surajjoshi's picture
End of training
1633a06
{
"best_metric": 0.9949179046129789,
"best_model_checkpoint": "Brain_Tumor_Classification_using_swin_transformer/checkpoint-540",
"epoch": 3.0,
"global_step": 540,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.06,
"learning_rate": 9.259259259259259e-06,
"loss": 1.1369,
"step": 10
},
{
"epoch": 0.11,
"learning_rate": 1.8518518518518518e-05,
"loss": 0.9198,
"step": 20
},
{
"epoch": 0.17,
"learning_rate": 2.777777777777778e-05,
"loss": 0.6536,
"step": 30
},
{
"epoch": 0.22,
"learning_rate": 3.7037037037037037e-05,
"loss": 0.454,
"step": 40
},
{
"epoch": 0.28,
"learning_rate": 4.62962962962963e-05,
"loss": 0.3619,
"step": 50
},
{
"epoch": 0.33,
"learning_rate": 4.938271604938271e-05,
"loss": 0.3045,
"step": 60
},
{
"epoch": 0.39,
"learning_rate": 4.835390946502058e-05,
"loss": 0.2705,
"step": 70
},
{
"epoch": 0.44,
"learning_rate": 4.732510288065844e-05,
"loss": 0.2496,
"step": 80
},
{
"epoch": 0.5,
"learning_rate": 4.62962962962963e-05,
"loss": 0.2166,
"step": 90
},
{
"epoch": 0.56,
"learning_rate": 4.5267489711934157e-05,
"loss": 0.2754,
"step": 100
},
{
"epoch": 0.61,
"learning_rate": 4.423868312757202e-05,
"loss": 0.1692,
"step": 110
},
{
"epoch": 0.67,
"learning_rate": 4.3209876543209875e-05,
"loss": 0.1326,
"step": 120
},
{
"epoch": 0.72,
"learning_rate": 4.2181069958847744e-05,
"loss": 0.1449,
"step": 130
},
{
"epoch": 0.78,
"learning_rate": 4.11522633744856e-05,
"loss": 0.1632,
"step": 140
},
{
"epoch": 0.83,
"learning_rate": 4.012345679012346e-05,
"loss": 0.1412,
"step": 150
},
{
"epoch": 0.89,
"learning_rate": 3.909465020576132e-05,
"loss": 0.1427,
"step": 160
},
{
"epoch": 0.94,
"learning_rate": 3.806584362139918e-05,
"loss": 0.127,
"step": 170
},
{
"epoch": 1.0,
"learning_rate": 3.7037037037037037e-05,
"loss": 0.081,
"step": 180
},
{
"epoch": 1.0,
"eval_accuracy": 0.9831899921813917,
"eval_f1": 0.9831899921813917,
"eval_loss": 0.055702339857816696,
"eval_precision": 0.9831899921813917,
"eval_recall": 0.9831899921813917,
"eval_runtime": 438.8809,
"eval_samples_per_second": 5.828,
"eval_steps_per_second": 0.182,
"step": 180
},
{
"epoch": 1.06,
"learning_rate": 3.60082304526749e-05,
"loss": 0.1341,
"step": 190
},
{
"epoch": 1.11,
"learning_rate": 3.497942386831276e-05,
"loss": 0.1193,
"step": 200
},
{
"epoch": 1.17,
"learning_rate": 3.395061728395062e-05,
"loss": 0.1082,
"step": 210
},
{
"epoch": 1.22,
"learning_rate": 3.292181069958848e-05,
"loss": 0.0874,
"step": 220
},
{
"epoch": 1.28,
"learning_rate": 3.1893004115226336e-05,
"loss": 0.0927,
"step": 230
},
{
"epoch": 1.33,
"learning_rate": 3.08641975308642e-05,
"loss": 0.1024,
"step": 240
},
{
"epoch": 1.39,
"learning_rate": 2.9835390946502057e-05,
"loss": 0.1166,
"step": 250
},
{
"epoch": 1.44,
"learning_rate": 2.880658436213992e-05,
"loss": 0.1043,
"step": 260
},
{
"epoch": 1.5,
"learning_rate": 2.777777777777778e-05,
"loss": 0.087,
"step": 270
},
{
"epoch": 1.56,
"learning_rate": 2.6748971193415638e-05,
"loss": 0.071,
"step": 280
},
{
"epoch": 1.61,
"learning_rate": 2.5720164609053497e-05,
"loss": 0.0836,
"step": 290
},
{
"epoch": 1.67,
"learning_rate": 2.4691358024691357e-05,
"loss": 0.0942,
"step": 300
},
{
"epoch": 1.72,
"learning_rate": 2.366255144032922e-05,
"loss": 0.118,
"step": 310
},
{
"epoch": 1.78,
"learning_rate": 2.2633744855967078e-05,
"loss": 0.069,
"step": 320
},
{
"epoch": 1.83,
"learning_rate": 2.1604938271604937e-05,
"loss": 0.072,
"step": 330
},
{
"epoch": 1.89,
"learning_rate": 2.05761316872428e-05,
"loss": 0.1062,
"step": 340
},
{
"epoch": 1.94,
"learning_rate": 1.954732510288066e-05,
"loss": 0.0815,
"step": 350
},
{
"epoch": 2.0,
"learning_rate": 1.8518518518518518e-05,
"loss": 0.0816,
"step": 360
},
{
"epoch": 2.0,
"eval_accuracy": 0.9937451133698202,
"eval_f1": 0.9937451133698202,
"eval_loss": 0.01865515671670437,
"eval_precision": 0.9937451133698202,
"eval_recall": 0.9937451133698202,
"eval_runtime": 444.8168,
"eval_samples_per_second": 5.751,
"eval_steps_per_second": 0.18,
"step": 360
},
{
"epoch": 2.06,
"learning_rate": 1.748971193415638e-05,
"loss": 0.0759,
"step": 370
},
{
"epoch": 2.11,
"learning_rate": 1.646090534979424e-05,
"loss": 0.0788,
"step": 380
},
{
"epoch": 2.17,
"learning_rate": 1.54320987654321e-05,
"loss": 0.0719,
"step": 390
},
{
"epoch": 2.22,
"learning_rate": 1.440329218106996e-05,
"loss": 0.069,
"step": 400
},
{
"epoch": 2.28,
"learning_rate": 1.3374485596707819e-05,
"loss": 0.0615,
"step": 410
},
{
"epoch": 2.33,
"learning_rate": 1.2345679012345678e-05,
"loss": 0.0574,
"step": 420
},
{
"epoch": 2.39,
"learning_rate": 1.1316872427983539e-05,
"loss": 0.0683,
"step": 430
},
{
"epoch": 2.44,
"learning_rate": 1.02880658436214e-05,
"loss": 0.0651,
"step": 440
},
{
"epoch": 2.5,
"learning_rate": 9.259259259259259e-06,
"loss": 0.0749,
"step": 450
},
{
"epoch": 2.56,
"learning_rate": 8.23045267489712e-06,
"loss": 0.052,
"step": 460
},
{
"epoch": 2.61,
"learning_rate": 7.20164609053498e-06,
"loss": 0.0653,
"step": 470
},
{
"epoch": 2.67,
"learning_rate": 6.172839506172839e-06,
"loss": 0.0687,
"step": 480
},
{
"epoch": 2.72,
"learning_rate": 5.1440329218107e-06,
"loss": 0.0538,
"step": 490
},
{
"epoch": 2.78,
"learning_rate": 4.11522633744856e-06,
"loss": 0.0577,
"step": 500
},
{
"epoch": 2.83,
"learning_rate": 3.0864197530864196e-06,
"loss": 0.0579,
"step": 510
},
{
"epoch": 2.89,
"learning_rate": 2.05761316872428e-06,
"loss": 0.0571,
"step": 520
},
{
"epoch": 2.94,
"learning_rate": 1.02880658436214e-06,
"loss": 0.0548,
"step": 530
},
{
"epoch": 3.0,
"learning_rate": 0.0,
"loss": 0.0543,
"step": 540
},
{
"epoch": 3.0,
"eval_accuracy": 0.9949179046129789,
"eval_f1": 0.9949179046129789,
"eval_loss": 0.011820005252957344,
"eval_precision": 0.9949179046129789,
"eval_recall": 0.9949179046129789,
"eval_runtime": 467.8126,
"eval_samples_per_second": 5.468,
"eval_steps_per_second": 0.171,
"step": 540
},
{
"epoch": 3.0,
"step": 540,
"total_flos": 5.409667845997019e+18,
"train_loss": 0.1632997931153686,
"train_runtime": 53634.0894,
"train_samples_per_second": 1.287,
"train_steps_per_second": 0.01
}
],
"max_steps": 540,
"num_train_epochs": 3,
"total_flos": 5.409667845997019e+18,
"trial_name": null,
"trial_params": null
}