Steve Chiou
commited on
Commit
·
cee0f66
1
Parent(s):
b10190c
End of training
Browse files- all_results.json +8 -0
- test_results.json +8 -0
- trainer_state.json +583 -0
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 19.02,
|
| 3 |
+
"eval_accuracy": 0.85,
|
| 4 |
+
"eval_loss": 0.49579015374183655,
|
| 5 |
+
"eval_runtime": 64.7531,
|
| 6 |
+
"eval_samples_per_second": 1.235,
|
| 7 |
+
"eval_steps_per_second": 0.216
|
| 8 |
+
}
|
test_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 19.02,
|
| 3 |
+
"eval_accuracy": 0.85,
|
| 4 |
+
"eval_loss": 0.49579015374183655,
|
| 5 |
+
"eval_runtime": 64.7531,
|
| 6 |
+
"eval_samples_per_second": 1.235,
|
| 7 |
+
"eval_steps_per_second": 0.216
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,583 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.85,
|
| 3 |
+
"best_model_checkpoint": "videomae-base-finetuned-engine-subset-20230310/checkpoint-372",
|
| 4 |
+
"epoch": 19.018333333333334,
|
| 5 |
+
"global_step": 600,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.02,
|
| 12 |
+
"learning_rate": 8.333333333333334e-06,
|
| 13 |
+
"loss": 2.8307,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.03,
|
| 18 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 19 |
+
"loss": 2.6132,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.05,
|
| 24 |
+
"learning_rate": 2.5e-05,
|
| 25 |
+
"loss": 2.5947,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.05,
|
| 30 |
+
"eval_accuracy": 0.15,
|
| 31 |
+
"eval_loss": 2.538285493850708,
|
| 32 |
+
"eval_runtime": 64.1944,
|
| 33 |
+
"eval_samples_per_second": 1.246,
|
| 34 |
+
"eval_steps_per_second": 0.218,
|
| 35 |
+
"step": 31
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 1.01,
|
| 39 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 40 |
+
"loss": 2.5062,
|
| 41 |
+
"step": 40
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 1.03,
|
| 45 |
+
"learning_rate": 4.166666666666667e-05,
|
| 46 |
+
"loss": 2.665,
|
| 47 |
+
"step": 50
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 1.05,
|
| 51 |
+
"learning_rate": 5e-05,
|
| 52 |
+
"loss": 2.4195,
|
| 53 |
+
"step": 60
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.05,
|
| 57 |
+
"eval_accuracy": 0.15,
|
| 58 |
+
"eval_loss": 2.5108141899108887,
|
| 59 |
+
"eval_runtime": 62.8569,
|
| 60 |
+
"eval_samples_per_second": 1.273,
|
| 61 |
+
"eval_steps_per_second": 0.223,
|
| 62 |
+
"step": 62
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 2.01,
|
| 66 |
+
"learning_rate": 4.9074074074074075e-05,
|
| 67 |
+
"loss": 2.3539,
|
| 68 |
+
"step": 70
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 2.03,
|
| 72 |
+
"learning_rate": 4.814814814814815e-05,
|
| 73 |
+
"loss": 2.3374,
|
| 74 |
+
"step": 80
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 2.05,
|
| 78 |
+
"learning_rate": 4.722222222222222e-05,
|
| 79 |
+
"loss": 2.2476,
|
| 80 |
+
"step": 90
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 2.05,
|
| 84 |
+
"eval_accuracy": 0.225,
|
| 85 |
+
"eval_loss": 2.0532896518707275,
|
| 86 |
+
"eval_runtime": 62.7732,
|
| 87 |
+
"eval_samples_per_second": 1.274,
|
| 88 |
+
"eval_steps_per_second": 0.223,
|
| 89 |
+
"step": 93
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 3.01,
|
| 93 |
+
"learning_rate": 4.62962962962963e-05,
|
| 94 |
+
"loss": 2.2757,
|
| 95 |
+
"step": 100
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 3.03,
|
| 99 |
+
"learning_rate": 4.5370370370370374e-05,
|
| 100 |
+
"loss": 1.9768,
|
| 101 |
+
"step": 110
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 3.04,
|
| 105 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 106 |
+
"loss": 1.9449,
|
| 107 |
+
"step": 120
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 3.05,
|
| 111 |
+
"eval_accuracy": 0.2375,
|
| 112 |
+
"eval_loss": 2.071887493133545,
|
| 113 |
+
"eval_runtime": 65.8713,
|
| 114 |
+
"eval_samples_per_second": 1.214,
|
| 115 |
+
"eval_steps_per_second": 0.213,
|
| 116 |
+
"step": 124
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 4.01,
|
| 120 |
+
"learning_rate": 4.351851851851852e-05,
|
| 121 |
+
"loss": 1.7178,
|
| 122 |
+
"step": 130
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 4.03,
|
| 126 |
+
"learning_rate": 4.259259259259259e-05,
|
| 127 |
+
"loss": 1.6786,
|
| 128 |
+
"step": 140
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 4.04,
|
| 132 |
+
"learning_rate": 4.166666666666667e-05,
|
| 133 |
+
"loss": 1.5724,
|
| 134 |
+
"step": 150
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 4.05,
|
| 138 |
+
"eval_accuracy": 0.475,
|
| 139 |
+
"eval_loss": 1.4755998849868774,
|
| 140 |
+
"eval_runtime": 62.826,
|
| 141 |
+
"eval_samples_per_second": 1.273,
|
| 142 |
+
"eval_steps_per_second": 0.223,
|
| 143 |
+
"step": 155
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 5.01,
|
| 147 |
+
"learning_rate": 4.074074074074074e-05,
|
| 148 |
+
"loss": 1.3465,
|
| 149 |
+
"step": 160
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 5.03,
|
| 153 |
+
"learning_rate": 3.981481481481482e-05,
|
| 154 |
+
"loss": 1.3641,
|
| 155 |
+
"step": 170
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 5.04,
|
| 159 |
+
"learning_rate": 3.888888888888889e-05,
|
| 160 |
+
"loss": 1.395,
|
| 161 |
+
"step": 180
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 5.05,
|
| 165 |
+
"eval_accuracy": 0.5,
|
| 166 |
+
"eval_loss": 1.2884117364883423,
|
| 167 |
+
"eval_runtime": 63.6279,
|
| 168 |
+
"eval_samples_per_second": 1.257,
|
| 169 |
+
"eval_steps_per_second": 0.22,
|
| 170 |
+
"step": 186
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 6.01,
|
| 174 |
+
"learning_rate": 3.7962962962962964e-05,
|
| 175 |
+
"loss": 1.1859,
|
| 176 |
+
"step": 190
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 6.02,
|
| 180 |
+
"learning_rate": 3.7037037037037037e-05,
|
| 181 |
+
"loss": 1.2224,
|
| 182 |
+
"step": 200
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 6.04,
|
| 186 |
+
"learning_rate": 3.611111111111111e-05,
|
| 187 |
+
"loss": 1.0822,
|
| 188 |
+
"step": 210
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 6.05,
|
| 192 |
+
"eval_accuracy": 0.575,
|
| 193 |
+
"eval_loss": 1.0678651332855225,
|
| 194 |
+
"eval_runtime": 63.3472,
|
| 195 |
+
"eval_samples_per_second": 1.263,
|
| 196 |
+
"eval_steps_per_second": 0.221,
|
| 197 |
+
"step": 217
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 7.0,
|
| 201 |
+
"learning_rate": 3.518518518518519e-05,
|
| 202 |
+
"loss": 0.9066,
|
| 203 |
+
"step": 220
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 7.02,
|
| 207 |
+
"learning_rate": 3.425925925925926e-05,
|
| 208 |
+
"loss": 0.7939,
|
| 209 |
+
"step": 230
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 7.04,
|
| 213 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 214 |
+
"loss": 1.0635,
|
| 215 |
+
"step": 240
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 7.05,
|
| 219 |
+
"eval_accuracy": 0.7,
|
| 220 |
+
"eval_loss": 0.8040415048599243,
|
| 221 |
+
"eval_runtime": 63.8719,
|
| 222 |
+
"eval_samples_per_second": 1.253,
|
| 223 |
+
"eval_steps_per_second": 0.219,
|
| 224 |
+
"step": 248
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 8.0,
|
| 228 |
+
"learning_rate": 3.240740740740741e-05,
|
| 229 |
+
"loss": 1.1727,
|
| 230 |
+
"step": 250
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 8.02,
|
| 234 |
+
"learning_rate": 3.148148148148148e-05,
|
| 235 |
+
"loss": 0.8291,
|
| 236 |
+
"step": 260
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 8.04,
|
| 240 |
+
"learning_rate": 3.055555555555556e-05,
|
| 241 |
+
"loss": 0.8707,
|
| 242 |
+
"step": 270
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 8.05,
|
| 246 |
+
"eval_accuracy": 0.525,
|
| 247 |
+
"eval_loss": 0.9334062337875366,
|
| 248 |
+
"eval_runtime": 64.0172,
|
| 249 |
+
"eval_samples_per_second": 1.25,
|
| 250 |
+
"eval_steps_per_second": 0.219,
|
| 251 |
+
"step": 279
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 9.0,
|
| 255 |
+
"learning_rate": 2.962962962962963e-05,
|
| 256 |
+
"loss": 0.8339,
|
| 257 |
+
"step": 280
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 9.02,
|
| 261 |
+
"learning_rate": 2.8703703703703706e-05,
|
| 262 |
+
"loss": 0.8118,
|
| 263 |
+
"step": 290
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 9.04,
|
| 267 |
+
"learning_rate": 2.777777777777778e-05,
|
| 268 |
+
"loss": 0.8068,
|
| 269 |
+
"step": 300
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 9.05,
|
| 273 |
+
"learning_rate": 2.6851851851851855e-05,
|
| 274 |
+
"loss": 0.7042,
|
| 275 |
+
"step": 310
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 9.05,
|
| 279 |
+
"eval_accuracy": 0.75,
|
| 280 |
+
"eval_loss": 0.6476640701293945,
|
| 281 |
+
"eval_runtime": 63.3755,
|
| 282 |
+
"eval_samples_per_second": 1.262,
|
| 283 |
+
"eval_steps_per_second": 0.221,
|
| 284 |
+
"step": 310
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 10.02,
|
| 288 |
+
"learning_rate": 2.5925925925925925e-05,
|
| 289 |
+
"loss": 0.6771,
|
| 290 |
+
"step": 320
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 10.03,
|
| 294 |
+
"learning_rate": 2.5e-05,
|
| 295 |
+
"loss": 0.7369,
|
| 296 |
+
"step": 330
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 10.05,
|
| 300 |
+
"learning_rate": 2.4074074074074074e-05,
|
| 301 |
+
"loss": 0.6543,
|
| 302 |
+
"step": 340
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 10.05,
|
| 306 |
+
"eval_accuracy": 0.7375,
|
| 307 |
+
"eval_loss": 0.6962689757347107,
|
| 308 |
+
"eval_runtime": 63.3385,
|
| 309 |
+
"eval_samples_per_second": 1.263,
|
| 310 |
+
"eval_steps_per_second": 0.221,
|
| 311 |
+
"step": 341
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 11.02,
|
| 315 |
+
"learning_rate": 2.314814814814815e-05,
|
| 316 |
+
"loss": 0.7641,
|
| 317 |
+
"step": 350
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 11.03,
|
| 321 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 322 |
+
"loss": 0.5571,
|
| 323 |
+
"step": 360
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 11.05,
|
| 327 |
+
"learning_rate": 2.1296296296296296e-05,
|
| 328 |
+
"loss": 0.6807,
|
| 329 |
+
"step": 370
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 11.05,
|
| 333 |
+
"eval_accuracy": 0.85,
|
| 334 |
+
"eval_loss": 0.49579018354415894,
|
| 335 |
+
"eval_runtime": 64.8171,
|
| 336 |
+
"eval_samples_per_second": 1.234,
|
| 337 |
+
"eval_steps_per_second": 0.216,
|
| 338 |
+
"step": 372
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 12.01,
|
| 342 |
+
"learning_rate": 2.037037037037037e-05,
|
| 343 |
+
"loss": 0.5435,
|
| 344 |
+
"step": 380
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 12.03,
|
| 348 |
+
"learning_rate": 1.9444444444444445e-05,
|
| 349 |
+
"loss": 0.5266,
|
| 350 |
+
"step": 390
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 12.05,
|
| 354 |
+
"learning_rate": 1.8518518518518518e-05,
|
| 355 |
+
"loss": 0.4924,
|
| 356 |
+
"step": 400
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 12.05,
|
| 360 |
+
"eval_accuracy": 0.775,
|
| 361 |
+
"eval_loss": 0.6373826265335083,
|
| 362 |
+
"eval_runtime": 67.4155,
|
| 363 |
+
"eval_samples_per_second": 1.187,
|
| 364 |
+
"eval_steps_per_second": 0.208,
|
| 365 |
+
"step": 403
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 13.01,
|
| 369 |
+
"learning_rate": 1.7592592592592595e-05,
|
| 370 |
+
"loss": 0.4775,
|
| 371 |
+
"step": 410
|
| 372 |
+
},
|
| 373 |
+
{
|
| 374 |
+
"epoch": 13.03,
|
| 375 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 376 |
+
"loss": 0.5071,
|
| 377 |
+
"step": 420
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 13.04,
|
| 381 |
+
"learning_rate": 1.574074074074074e-05,
|
| 382 |
+
"loss": 0.4822,
|
| 383 |
+
"step": 430
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 13.05,
|
| 387 |
+
"eval_accuracy": 0.75,
|
| 388 |
+
"eval_loss": 0.6144760847091675,
|
| 389 |
+
"eval_runtime": 64.8207,
|
| 390 |
+
"eval_samples_per_second": 1.234,
|
| 391 |
+
"eval_steps_per_second": 0.216,
|
| 392 |
+
"step": 434
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 14.01,
|
| 396 |
+
"learning_rate": 1.4814814814814815e-05,
|
| 397 |
+
"loss": 0.3259,
|
| 398 |
+
"step": 440
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 14.03,
|
| 402 |
+
"learning_rate": 1.388888888888889e-05,
|
| 403 |
+
"loss": 0.5054,
|
| 404 |
+
"step": 450
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 14.04,
|
| 408 |
+
"learning_rate": 1.2962962962962962e-05,
|
| 409 |
+
"loss": 0.4878,
|
| 410 |
+
"step": 460
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 14.05,
|
| 414 |
+
"eval_accuracy": 0.7625,
|
| 415 |
+
"eval_loss": 0.6274302005767822,
|
| 416 |
+
"eval_runtime": 62.4581,
|
| 417 |
+
"eval_samples_per_second": 1.281,
|
| 418 |
+
"eval_steps_per_second": 0.224,
|
| 419 |
+
"step": 465
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 15.01,
|
| 423 |
+
"learning_rate": 1.2037037037037037e-05,
|
| 424 |
+
"loss": 0.3902,
|
| 425 |
+
"step": 470
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 15.03,
|
| 429 |
+
"learning_rate": 1.1111111111111112e-05,
|
| 430 |
+
"loss": 0.4728,
|
| 431 |
+
"step": 480
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 15.04,
|
| 435 |
+
"learning_rate": 1.0185185185185185e-05,
|
| 436 |
+
"loss": 0.4442,
|
| 437 |
+
"step": 490
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 15.05,
|
| 441 |
+
"eval_accuracy": 0.85,
|
| 442 |
+
"eval_loss": 0.42305102944374084,
|
| 443 |
+
"eval_runtime": 62.8623,
|
| 444 |
+
"eval_samples_per_second": 1.273,
|
| 445 |
+
"eval_steps_per_second": 0.223,
|
| 446 |
+
"step": 496
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 16.01,
|
| 450 |
+
"learning_rate": 9.259259259259259e-06,
|
| 451 |
+
"loss": 0.443,
|
| 452 |
+
"step": 500
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 16.02,
|
| 456 |
+
"learning_rate": 8.333333333333334e-06,
|
| 457 |
+
"loss": 0.3427,
|
| 458 |
+
"step": 510
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 16.04,
|
| 462 |
+
"learning_rate": 7.4074074074074075e-06,
|
| 463 |
+
"loss": 0.2739,
|
| 464 |
+
"step": 520
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 16.05,
|
| 468 |
+
"eval_accuracy": 0.85,
|
| 469 |
+
"eval_loss": 0.4999323785305023,
|
| 470 |
+
"eval_runtime": 63.9597,
|
| 471 |
+
"eval_samples_per_second": 1.251,
|
| 472 |
+
"eval_steps_per_second": 0.219,
|
| 473 |
+
"step": 527
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 17.0,
|
| 477 |
+
"learning_rate": 6.481481481481481e-06,
|
| 478 |
+
"loss": 0.3676,
|
| 479 |
+
"step": 530
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 17.02,
|
| 483 |
+
"learning_rate": 5.555555555555556e-06,
|
| 484 |
+
"loss": 0.5171,
|
| 485 |
+
"step": 540
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 17.04,
|
| 489 |
+
"learning_rate": 4.6296296296296296e-06,
|
| 490 |
+
"loss": 0.3514,
|
| 491 |
+
"step": 550
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 17.05,
|
| 495 |
+
"eval_accuracy": 0.8375,
|
| 496 |
+
"eval_loss": 0.4638718068599701,
|
| 497 |
+
"eval_runtime": 65.0616,
|
| 498 |
+
"eval_samples_per_second": 1.23,
|
| 499 |
+
"eval_steps_per_second": 0.215,
|
| 500 |
+
"step": 558
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 18.0,
|
| 504 |
+
"learning_rate": 3.7037037037037037e-06,
|
| 505 |
+
"loss": 0.2877,
|
| 506 |
+
"step": 560
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 18.02,
|
| 510 |
+
"learning_rate": 2.777777777777778e-06,
|
| 511 |
+
"loss": 0.3558,
|
| 512 |
+
"step": 570
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 18.04,
|
| 516 |
+
"learning_rate": 1.8518518518518519e-06,
|
| 517 |
+
"loss": 0.4158,
|
| 518 |
+
"step": 580
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 18.05,
|
| 522 |
+
"eval_accuracy": 0.85,
|
| 523 |
+
"eval_loss": 0.42912358045578003,
|
| 524 |
+
"eval_runtime": 64.0669,
|
| 525 |
+
"eval_samples_per_second": 1.249,
|
| 526 |
+
"eval_steps_per_second": 0.219,
|
| 527 |
+
"step": 589
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 19.0,
|
| 531 |
+
"learning_rate": 9.259259259259259e-07,
|
| 532 |
+
"loss": 0.3158,
|
| 533 |
+
"step": 590
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 19.02,
|
| 537 |
+
"learning_rate": 0.0,
|
| 538 |
+
"loss": 0.2689,
|
| 539 |
+
"step": 600
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 19.02,
|
| 543 |
+
"eval_accuracy": 0.85,
|
| 544 |
+
"eval_loss": 0.42943257093429565,
|
| 545 |
+
"eval_runtime": 64.3343,
|
| 546 |
+
"eval_samples_per_second": 1.244,
|
| 547 |
+
"eval_steps_per_second": 0.218,
|
| 548 |
+
"step": 600
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 19.02,
|
| 552 |
+
"step": 600,
|
| 553 |
+
"total_flos": 4.415279302910214e+18,
|
| 554 |
+
"train_loss": 1.063151851495107,
|
| 555 |
+
"train_runtime": 5413.8579,
|
| 556 |
+
"train_samples_per_second": 0.665,
|
| 557 |
+
"train_steps_per_second": 0.111
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 19.02,
|
| 561 |
+
"eval_accuracy": 0.85,
|
| 562 |
+
"eval_loss": 0.49579018354415894,
|
| 563 |
+
"eval_runtime": 66.0758,
|
| 564 |
+
"eval_samples_per_second": 1.211,
|
| 565 |
+
"eval_steps_per_second": 0.212,
|
| 566 |
+
"step": 600
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 19.02,
|
| 570 |
+
"eval_accuracy": 0.85,
|
| 571 |
+
"eval_loss": 0.49579015374183655,
|
| 572 |
+
"eval_runtime": 64.7531,
|
| 573 |
+
"eval_samples_per_second": 1.235,
|
| 574 |
+
"eval_steps_per_second": 0.216,
|
| 575 |
+
"step": 600
|
| 576 |
+
}
|
| 577 |
+
],
|
| 578 |
+
"max_steps": 600,
|
| 579 |
+
"num_train_epochs": 9223372036854775807,
|
| 580 |
+
"total_flos": 4.415279302910214e+18,
|
| 581 |
+
"trial_name": null,
|
| 582 |
+
"trial_params": null
|
| 583 |
+
}
|