Commit ·
4ec609e
1
Parent(s): 7e013ad
First model version
Browse files- all_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +199 -0
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"total_flos": 3.205097416476426e+17,
|
| 4 |
+
"train_loss": 0.14605647795475446,
|
| 5 |
+
"train_runtime": 150.6081,
|
| 6 |
+
"train_samples_per_second": 27.462,
|
| 7 |
+
"train_steps_per_second": 1.726
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"total_flos": 3.205097416476426e+17,
|
| 4 |
+
"train_loss": 0.14605647795475446,
|
| 5 |
+
"train_runtime": 150.6081,
|
| 6 |
+
"train_samples_per_second": 27.462,
|
| 7 |
+
"train_steps_per_second": 1.726
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,199 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.03212887421250343,
|
| 3 |
+
"best_model_checkpoint": "./my_bean_VIT/checkpoint-200",
|
| 4 |
+
"epoch": 4.0,
|
| 5 |
+
"global_step": 260,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.15,
|
| 12 |
+
"learning_rate": 0.00019230769230769233,
|
| 13 |
+
"loss": 0.8792,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.31,
|
| 18 |
+
"learning_rate": 0.00018461538461538463,
|
| 19 |
+
"loss": 0.4134,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.46,
|
| 24 |
+
"learning_rate": 0.00017692307692307693,
|
| 25 |
+
"loss": 0.3521,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.62,
|
| 30 |
+
"learning_rate": 0.00016923076923076923,
|
| 31 |
+
"loss": 0.3909,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.77,
|
| 36 |
+
"learning_rate": 0.00016153846153846155,
|
| 37 |
+
"loss": 0.2881,
|
| 38 |
+
"step": 50
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.92,
|
| 42 |
+
"learning_rate": 0.00015384615384615385,
|
| 43 |
+
"loss": 0.2193,
|
| 44 |
+
"step": 60
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 1.08,
|
| 48 |
+
"learning_rate": 0.00014615384615384615,
|
| 49 |
+
"loss": 0.1367,
|
| 50 |
+
"step": 70
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 1.23,
|
| 54 |
+
"learning_rate": 0.00013846153846153847,
|
| 55 |
+
"loss": 0.0824,
|
| 56 |
+
"step": 80
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 1.38,
|
| 60 |
+
"learning_rate": 0.00013076923076923077,
|
| 61 |
+
"loss": 0.084,
|
| 62 |
+
"step": 90
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 1.54,
|
| 66 |
+
"learning_rate": 0.0001230769230769231,
|
| 67 |
+
"loss": 0.2698,
|
| 68 |
+
"step": 100
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 1.54,
|
| 72 |
+
"eval_accuracy": 0.9548872180451128,
|
| 73 |
+
"eval_loss": 0.1349506825208664,
|
| 74 |
+
"eval_runtime": 1.7784,
|
| 75 |
+
"eval_samples_per_second": 74.788,
|
| 76 |
+
"eval_steps_per_second": 9.559,
|
| 77 |
+
"step": 100
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 1.69,
|
| 81 |
+
"learning_rate": 0.00011538461538461538,
|
| 82 |
+
"loss": 0.1188,
|
| 83 |
+
"step": 110
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 1.85,
|
| 87 |
+
"learning_rate": 0.0001076923076923077,
|
| 88 |
+
"loss": 0.0767,
|
| 89 |
+
"step": 120
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 2.0,
|
| 93 |
+
"learning_rate": 0.0001,
|
| 94 |
+
"loss": 0.059,
|
| 95 |
+
"step": 130
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 2.15,
|
| 99 |
+
"learning_rate": 9.230769230769232e-05,
|
| 100 |
+
"loss": 0.0507,
|
| 101 |
+
"step": 140
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 2.31,
|
| 105 |
+
"learning_rate": 8.461538461538461e-05,
|
| 106 |
+
"loss": 0.023,
|
| 107 |
+
"step": 150
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 2.46,
|
| 111 |
+
"learning_rate": 7.692307692307693e-05,
|
| 112 |
+
"loss": 0.1037,
|
| 113 |
+
"step": 160
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 2.62,
|
| 117 |
+
"learning_rate": 7e-05,
|
| 118 |
+
"loss": 0.0724,
|
| 119 |
+
"step": 170
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 2.77,
|
| 123 |
+
"learning_rate": 6.23076923076923e-05,
|
| 124 |
+
"loss": 0.0279,
|
| 125 |
+
"step": 180
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 2.92,
|
| 129 |
+
"learning_rate": 5.461538461538461e-05,
|
| 130 |
+
"loss": 0.0411,
|
| 131 |
+
"step": 190
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 3.08,
|
| 135 |
+
"learning_rate": 4.692307692307693e-05,
|
| 136 |
+
"loss": 0.0147,
|
| 137 |
+
"step": 200
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 3.08,
|
| 141 |
+
"eval_accuracy": 0.9924812030075187,
|
| 142 |
+
"eval_loss": 0.03212887421250343,
|
| 143 |
+
"eval_runtime": 1.8163,
|
| 144 |
+
"eval_samples_per_second": 73.225,
|
| 145 |
+
"eval_steps_per_second": 9.36,
|
| 146 |
+
"step": 200
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 3.23,
|
| 150 |
+
"learning_rate": 3.923076923076923e-05,
|
| 151 |
+
"loss": 0.0137,
|
| 152 |
+
"step": 210
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 3.38,
|
| 156 |
+
"learning_rate": 3.153846153846154e-05,
|
| 157 |
+
"loss": 0.0134,
|
| 158 |
+
"step": 220
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 3.54,
|
| 162 |
+
"learning_rate": 2.384615384615385e-05,
|
| 163 |
+
"loss": 0.0145,
|
| 164 |
+
"step": 230
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 3.69,
|
| 168 |
+
"learning_rate": 1.6153846153846154e-05,
|
| 169 |
+
"loss": 0.0167,
|
| 170 |
+
"step": 240
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 3.85,
|
| 174 |
+
"learning_rate": 8.461538461538462e-06,
|
| 175 |
+
"loss": 0.0142,
|
| 176 |
+
"step": 250
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 4.0,
|
| 180 |
+
"learning_rate": 7.692307692307694e-07,
|
| 181 |
+
"loss": 0.0212,
|
| 182 |
+
"step": 260
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 4.0,
|
| 186 |
+
"step": 260,
|
| 187 |
+
"total_flos": 3.205097416476426e+17,
|
| 188 |
+
"train_loss": 0.14605647795475446,
|
| 189 |
+
"train_runtime": 150.6081,
|
| 190 |
+
"train_samples_per_second": 27.462,
|
| 191 |
+
"train_steps_per_second": 1.726
|
| 192 |
+
}
|
| 193 |
+
],
|
| 194 |
+
"max_steps": 260,
|
| 195 |
+
"num_train_epochs": 4,
|
| 196 |
+
"total_flos": 3.205097416476426e+17,
|
| 197 |
+
"trial_name": null,
|
| 198 |
+
"trial_params": null
|
| 199 |
+
}
|