ernie-ai's picture
🍻 cheers
f8ea65c
raw
history blame
5.86 kB
{
"best_metric": 0.310745507478714,
"best_model_checkpoint": "finetuned-vit-doc-text-classifer/checkpoint-300",
"epoch": 8.0,
"global_step": 384,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.21,
"learning_rate": 0.00019479166666666668,
"loss": 1.2187,
"step": 10
},
{
"epoch": 0.42,
"learning_rate": 0.00018958333333333332,
"loss": 0.8318,
"step": 20
},
{
"epoch": 0.62,
"learning_rate": 0.000184375,
"loss": 0.5961,
"step": 30
},
{
"epoch": 0.83,
"learning_rate": 0.0001791666666666667,
"loss": 0.5924,
"step": 40
},
{
"epoch": 1.04,
"learning_rate": 0.00017395833333333334,
"loss": 0.5361,
"step": 50
},
{
"epoch": 1.25,
"learning_rate": 0.00016875,
"loss": 0.4357,
"step": 60
},
{
"epoch": 1.46,
"learning_rate": 0.00016354166666666668,
"loss": 0.3754,
"step": 70
},
{
"epoch": 1.67,
"learning_rate": 0.00015833333333333332,
"loss": 0.3348,
"step": 80
},
{
"epoch": 1.88,
"learning_rate": 0.000153125,
"loss": 0.315,
"step": 90
},
{
"epoch": 2.08,
"learning_rate": 0.0001484375,
"loss": 0.2719,
"step": 100
},
{
"epoch": 2.08,
"eval_accuracy": 0.8656716417910447,
"eval_loss": 0.41197091341018677,
"eval_runtime": 1.3481,
"eval_samples_per_second": 99.4,
"eval_steps_per_second": 12.611,
"step": 100
},
{
"epoch": 2.29,
"learning_rate": 0.00014322916666666667,
"loss": 0.1733,
"step": 110
},
{
"epoch": 2.5,
"learning_rate": 0.00013802083333333334,
"loss": 0.3338,
"step": 120
},
{
"epoch": 2.71,
"learning_rate": 0.0001328125,
"loss": 0.2939,
"step": 130
},
{
"epoch": 2.92,
"learning_rate": 0.00012760416666666668,
"loss": 0.3062,
"step": 140
},
{
"epoch": 3.12,
"learning_rate": 0.00012239583333333332,
"loss": 0.203,
"step": 150
},
{
"epoch": 3.33,
"learning_rate": 0.00011718750000000001,
"loss": 0.1359,
"step": 160
},
{
"epoch": 3.54,
"learning_rate": 0.00011197916666666668,
"loss": 0.1954,
"step": 170
},
{
"epoch": 3.75,
"learning_rate": 0.00010677083333333334,
"loss": 0.123,
"step": 180
},
{
"epoch": 3.96,
"learning_rate": 0.00010156250000000001,
"loss": 0.1509,
"step": 190
},
{
"epoch": 4.17,
"learning_rate": 9.635416666666667e-05,
"loss": 0.1027,
"step": 200
},
{
"epoch": 4.17,
"eval_accuracy": 0.8880597014925373,
"eval_loss": 0.3907298743724823,
"eval_runtime": 2.1054,
"eval_samples_per_second": 63.647,
"eval_steps_per_second": 8.075,
"step": 200
},
{
"epoch": 4.38,
"learning_rate": 9.114583333333334e-05,
"loss": 0.1048,
"step": 210
},
{
"epoch": 4.58,
"learning_rate": 8.593750000000001e-05,
"loss": 0.1449,
"step": 220
},
{
"epoch": 4.79,
"learning_rate": 8.072916666666667e-05,
"loss": 0.1575,
"step": 230
},
{
"epoch": 5.0,
"learning_rate": 7.552083333333334e-05,
"loss": 0.267,
"step": 240
},
{
"epoch": 5.21,
"learning_rate": 7.031250000000001e-05,
"loss": 0.1269,
"step": 250
},
{
"epoch": 5.42,
"learning_rate": 6.510416666666667e-05,
"loss": 0.0954,
"step": 260
},
{
"epoch": 5.62,
"learning_rate": 5.989583333333334e-05,
"loss": 0.0505,
"step": 270
},
{
"epoch": 5.83,
"learning_rate": 5.46875e-05,
"loss": 0.0379,
"step": 280
},
{
"epoch": 6.04,
"learning_rate": 4.947916666666667e-05,
"loss": 0.13,
"step": 290
},
{
"epoch": 6.25,
"learning_rate": 4.4270833333333337e-05,
"loss": 0.0723,
"step": 300
},
{
"epoch": 6.25,
"eval_accuracy": 0.9029850746268657,
"eval_loss": 0.310745507478714,
"eval_runtime": 1.3337,
"eval_samples_per_second": 100.474,
"eval_steps_per_second": 12.747,
"step": 300
},
{
"epoch": 6.46,
"learning_rate": 3.90625e-05,
"loss": 0.1076,
"step": 310
},
{
"epoch": 6.67,
"learning_rate": 3.385416666666667e-05,
"loss": 0.1555,
"step": 320
},
{
"epoch": 6.88,
"learning_rate": 2.8645833333333333e-05,
"loss": 0.0671,
"step": 330
},
{
"epoch": 7.08,
"learning_rate": 2.34375e-05,
"loss": 0.082,
"step": 340
},
{
"epoch": 7.29,
"learning_rate": 1.8229166666666668e-05,
"loss": 0.0557,
"step": 350
},
{
"epoch": 7.5,
"learning_rate": 1.3020833333333334e-05,
"loss": 0.0301,
"step": 360
},
{
"epoch": 7.71,
"learning_rate": 7.8125e-06,
"loss": 0.0318,
"step": 370
},
{
"epoch": 7.92,
"learning_rate": 2.604166666666667e-06,
"loss": 0.0667,
"step": 380
},
{
"epoch": 8.0,
"step": 384,
"total_flos": 4.674400602398392e+17,
"train_loss": 0.24295292422175407,
"train_runtime": 151.2132,
"train_samples_per_second": 39.891,
"train_steps_per_second": 2.539
}
],
"max_steps": 384,
"num_train_epochs": 8,
"total_flos": 4.674400602398392e+17,
"trial_name": null,
"trial_params": null
}