Commit
•
dd9b62a
1
Parent(s):
650673e
End of training
Browse files- trainer_state.json +165 -0
trainer_state.json
ADDED
|
@@ -0,0 +1,165 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 1.8622814416885376,
|
| 3 |
+
"best_model_checkpoint": "detr-r50-cd45rb-2ah-6l-corrected\\checkpoint-46060",
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"global_step": 46060,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 1.0,
|
| 12 |
+
"learning_rate": 9.002171081198437e-06,
|
| 13 |
+
"loss": 3.4312,
|
| 14 |
+
"step": 4606
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 1.0,
|
| 18 |
+
"eval_loss": 2.442598342895508,
|
| 19 |
+
"eval_runtime": 204.3572,
|
| 20 |
+
"eval_samples_per_second": 8.715,
|
| 21 |
+
"eval_steps_per_second": 1.091,
|
| 22 |
+
"step": 4606
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"epoch": 2.0,
|
| 26 |
+
"learning_rate": 8.002388189318281e-06,
|
| 27 |
+
"loss": 2.9035,
|
| 28 |
+
"step": 9212
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 2.0,
|
| 32 |
+
"eval_loss": 2.1671295166015625,
|
| 33 |
+
"eval_runtime": 204.0022,
|
| 34 |
+
"eval_samples_per_second": 8.73,
|
| 35 |
+
"eval_steps_per_second": 1.093,
|
| 36 |
+
"step": 9212
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 3.0,
|
| 40 |
+
"learning_rate": 7.0028224055579685e-06,
|
| 41 |
+
"loss": 2.6788,
|
| 42 |
+
"step": 13818
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 3.0,
|
| 46 |
+
"eval_loss": 2.0924932956695557,
|
| 47 |
+
"eval_runtime": 203.9348,
|
| 48 |
+
"eval_samples_per_second": 8.733,
|
| 49 |
+
"eval_steps_per_second": 1.093,
|
| 50 |
+
"step": 13818
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 4.0,
|
| 54 |
+
"learning_rate": 6.003256621797656e-06,
|
| 55 |
+
"loss": 2.5847,
|
| 56 |
+
"step": 18424
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 4.0,
|
| 60 |
+
"eval_loss": 2.021491765975952,
|
| 61 |
+
"eval_runtime": 201.6799,
|
| 62 |
+
"eval_samples_per_second": 8.831,
|
| 63 |
+
"eval_steps_per_second": 1.106,
|
| 64 |
+
"step": 18424
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 5.0,
|
| 68 |
+
"learning_rate": 5.0034737299175e-06,
|
| 69 |
+
"loss": 2.5075,
|
| 70 |
+
"step": 23030
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 5.0,
|
| 74 |
+
"eval_loss": 1.9688512086868286,
|
| 75 |
+
"eval_runtime": 205.2053,
|
| 76 |
+
"eval_samples_per_second": 8.679,
|
| 77 |
+
"eval_steps_per_second": 1.087,
|
| 78 |
+
"step": 23030
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 6.0,
|
| 82 |
+
"learning_rate": 4.0041250542770305e-06,
|
| 83 |
+
"loss": 2.445,
|
| 84 |
+
"step": 27636
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 6.0,
|
| 88 |
+
"eval_loss": 1.9527798891067505,
|
| 89 |
+
"eval_runtime": 205.1637,
|
| 90 |
+
"eval_samples_per_second": 8.681,
|
| 91 |
+
"eval_steps_per_second": 1.087,
|
| 92 |
+
"step": 27636
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 7.0,
|
| 96 |
+
"learning_rate": 3.0045592705167175e-06,
|
| 97 |
+
"loss": 2.4078,
|
| 98 |
+
"step": 32242
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 7.0,
|
| 102 |
+
"eval_loss": 1.9058469533920288,
|
| 103 |
+
"eval_runtime": 205.1138,
|
| 104 |
+
"eval_samples_per_second": 8.683,
|
| 105 |
+
"eval_steps_per_second": 1.087,
|
| 106 |
+
"step": 32242
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 8.0,
|
| 110 |
+
"learning_rate": 2.004993486756405e-06,
|
| 111 |
+
"loss": 2.3794,
|
| 112 |
+
"step": 36848
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 8.0,
|
| 116 |
+
"eval_loss": 1.888853907585144,
|
| 117 |
+
"eval_runtime": 205.5976,
|
| 118 |
+
"eval_samples_per_second": 8.663,
|
| 119 |
+
"eval_steps_per_second": 1.085,
|
| 120 |
+
"step": 36848
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 9.0,
|
| 124 |
+
"learning_rate": 1.0049934867564047e-06,
|
| 125 |
+
"loss": 2.3572,
|
| 126 |
+
"step": 41454
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 9.0,
|
| 130 |
+
"eval_loss": 1.875107765197754,
|
| 131 |
+
"eval_runtime": 202.1582,
|
| 132 |
+
"eval_samples_per_second": 8.81,
|
| 133 |
+
"eval_steps_per_second": 1.103,
|
| 134 |
+
"step": 41454
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 10.0,
|
| 138 |
+
"learning_rate": 5.427702996092055e-09,
|
| 139 |
+
"loss": 2.3374,
|
| 140 |
+
"step": 46060
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 10.0,
|
| 144 |
+
"eval_loss": 1.8622814416885376,
|
| 145 |
+
"eval_runtime": 202.8571,
|
| 146 |
+
"eval_samples_per_second": 8.78,
|
| 147 |
+
"eval_steps_per_second": 1.099,
|
| 148 |
+
"step": 46060
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 10.0,
|
| 152 |
+
"step": 46060,
|
| 153 |
+
"total_flos": 8.803256478082562e+19,
|
| 154 |
+
"train_loss": 2.603237603635204,
|
| 155 |
+
"train_runtime": 29457.3491,
|
| 156 |
+
"train_samples_per_second": 6.253,
|
| 157 |
+
"train_steps_per_second": 1.564
|
| 158 |
+
}
|
| 159 |
+
],
|
| 160 |
+
"max_steps": 46060,
|
| 161 |
+
"num_train_epochs": 10,
|
| 162 |
+
"total_flos": 8.803256478082562e+19,
|
| 163 |
+
"trial_name": null,
|
| 164 |
+
"trial_params": null
|
| 165 |
+
}
|