|
{ |
|
"best_metric": 0.28963080048561096, |
|
"best_model_checkpoint": "food_type_image_detection/checkpoint-8940", |
|
"epoch": 15.0, |
|
"eval_steps": 500, |
|
"global_step": 8940, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.84, |
|
"learning_rate": 9.62089300758214e-06, |
|
"loss": 2.844, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.7601762484263533, |
|
"eval_loss": 2.02563738822937, |
|
"eval_runtime": 74.4294, |
|
"eval_samples_per_second": 64.034, |
|
"eval_steps_per_second": 8.008, |
|
"step": 596 |
|
}, |
|
{ |
|
"epoch": 1.68, |
|
"learning_rate": 9.19966301600674e-06, |
|
"loss": 1.7896, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.8583718002517835, |
|
"eval_loss": 1.3447567224502563, |
|
"eval_runtime": 74.2376, |
|
"eval_samples_per_second": 64.199, |
|
"eval_steps_per_second": 8.028, |
|
"step": 1192 |
|
}, |
|
{ |
|
"epoch": 2.52, |
|
"learning_rate": 8.77843302443134e-06, |
|
"loss": 1.2585, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy": 0.8927822073017205, |
|
"eval_loss": 0.9783705472946167, |
|
"eval_runtime": 77.5013, |
|
"eval_samples_per_second": 61.496, |
|
"eval_steps_per_second": 7.69, |
|
"step": 1788 |
|
}, |
|
{ |
|
"epoch": 3.36, |
|
"learning_rate": 8.357203032855939e-06, |
|
"loss": 0.9277, |
|
"step": 2000 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"eval_accuracy": 0.9175409148132606, |
|
"eval_loss": 0.748791515827179, |
|
"eval_runtime": 73.4549, |
|
"eval_samples_per_second": 64.883, |
|
"eval_steps_per_second": 8.114, |
|
"step": 2384 |
|
}, |
|
{ |
|
"epoch": 4.19, |
|
"learning_rate": 7.93597304128054e-06, |
|
"loss": 0.7029, |
|
"step": 2500 |
|
}, |
|
{ |
|
"epoch": 5.0, |
|
"eval_accuracy": 0.9267729752412925, |
|
"eval_loss": 0.5919772386550903, |
|
"eval_runtime": 75.09, |
|
"eval_samples_per_second": 63.471, |
|
"eval_steps_per_second": 7.937, |
|
"step": 2980 |
|
}, |
|
{ |
|
"epoch": 5.03, |
|
"learning_rate": 7.51474304970514e-06, |
|
"loss": 0.5413, |
|
"step": 3000 |
|
}, |
|
{ |
|
"epoch": 5.87, |
|
"learning_rate": 7.09351305812974e-06, |
|
"loss": 0.4146, |
|
"step": 3500 |
|
}, |
|
{ |
|
"epoch": 6.0, |
|
"eval_accuracy": 0.9313890054553084, |
|
"eval_loss": 0.4901541471481323, |
|
"eval_runtime": 78.8726, |
|
"eval_samples_per_second": 60.427, |
|
"eval_steps_per_second": 7.556, |
|
"step": 3576 |
|
}, |
|
{ |
|
"epoch": 6.71, |
|
"learning_rate": 6.672283066554339e-06, |
|
"loss": 0.3251, |
|
"step": 4000 |
|
}, |
|
{ |
|
"epoch": 7.0, |
|
"eval_accuracy": 0.9318086445656735, |
|
"eval_loss": 0.4218488931655884, |
|
"eval_runtime": 73.7814, |
|
"eval_samples_per_second": 64.596, |
|
"eval_steps_per_second": 8.078, |
|
"step": 4172 |
|
}, |
|
{ |
|
"epoch": 7.55, |
|
"learning_rate": 6.251053074978939e-06, |
|
"loss": 0.2602, |
|
"step": 4500 |
|
}, |
|
{ |
|
"epoch": 8.0, |
|
"eval_accuracy": 0.9362148552245069, |
|
"eval_loss": 0.3739270269870758, |
|
"eval_runtime": 75.0777, |
|
"eval_samples_per_second": 63.481, |
|
"eval_steps_per_second": 7.938, |
|
"step": 4768 |
|
}, |
|
{ |
|
"epoch": 8.39, |
|
"learning_rate": 5.829823083403539e-06, |
|
"loss": 0.2034, |
|
"step": 5000 |
|
}, |
|
{ |
|
"epoch": 9.0, |
|
"eval_accuracy": 0.9336970205623164, |
|
"eval_loss": 0.3482232093811035, |
|
"eval_runtime": 74.9107, |
|
"eval_samples_per_second": 63.622, |
|
"eval_steps_per_second": 7.956, |
|
"step": 5364 |
|
}, |
|
{ |
|
"epoch": 9.23, |
|
"learning_rate": 5.408593091828138e-06, |
|
"loss": 0.1691, |
|
"step": 5500 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"eval_accuracy": 0.9376835921107848, |
|
"eval_loss": 0.3193051517009735, |
|
"eval_runtime": 73.7221, |
|
"eval_samples_per_second": 64.648, |
|
"eval_steps_per_second": 8.084, |
|
"step": 5960 |
|
}, |
|
{ |
|
"epoch": 10.07, |
|
"learning_rate": 4.987363100252738e-06, |
|
"loss": 0.1355, |
|
"step": 6000 |
|
}, |
|
{ |
|
"epoch": 10.91, |
|
"learning_rate": 4.566133108677338e-06, |
|
"loss": 0.1122, |
|
"step": 6500 |
|
}, |
|
{ |
|
"epoch": 11.0, |
|
"eval_accuracy": 0.9376835921107848, |
|
"eval_loss": 0.3138887286186218, |
|
"eval_runtime": 73.4016, |
|
"eval_samples_per_second": 64.93, |
|
"eval_steps_per_second": 8.12, |
|
"step": 6556 |
|
}, |
|
{ |
|
"epoch": 11.74, |
|
"learning_rate": 4.144903117101938e-06, |
|
"loss": 0.0917, |
|
"step": 7000 |
|
}, |
|
{ |
|
"epoch": 12.0, |
|
"eval_accuracy": 0.9372639530004196, |
|
"eval_loss": 0.3041510581970215, |
|
"eval_runtime": 73.7867, |
|
"eval_samples_per_second": 64.592, |
|
"eval_steps_per_second": 8.077, |
|
"step": 7152 |
|
}, |
|
{ |
|
"epoch": 12.58, |
|
"learning_rate": 3.723673125526538e-06, |
|
"loss": 0.081, |
|
"step": 7500 |
|
}, |
|
{ |
|
"epoch": 13.0, |
|
"eval_accuracy": 0.9368443138900545, |
|
"eval_loss": 0.2987290918827057, |
|
"eval_runtime": 73.3422, |
|
"eval_samples_per_second": 64.983, |
|
"eval_steps_per_second": 8.126, |
|
"step": 7748 |
|
}, |
|
{ |
|
"epoch": 13.42, |
|
"learning_rate": 3.3024431339511375e-06, |
|
"loss": 0.0712, |
|
"step": 8000 |
|
}, |
|
{ |
|
"epoch": 14.0, |
|
"eval_accuracy": 0.936634494334872, |
|
"eval_loss": 0.2976749837398529, |
|
"eval_runtime": 74.536, |
|
"eval_samples_per_second": 63.942, |
|
"eval_steps_per_second": 7.996, |
|
"step": 8344 |
|
}, |
|
{ |
|
"epoch": 14.26, |
|
"learning_rate": 2.881213142375737e-06, |
|
"loss": 0.0571, |
|
"step": 8500 |
|
}, |
|
{ |
|
"epoch": 15.0, |
|
"eval_accuracy": 0.9385228703315149, |
|
"eval_loss": 0.28963080048561096, |
|
"eval_runtime": 75.6121, |
|
"eval_samples_per_second": 63.032, |
|
"eval_steps_per_second": 7.882, |
|
"step": 8940 |
|
} |
|
], |
|
"logging_steps": 500, |
|
"max_steps": 11920, |
|
"num_train_epochs": 20, |
|
"save_steps": 500, |
|
"total_flos": 2.2162476716746813e+19, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|