deberta-v3-xsmall-CoLA / trainer_state.json
pszemraj's picture
End of training
0d72eaf
raw
history blame
9.43 kB
{
"best_metric": 0.5894856058137782,
"best_model_checkpoint": "./runtime-text-classification/deberta-v3-xsmall-CoLA/checkpoint-134",
"epoch": 3.0,
"global_step": 201,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.04,
"learning_rate": 2.5714285714285714e-05,
"loss": 0.6778,
"step": 3
},
{
"epoch": 0.09,
"learning_rate": 5.142857142857143e-05,
"loss": 0.6609,
"step": 6
},
{
"epoch": 0.13,
"learning_rate": 5.998426707136545e-05,
"loss": 0.6124,
"step": 9
},
{
"epoch": 0.18,
"learning_rate": 5.990171431423709e-05,
"loss": 0.5988,
"step": 12
},
{
"epoch": 0.22,
"learning_rate": 5.9748603036792754e-05,
"loss": 0.5704,
"step": 15
},
{
"epoch": 0.27,
"learning_rate": 5.952529453287223e-05,
"loss": 0.5851,
"step": 18
},
{
"epoch": 0.31,
"learning_rate": 5.923231573940778e-05,
"loss": 0.5969,
"step": 21
},
{
"epoch": 0.36,
"learning_rate": 5.887035799302091e-05,
"loss": 0.5408,
"step": 24
},
{
"epoch": 0.4,
"learning_rate": 5.844027539868812e-05,
"loss": 0.5148,
"step": 27
},
{
"epoch": 0.45,
"learning_rate": 5.794308281432527e-05,
"loss": 0.4907,
"step": 30
},
{
"epoch": 0.49,
"learning_rate": 5.7379953456046206e-05,
"loss": 0.4503,
"step": 33
},
{
"epoch": 0.54,
"learning_rate": 5.675221612974643e-05,
"loss": 0.4701,
"step": 36
},
{
"epoch": 0.58,
"learning_rate": 5.606135209554454e-05,
"loss": 0.449,
"step": 39
},
{
"epoch": 0.63,
"learning_rate": 5.530899157248014e-05,
"loss": 0.4227,
"step": 42
},
{
"epoch": 0.67,
"learning_rate": 5.449690989171633e-05,
"loss": 0.3914,
"step": 45
},
{
"epoch": 0.72,
"learning_rate": 5.3627023307323634e-05,
"loss": 0.4338,
"step": 48
},
{
"epoch": 0.76,
"learning_rate": 5.2701384474530855e-05,
"loss": 0.3944,
"step": 51
},
{
"epoch": 0.81,
"learning_rate": 5.172217760611265e-05,
"loss": 0.4302,
"step": 54
},
{
"epoch": 0.85,
"learning_rate": 5.0691713318343134e-05,
"loss": 0.4251,
"step": 57
},
{
"epoch": 0.9,
"learning_rate": 4.961242317867758e-05,
"loss": 0.3713,
"step": 60
},
{
"epoch": 0.94,
"learning_rate": 4.848685396802782e-05,
"loss": 0.4146,
"step": 63
},
{
"epoch": 0.99,
"learning_rate": 4.731766167117059e-05,
"loss": 0.3945,
"step": 66
},
{
"epoch": 1.0,
"eval_loss": 0.43225371837615967,
"eval_matthews_correlation": 0.5778184033685675,
"eval_runtime": 5.0494,
"eval_samples_per_second": 206.558,
"eval_steps_per_second": 51.689,
"step": 67
},
{
"epoch": 1.03,
"learning_rate": 4.610760520946952e-05,
"loss": 0.3873,
"step": 69
},
{
"epoch": 1.07,
"learning_rate": 4.4859539930699636e-05,
"loss": 0.33,
"step": 72
},
{
"epoch": 1.12,
"learning_rate": 4.357641087133617e-05,
"loss": 0.3265,
"step": 75
},
{
"epoch": 1.16,
"learning_rate": 4.226124580720669e-05,
"loss": 0.3306,
"step": 78
},
{
"epoch": 1.21,
"learning_rate": 4.091714810890463e-05,
"loss": 0.3695,
"step": 81
},
{
"epoch": 1.25,
"learning_rate": 3.9547289418823325e-05,
"loss": 0.3862,
"step": 84
},
{
"epoch": 1.3,
"learning_rate": 3.815490216709029e-05,
"loss": 0.3557,
"step": 87
},
{
"epoch": 1.34,
"learning_rate": 3.6743271944061884e-05,
"loss": 0.2995,
"step": 90
},
{
"epoch": 1.39,
"learning_rate": 3.5315729747376686e-05,
"loss": 0.2953,
"step": 93
},
{
"epoch": 1.43,
"learning_rate": 3.3875644121862195e-05,
"loss": 0.3356,
"step": 96
},
{
"epoch": 1.48,
"learning_rate": 3.242641321084213e-05,
"loss": 0.3182,
"step": 99
},
{
"epoch": 1.52,
"learning_rate": 3.097145673760068e-05,
"loss": 0.358,
"step": 102
},
{
"epoch": 1.57,
"learning_rate": 2.9514207935924957e-05,
"loss": 0.3351,
"step": 105
},
{
"epoch": 1.61,
"learning_rate": 2.805810544876693e-05,
"loss": 0.3057,
"step": 108
},
{
"epoch": 1.66,
"learning_rate": 2.6606585214141475e-05,
"loss": 0.3138,
"step": 111
},
{
"epoch": 1.7,
"learning_rate": 2.5163072357407205e-05,
"loss": 0.335,
"step": 114
},
{
"epoch": 1.75,
"learning_rate": 2.373097310906175e-05,
"loss": 0.3209,
"step": 117
},
{
"epoch": 1.79,
"learning_rate": 2.231366676712283e-05,
"loss": 0.3483,
"step": 120
},
{
"epoch": 1.84,
"learning_rate": 2.0914497723061486e-05,
"loss": 0.2824,
"step": 123
},
{
"epoch": 1.88,
"learning_rate": 1.953676757010353e-05,
"loss": 0.2709,
"step": 126
},
{
"epoch": 1.93,
"learning_rate": 1.8183727312521392e-05,
"loss": 0.3404,
"step": 129
},
{
"epoch": 1.97,
"learning_rate": 1.685856969429953e-05,
"loss": 0.3214,
"step": 132
},
{
"epoch": 2.0,
"eval_loss": 0.423650860786438,
"eval_matthews_correlation": 0.5894856058137782,
"eval_runtime": 5.1798,
"eval_samples_per_second": 201.361,
"eval_steps_per_second": 50.388,
"step": 134
},
{
"epoch": 2.01,
"learning_rate": 1.556442166527585e-05,
"loss": 0.2703,
"step": 135
},
{
"epoch": 2.06,
"learning_rate": 1.4304337002536097e-05,
"loss": 0.2356,
"step": 138
},
{
"epoch": 2.1,
"learning_rate": 1.3081289104472803e-05,
"loss": 0.2812,
"step": 141
},
{
"epoch": 2.15,
"learning_rate": 1.189816397451226e-05,
"loss": 0.2759,
"step": 144
},
{
"epoch": 2.19,
"learning_rate": 1.075775341106581e-05,
"loss": 0.25,
"step": 147
},
{
"epoch": 2.24,
"learning_rate": 9.662748419774903e-06,
"loss": 0.3233,
"step": 150
},
{
"epoch": 2.28,
"learning_rate": 8.61573286359513e-06,
"loss": 0.2499,
"step": 153
},
{
"epoch": 2.33,
"learning_rate": 7.619177365702619e-06,
"loss": 0.2734,
"step": 156
},
{
"epoch": 2.37,
"learning_rate": 6.6754334796105274e-06,
"loss": 0.2456,
"step": 159
},
{
"epoch": 2.42,
"learning_rate": 5.786728140251719e-06,
"loss": 0.2782,
"step": 162
},
{
"epoch": 2.46,
"learning_rate": 4.95515840912187e-06,
"loss": 0.248,
"step": 165
},
{
"epoch": 2.51,
"learning_rate": 4.182686525882242e-06,
"loss": 0.2766,
"step": 168
},
{
"epoch": 2.55,
"learning_rate": 3.4711352780992968e-06,
"loss": 0.2192,
"step": 171
},
{
"epoch": 2.6,
"learning_rate": 2.822183700046593e-06,
"loss": 0.1969,
"step": 174
},
{
"epoch": 2.64,
"learning_rate": 2.237363110718876e-06,
"loss": 0.2931,
"step": 177
},
{
"epoch": 2.69,
"learning_rate": 1.718053500406972e-06,
"loss": 0.2551,
"step": 180
},
{
"epoch": 2.73,
"learning_rate": 1.2654802743603888e-06,
"loss": 0.246,
"step": 183
},
{
"epoch": 2.78,
"learning_rate": 8.807113612212248e-07,
"loss": 0.2474,
"step": 186
},
{
"epoch": 2.82,
"learning_rate": 5.646546930528284e-07,
"loss": 0.2691,
"step": 189
},
{
"epoch": 2.87,
"learning_rate": 3.1805606290931987e-07,
"loss": 0.3154,
"step": 192
},
{
"epoch": 2.91,
"learning_rate": 1.414973650015816e-07,
"loss": 0.2823,
"step": 195
},
{
"epoch": 2.96,
"learning_rate": 3.5395221612279924e-08,
"loss": 0.2711,
"step": 198
},
{
"epoch": 3.0,
"learning_rate": 0.0,
"loss": 0.3059,
"step": 201
},
{
"epoch": 3.0,
"eval_loss": 0.4636174738407135,
"eval_matthews_correlation": 0.5794528111058918,
"eval_runtime": 5.0259,
"eval_samples_per_second": 207.524,
"eval_steps_per_second": 51.931,
"step": 201
},
{
"epoch": 3.0,
"step": 201,
"total_flos": 422473413165056.0,
"train_loss": 0.3622648042825917,
"train_runtime": 157.4171,
"train_samples_per_second": 162.962,
"train_steps_per_second": 1.277
}
],
"max_steps": 201,
"num_train_epochs": 3,
"total_flos": 422473413165056.0,
"trial_name": null,
"trial_params": null
}