|
{ |
|
"best_metric": 0.8158415841584158, |
|
"best_model_checkpoint": "food-image-classification/checkpoint-16000", |
|
"epoch": 17.95142555438226, |
|
"eval_steps": 1000, |
|
"global_step": 17000, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 1.06, |
|
"learning_rate": 1.0559662090813095e-06, |
|
"loss": 4.6112, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 1.06, |
|
"eval_accuracy": 0.034851485148514855, |
|
"eval_loss": 4.575930118560791, |
|
"eval_runtime": 157.053, |
|
"eval_samples_per_second": 96.464, |
|
"eval_steps_per_second": 6.03, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 2.11, |
|
"learning_rate": 2.111932418162619e-06, |
|
"loss": 4.4899, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 2.11, |
|
"eval_accuracy": 0.3103630363036304, |
|
"eval_loss": 4.3788862228393555, |
|
"eval_runtime": 154.3948, |
|
"eval_samples_per_second": 98.125, |
|
"eval_steps_per_second": 6.134, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 3.17, |
|
"learning_rate": 3.167898627243928e-06, |
|
"loss": 4.2111, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 3.17, |
|
"eval_accuracy": 0.5498349834983498, |
|
"eval_loss": 4.030922889709473, |
|
"eval_runtime": 155.2257, |
|
"eval_samples_per_second": 97.6, |
|
"eval_steps_per_second": 6.101, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 4.22, |
|
"learning_rate": 4.223864836325238e-06, |
|
"loss": 3.8257, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 4.22, |
|
"eval_accuracy": 0.6111551155115511, |
|
"eval_loss": 3.634243965148926, |
|
"eval_runtime": 156.2293, |
|
"eval_samples_per_second": 96.973, |
|
"eval_steps_per_second": 6.062, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 5.28, |
|
"learning_rate": 5.279831045406547e-06, |
|
"loss": 3.4182, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 5.28, |
|
"eval_accuracy": 0.6514851485148515, |
|
"eval_loss": 3.225186586380005, |
|
"eval_runtime": 154.6911, |
|
"eval_samples_per_second": 97.937, |
|
"eval_steps_per_second": 6.122, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 6.34, |
|
"learning_rate": 6.335797254487856e-06, |
|
"loss": 2.9962, |
|
"step": 6000 |
|
}, |
|
{ |
|
"epoch": 6.34, |
|
"eval_accuracy": 0.687062706270627, |
|
"eval_loss": 2.805878162384033, |
|
"eval_runtime": 154.6826, |
|
"eval_samples_per_second": 97.943, |
|
"eval_steps_per_second": 6.122, |
|
"step": 6000 |
|
}, |
|
{ |
|
"epoch": 7.39, |
|
"learning_rate": 7.3917634635691666e-06, |
|
"loss": 2.5605, |
|
"step": 7000 |
|
}, |
|
{ |
|
"epoch": 7.39, |
|
"eval_accuracy": 0.7070627062706271, |
|
"eval_loss": 2.382246494293213, |
|
"eval_runtime": 155.5967, |
|
"eval_samples_per_second": 97.367, |
|
"eval_steps_per_second": 6.086, |
|
"step": 7000 |
|
}, |
|
{ |
|
"epoch": 8.45, |
|
"learning_rate": 8.447729672650476e-06, |
|
"loss": 2.1397, |
|
"step": 8000 |
|
}, |
|
{ |
|
"epoch": 8.45, |
|
"eval_accuracy": 0.7335973597359736, |
|
"eval_loss": 1.975380539894104, |
|
"eval_runtime": 156.8793, |
|
"eval_samples_per_second": 96.571, |
|
"eval_steps_per_second": 6.036, |
|
"step": 8000 |
|
}, |
|
{ |
|
"epoch": 9.5, |
|
"learning_rate": 9.503695881731786e-06, |
|
"loss": 1.7383, |
|
"step": 9000 |
|
}, |
|
{ |
|
"epoch": 9.5, |
|
"eval_accuracy": 0.7576897689768977, |
|
"eval_loss": 1.608676552772522, |
|
"eval_runtime": 154.5661, |
|
"eval_samples_per_second": 98.016, |
|
"eval_steps_per_second": 6.127, |
|
"step": 9000 |
|
}, |
|
{ |
|
"epoch": 10.56, |
|
"learning_rate": 1.0559662090813093e-05, |
|
"loss": 1.3909, |
|
"step": 10000 |
|
}, |
|
{ |
|
"epoch": 10.56, |
|
"eval_accuracy": 0.7758415841584159, |
|
"eval_loss": 1.3203929662704468, |
|
"eval_runtime": 155.6061, |
|
"eval_samples_per_second": 97.361, |
|
"eval_steps_per_second": 6.086, |
|
"step": 10000 |
|
}, |
|
{ |
|
"epoch": 11.62, |
|
"learning_rate": 1.1615628299894405e-05, |
|
"loss": 1.1223, |
|
"step": 11000 |
|
}, |
|
{ |
|
"epoch": 11.62, |
|
"eval_accuracy": 0.7831683168316832, |
|
"eval_loss": 1.1283260583877563, |
|
"eval_runtime": 153.8564, |
|
"eval_samples_per_second": 98.468, |
|
"eval_steps_per_second": 6.155, |
|
"step": 11000 |
|
}, |
|
{ |
|
"epoch": 12.67, |
|
"learning_rate": 1.2671594508975712e-05, |
|
"loss": 0.9312, |
|
"step": 12000 |
|
}, |
|
{ |
|
"epoch": 12.67, |
|
"eval_accuracy": 0.7946534653465347, |
|
"eval_loss": 0.9766868352890015, |
|
"eval_runtime": 155.0819, |
|
"eval_samples_per_second": 97.69, |
|
"eval_steps_per_second": 6.106, |
|
"step": 12000 |
|
}, |
|
{ |
|
"epoch": 13.73, |
|
"learning_rate": 1.3727560718057022e-05, |
|
"loss": 0.7817, |
|
"step": 13000 |
|
}, |
|
{ |
|
"epoch": 13.73, |
|
"eval_accuracy": 0.7984158415841585, |
|
"eval_loss": 0.8917332887649536, |
|
"eval_runtime": 154.0819, |
|
"eval_samples_per_second": 98.324, |
|
"eval_steps_per_second": 6.146, |
|
"step": 13000 |
|
}, |
|
{ |
|
"epoch": 14.78, |
|
"learning_rate": 1.4783526927138333e-05, |
|
"loss": 0.697, |
|
"step": 14000 |
|
}, |
|
{ |
|
"epoch": 14.78, |
|
"eval_accuracy": 0.8043564356435644, |
|
"eval_loss": 0.8234559297561646, |
|
"eval_runtime": 153.6348, |
|
"eval_samples_per_second": 98.61, |
|
"eval_steps_per_second": 6.164, |
|
"step": 14000 |
|
}, |
|
{ |
|
"epoch": 15.84, |
|
"learning_rate": 1.583949313621964e-05, |
|
"loss": 0.6281, |
|
"step": 15000 |
|
}, |
|
{ |
|
"epoch": 15.84, |
|
"eval_accuracy": 0.8075247524752476, |
|
"eval_loss": 0.7959182858467102, |
|
"eval_runtime": 153.9402, |
|
"eval_samples_per_second": 98.415, |
|
"eval_steps_per_second": 6.152, |
|
"step": 15000 |
|
}, |
|
{ |
|
"epoch": 16.9, |
|
"learning_rate": 1.6895459345300952e-05, |
|
"loss": 0.5659, |
|
"step": 16000 |
|
}, |
|
{ |
|
"epoch": 16.9, |
|
"eval_accuracy": 0.8158415841584158, |
|
"eval_loss": 0.7547946572303772, |
|
"eval_runtime": 155.6477, |
|
"eval_samples_per_second": 97.335, |
|
"eval_steps_per_second": 6.084, |
|
"step": 16000 |
|
}, |
|
{ |
|
"epoch": 17.95, |
|
"learning_rate": 1.795142555438226e-05, |
|
"loss": 0.5198, |
|
"step": 17000 |
|
}, |
|
{ |
|
"epoch": 17.95, |
|
"eval_accuracy": 0.8104950495049505, |
|
"eval_loss": 0.7739244103431702, |
|
"eval_runtime": 155.4026, |
|
"eval_samples_per_second": 97.489, |
|
"eval_steps_per_second": 6.094, |
|
"step": 17000 |
|
} |
|
], |
|
"logging_steps": 1000, |
|
"max_steps": 473500, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 500, |
|
"save_steps": 1000, |
|
"total_flos": 8.437554632116593e+19, |
|
"train_batch_size": 16, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|