google-vit-base-patch16-224-face / trainer_state.json
paul
End of training
884b0dc
raw
history blame
31.6 kB
{
"best_metric": 0.728460793804453,
"best_model_checkpoint": "google-vit-base-patch16-224-face/checkpoint-870",
"epoch": 7.9974204643164235,
"global_step": 2320,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.03,
"learning_rate": 5.172413793103449e-06,
"loss": 2.0401,
"step": 10
},
{
"epoch": 0.07,
"learning_rate": 1.0344827586206898e-05,
"loss": 1.8797,
"step": 20
},
{
"epoch": 0.1,
"learning_rate": 1.5517241379310346e-05,
"loss": 1.6439,
"step": 30
},
{
"epoch": 0.14,
"learning_rate": 2.0689655172413797e-05,
"loss": 1.4464,
"step": 40
},
{
"epoch": 0.17,
"learning_rate": 2.586206896551724e-05,
"loss": 1.26,
"step": 50
},
{
"epoch": 0.21,
"learning_rate": 3.103448275862069e-05,
"loss": 1.1469,
"step": 60
},
{
"epoch": 0.24,
"learning_rate": 3.6206896551724136e-05,
"loss": 1.0549,
"step": 70
},
{
"epoch": 0.28,
"learning_rate": 4.137931034482759e-05,
"loss": 1.047,
"step": 80
},
{
"epoch": 0.31,
"learning_rate": 4.655172413793104e-05,
"loss": 0.9929,
"step": 90
},
{
"epoch": 0.34,
"learning_rate": 5.172413793103448e-05,
"loss": 0.9421,
"step": 100
},
{
"epoch": 0.38,
"learning_rate": 5.689655172413793e-05,
"loss": 1.0011,
"step": 110
},
{
"epoch": 0.41,
"learning_rate": 6.206896551724138e-05,
"loss": 0.9637,
"step": 120
},
{
"epoch": 0.45,
"learning_rate": 6.724137931034483e-05,
"loss": 0.9283,
"step": 130
},
{
"epoch": 0.48,
"learning_rate": 7.241379310344827e-05,
"loss": 0.9275,
"step": 140
},
{
"epoch": 0.52,
"learning_rate": 7.758620689655173e-05,
"loss": 0.9041,
"step": 150
},
{
"epoch": 0.55,
"learning_rate": 8.275862068965519e-05,
"loss": 0.8695,
"step": 160
},
{
"epoch": 0.58,
"learning_rate": 8.793103448275862e-05,
"loss": 0.9382,
"step": 170
},
{
"epoch": 0.62,
"learning_rate": 9.310344827586207e-05,
"loss": 0.8939,
"step": 180
},
{
"epoch": 0.65,
"learning_rate": 9.827586206896552e-05,
"loss": 0.8945,
"step": 190
},
{
"epoch": 0.69,
"learning_rate": 0.00010344827586206896,
"loss": 0.8488,
"step": 200
},
{
"epoch": 0.72,
"learning_rate": 0.00010862068965517241,
"loss": 0.9054,
"step": 210
},
{
"epoch": 0.76,
"learning_rate": 0.00011379310344827586,
"loss": 0.8823,
"step": 220
},
{
"epoch": 0.79,
"learning_rate": 0.00011896551724137932,
"loss": 0.8898,
"step": 230
},
{
"epoch": 0.83,
"learning_rate": 0.00011954022988505747,
"loss": 0.8953,
"step": 240
},
{
"epoch": 0.86,
"learning_rate": 0.00011896551724137932,
"loss": 0.8591,
"step": 250
},
{
"epoch": 0.89,
"learning_rate": 0.00011839080459770115,
"loss": 0.8697,
"step": 260
},
{
"epoch": 0.93,
"learning_rate": 0.00011781609195402298,
"loss": 0.8245,
"step": 270
},
{
"epoch": 0.96,
"learning_rate": 0.00011724137931034484,
"loss": 0.8431,
"step": 280
},
{
"epoch": 1.0,
"learning_rate": 0.00011666666666666667,
"loss": 0.8514,
"step": 290
},
{
"epoch": 1.0,
"eval_accuracy": 0.7047972464235776,
"eval_f1": 0.6909131065764619,
"eval_loss": 0.8464350700378418,
"eval_precision": 0.7034765556664452,
"eval_recall": 0.7047972464235776,
"eval_runtime": 103.9481,
"eval_samples_per_second": 178.878,
"eval_steps_per_second": 2.799,
"step": 290
},
{
"epoch": 1.03,
"learning_rate": 0.00011609195402298851,
"loss": 0.8106,
"step": 300
},
{
"epoch": 1.07,
"learning_rate": 0.00011551724137931035,
"loss": 0.766,
"step": 310
},
{
"epoch": 1.1,
"learning_rate": 0.00011494252873563218,
"loss": 0.7461,
"step": 320
},
{
"epoch": 1.14,
"learning_rate": 0.00011436781609195404,
"loss": 0.7814,
"step": 330
},
{
"epoch": 1.17,
"learning_rate": 0.00011379310344827586,
"loss": 0.741,
"step": 340
},
{
"epoch": 1.21,
"learning_rate": 0.0001132183908045977,
"loss": 0.7118,
"step": 350
},
{
"epoch": 1.24,
"learning_rate": 0.00011264367816091955,
"loss": 0.7539,
"step": 360
},
{
"epoch": 1.28,
"learning_rate": 0.00011206896551724138,
"loss": 0.7633,
"step": 370
},
{
"epoch": 1.31,
"learning_rate": 0.00011149425287356322,
"loss": 0.7305,
"step": 380
},
{
"epoch": 1.34,
"learning_rate": 0.00011091954022988506,
"loss": 0.7455,
"step": 390
},
{
"epoch": 1.38,
"learning_rate": 0.00011034482758620689,
"loss": 0.724,
"step": 400
},
{
"epoch": 1.41,
"learning_rate": 0.00010977011494252874,
"loss": 0.7825,
"step": 410
},
{
"epoch": 1.45,
"learning_rate": 0.00010919540229885058,
"loss": 0.7558,
"step": 420
},
{
"epoch": 1.48,
"learning_rate": 0.00010862068965517241,
"loss": 0.744,
"step": 430
},
{
"epoch": 1.52,
"learning_rate": 0.00010804597701149426,
"loss": 0.7328,
"step": 440
},
{
"epoch": 1.55,
"learning_rate": 0.00010747126436781609,
"loss": 0.7385,
"step": 450
},
{
"epoch": 1.58,
"learning_rate": 0.00010689655172413793,
"loss": 0.7443,
"step": 460
},
{
"epoch": 1.62,
"learning_rate": 0.00010632183908045978,
"loss": 0.7126,
"step": 470
},
{
"epoch": 1.65,
"learning_rate": 0.0001057471264367816,
"loss": 0.8166,
"step": 480
},
{
"epoch": 1.69,
"learning_rate": 0.00010517241379310345,
"loss": 0.7516,
"step": 490
},
{
"epoch": 1.72,
"learning_rate": 0.00010459770114942529,
"loss": 0.7389,
"step": 500
},
{
"epoch": 1.76,
"learning_rate": 0.00010402298850574713,
"loss": 0.766,
"step": 510
},
{
"epoch": 1.79,
"learning_rate": 0.00010344827586206896,
"loss": 0.7343,
"step": 520
},
{
"epoch": 1.83,
"learning_rate": 0.0001028735632183908,
"loss": 0.7132,
"step": 530
},
{
"epoch": 1.86,
"learning_rate": 0.00010229885057471265,
"loss": 0.7494,
"step": 540
},
{
"epoch": 1.89,
"learning_rate": 0.00010172413793103449,
"loss": 0.7426,
"step": 550
},
{
"epoch": 1.93,
"learning_rate": 0.00010114942528735632,
"loss": 0.7305,
"step": 560
},
{
"epoch": 1.96,
"learning_rate": 0.00010057471264367816,
"loss": 0.7489,
"step": 570
},
{
"epoch": 2.0,
"learning_rate": 0.0001,
"loss": 0.7202,
"step": 580
},
{
"epoch": 2.0,
"eval_accuracy": 0.7282994514359471,
"eval_f1": 0.7111205455958668,
"eval_loss": 0.7790911793708801,
"eval_precision": 0.7297231799712814,
"eval_recall": 0.7282994514359471,
"eval_runtime": 104.6701,
"eval_samples_per_second": 177.644,
"eval_steps_per_second": 2.78,
"step": 580
},
{
"epoch": 2.03,
"learning_rate": 9.942528735632185e-05,
"loss": 0.6398,
"step": 590
},
{
"epoch": 2.07,
"learning_rate": 9.885057471264368e-05,
"loss": 0.5381,
"step": 600
},
{
"epoch": 2.1,
"learning_rate": 9.827586206896552e-05,
"loss": 0.5646,
"step": 610
},
{
"epoch": 2.14,
"learning_rate": 9.770114942528736e-05,
"loss": 0.5356,
"step": 620
},
{
"epoch": 2.17,
"learning_rate": 9.712643678160919e-05,
"loss": 0.5734,
"step": 630
},
{
"epoch": 2.21,
"learning_rate": 9.655172413793103e-05,
"loss": 0.5097,
"step": 640
},
{
"epoch": 2.24,
"learning_rate": 9.597701149425288e-05,
"loss": 0.5701,
"step": 650
},
{
"epoch": 2.28,
"learning_rate": 9.540229885057472e-05,
"loss": 0.5802,
"step": 660
},
{
"epoch": 2.31,
"learning_rate": 9.482758620689656e-05,
"loss": 0.5794,
"step": 670
},
{
"epoch": 2.34,
"learning_rate": 9.425287356321839e-05,
"loss": 0.5845,
"step": 680
},
{
"epoch": 2.38,
"learning_rate": 9.367816091954023e-05,
"loss": 0.5591,
"step": 690
},
{
"epoch": 2.41,
"learning_rate": 9.310344827586207e-05,
"loss": 0.5567,
"step": 700
},
{
"epoch": 2.45,
"learning_rate": 9.25287356321839e-05,
"loss": 0.5524,
"step": 710
},
{
"epoch": 2.48,
"learning_rate": 9.195402298850575e-05,
"loss": 0.5569,
"step": 720
},
{
"epoch": 2.52,
"learning_rate": 9.137931034482759e-05,
"loss": 0.5582,
"step": 730
},
{
"epoch": 2.55,
"learning_rate": 9.080459770114942e-05,
"loss": 0.5819,
"step": 740
},
{
"epoch": 2.58,
"learning_rate": 9.022988505747127e-05,
"loss": 0.5735,
"step": 750
},
{
"epoch": 2.62,
"learning_rate": 8.96551724137931e-05,
"loss": 0.5534,
"step": 760
},
{
"epoch": 2.65,
"learning_rate": 8.908045977011495e-05,
"loss": 0.5936,
"step": 770
},
{
"epoch": 2.69,
"learning_rate": 8.850574712643679e-05,
"loss": 0.5248,
"step": 780
},
{
"epoch": 2.72,
"learning_rate": 8.793103448275862e-05,
"loss": 0.5574,
"step": 790
},
{
"epoch": 2.76,
"learning_rate": 8.735632183908046e-05,
"loss": 0.5148,
"step": 800
},
{
"epoch": 2.79,
"learning_rate": 8.67816091954023e-05,
"loss": 0.5914,
"step": 810
},
{
"epoch": 2.83,
"learning_rate": 8.620689655172413e-05,
"loss": 0.6162,
"step": 820
},
{
"epoch": 2.86,
"learning_rate": 8.563218390804599e-05,
"loss": 0.5406,
"step": 830
},
{
"epoch": 2.89,
"learning_rate": 8.505747126436782e-05,
"loss": 0.5718,
"step": 840
},
{
"epoch": 2.93,
"learning_rate": 8.448275862068965e-05,
"loss": 0.5683,
"step": 850
},
{
"epoch": 2.96,
"learning_rate": 8.39080459770115e-05,
"loss": 0.5374,
"step": 860
},
{
"epoch": 3.0,
"learning_rate": 8.333333333333333e-05,
"loss": 0.5455,
"step": 870
},
{
"epoch": 3.0,
"eval_accuracy": 0.728460793804453,
"eval_f1": 0.717108375092518,
"eval_loss": 0.7950025200843811,
"eval_precision": 0.7174203394442972,
"eval_recall": 0.728460793804453,
"eval_runtime": 109.0183,
"eval_samples_per_second": 170.559,
"eval_steps_per_second": 2.669,
"step": 870
},
{
"epoch": 3.03,
"learning_rate": 8.275862068965519e-05,
"loss": 0.4109,
"step": 880
},
{
"epoch": 3.07,
"learning_rate": 8.218390804597702e-05,
"loss": 0.3615,
"step": 890
},
{
"epoch": 3.1,
"learning_rate": 8.160919540229884e-05,
"loss": 0.3445,
"step": 900
},
{
"epoch": 3.14,
"learning_rate": 8.10344827586207e-05,
"loss": 0.3407,
"step": 910
},
{
"epoch": 3.17,
"learning_rate": 8.045977011494253e-05,
"loss": 0.3228,
"step": 920
},
{
"epoch": 3.21,
"learning_rate": 7.988505747126436e-05,
"loss": 0.3376,
"step": 930
},
{
"epoch": 3.24,
"learning_rate": 7.931034482758621e-05,
"loss": 0.3616,
"step": 940
},
{
"epoch": 3.28,
"learning_rate": 7.873563218390804e-05,
"loss": 0.3393,
"step": 950
},
{
"epoch": 3.31,
"learning_rate": 7.816091954022989e-05,
"loss": 0.324,
"step": 960
},
{
"epoch": 3.34,
"learning_rate": 7.758620689655173e-05,
"loss": 0.3391,
"step": 970
},
{
"epoch": 3.38,
"learning_rate": 7.701149425287356e-05,
"loss": 0.3589,
"step": 980
},
{
"epoch": 3.41,
"learning_rate": 7.643678160919541e-05,
"loss": 0.328,
"step": 990
},
{
"epoch": 3.45,
"learning_rate": 7.586206896551724e-05,
"loss": 0.3256,
"step": 1000
},
{
"epoch": 3.48,
"learning_rate": 7.528735632183907e-05,
"loss": 0.338,
"step": 1010
},
{
"epoch": 3.52,
"learning_rate": 7.471264367816093e-05,
"loss": 0.3353,
"step": 1020
},
{
"epoch": 3.55,
"learning_rate": 7.413793103448276e-05,
"loss": 0.3147,
"step": 1030
},
{
"epoch": 3.58,
"learning_rate": 7.35632183908046e-05,
"loss": 0.3405,
"step": 1040
},
{
"epoch": 3.62,
"learning_rate": 7.298850574712644e-05,
"loss": 0.3409,
"step": 1050
},
{
"epoch": 3.65,
"learning_rate": 7.241379310344827e-05,
"loss": 0.3611,
"step": 1060
},
{
"epoch": 3.69,
"learning_rate": 7.183908045977013e-05,
"loss": 0.3501,
"step": 1070
},
{
"epoch": 3.72,
"learning_rate": 7.126436781609196e-05,
"loss": 0.3594,
"step": 1080
},
{
"epoch": 3.76,
"learning_rate": 7.068965517241379e-05,
"loss": 0.3507,
"step": 1090
},
{
"epoch": 3.79,
"learning_rate": 7.011494252873564e-05,
"loss": 0.3376,
"step": 1100
},
{
"epoch": 3.83,
"learning_rate": 6.954022988505747e-05,
"loss": 0.3355,
"step": 1110
},
{
"epoch": 3.86,
"learning_rate": 6.896551724137931e-05,
"loss": 0.3356,
"step": 1120
},
{
"epoch": 3.89,
"learning_rate": 6.839080459770116e-05,
"loss": 0.3489,
"step": 1130
},
{
"epoch": 3.93,
"learning_rate": 6.781609195402298e-05,
"loss": 0.3348,
"step": 1140
},
{
"epoch": 3.96,
"learning_rate": 6.724137931034483e-05,
"loss": 0.3436,
"step": 1150
},
{
"epoch": 4.0,
"learning_rate": 6.666666666666667e-05,
"loss": 0.334,
"step": 1160
},
{
"epoch": 4.0,
"eval_accuracy": 0.7154996235344735,
"eval_f1": 0.7144950276664553,
"eval_loss": 0.894753098487854,
"eval_precision": 0.715167448955365,
"eval_recall": 0.7154996235344735,
"eval_runtime": 102.1557,
"eval_samples_per_second": 182.016,
"eval_steps_per_second": 2.849,
"step": 1160
},
{
"epoch": 4.03,
"learning_rate": 6.60919540229885e-05,
"loss": 0.2132,
"step": 1170
},
{
"epoch": 4.07,
"learning_rate": 6.551724137931035e-05,
"loss": 0.1556,
"step": 1180
},
{
"epoch": 4.1,
"learning_rate": 6.494252873563218e-05,
"loss": 0.1607,
"step": 1190
},
{
"epoch": 4.14,
"learning_rate": 6.436781609195403e-05,
"loss": 0.1572,
"step": 1200
},
{
"epoch": 4.17,
"learning_rate": 6.379310344827587e-05,
"loss": 0.1554,
"step": 1210
},
{
"epoch": 4.21,
"learning_rate": 6.32183908045977e-05,
"loss": 0.156,
"step": 1220
},
{
"epoch": 4.24,
"learning_rate": 6.264367816091954e-05,
"loss": 0.1339,
"step": 1230
},
{
"epoch": 4.28,
"learning_rate": 6.206896551724138e-05,
"loss": 0.1565,
"step": 1240
},
{
"epoch": 4.31,
"learning_rate": 6.149425287356323e-05,
"loss": 0.1582,
"step": 1250
},
{
"epoch": 4.34,
"learning_rate": 6.0919540229885055e-05,
"loss": 0.1466,
"step": 1260
},
{
"epoch": 4.38,
"learning_rate": 6.0344827586206904e-05,
"loss": 0.1557,
"step": 1270
},
{
"epoch": 4.41,
"learning_rate": 5.977011494252873e-05,
"loss": 0.1522,
"step": 1280
},
{
"epoch": 4.45,
"learning_rate": 5.9195402298850576e-05,
"loss": 0.1373,
"step": 1290
},
{
"epoch": 4.48,
"learning_rate": 5.862068965517242e-05,
"loss": 0.1515,
"step": 1300
},
{
"epoch": 4.52,
"learning_rate": 5.8045977011494254e-05,
"loss": 0.1473,
"step": 1310
},
{
"epoch": 4.55,
"learning_rate": 5.747126436781609e-05,
"loss": 0.1513,
"step": 1320
},
{
"epoch": 4.58,
"learning_rate": 5.689655172413793e-05,
"loss": 0.1553,
"step": 1330
},
{
"epoch": 4.62,
"learning_rate": 5.6321839080459775e-05,
"loss": 0.158,
"step": 1340
},
{
"epoch": 4.65,
"learning_rate": 5.574712643678161e-05,
"loss": 0.1603,
"step": 1350
},
{
"epoch": 4.69,
"learning_rate": 5.5172413793103446e-05,
"loss": 0.1488,
"step": 1360
},
{
"epoch": 4.72,
"learning_rate": 5.459770114942529e-05,
"loss": 0.1495,
"step": 1370
},
{
"epoch": 4.76,
"learning_rate": 5.402298850574713e-05,
"loss": 0.1598,
"step": 1380
},
{
"epoch": 4.79,
"learning_rate": 5.344827586206897e-05,
"loss": 0.1468,
"step": 1390
},
{
"epoch": 4.83,
"learning_rate": 5.28735632183908e-05,
"loss": 0.1535,
"step": 1400
},
{
"epoch": 4.86,
"learning_rate": 5.2298850574712646e-05,
"loss": 0.1491,
"step": 1410
},
{
"epoch": 4.89,
"learning_rate": 5.172413793103448e-05,
"loss": 0.1436,
"step": 1420
},
{
"epoch": 4.93,
"learning_rate": 5.1149425287356324e-05,
"loss": 0.138,
"step": 1430
},
{
"epoch": 4.96,
"learning_rate": 5.057471264367816e-05,
"loss": 0.1439,
"step": 1440
},
{
"epoch": 5.0,
"learning_rate": 5e-05,
"loss": 0.1644,
"step": 1450
},
{
"epoch": 5.0,
"eval_accuracy": 0.7238894266967839,
"eval_f1": 0.7194230600781256,
"eval_loss": 1.0820467472076416,
"eval_precision": 0.7188913807380165,
"eval_recall": 0.7238894266967839,
"eval_runtime": 105.8352,
"eval_samples_per_second": 175.688,
"eval_steps_per_second": 2.75,
"step": 1450
},
{
"epoch": 5.03,
"learning_rate": 4.942528735632184e-05,
"loss": 0.0688,
"step": 1460
},
{
"epoch": 5.07,
"learning_rate": 4.885057471264368e-05,
"loss": 0.0619,
"step": 1470
},
{
"epoch": 5.1,
"learning_rate": 4.8275862068965517e-05,
"loss": 0.0502,
"step": 1480
},
{
"epoch": 5.14,
"learning_rate": 4.770114942528736e-05,
"loss": 0.0503,
"step": 1490
},
{
"epoch": 5.17,
"learning_rate": 4.7126436781609195e-05,
"loss": 0.0546,
"step": 1500
},
{
"epoch": 5.21,
"learning_rate": 4.655172413793104e-05,
"loss": 0.0529,
"step": 1510
},
{
"epoch": 5.24,
"learning_rate": 4.597701149425287e-05,
"loss": 0.0529,
"step": 1520
},
{
"epoch": 5.28,
"learning_rate": 4.540229885057471e-05,
"loss": 0.0499,
"step": 1530
},
{
"epoch": 5.31,
"learning_rate": 4.482758620689655e-05,
"loss": 0.0525,
"step": 1540
},
{
"epoch": 5.34,
"learning_rate": 4.4252873563218394e-05,
"loss": 0.0582,
"step": 1550
},
{
"epoch": 5.38,
"learning_rate": 4.367816091954023e-05,
"loss": 0.0598,
"step": 1560
},
{
"epoch": 5.41,
"learning_rate": 4.3103448275862066e-05,
"loss": 0.0587,
"step": 1570
},
{
"epoch": 5.45,
"learning_rate": 4.252873563218391e-05,
"loss": 0.0507,
"step": 1580
},
{
"epoch": 5.48,
"learning_rate": 4.195402298850575e-05,
"loss": 0.0565,
"step": 1590
},
{
"epoch": 5.52,
"learning_rate": 4.137931034482759e-05,
"loss": 0.0609,
"step": 1600
},
{
"epoch": 5.55,
"learning_rate": 4.080459770114942e-05,
"loss": 0.0459,
"step": 1610
},
{
"epoch": 5.58,
"learning_rate": 4.0229885057471265e-05,
"loss": 0.0647,
"step": 1620
},
{
"epoch": 5.62,
"learning_rate": 3.965517241379311e-05,
"loss": 0.0604,
"step": 1630
},
{
"epoch": 5.65,
"learning_rate": 3.908045977011494e-05,
"loss": 0.0648,
"step": 1640
},
{
"epoch": 5.69,
"learning_rate": 3.850574712643678e-05,
"loss": 0.0581,
"step": 1650
},
{
"epoch": 5.72,
"learning_rate": 3.793103448275862e-05,
"loss": 0.0517,
"step": 1660
},
{
"epoch": 5.76,
"learning_rate": 3.7356321839080464e-05,
"loss": 0.0558,
"step": 1670
},
{
"epoch": 5.79,
"learning_rate": 3.67816091954023e-05,
"loss": 0.0464,
"step": 1680
},
{
"epoch": 5.83,
"learning_rate": 3.6206896551724136e-05,
"loss": 0.054,
"step": 1690
},
{
"epoch": 5.86,
"learning_rate": 3.563218390804598e-05,
"loss": 0.0609,
"step": 1700
},
{
"epoch": 5.89,
"learning_rate": 3.505747126436782e-05,
"loss": 0.0507,
"step": 1710
},
{
"epoch": 5.93,
"learning_rate": 3.4482758620689657e-05,
"loss": 0.0521,
"step": 1720
},
{
"epoch": 5.96,
"learning_rate": 3.390804597701149e-05,
"loss": 0.0572,
"step": 1730
},
{
"epoch": 6.0,
"learning_rate": 3.3333333333333335e-05,
"loss": 0.0482,
"step": 1740
},
{
"epoch": 6.0,
"eval_accuracy": 0.7204474561686566,
"eval_f1": 0.71604121882508,
"eval_loss": 1.2792034149169922,
"eval_precision": 0.7144269552366298,
"eval_recall": 0.7204474561686566,
"eval_runtime": 99.5538,
"eval_samples_per_second": 186.773,
"eval_steps_per_second": 2.923,
"step": 1740
},
{
"epoch": 6.03,
"learning_rate": 3.275862068965518e-05,
"loss": 0.0261,
"step": 1750
},
{
"epoch": 6.07,
"learning_rate": 3.218390804597701e-05,
"loss": 0.0242,
"step": 1760
},
{
"epoch": 6.1,
"learning_rate": 3.160919540229885e-05,
"loss": 0.0261,
"step": 1770
},
{
"epoch": 6.14,
"learning_rate": 3.103448275862069e-05,
"loss": 0.0174,
"step": 1780
},
{
"epoch": 6.17,
"learning_rate": 3.0459770114942527e-05,
"loss": 0.0174,
"step": 1790
},
{
"epoch": 6.21,
"learning_rate": 2.9885057471264367e-05,
"loss": 0.0261,
"step": 1800
},
{
"epoch": 6.24,
"learning_rate": 2.931034482758621e-05,
"loss": 0.0233,
"step": 1810
},
{
"epoch": 6.28,
"learning_rate": 2.8735632183908045e-05,
"loss": 0.022,
"step": 1820
},
{
"epoch": 6.31,
"learning_rate": 2.8160919540229887e-05,
"loss": 0.0264,
"step": 1830
},
{
"epoch": 6.34,
"learning_rate": 2.7586206896551723e-05,
"loss": 0.0222,
"step": 1840
},
{
"epoch": 6.38,
"learning_rate": 2.7011494252873566e-05,
"loss": 0.0216,
"step": 1850
},
{
"epoch": 6.41,
"learning_rate": 2.64367816091954e-05,
"loss": 0.0193,
"step": 1860
},
{
"epoch": 6.45,
"learning_rate": 2.586206896551724e-05,
"loss": 0.0222,
"step": 1870
},
{
"epoch": 6.48,
"learning_rate": 2.528735632183908e-05,
"loss": 0.0153,
"step": 1880
},
{
"epoch": 6.52,
"learning_rate": 2.471264367816092e-05,
"loss": 0.0142,
"step": 1890
},
{
"epoch": 6.55,
"learning_rate": 2.4137931034482758e-05,
"loss": 0.025,
"step": 1900
},
{
"epoch": 6.58,
"learning_rate": 2.3563218390804597e-05,
"loss": 0.0197,
"step": 1910
},
{
"epoch": 6.62,
"learning_rate": 2.2988505747126437e-05,
"loss": 0.0212,
"step": 1920
},
{
"epoch": 6.65,
"learning_rate": 2.2413793103448276e-05,
"loss": 0.0181,
"step": 1930
},
{
"epoch": 6.69,
"learning_rate": 2.1839080459770115e-05,
"loss": 0.0228,
"step": 1940
},
{
"epoch": 6.72,
"learning_rate": 2.1264367816091954e-05,
"loss": 0.015,
"step": 1950
},
{
"epoch": 6.76,
"learning_rate": 2.0689655172413797e-05,
"loss": 0.0188,
"step": 1960
},
{
"epoch": 6.79,
"learning_rate": 2.0114942528735632e-05,
"loss": 0.0231,
"step": 1970
},
{
"epoch": 6.83,
"learning_rate": 1.954022988505747e-05,
"loss": 0.0157,
"step": 1980
},
{
"epoch": 6.86,
"learning_rate": 1.896551724137931e-05,
"loss": 0.0155,
"step": 1990
},
{
"epoch": 6.89,
"learning_rate": 1.839080459770115e-05,
"loss": 0.0214,
"step": 2000
},
{
"epoch": 6.93,
"learning_rate": 1.781609195402299e-05,
"loss": 0.0154,
"step": 2010
},
{
"epoch": 6.96,
"learning_rate": 1.7241379310344828e-05,
"loss": 0.0186,
"step": 2020
},
{
"epoch": 7.0,
"learning_rate": 1.6666666666666667e-05,
"loss": 0.0236,
"step": 2030
},
{
"epoch": 7.0,
"eval_accuracy": 0.7278692051199311,
"eval_f1": 0.7209486341359743,
"eval_loss": 1.4161986112594604,
"eval_precision": 0.7194977934958102,
"eval_recall": 0.7278692051199311,
"eval_runtime": 103.3877,
"eval_samples_per_second": 179.847,
"eval_steps_per_second": 2.815,
"step": 2030
},
{
"epoch": 7.03,
"learning_rate": 1.6091954022988507e-05,
"loss": 0.0094,
"step": 2040
},
{
"epoch": 7.07,
"learning_rate": 1.5517241379310346e-05,
"loss": 0.0117,
"step": 2050
},
{
"epoch": 7.1,
"learning_rate": 1.4942528735632183e-05,
"loss": 0.0049,
"step": 2060
},
{
"epoch": 7.14,
"learning_rate": 1.4367816091954022e-05,
"loss": 0.0097,
"step": 2070
},
{
"epoch": 7.17,
"learning_rate": 1.3793103448275862e-05,
"loss": 0.0092,
"step": 2080
},
{
"epoch": 7.21,
"learning_rate": 1.32183908045977e-05,
"loss": 0.0086,
"step": 2090
},
{
"epoch": 7.24,
"learning_rate": 1.264367816091954e-05,
"loss": 0.008,
"step": 2100
},
{
"epoch": 7.28,
"learning_rate": 1.2068965517241379e-05,
"loss": 0.01,
"step": 2110
},
{
"epoch": 7.31,
"learning_rate": 1.1494252873563218e-05,
"loss": 0.0078,
"step": 2120
},
{
"epoch": 7.34,
"learning_rate": 1.0919540229885057e-05,
"loss": 0.0093,
"step": 2130
},
{
"epoch": 7.38,
"learning_rate": 1.0344827586206898e-05,
"loss": 0.0081,
"step": 2140
},
{
"epoch": 7.41,
"learning_rate": 9.770114942528736e-06,
"loss": 0.0057,
"step": 2150
},
{
"epoch": 7.45,
"learning_rate": 9.195402298850575e-06,
"loss": 0.0095,
"step": 2160
},
{
"epoch": 7.48,
"learning_rate": 8.620689655172414e-06,
"loss": 0.0062,
"step": 2170
},
{
"epoch": 7.52,
"learning_rate": 8.045977011494253e-06,
"loss": 0.0124,
"step": 2180
},
{
"epoch": 7.55,
"learning_rate": 7.471264367816092e-06,
"loss": 0.008,
"step": 2190
},
{
"epoch": 7.58,
"learning_rate": 6.896551724137931e-06,
"loss": 0.0067,
"step": 2200
},
{
"epoch": 7.62,
"learning_rate": 6.32183908045977e-06,
"loss": 0.0085,
"step": 2210
},
{
"epoch": 7.65,
"learning_rate": 5.747126436781609e-06,
"loss": 0.0053,
"step": 2220
},
{
"epoch": 7.69,
"learning_rate": 5.172413793103449e-06,
"loss": 0.007,
"step": 2230
},
{
"epoch": 7.72,
"learning_rate": 4.5977011494252875e-06,
"loss": 0.0061,
"step": 2240
},
{
"epoch": 7.76,
"learning_rate": 4.022988505747127e-06,
"loss": 0.0054,
"step": 2250
},
{
"epoch": 7.79,
"learning_rate": 3.4482758620689654e-06,
"loss": 0.0103,
"step": 2260
},
{
"epoch": 7.83,
"learning_rate": 2.8735632183908046e-06,
"loss": 0.0131,
"step": 2270
},
{
"epoch": 7.86,
"learning_rate": 2.2988505747126437e-06,
"loss": 0.0143,
"step": 2280
},
{
"epoch": 7.89,
"learning_rate": 1.7241379310344827e-06,
"loss": 0.0075,
"step": 2290
},
{
"epoch": 7.93,
"learning_rate": 1.1494252873563219e-06,
"loss": 0.0079,
"step": 2300
},
{
"epoch": 7.96,
"learning_rate": 5.747126436781609e-07,
"loss": 0.0077,
"step": 2310
},
{
"epoch": 8.0,
"learning_rate": 0.0,
"loss": 0.0049,
"step": 2320
},
{
"epoch": 8.0,
"eval_accuracy": 0.7248574809078198,
"eval_f1": 0.7195690317790054,
"eval_loss": 1.4531140327453613,
"eval_precision": 0.717172031675939,
"eval_recall": 0.7248574809078198,
"eval_runtime": 105.0016,
"eval_samples_per_second": 177.083,
"eval_steps_per_second": 2.771,
"step": 2320
},
{
"epoch": 8.0,
"step": 2320,
"total_flos": 4.6098795710207476e+19,
"train_loss": 0.3678270006343594,
"train_runtime": 6722.1212,
"train_samples_per_second": 88.513,
"train_steps_per_second": 0.345
}
],
"max_steps": 2320,
"num_train_epochs": 8,
"total_flos": 4.6098795710207476e+19,
"trial_name": null,
"trial_params": null
}