vit-focal-skin / trainer_state.json
khleeloo's picture
Training in progress, epoch 1
043aa4a
{
"best_metric": 0.8704663212435233,
"best_model_checkpoint": "./vit-focal-skin/checkpoint-3130",
"epoch": 6.0,
"global_step": 3756,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.08,
"learning_rate": 0.00019733759318423856,
"loss": 0.325,
"step": 50
},
{
"epoch": 0.16,
"learning_rate": 0.0001946751863684771,
"loss": 0.2897,
"step": 100
},
{
"epoch": 0.24,
"learning_rate": 0.00019201277955271565,
"loss": 0.2508,
"step": 150
},
{
"epoch": 0.32,
"learning_rate": 0.00018935037273695422,
"loss": 0.2626,
"step": 200
},
{
"epoch": 0.4,
"learning_rate": 0.00018668796592119277,
"loss": 0.2308,
"step": 250
},
{
"epoch": 0.48,
"learning_rate": 0.00018402555910543132,
"loss": 0.2437,
"step": 300
},
{
"epoch": 0.56,
"learning_rate": 0.00018136315228966987,
"loss": 0.2594,
"step": 350
},
{
"epoch": 0.64,
"learning_rate": 0.0001787007454739084,
"loss": 0.2355,
"step": 400
},
{
"epoch": 0.72,
"learning_rate": 0.000176038338658147,
"loss": 0.2457,
"step": 450
},
{
"epoch": 0.8,
"learning_rate": 0.0001733759318423855,
"loss": 0.2136,
"step": 500
},
{
"epoch": 0.88,
"learning_rate": 0.00017071352502662408,
"loss": 0.2622,
"step": 550
},
{
"epoch": 0.96,
"learning_rate": 0.00016805111821086263,
"loss": 0.1966,
"step": 600
},
{
"epoch": 1.0,
"eval_accuracy": 0.8290155440414507,
"eval_f1": 0.8306733167328754,
"eval_loss": 0.3647349774837494,
"eval_precision": 0.8430511449105799,
"eval_recall": 0.8290155440414507,
"eval_runtime": 1.4278,
"eval_samples_per_second": 135.17,
"eval_steps_per_second": 17.509,
"step": 626
},
{
"epoch": 1.04,
"learning_rate": 0.00016538871139510118,
"loss": 0.1998,
"step": 650
},
{
"epoch": 1.12,
"learning_rate": 0.00016272630457933972,
"loss": 0.1312,
"step": 700
},
{
"epoch": 1.2,
"learning_rate": 0.0001600638977635783,
"loss": 0.1477,
"step": 750
},
{
"epoch": 1.28,
"learning_rate": 0.00015740149094781684,
"loss": 0.1155,
"step": 800
},
{
"epoch": 1.36,
"learning_rate": 0.0001547390841320554,
"loss": 0.1591,
"step": 850
},
{
"epoch": 1.44,
"learning_rate": 0.00015207667731629394,
"loss": 0.1568,
"step": 900
},
{
"epoch": 1.52,
"learning_rate": 0.00014941427050053249,
"loss": 0.1322,
"step": 950
},
{
"epoch": 1.6,
"learning_rate": 0.00014675186368477103,
"loss": 0.1478,
"step": 1000
},
{
"epoch": 1.68,
"learning_rate": 0.00014408945686900958,
"loss": 0.1145,
"step": 1050
},
{
"epoch": 1.76,
"learning_rate": 0.00014142705005324815,
"loss": 0.1483,
"step": 1100
},
{
"epoch": 1.84,
"learning_rate": 0.0001387646432374867,
"loss": 0.1517,
"step": 1150
},
{
"epoch": 1.92,
"learning_rate": 0.00013610223642172525,
"loss": 0.1294,
"step": 1200
},
{
"epoch": 2.0,
"learning_rate": 0.0001334398296059638,
"loss": 0.1434,
"step": 1250
},
{
"epoch": 2.0,
"eval_accuracy": 0.8238341968911918,
"eval_f1": 0.825913518485432,
"eval_loss": 0.38836923241615295,
"eval_precision": 0.8418087741274269,
"eval_recall": 0.8238341968911918,
"eval_runtime": 1.4008,
"eval_samples_per_second": 137.782,
"eval_steps_per_second": 17.847,
"step": 1252
},
{
"epoch": 2.08,
"learning_rate": 0.00013077742279020234,
"loss": 0.0846,
"step": 1300
},
{
"epoch": 2.16,
"learning_rate": 0.00012811501597444092,
"loss": 0.0635,
"step": 1350
},
{
"epoch": 2.24,
"learning_rate": 0.00012545260915867944,
"loss": 0.0466,
"step": 1400
},
{
"epoch": 2.32,
"learning_rate": 0.000122790202342918,
"loss": 0.0757,
"step": 1450
},
{
"epoch": 2.4,
"learning_rate": 0.00012012779552715656,
"loss": 0.0996,
"step": 1500
},
{
"epoch": 2.48,
"learning_rate": 0.00011746538871139509,
"loss": 0.089,
"step": 1550
},
{
"epoch": 2.56,
"learning_rate": 0.00011480298189563365,
"loss": 0.0679,
"step": 1600
},
{
"epoch": 2.64,
"learning_rate": 0.00011214057507987221,
"loss": 0.0747,
"step": 1650
},
{
"epoch": 2.72,
"learning_rate": 0.00010947816826411077,
"loss": 0.066,
"step": 1700
},
{
"epoch": 2.8,
"learning_rate": 0.0001068157614483493,
"loss": 0.0714,
"step": 1750
},
{
"epoch": 2.88,
"learning_rate": 0.00010415335463258787,
"loss": 0.0776,
"step": 1800
},
{
"epoch": 2.96,
"learning_rate": 0.00010149094781682643,
"loss": 0.058,
"step": 1850
},
{
"epoch": 3.0,
"eval_accuracy": 0.8186528497409327,
"eval_f1": 0.8136687258538028,
"eval_loss": 0.5063741207122803,
"eval_precision": 0.8183198316271921,
"eval_recall": 0.8186528497409327,
"eval_runtime": 1.4832,
"eval_samples_per_second": 130.125,
"eval_steps_per_second": 16.856,
"step": 1878
},
{
"epoch": 3.04,
"learning_rate": 9.882854100106496e-05,
"loss": 0.0361,
"step": 1900
},
{
"epoch": 3.12,
"learning_rate": 9.616613418530351e-05,
"loss": 0.0305,
"step": 1950
},
{
"epoch": 3.19,
"learning_rate": 9.350372736954207e-05,
"loss": 0.0224,
"step": 2000
},
{
"epoch": 3.27,
"learning_rate": 9.084132055378062e-05,
"loss": 0.0186,
"step": 2050
},
{
"epoch": 3.35,
"learning_rate": 8.817891373801918e-05,
"loss": 0.0138,
"step": 2100
},
{
"epoch": 3.43,
"learning_rate": 8.556975505857295e-05,
"loss": 0.0344,
"step": 2150
},
{
"epoch": 3.51,
"learning_rate": 8.29073482428115e-05,
"loss": 0.0158,
"step": 2200
},
{
"epoch": 3.59,
"learning_rate": 8.029818956336529e-05,
"loss": 0.0557,
"step": 2250
},
{
"epoch": 3.67,
"learning_rate": 7.763578274760383e-05,
"loss": 0.0409,
"step": 2300
},
{
"epoch": 3.75,
"learning_rate": 7.49733759318424e-05,
"loss": 0.0374,
"step": 2350
},
{
"epoch": 3.83,
"learning_rate": 7.231096911608094e-05,
"loss": 0.0301,
"step": 2400
},
{
"epoch": 3.91,
"learning_rate": 6.96485623003195e-05,
"loss": 0.02,
"step": 2450
},
{
"epoch": 3.99,
"learning_rate": 6.698615548455805e-05,
"loss": 0.02,
"step": 2500
},
{
"epoch": 4.0,
"eval_accuracy": 0.8393782383419689,
"eval_f1": 0.8431497010182372,
"eval_loss": 0.5476517081260681,
"eval_precision": 0.8537832767554047,
"eval_recall": 0.8393782383419689,
"eval_runtime": 1.5001,
"eval_samples_per_second": 128.658,
"eval_steps_per_second": 16.666,
"step": 2504
},
{
"epoch": 4.07,
"learning_rate": 6.43237486687966e-05,
"loss": 0.0096,
"step": 2550
},
{
"epoch": 4.15,
"learning_rate": 6.166134185303514e-05,
"loss": 0.0025,
"step": 2600
},
{
"epoch": 4.23,
"learning_rate": 5.8998935037273696e-05,
"loss": 0.0031,
"step": 2650
},
{
"epoch": 4.31,
"learning_rate": 5.633652822151225e-05,
"loss": 0.0018,
"step": 2700
},
{
"epoch": 4.39,
"learning_rate": 5.36741214057508e-05,
"loss": 0.0035,
"step": 2750
},
{
"epoch": 4.47,
"learning_rate": 5.101171458998936e-05,
"loss": 0.0042,
"step": 2800
},
{
"epoch": 4.55,
"learning_rate": 4.8349307774227905e-05,
"loss": 0.0046,
"step": 2850
},
{
"epoch": 4.63,
"learning_rate": 4.568690095846646e-05,
"loss": 0.002,
"step": 2900
},
{
"epoch": 4.71,
"learning_rate": 4.3024494142705005e-05,
"loss": 0.0039,
"step": 2950
},
{
"epoch": 4.79,
"learning_rate": 4.036208732694356e-05,
"loss": 0.0011,
"step": 3000
},
{
"epoch": 4.87,
"learning_rate": 3.769968051118211e-05,
"loss": 0.0106,
"step": 3050
},
{
"epoch": 4.95,
"learning_rate": 3.503727369542067e-05,
"loss": 0.0018,
"step": 3100
},
{
"epoch": 5.0,
"eval_accuracy": 0.8704663212435233,
"eval_f1": 0.8749284792904506,
"eval_loss": 0.48757874965667725,
"eval_precision": 0.8863835430545274,
"eval_recall": 0.8704663212435233,
"eval_runtime": 1.4425,
"eval_samples_per_second": 133.795,
"eval_steps_per_second": 17.331,
"step": 3130
},
{
"epoch": 5.03,
"learning_rate": 3.2374866879659214e-05,
"loss": 0.0037,
"step": 3150
},
{
"epoch": 5.11,
"learning_rate": 2.971246006389776e-05,
"loss": 0.0004,
"step": 3200
},
{
"epoch": 5.19,
"learning_rate": 2.7050053248136315e-05,
"loss": 0.0004,
"step": 3250
},
{
"epoch": 5.27,
"learning_rate": 2.438764643237487e-05,
"loss": 0.0004,
"step": 3300
},
{
"epoch": 5.35,
"learning_rate": 2.172523961661342e-05,
"loss": 0.0003,
"step": 3350
},
{
"epoch": 5.43,
"learning_rate": 1.906283280085197e-05,
"loss": 0.0014,
"step": 3400
},
{
"epoch": 5.51,
"learning_rate": 1.6400425985090524e-05,
"loss": 0.0004,
"step": 3450
},
{
"epoch": 5.59,
"learning_rate": 1.3738019169329076e-05,
"loss": 0.0004,
"step": 3500
},
{
"epoch": 5.67,
"learning_rate": 1.1075612353567626e-05,
"loss": 0.0014,
"step": 3550
},
{
"epoch": 5.75,
"learning_rate": 8.413205537806178e-06,
"loss": 0.0004,
"step": 3600
},
{
"epoch": 5.83,
"learning_rate": 5.750798722044729e-06,
"loss": 0.0003,
"step": 3650
},
{
"epoch": 5.91,
"learning_rate": 3.08839190628328e-06,
"loss": 0.0003,
"step": 3700
},
{
"epoch": 5.99,
"learning_rate": 4.259850905218318e-07,
"loss": 0.0003,
"step": 3750
},
{
"epoch": 6.0,
"eval_accuracy": 0.8704663212435233,
"eval_f1": 0.8761642097552051,
"eval_loss": 0.4871196746826172,
"eval_precision": 0.8862443826726218,
"eval_recall": 0.8704663212435233,
"eval_runtime": 1.5001,
"eval_samples_per_second": 128.657,
"eval_steps_per_second": 16.665,
"step": 3756
},
{
"epoch": 6.0,
"step": 3756,
"total_flos": 4.65670232933972e+18,
"train_loss": 0.08256206865654514,
"train_runtime": 1281.0434,
"train_samples_per_second": 46.907,
"train_steps_per_second": 2.932
}
],
"max_steps": 3756,
"num_train_epochs": 6,
"total_flos": 4.65670232933972e+18,
"trial_name": null,
"trial_params": null
}