End of training
Browse files- README.md +2 -2
- all_results.json +13 -0
- eval_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +460 -0
README.md
CHANGED
|
@@ -17,8 +17,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [facebook/convnextv2-base-1k-224](https://huggingface.co/facebook/convnextv2-base-1k-224) on the None dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss: 1.
|
| 21 |
-
- Accuracy: 0.
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [facebook/convnextv2-base-1k-224](https://huggingface.co/facebook/convnextv2-base-1k-224) on the None dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 1.4479
|
| 21 |
+
- Accuracy: 0.4382
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 27.91,
|
| 3 |
+
"eval_accuracy": 0.4382284382284382,
|
| 4 |
+
"eval_loss": 1.447946548461914,
|
| 5 |
+
"eval_runtime": 8.404,
|
| 6 |
+
"eval_samples_per_second": 51.047,
|
| 7 |
+
"eval_steps_per_second": 1.666,
|
| 8 |
+
"total_flos": 3.007291871298355e+18,
|
| 9 |
+
"train_loss": 1.1542485936482747,
|
| 10 |
+
"train_runtime": 2419.37,
|
| 11 |
+
"train_samples_per_second": 16.864,
|
| 12 |
+
"train_steps_per_second": 0.124
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 27.91,
|
| 3 |
+
"eval_accuracy": 0.4382284382284382,
|
| 4 |
+
"eval_loss": 1.447946548461914,
|
| 5 |
+
"eval_runtime": 8.404,
|
| 6 |
+
"eval_samples_per_second": 51.047,
|
| 7 |
+
"eval_steps_per_second": 1.666
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 27.91,
|
| 3 |
+
"total_flos": 3.007291871298355e+18,
|
| 4 |
+
"train_loss": 1.1542485936482747,
|
| 5 |
+
"train_runtime": 2419.37,
|
| 6 |
+
"train_samples_per_second": 16.864,
|
| 7 |
+
"train_steps_per_second": 0.124
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,460 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.4382284382284382,
|
| 3 |
+
"best_model_checkpoint": "convnextv2-base-1k-224-for-pre_evaluation/checkpoint-268",
|
| 4 |
+
"epoch": 27.906976744186046,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 300,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.93,
|
| 13 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 14 |
+
"loss": 1.5952,
|
| 15 |
+
"step": 10
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.93,
|
| 19 |
+
"eval_accuracy": 0.29603729603729606,
|
| 20 |
+
"eval_loss": 1.5510554313659668,
|
| 21 |
+
"eval_runtime": 8.4324,
|
| 22 |
+
"eval_samples_per_second": 50.875,
|
| 23 |
+
"eval_steps_per_second": 1.66,
|
| 24 |
+
"step": 10
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 1.86,
|
| 28 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 29 |
+
"loss": 1.5238,
|
| 30 |
+
"step": 20
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 1.95,
|
| 34 |
+
"eval_accuracy": 0.34265734265734266,
|
| 35 |
+
"eval_loss": 1.5091006755828857,
|
| 36 |
+
"eval_runtime": 8.9811,
|
| 37 |
+
"eval_samples_per_second": 47.767,
|
| 38 |
+
"eval_steps_per_second": 1.559,
|
| 39 |
+
"step": 21
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 2.79,
|
| 43 |
+
"learning_rate": 5e-05,
|
| 44 |
+
"loss": 1.4881,
|
| 45 |
+
"step": 30
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 2.98,
|
| 49 |
+
"eval_accuracy": 0.34498834498834496,
|
| 50 |
+
"eval_loss": 1.4853538274765015,
|
| 51 |
+
"eval_runtime": 8.6991,
|
| 52 |
+
"eval_samples_per_second": 49.315,
|
| 53 |
+
"eval_steps_per_second": 1.609,
|
| 54 |
+
"step": 32
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 3.72,
|
| 58 |
+
"learning_rate": 4.814814814814815e-05,
|
| 59 |
+
"loss": 1.4708,
|
| 60 |
+
"step": 40
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 4.0,
|
| 64 |
+
"eval_accuracy": 0.3473193473193473,
|
| 65 |
+
"eval_loss": 1.4616328477859497,
|
| 66 |
+
"eval_runtime": 8.3701,
|
| 67 |
+
"eval_samples_per_second": 51.254,
|
| 68 |
+
"eval_steps_per_second": 1.673,
|
| 69 |
+
"step": 43
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 4.65,
|
| 73 |
+
"learning_rate": 4.62962962962963e-05,
|
| 74 |
+
"loss": 1.4361,
|
| 75 |
+
"step": 50
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 4.93,
|
| 79 |
+
"eval_accuracy": 0.34498834498834496,
|
| 80 |
+
"eval_loss": 1.4416619539260864,
|
| 81 |
+
"eval_runtime": 8.7032,
|
| 82 |
+
"eval_samples_per_second": 49.292,
|
| 83 |
+
"eval_steps_per_second": 1.609,
|
| 84 |
+
"step": 53
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 5.58,
|
| 88 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 89 |
+
"loss": 1.3764,
|
| 90 |
+
"step": 60
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 5.95,
|
| 94 |
+
"eval_accuracy": 0.3752913752913753,
|
| 95 |
+
"eval_loss": 1.4134629964828491,
|
| 96 |
+
"eval_runtime": 8.9398,
|
| 97 |
+
"eval_samples_per_second": 47.988,
|
| 98 |
+
"eval_steps_per_second": 1.566,
|
| 99 |
+
"step": 64
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 6.51,
|
| 103 |
+
"learning_rate": 4.259259259259259e-05,
|
| 104 |
+
"loss": 1.3333,
|
| 105 |
+
"step": 70
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 6.98,
|
| 109 |
+
"eval_accuracy": 0.3986013986013986,
|
| 110 |
+
"eval_loss": 1.3822472095489502,
|
| 111 |
+
"eval_runtime": 8.7499,
|
| 112 |
+
"eval_samples_per_second": 49.029,
|
| 113 |
+
"eval_steps_per_second": 1.6,
|
| 114 |
+
"step": 75
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 7.44,
|
| 118 |
+
"learning_rate": 4.074074074074074e-05,
|
| 119 |
+
"loss": 1.3296,
|
| 120 |
+
"step": 80
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 8.0,
|
| 124 |
+
"eval_accuracy": 0.36363636363636365,
|
| 125 |
+
"eval_loss": 1.4111592769622803,
|
| 126 |
+
"eval_runtime": 8.8496,
|
| 127 |
+
"eval_samples_per_second": 48.477,
|
| 128 |
+
"eval_steps_per_second": 1.582,
|
| 129 |
+
"step": 86
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 8.37,
|
| 133 |
+
"learning_rate": 3.888888888888889e-05,
|
| 134 |
+
"loss": 1.2798,
|
| 135 |
+
"step": 90
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 8.93,
|
| 139 |
+
"eval_accuracy": 0.38927738927738925,
|
| 140 |
+
"eval_loss": 1.4038037061691284,
|
| 141 |
+
"eval_runtime": 8.5853,
|
| 142 |
+
"eval_samples_per_second": 49.969,
|
| 143 |
+
"eval_steps_per_second": 1.631,
|
| 144 |
+
"step": 96
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 9.3,
|
| 148 |
+
"learning_rate": 3.7037037037037037e-05,
|
| 149 |
+
"loss": 1.3129,
|
| 150 |
+
"step": 100
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 9.95,
|
| 154 |
+
"eval_accuracy": 0.3776223776223776,
|
| 155 |
+
"eval_loss": 1.424072265625,
|
| 156 |
+
"eval_runtime": 8.6578,
|
| 157 |
+
"eval_samples_per_second": 49.551,
|
| 158 |
+
"eval_steps_per_second": 1.617,
|
| 159 |
+
"step": 107
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 10.23,
|
| 163 |
+
"learning_rate": 3.518518518518519e-05,
|
| 164 |
+
"loss": 1.3014,
|
| 165 |
+
"step": 110
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 10.98,
|
| 169 |
+
"eval_accuracy": 0.38927738927738925,
|
| 170 |
+
"eval_loss": 1.356952428817749,
|
| 171 |
+
"eval_runtime": 8.5422,
|
| 172 |
+
"eval_samples_per_second": 50.221,
|
| 173 |
+
"eval_steps_per_second": 1.639,
|
| 174 |
+
"step": 118
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 11.16,
|
| 178 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 179 |
+
"loss": 1.2332,
|
| 180 |
+
"step": 120
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 12.0,
|
| 184 |
+
"eval_accuracy": 0.38927738927738925,
|
| 185 |
+
"eval_loss": 1.4072706699371338,
|
| 186 |
+
"eval_runtime": 8.3607,
|
| 187 |
+
"eval_samples_per_second": 51.312,
|
| 188 |
+
"eval_steps_per_second": 1.675,
|
| 189 |
+
"step": 129
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 12.09,
|
| 193 |
+
"learning_rate": 3.148148148148148e-05,
|
| 194 |
+
"loss": 1.212,
|
| 195 |
+
"step": 130
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 12.93,
|
| 199 |
+
"eval_accuracy": 0.40326340326340326,
|
| 200 |
+
"eval_loss": 1.376956820487976,
|
| 201 |
+
"eval_runtime": 8.3675,
|
| 202 |
+
"eval_samples_per_second": 51.27,
|
| 203 |
+
"eval_steps_per_second": 1.673,
|
| 204 |
+
"step": 139
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 13.02,
|
| 208 |
+
"learning_rate": 2.962962962962963e-05,
|
| 209 |
+
"loss": 1.1844,
|
| 210 |
+
"step": 140
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 13.95,
|
| 214 |
+
"learning_rate": 2.777777777777778e-05,
|
| 215 |
+
"loss": 1.1763,
|
| 216 |
+
"step": 150
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 13.95,
|
| 220 |
+
"eval_accuracy": 0.3962703962703963,
|
| 221 |
+
"eval_loss": 1.3891488313674927,
|
| 222 |
+
"eval_runtime": 8.5717,
|
| 223 |
+
"eval_samples_per_second": 50.049,
|
| 224 |
+
"eval_steps_per_second": 1.633,
|
| 225 |
+
"step": 150
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 14.88,
|
| 229 |
+
"learning_rate": 2.5925925925925925e-05,
|
| 230 |
+
"loss": 1.124,
|
| 231 |
+
"step": 160
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 14.98,
|
| 235 |
+
"eval_accuracy": 0.4125874125874126,
|
| 236 |
+
"eval_loss": 1.3915237188339233,
|
| 237 |
+
"eval_runtime": 8.5638,
|
| 238 |
+
"eval_samples_per_second": 50.095,
|
| 239 |
+
"eval_steps_per_second": 1.635,
|
| 240 |
+
"step": 161
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 15.81,
|
| 244 |
+
"learning_rate": 2.4074074074074074e-05,
|
| 245 |
+
"loss": 1.0963,
|
| 246 |
+
"step": 170
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 16.0,
|
| 250 |
+
"eval_accuracy": 0.4149184149184149,
|
| 251 |
+
"eval_loss": 1.4098657369613647,
|
| 252 |
+
"eval_runtime": 8.8116,
|
| 253 |
+
"eval_samples_per_second": 48.686,
|
| 254 |
+
"eval_steps_per_second": 1.589,
|
| 255 |
+
"step": 172
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 16.74,
|
| 259 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 260 |
+
"loss": 1.0547,
|
| 261 |
+
"step": 180
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 16.93,
|
| 265 |
+
"eval_accuracy": 0.40326340326340326,
|
| 266 |
+
"eval_loss": 1.4206278324127197,
|
| 267 |
+
"eval_runtime": 8.7717,
|
| 268 |
+
"eval_samples_per_second": 48.907,
|
| 269 |
+
"eval_steps_per_second": 1.596,
|
| 270 |
+
"step": 182
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 17.67,
|
| 274 |
+
"learning_rate": 2.037037037037037e-05,
|
| 275 |
+
"loss": 1.0631,
|
| 276 |
+
"step": 190
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 17.95,
|
| 280 |
+
"eval_accuracy": 0.4195804195804196,
|
| 281 |
+
"eval_loss": 1.4040827751159668,
|
| 282 |
+
"eval_runtime": 8.3983,
|
| 283 |
+
"eval_samples_per_second": 51.082,
|
| 284 |
+
"eval_steps_per_second": 1.667,
|
| 285 |
+
"step": 193
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 18.6,
|
| 289 |
+
"learning_rate": 1.8518518518518518e-05,
|
| 290 |
+
"loss": 0.9911,
|
| 291 |
+
"step": 200
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 18.98,
|
| 295 |
+
"eval_accuracy": 0.4149184149184149,
|
| 296 |
+
"eval_loss": 1.4271957874298096,
|
| 297 |
+
"eval_runtime": 8.2919,
|
| 298 |
+
"eval_samples_per_second": 51.737,
|
| 299 |
+
"eval_steps_per_second": 1.688,
|
| 300 |
+
"step": 204
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 19.53,
|
| 304 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 305 |
+
"loss": 1.005,
|
| 306 |
+
"step": 210
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 20.0,
|
| 310 |
+
"eval_accuracy": 0.4219114219114219,
|
| 311 |
+
"eval_loss": 1.42105233669281,
|
| 312 |
+
"eval_runtime": 8.2769,
|
| 313 |
+
"eval_samples_per_second": 51.831,
|
| 314 |
+
"eval_steps_per_second": 1.691,
|
| 315 |
+
"step": 215
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 20.47,
|
| 319 |
+
"learning_rate": 1.4814814814814815e-05,
|
| 320 |
+
"loss": 0.9663,
|
| 321 |
+
"step": 220
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 20.93,
|
| 325 |
+
"eval_accuracy": 0.40093240093240096,
|
| 326 |
+
"eval_loss": 1.466171145439148,
|
| 327 |
+
"eval_runtime": 9.4718,
|
| 328 |
+
"eval_samples_per_second": 45.292,
|
| 329 |
+
"eval_steps_per_second": 1.478,
|
| 330 |
+
"step": 225
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 21.4,
|
| 334 |
+
"learning_rate": 1.2962962962962962e-05,
|
| 335 |
+
"loss": 0.9533,
|
| 336 |
+
"step": 230
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 21.95,
|
| 340 |
+
"eval_accuracy": 0.43356643356643354,
|
| 341 |
+
"eval_loss": 1.428614616394043,
|
| 342 |
+
"eval_runtime": 8.343,
|
| 343 |
+
"eval_samples_per_second": 51.42,
|
| 344 |
+
"eval_steps_per_second": 1.678,
|
| 345 |
+
"step": 236
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 22.33,
|
| 349 |
+
"learning_rate": 1.1111111111111112e-05,
|
| 350 |
+
"loss": 0.9506,
|
| 351 |
+
"step": 240
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 22.98,
|
| 355 |
+
"eval_accuracy": 0.43123543123543123,
|
| 356 |
+
"eval_loss": 1.413465976715088,
|
| 357 |
+
"eval_runtime": 8.7694,
|
| 358 |
+
"eval_samples_per_second": 48.92,
|
| 359 |
+
"eval_steps_per_second": 1.596,
|
| 360 |
+
"step": 247
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 23.26,
|
| 364 |
+
"learning_rate": 9.259259259259259e-06,
|
| 365 |
+
"loss": 0.8973,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 24.0,
|
| 370 |
+
"eval_accuracy": 0.42657342657342656,
|
| 371 |
+
"eval_loss": 1.442847728729248,
|
| 372 |
+
"eval_runtime": 8.7464,
|
| 373 |
+
"eval_samples_per_second": 49.049,
|
| 374 |
+
"eval_steps_per_second": 1.601,
|
| 375 |
+
"step": 258
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 24.19,
|
| 379 |
+
"learning_rate": 7.4074074074074075e-06,
|
| 380 |
+
"loss": 0.8807,
|
| 381 |
+
"step": 260
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 24.93,
|
| 385 |
+
"eval_accuracy": 0.4382284382284382,
|
| 386 |
+
"eval_loss": 1.447946548461914,
|
| 387 |
+
"eval_runtime": 8.3339,
|
| 388 |
+
"eval_samples_per_second": 51.476,
|
| 389 |
+
"eval_steps_per_second": 1.68,
|
| 390 |
+
"step": 268
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 25.12,
|
| 394 |
+
"learning_rate": 5.555555555555556e-06,
|
| 395 |
+
"loss": 0.8731,
|
| 396 |
+
"step": 270
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 25.95,
|
| 400 |
+
"eval_accuracy": 0.4289044289044289,
|
| 401 |
+
"eval_loss": 1.4429428577423096,
|
| 402 |
+
"eval_runtime": 8.7998,
|
| 403 |
+
"eval_samples_per_second": 48.751,
|
| 404 |
+
"eval_steps_per_second": 1.591,
|
| 405 |
+
"step": 279
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 26.05,
|
| 409 |
+
"learning_rate": 3.7037037037037037e-06,
|
| 410 |
+
"loss": 0.8366,
|
| 411 |
+
"step": 280
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 26.98,
|
| 415 |
+
"learning_rate": 1.8518518518518519e-06,
|
| 416 |
+
"loss": 0.8472,
|
| 417 |
+
"step": 290
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 26.98,
|
| 421 |
+
"eval_accuracy": 0.43123543123543123,
|
| 422 |
+
"eval_loss": 1.4461231231689453,
|
| 423 |
+
"eval_runtime": 8.7766,
|
| 424 |
+
"eval_samples_per_second": 48.88,
|
| 425 |
+
"eval_steps_per_second": 1.595,
|
| 426 |
+
"step": 290
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 27.91,
|
| 430 |
+
"learning_rate": 0.0,
|
| 431 |
+
"loss": 0.8348,
|
| 432 |
+
"step": 300
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 27.91,
|
| 436 |
+
"eval_accuracy": 0.43356643356643354,
|
| 437 |
+
"eval_loss": 1.453087568283081,
|
| 438 |
+
"eval_runtime": 8.3523,
|
| 439 |
+
"eval_samples_per_second": 51.363,
|
| 440 |
+
"eval_steps_per_second": 1.676,
|
| 441 |
+
"step": 300
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 27.91,
|
| 445 |
+
"step": 300,
|
| 446 |
+
"total_flos": 3.007291871298355e+18,
|
| 447 |
+
"train_loss": 1.1542485936482747,
|
| 448 |
+
"train_runtime": 2419.37,
|
| 449 |
+
"train_samples_per_second": 16.864,
|
| 450 |
+
"train_steps_per_second": 0.124
|
| 451 |
+
}
|
| 452 |
+
],
|
| 453 |
+
"logging_steps": 10,
|
| 454 |
+
"max_steps": 300,
|
| 455 |
+
"num_train_epochs": 30,
|
| 456 |
+
"save_steps": 500,
|
| 457 |
+
"total_flos": 3.007291871298355e+18,
|
| 458 |
+
"trial_name": null,
|
| 459 |
+
"trial_params": null
|
| 460 |
+
}
|