djbp's picture
Training in progress, epoch 0
a983fe2 verified
raw
history blame
14.5 kB
{
"best_metric": 0.8425624321389794,
"best_model_checkpoint": "swin-tiny-patch4-window7-224-Mid-NonMidMarket-Classification/checkpoint-453",
"epoch": 9.884169884169884,
"eval_steps": 500,
"global_step": 640,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.15444015444015444,
"grad_norm": 6.145582675933838,
"learning_rate": 7.8125e-06,
"loss": 1.1143,
"step": 10
},
{
"epoch": 0.3088803088803089,
"grad_norm": 4.804866313934326,
"learning_rate": 1.5625e-05,
"loss": 0.9049,
"step": 20
},
{
"epoch": 0.46332046332046334,
"grad_norm": 4.023855686187744,
"learning_rate": 2.34375e-05,
"loss": 0.8321,
"step": 30
},
{
"epoch": 0.6177606177606177,
"grad_norm": 5.751364707946777,
"learning_rate": 3.125e-05,
"loss": 0.7323,
"step": 40
},
{
"epoch": 0.7722007722007722,
"grad_norm": 4.711376667022705,
"learning_rate": 3.90625e-05,
"loss": 0.6296,
"step": 50
},
{
"epoch": 0.9266409266409267,
"grad_norm": 5.55043363571167,
"learning_rate": 4.6875e-05,
"loss": 0.5809,
"step": 60
},
{
"epoch": 0.9884169884169884,
"eval_accuracy": 0.7937024972855592,
"eval_loss": 0.5023894309997559,
"eval_runtime": 107.4922,
"eval_samples_per_second": 8.568,
"eval_steps_per_second": 0.27,
"step": 64
},
{
"epoch": 1.0810810810810811,
"grad_norm": 4.052460670471191,
"learning_rate": 4.947916666666667e-05,
"loss": 0.5564,
"step": 70
},
{
"epoch": 1.2355212355212355,
"grad_norm": 4.5774359703063965,
"learning_rate": 4.8611111111111115e-05,
"loss": 0.5529,
"step": 80
},
{
"epoch": 1.3899613899613898,
"grad_norm": 7.030020713806152,
"learning_rate": 4.774305555555556e-05,
"loss": 0.5762,
"step": 90
},
{
"epoch": 1.5444015444015444,
"grad_norm": 4.231429100036621,
"learning_rate": 4.6875e-05,
"loss": 0.5017,
"step": 100
},
{
"epoch": 1.698841698841699,
"grad_norm": 3.0762908458709717,
"learning_rate": 4.6006944444444444e-05,
"loss": 0.5173,
"step": 110
},
{
"epoch": 1.8532818532818531,
"grad_norm": 3.419095039367676,
"learning_rate": 4.5138888888888894e-05,
"loss": 0.5326,
"step": 120
},
{
"epoch": 1.9922779922779923,
"eval_accuracy": 0.8132464712269273,
"eval_loss": 0.4402076005935669,
"eval_runtime": 106.2614,
"eval_samples_per_second": 8.667,
"eval_steps_per_second": 0.273,
"step": 129
},
{
"epoch": 2.0077220077220077,
"grad_norm": 4.168376445770264,
"learning_rate": 4.4270833333333337e-05,
"loss": 0.5082,
"step": 130
},
{
"epoch": 2.1621621621621623,
"grad_norm": 4.488511562347412,
"learning_rate": 4.340277777777778e-05,
"loss": 0.4642,
"step": 140
},
{
"epoch": 2.3166023166023164,
"grad_norm": 3.359255313873291,
"learning_rate": 4.253472222222222e-05,
"loss": 0.4952,
"step": 150
},
{
"epoch": 2.471042471042471,
"grad_norm": 4.701355457305908,
"learning_rate": 4.166666666666667e-05,
"loss": 0.4925,
"step": 160
},
{
"epoch": 2.6254826254826256,
"grad_norm": 4.631283283233643,
"learning_rate": 4.0798611111111115e-05,
"loss": 0.4795,
"step": 170
},
{
"epoch": 2.7799227799227797,
"grad_norm": 4.823609352111816,
"learning_rate": 3.993055555555556e-05,
"loss": 0.4878,
"step": 180
},
{
"epoch": 2.9343629343629343,
"grad_norm": 3.890035390853882,
"learning_rate": 3.90625e-05,
"loss": 0.4626,
"step": 190
},
{
"epoch": 2.9961389961389964,
"eval_accuracy": 0.8284473398479913,
"eval_loss": 0.4243987202644348,
"eval_runtime": 107.2966,
"eval_samples_per_second": 8.584,
"eval_steps_per_second": 0.27,
"step": 194
},
{
"epoch": 3.088803088803089,
"grad_norm": 3.165153741836548,
"learning_rate": 3.8194444444444444e-05,
"loss": 0.4434,
"step": 200
},
{
"epoch": 3.2432432432432434,
"grad_norm": 3.860111713409424,
"learning_rate": 3.7326388888888893e-05,
"loss": 0.4383,
"step": 210
},
{
"epoch": 3.3976833976833976,
"grad_norm": 7.453273296356201,
"learning_rate": 3.6458333333333336e-05,
"loss": 0.4432,
"step": 220
},
{
"epoch": 3.552123552123552,
"grad_norm": 5.200356483459473,
"learning_rate": 3.559027777777778e-05,
"loss": 0.4766,
"step": 230
},
{
"epoch": 3.7065637065637067,
"grad_norm": 4.4838433265686035,
"learning_rate": 3.472222222222222e-05,
"loss": 0.4428,
"step": 240
},
{
"epoch": 3.861003861003861,
"grad_norm": 3.7621071338653564,
"learning_rate": 3.385416666666667e-05,
"loss": 0.4778,
"step": 250
},
{
"epoch": 4.0,
"eval_accuracy": 0.8273615635179153,
"eval_loss": 0.42337119579315186,
"eval_runtime": 106.8233,
"eval_samples_per_second": 8.622,
"eval_steps_per_second": 0.271,
"step": 259
},
{
"epoch": 4.015444015444015,
"grad_norm": 4.093021869659424,
"learning_rate": 3.2986111111111115e-05,
"loss": 0.4306,
"step": 260
},
{
"epoch": 4.1698841698841695,
"grad_norm": 3.669459581375122,
"learning_rate": 3.211805555555556e-05,
"loss": 0.4606,
"step": 270
},
{
"epoch": 4.324324324324325,
"grad_norm": 4.8231892585754395,
"learning_rate": 3.125e-05,
"loss": 0.3992,
"step": 280
},
{
"epoch": 4.478764478764479,
"grad_norm": 4.587674617767334,
"learning_rate": 3.0381944444444444e-05,
"loss": 0.4024,
"step": 290
},
{
"epoch": 4.633204633204633,
"grad_norm": 9.029142379760742,
"learning_rate": 2.951388888888889e-05,
"loss": 0.4408,
"step": 300
},
{
"epoch": 4.787644787644788,
"grad_norm": 3.7836337089538574,
"learning_rate": 2.8645833333333333e-05,
"loss": 0.4332,
"step": 310
},
{
"epoch": 4.942084942084942,
"grad_norm": 5.781239032745361,
"learning_rate": 2.777777777777778e-05,
"loss": 0.4109,
"step": 320
},
{
"epoch": 4.988416988416988,
"eval_accuracy": 0.8306188925081434,
"eval_loss": 0.4197309613227844,
"eval_runtime": 106.1724,
"eval_samples_per_second": 8.675,
"eval_steps_per_second": 0.273,
"step": 323
},
{
"epoch": 5.096525096525096,
"grad_norm": 4.933718204498291,
"learning_rate": 2.6909722222222222e-05,
"loss": 0.4144,
"step": 330
},
{
"epoch": 5.250965250965251,
"grad_norm": 4.699091911315918,
"learning_rate": 2.604166666666667e-05,
"loss": 0.4191,
"step": 340
},
{
"epoch": 5.405405405405405,
"grad_norm": 5.412081718444824,
"learning_rate": 2.517361111111111e-05,
"loss": 0.4314,
"step": 350
},
{
"epoch": 5.559845559845559,
"grad_norm": 3.4998362064361572,
"learning_rate": 2.4305555555555558e-05,
"loss": 0.3793,
"step": 360
},
{
"epoch": 5.714285714285714,
"grad_norm": 3.954893112182617,
"learning_rate": 2.34375e-05,
"loss": 0.3815,
"step": 370
},
{
"epoch": 5.8687258687258685,
"grad_norm": 4.797443866729736,
"learning_rate": 2.2569444444444447e-05,
"loss": 0.3764,
"step": 380
},
{
"epoch": 5.992277992277993,
"eval_accuracy": 0.8295331161780674,
"eval_loss": 0.4095376133918762,
"eval_runtime": 106.8074,
"eval_samples_per_second": 8.623,
"eval_steps_per_second": 0.272,
"step": 388
},
{
"epoch": 6.023166023166024,
"grad_norm": 6.05122709274292,
"learning_rate": 2.170138888888889e-05,
"loss": 0.3947,
"step": 390
},
{
"epoch": 6.177606177606178,
"grad_norm": 4.6088666915893555,
"learning_rate": 2.0833333333333336e-05,
"loss": 0.3712,
"step": 400
},
{
"epoch": 6.332046332046332,
"grad_norm": 3.9029107093811035,
"learning_rate": 1.996527777777778e-05,
"loss": 0.359,
"step": 410
},
{
"epoch": 6.486486486486487,
"grad_norm": 4.032750129699707,
"learning_rate": 1.9097222222222222e-05,
"loss": 0.4075,
"step": 420
},
{
"epoch": 6.640926640926641,
"grad_norm": 4.65377140045166,
"learning_rate": 1.8229166666666668e-05,
"loss": 0.3921,
"step": 430
},
{
"epoch": 6.795366795366795,
"grad_norm": 5.404110908508301,
"learning_rate": 1.736111111111111e-05,
"loss": 0.3905,
"step": 440
},
{
"epoch": 6.94980694980695,
"grad_norm": 4.541908264160156,
"learning_rate": 1.6493055555555557e-05,
"loss": 0.3725,
"step": 450
},
{
"epoch": 6.996138996138996,
"eval_accuracy": 0.8425624321389794,
"eval_loss": 0.4046495258808136,
"eval_runtime": 105.8848,
"eval_samples_per_second": 8.698,
"eval_steps_per_second": 0.274,
"step": 453
},
{
"epoch": 7.104247104247104,
"grad_norm": 4.969923496246338,
"learning_rate": 1.5625e-05,
"loss": 0.375,
"step": 460
},
{
"epoch": 7.258687258687258,
"grad_norm": 4.746829986572266,
"learning_rate": 1.4756944444444445e-05,
"loss": 0.3536,
"step": 470
},
{
"epoch": 7.413127413127413,
"grad_norm": 6.098570823669434,
"learning_rate": 1.388888888888889e-05,
"loss": 0.3418,
"step": 480
},
{
"epoch": 7.5675675675675675,
"grad_norm": 9.79119873046875,
"learning_rate": 1.3020833333333334e-05,
"loss": 0.3769,
"step": 490
},
{
"epoch": 7.722007722007722,
"grad_norm": 7.424502849578857,
"learning_rate": 1.2152777777777779e-05,
"loss": 0.3598,
"step": 500
},
{
"epoch": 7.876447876447877,
"grad_norm": 3.9304542541503906,
"learning_rate": 1.1284722222222223e-05,
"loss": 0.3583,
"step": 510
},
{
"epoch": 8.0,
"eval_accuracy": 0.8371335504885994,
"eval_loss": 0.4108859896659851,
"eval_runtime": 107.4011,
"eval_samples_per_second": 8.575,
"eval_steps_per_second": 0.27,
"step": 518
},
{
"epoch": 8.03088803088803,
"grad_norm": 4.0118632316589355,
"learning_rate": 1.0416666666666668e-05,
"loss": 0.36,
"step": 520
},
{
"epoch": 8.185328185328185,
"grad_norm": 6.711050510406494,
"learning_rate": 9.548611111111111e-06,
"loss": 0.3427,
"step": 530
},
{
"epoch": 8.339768339768339,
"grad_norm": 4.516994476318359,
"learning_rate": 8.680555555555556e-06,
"loss": 0.3335,
"step": 540
},
{
"epoch": 8.494208494208495,
"grad_norm": 7.737695217132568,
"learning_rate": 7.8125e-06,
"loss": 0.3658,
"step": 550
},
{
"epoch": 8.64864864864865,
"grad_norm": 5.0886759757995605,
"learning_rate": 6.944444444444445e-06,
"loss": 0.3635,
"step": 560
},
{
"epoch": 8.803088803088803,
"grad_norm": 3.6143264770507812,
"learning_rate": 6.076388888888889e-06,
"loss": 0.3493,
"step": 570
},
{
"epoch": 8.957528957528957,
"grad_norm": 4.734116554260254,
"learning_rate": 5.208333333333334e-06,
"loss": 0.3451,
"step": 580
},
{
"epoch": 8.988416988416988,
"eval_accuracy": 0.8349619978284474,
"eval_loss": 0.4170722961425781,
"eval_runtime": 108.6012,
"eval_samples_per_second": 8.481,
"eval_steps_per_second": 0.267,
"step": 582
},
{
"epoch": 9.111969111969112,
"grad_norm": 4.281556129455566,
"learning_rate": 4.340277777777778e-06,
"loss": 0.3268,
"step": 590
},
{
"epoch": 9.266409266409266,
"grad_norm": 3.9609580039978027,
"learning_rate": 3.4722222222222224e-06,
"loss": 0.371,
"step": 600
},
{
"epoch": 9.420849420849422,
"grad_norm": 5.559605598449707,
"learning_rate": 2.604166666666667e-06,
"loss": 0.3446,
"step": 610
},
{
"epoch": 9.575289575289576,
"grad_norm": 4.212625980377197,
"learning_rate": 1.7361111111111112e-06,
"loss": 0.3206,
"step": 620
},
{
"epoch": 9.72972972972973,
"grad_norm": 3.9171366691589355,
"learning_rate": 8.680555555555556e-07,
"loss": 0.3654,
"step": 630
},
{
"epoch": 9.884169884169884,
"grad_norm": 5.055610179901123,
"learning_rate": 0.0,
"loss": 0.3351,
"step": 640
},
{
"epoch": 9.884169884169884,
"eval_accuracy": 0.8403908794788274,
"eval_loss": 0.41527804732322693,
"eval_runtime": 106.36,
"eval_samples_per_second": 8.659,
"eval_steps_per_second": 0.273,
"step": 640
},
{
"epoch": 9.884169884169884,
"step": 640,
"total_flos": 2.0360358039744737e+18,
"train_loss": 0.45154881179332734,
"train_runtime": 9581.2043,
"train_samples_per_second": 8.649,
"train_steps_per_second": 0.067
}
],
"logging_steps": 10,
"max_steps": 640,
"num_input_tokens_seen": 0,
"num_train_epochs": 10,
"save_steps": 500,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 2.0360358039744737e+18,
"train_batch_size": 32,
"trial_name": null,
"trial_params": null
}