |
|
"total_flos": 3.450110813048955e+18, |
|
"train_loss": 0.08533312985138304, |
|
"train_runtime": 324.7828, |
|
"train_samples_per_second": 137.138, |
|
"train_steps_per_second": 1.071 |
|
} |
|
], |
|
"max_steps": 348, |
|
"num_train_epochs": 4, |
|
"total_flos": 3.450110813048955e+18, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|