moock's picture
End of training
23bc122
raw
history blame
8.06 kB
{
"best_metric": 0.7951807228915663,
"best_model_checkpoint": "swinv2-tiny-patch4-window8-256-finetuned-gardner-icm-max/checkpoint-11",
"epoch": 18.72340425531915,
"eval_steps": 500,
"global_step": 220,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.85,
"learning_rate": 2.272727272727273e-05,
"loss": 1.0925,
"step": 10
},
{
"epoch": 0.94,
"eval_accuracy": 0.7951807228915663,
"eval_loss": 1.0630700588226318,
"eval_runtime": 8.4566,
"eval_samples_per_second": 19.63,
"eval_steps_per_second": 0.71,
"step": 11
},
{
"epoch": 1.7,
"learning_rate": 4.545454545454546e-05,
"loss": 0.9552,
"step": 20
},
{
"epoch": 1.96,
"eval_accuracy": 0.7951807228915663,
"eval_loss": 0.633553683757782,
"eval_runtime": 2.9014,
"eval_samples_per_second": 57.213,
"eval_steps_per_second": 2.068,
"step": 23
},
{
"epoch": 2.55,
"learning_rate": 4.797979797979798e-05,
"loss": 0.6566,
"step": 30
},
{
"epoch": 2.98,
"eval_accuracy": 0.7951807228915663,
"eval_loss": 0.5356065630912781,
"eval_runtime": 2.9308,
"eval_samples_per_second": 56.639,
"eval_steps_per_second": 2.047,
"step": 35
},
{
"epoch": 3.4,
"learning_rate": 4.545454545454546e-05,
"loss": 0.5686,
"step": 40
},
{
"epoch": 4.0,
"eval_accuracy": 0.7951807228915663,
"eval_loss": 0.5150089263916016,
"eval_runtime": 3.1556,
"eval_samples_per_second": 52.605,
"eval_steps_per_second": 1.901,
"step": 47
},
{
"epoch": 4.26,
"learning_rate": 4.292929292929293e-05,
"loss": 0.5703,
"step": 50
},
{
"epoch": 4.94,
"eval_accuracy": 0.7951807228915663,
"eval_loss": 0.5129297971725464,
"eval_runtime": 2.9359,
"eval_samples_per_second": 56.542,
"eval_steps_per_second": 2.044,
"step": 58
},
{
"epoch": 5.11,
"learning_rate": 4.0404040404040405e-05,
"loss": 0.5361,
"step": 60
},
{
"epoch": 5.96,
"learning_rate": 3.787878787878788e-05,
"loss": 0.5726,
"step": 70
},
{
"epoch": 5.96,
"eval_accuracy": 0.7951807228915663,
"eval_loss": 0.5154421925544739,
"eval_runtime": 2.9132,
"eval_samples_per_second": 56.982,
"eval_steps_per_second": 2.06,
"step": 70
},
{
"epoch": 6.81,
"learning_rate": 3.535353535353535e-05,
"loss": 0.5482,
"step": 80
},
{
"epoch": 6.98,
"eval_accuracy": 0.7951807228915663,
"eval_loss": 0.5141979455947876,
"eval_runtime": 2.9168,
"eval_samples_per_second": 56.912,
"eval_steps_per_second": 2.057,
"step": 82
},
{
"epoch": 7.66,
"learning_rate": 3.282828282828283e-05,
"loss": 0.568,
"step": 90
},
{
"epoch": 8.0,
"eval_accuracy": 0.7951807228915663,
"eval_loss": 0.5108516812324524,
"eval_runtime": 2.9147,
"eval_samples_per_second": 56.952,
"eval_steps_per_second": 2.059,
"step": 94
},
{
"epoch": 8.51,
"learning_rate": 3.0303030303030306e-05,
"loss": 0.5245,
"step": 100
},
{
"epoch": 8.94,
"eval_accuracy": 0.7951807228915663,
"eval_loss": 0.5134214162826538,
"eval_runtime": 2.9007,
"eval_samples_per_second": 57.227,
"eval_steps_per_second": 2.068,
"step": 105
},
{
"epoch": 9.36,
"learning_rate": 2.777777777777778e-05,
"loss": 0.5979,
"step": 110
},
{
"epoch": 9.96,
"eval_accuracy": 0.7951807228915663,
"eval_loss": 0.5237765908241272,
"eval_runtime": 2.9436,
"eval_samples_per_second": 56.394,
"eval_steps_per_second": 2.038,
"step": 117
},
{
"epoch": 10.21,
"learning_rate": 2.5252525252525256e-05,
"loss": 0.5442,
"step": 120
},
{
"epoch": 10.98,
"eval_accuracy": 0.7951807228915663,
"eval_loss": 0.507612943649292,
"eval_runtime": 2.9562,
"eval_samples_per_second": 56.154,
"eval_steps_per_second": 2.03,
"step": 129
},
{
"epoch": 11.06,
"learning_rate": 2.272727272727273e-05,
"loss": 0.5451,
"step": 130
},
{
"epoch": 11.91,
"learning_rate": 2.0202020202020203e-05,
"loss": 0.545,
"step": 140
},
{
"epoch": 12.0,
"eval_accuracy": 0.7951807228915663,
"eval_loss": 0.5061953067779541,
"eval_runtime": 2.9818,
"eval_samples_per_second": 55.671,
"eval_steps_per_second": 2.012,
"step": 141
},
{
"epoch": 12.77,
"learning_rate": 1.7676767676767676e-05,
"loss": 0.5514,
"step": 150
},
{
"epoch": 12.94,
"eval_accuracy": 0.7951807228915663,
"eval_loss": 0.5012802481651306,
"eval_runtime": 3.0403,
"eval_samples_per_second": 54.599,
"eval_steps_per_second": 1.973,
"step": 152
},
{
"epoch": 13.62,
"learning_rate": 1.5151515151515153e-05,
"loss": 0.5377,
"step": 160
},
{
"epoch": 13.96,
"eval_accuracy": 0.7951807228915663,
"eval_loss": 0.5044662356376648,
"eval_runtime": 3.0226,
"eval_samples_per_second": 54.919,
"eval_steps_per_second": 1.985,
"step": 164
},
{
"epoch": 14.47,
"learning_rate": 1.2626262626262628e-05,
"loss": 0.5282,
"step": 170
},
{
"epoch": 14.98,
"eval_accuracy": 0.7951807228915663,
"eval_loss": 0.5037835836410522,
"eval_runtime": 2.9367,
"eval_samples_per_second": 56.525,
"eval_steps_per_second": 2.043,
"step": 176
},
{
"epoch": 15.32,
"learning_rate": 1.0101010101010101e-05,
"loss": 0.5389,
"step": 180
},
{
"epoch": 16.0,
"eval_accuracy": 0.7951807228915663,
"eval_loss": 0.4994117021560669,
"eval_runtime": 2.9413,
"eval_samples_per_second": 56.437,
"eval_steps_per_second": 2.04,
"step": 188
},
{
"epoch": 16.17,
"learning_rate": 7.5757575757575764e-06,
"loss": 0.5039,
"step": 190
},
{
"epoch": 16.94,
"eval_accuracy": 0.7951807228915663,
"eval_loss": 0.49963268637657166,
"eval_runtime": 3.0102,
"eval_samples_per_second": 55.146,
"eval_steps_per_second": 1.993,
"step": 199
},
{
"epoch": 17.02,
"learning_rate": 5.050505050505051e-06,
"loss": 0.5449,
"step": 200
},
{
"epoch": 17.87,
"learning_rate": 2.5252525252525253e-06,
"loss": 0.5348,
"step": 210
},
{
"epoch": 17.96,
"eval_accuracy": 0.7951807228915663,
"eval_loss": 0.493960976600647,
"eval_runtime": 2.9022,
"eval_samples_per_second": 57.197,
"eval_steps_per_second": 2.067,
"step": 211
},
{
"epoch": 18.72,
"learning_rate": 0.0,
"loss": 0.5426,
"step": 220
},
{
"epoch": 18.72,
"eval_accuracy": 0.7951807228915663,
"eval_loss": 0.49473580718040466,
"eval_runtime": 2.8726,
"eval_samples_per_second": 57.787,
"eval_steps_per_second": 2.089,
"step": 220
},
{
"epoch": 18.72,
"step": 220,
"total_flos": 9.056508197685166e+17,
"train_loss": 0.595783019065857,
"train_runtime": 1248.9487,
"train_samples_per_second": 23.796,
"train_steps_per_second": 0.176
}
],
"logging_steps": 10,
"max_steps": 220,
"num_input_tokens_seen": 0,
"num_train_epochs": 20,
"save_steps": 500,
"total_flos": 9.056508197685166e+17,
"train_batch_size": 32,
"trial_name": null,
"trial_params": null
}