Commit
•
3758f77
1
Parent(s):
00dccdf
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 +421 -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 [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) on an unknown dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss: 0.
|
| 21 |
-
- Accuracy: 0.
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) on an unknown dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 0.3404
|
| 21 |
+
- Accuracy: 0.9302
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 36.92,
|
| 3 |
+
"eval_accuracy": 0.9302325581395349,
|
| 4 |
+
"eval_loss": 0.340448260307312,
|
| 5 |
+
"eval_runtime": 0.6612,
|
| 6 |
+
"eval_samples_per_second": 65.037,
|
| 7 |
+
"eval_steps_per_second": 3.025,
|
| 8 |
+
"total_flos": 1.1038156860172861e+18,
|
| 9 |
+
"train_loss": 0.3578883737325668,
|
| 10 |
+
"train_runtime": 762.5104,
|
| 11 |
+
"train_samples_per_second": 20.196,
|
| 12 |
+
"train_steps_per_second": 0.157
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 36.92,
|
| 3 |
+
"eval_accuracy": 0.9302325581395349,
|
| 4 |
+
"eval_loss": 0.340448260307312,
|
| 5 |
+
"eval_runtime": 0.6612,
|
| 6 |
+
"eval_samples_per_second": 65.037,
|
| 7 |
+
"eval_steps_per_second": 3.025
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 36.92,
|
| 3 |
+
"total_flos": 1.1038156860172861e+18,
|
| 4 |
+
"train_loss": 0.3578883737325668,
|
| 5 |
+
"train_runtime": 762.5104,
|
| 6 |
+
"train_samples_per_second": 20.196,
|
| 7 |
+
"train_steps_per_second": 0.157
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,421 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9302325581395349,
|
| 3 |
+
"best_model_checkpoint": "vit-base-patch16-224-dmae-va-da-40D/checkpoint-91",
|
| 4 |
+
"epoch": 36.92307692307692,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 120,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.92,
|
| 13 |
+
"eval_accuracy": 0.4186046511627907,
|
| 14 |
+
"eval_loss": 1.2900550365447998,
|
| 15 |
+
"eval_runtime": 1.1372,
|
| 16 |
+
"eval_samples_per_second": 37.812,
|
| 17 |
+
"eval_steps_per_second": 1.759,
|
| 18 |
+
"step": 3
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 1.85,
|
| 22 |
+
"eval_accuracy": 0.4418604651162791,
|
| 23 |
+
"eval_loss": 1.2313674688339233,
|
| 24 |
+
"eval_runtime": 0.5544,
|
| 25 |
+
"eval_samples_per_second": 77.555,
|
| 26 |
+
"eval_steps_per_second": 3.607,
|
| 27 |
+
"step": 6
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 2.77,
|
| 31 |
+
"eval_accuracy": 0.46511627906976744,
|
| 32 |
+
"eval_loss": 1.1530039310455322,
|
| 33 |
+
"eval_runtime": 0.558,
|
| 34 |
+
"eval_samples_per_second": 77.054,
|
| 35 |
+
"eval_steps_per_second": 3.584,
|
| 36 |
+
"step": 9
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 3.69,
|
| 40 |
+
"learning_rate": 5e-05,
|
| 41 |
+
"loss": 1.2976,
|
| 42 |
+
"step": 12
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 4.0,
|
| 46 |
+
"eval_accuracy": 0.6046511627906976,
|
| 47 |
+
"eval_loss": 0.9851991534233093,
|
| 48 |
+
"eval_runtime": 0.5772,
|
| 49 |
+
"eval_samples_per_second": 74.501,
|
| 50 |
+
"eval_steps_per_second": 3.465,
|
| 51 |
+
"step": 13
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 4.92,
|
| 55 |
+
"eval_accuracy": 0.7674418604651163,
|
| 56 |
+
"eval_loss": 0.8450184464454651,
|
| 57 |
+
"eval_runtime": 0.6419,
|
| 58 |
+
"eval_samples_per_second": 66.992,
|
| 59 |
+
"eval_steps_per_second": 3.116,
|
| 60 |
+
"step": 16
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 5.85,
|
| 64 |
+
"eval_accuracy": 0.6511627906976745,
|
| 65 |
+
"eval_loss": 0.8366963863372803,
|
| 66 |
+
"eval_runtime": 0.6054,
|
| 67 |
+
"eval_samples_per_second": 71.03,
|
| 68 |
+
"eval_steps_per_second": 3.304,
|
| 69 |
+
"step": 19
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 6.77,
|
| 73 |
+
"eval_accuracy": 0.7209302325581395,
|
| 74 |
+
"eval_loss": 0.7544969916343689,
|
| 75 |
+
"eval_runtime": 0.6137,
|
| 76 |
+
"eval_samples_per_second": 70.066,
|
| 77 |
+
"eval_steps_per_second": 3.259,
|
| 78 |
+
"step": 22
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 7.38,
|
| 82 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 83 |
+
"loss": 0.8294,
|
| 84 |
+
"step": 24
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 8.0,
|
| 88 |
+
"eval_accuracy": 0.7906976744186046,
|
| 89 |
+
"eval_loss": 0.6711422204971313,
|
| 90 |
+
"eval_runtime": 0.6274,
|
| 91 |
+
"eval_samples_per_second": 68.534,
|
| 92 |
+
"eval_steps_per_second": 3.188,
|
| 93 |
+
"step": 26
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 8.92,
|
| 97 |
+
"eval_accuracy": 0.7209302325581395,
|
| 98 |
+
"eval_loss": 0.6739276051521301,
|
| 99 |
+
"eval_runtime": 0.6711,
|
| 100 |
+
"eval_samples_per_second": 64.073,
|
| 101 |
+
"eval_steps_per_second": 2.98,
|
| 102 |
+
"step": 29
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 9.85,
|
| 106 |
+
"eval_accuracy": 0.7441860465116279,
|
| 107 |
+
"eval_loss": 0.6009616851806641,
|
| 108 |
+
"eval_runtime": 0.6278,
|
| 109 |
+
"eval_samples_per_second": 68.489,
|
| 110 |
+
"eval_steps_per_second": 3.186,
|
| 111 |
+
"step": 32
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 10.77,
|
| 115 |
+
"eval_accuracy": 0.7441860465116279,
|
| 116 |
+
"eval_loss": 0.5369246602058411,
|
| 117 |
+
"eval_runtime": 0.6174,
|
| 118 |
+
"eval_samples_per_second": 69.646,
|
| 119 |
+
"eval_steps_per_second": 3.239,
|
| 120 |
+
"step": 35
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 11.08,
|
| 124 |
+
"learning_rate": 3.888888888888889e-05,
|
| 125 |
+
"loss": 0.4293,
|
| 126 |
+
"step": 36
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 12.0,
|
| 130 |
+
"eval_accuracy": 0.7906976744186046,
|
| 131 |
+
"eval_loss": 0.5271600484848022,
|
| 132 |
+
"eval_runtime": 0.6878,
|
| 133 |
+
"eval_samples_per_second": 62.521,
|
| 134 |
+
"eval_steps_per_second": 2.908,
|
| 135 |
+
"step": 39
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 12.92,
|
| 139 |
+
"eval_accuracy": 0.7441860465116279,
|
| 140 |
+
"eval_loss": 0.5216893553733826,
|
| 141 |
+
"eval_runtime": 0.6118,
|
| 142 |
+
"eval_samples_per_second": 70.282,
|
| 143 |
+
"eval_steps_per_second": 3.269,
|
| 144 |
+
"step": 42
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 13.85,
|
| 148 |
+
"eval_accuracy": 0.7674418604651163,
|
| 149 |
+
"eval_loss": 0.4843863546848297,
|
| 150 |
+
"eval_runtime": 0.605,
|
| 151 |
+
"eval_samples_per_second": 71.078,
|
| 152 |
+
"eval_steps_per_second": 3.306,
|
| 153 |
+
"step": 45
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 14.77,
|
| 157 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 158 |
+
"loss": 0.2695,
|
| 159 |
+
"step": 48
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 14.77,
|
| 163 |
+
"eval_accuracy": 0.7906976744186046,
|
| 164 |
+
"eval_loss": 0.4947708547115326,
|
| 165 |
+
"eval_runtime": 0.664,
|
| 166 |
+
"eval_samples_per_second": 64.758,
|
| 167 |
+
"eval_steps_per_second": 3.012,
|
| 168 |
+
"step": 48
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 16.0,
|
| 172 |
+
"eval_accuracy": 0.7674418604651163,
|
| 173 |
+
"eval_loss": 0.4775972068309784,
|
| 174 |
+
"eval_runtime": 0.6155,
|
| 175 |
+
"eval_samples_per_second": 69.858,
|
| 176 |
+
"eval_steps_per_second": 3.249,
|
| 177 |
+
"step": 52
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 16.92,
|
| 181 |
+
"eval_accuracy": 0.7906976744186046,
|
| 182 |
+
"eval_loss": 0.44103220105171204,
|
| 183 |
+
"eval_runtime": 0.6148,
|
| 184 |
+
"eval_samples_per_second": 69.936,
|
| 185 |
+
"eval_steps_per_second": 3.253,
|
| 186 |
+
"step": 55
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 17.85,
|
| 190 |
+
"eval_accuracy": 0.7441860465116279,
|
| 191 |
+
"eval_loss": 0.48707905411720276,
|
| 192 |
+
"eval_runtime": 0.6184,
|
| 193 |
+
"eval_samples_per_second": 69.53,
|
| 194 |
+
"eval_steps_per_second": 3.234,
|
| 195 |
+
"step": 58
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 18.46,
|
| 199 |
+
"learning_rate": 2.777777777777778e-05,
|
| 200 |
+
"loss": 0.1905,
|
| 201 |
+
"step": 60
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 18.77,
|
| 205 |
+
"eval_accuracy": 0.7906976744186046,
|
| 206 |
+
"eval_loss": 0.4374682903289795,
|
| 207 |
+
"eval_runtime": 0.6741,
|
| 208 |
+
"eval_samples_per_second": 63.79,
|
| 209 |
+
"eval_steps_per_second": 2.967,
|
| 210 |
+
"step": 61
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 20.0,
|
| 214 |
+
"eval_accuracy": 0.813953488372093,
|
| 215 |
+
"eval_loss": 0.45782390236854553,
|
| 216 |
+
"eval_runtime": 0.6053,
|
| 217 |
+
"eval_samples_per_second": 71.04,
|
| 218 |
+
"eval_steps_per_second": 3.304,
|
| 219 |
+
"step": 65
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 20.92,
|
| 223 |
+
"eval_accuracy": 0.813953488372093,
|
| 224 |
+
"eval_loss": 0.4955892264842987,
|
| 225 |
+
"eval_runtime": 0.6057,
|
| 226 |
+
"eval_samples_per_second": 70.987,
|
| 227 |
+
"eval_steps_per_second": 3.302,
|
| 228 |
+
"step": 68
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 21.85,
|
| 232 |
+
"eval_accuracy": 0.813953488372093,
|
| 233 |
+
"eval_loss": 0.4500349164009094,
|
| 234 |
+
"eval_runtime": 0.6339,
|
| 235 |
+
"eval_samples_per_second": 67.831,
|
| 236 |
+
"eval_steps_per_second": 3.155,
|
| 237 |
+
"step": 71
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 22.15,
|
| 241 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 242 |
+
"loss": 0.135,
|
| 243 |
+
"step": 72
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 22.77,
|
| 247 |
+
"eval_accuracy": 0.8604651162790697,
|
| 248 |
+
"eval_loss": 0.40709012746810913,
|
| 249 |
+
"eval_runtime": 0.6806,
|
| 250 |
+
"eval_samples_per_second": 63.179,
|
| 251 |
+
"eval_steps_per_second": 2.939,
|
| 252 |
+
"step": 74
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 24.0,
|
| 256 |
+
"eval_accuracy": 0.8604651162790697,
|
| 257 |
+
"eval_loss": 0.41575565934181213,
|
| 258 |
+
"eval_runtime": 0.6102,
|
| 259 |
+
"eval_samples_per_second": 70.471,
|
| 260 |
+
"eval_steps_per_second": 3.278,
|
| 261 |
+
"step": 78
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 24.92,
|
| 265 |
+
"eval_accuracy": 0.8372093023255814,
|
| 266 |
+
"eval_loss": 0.43799346685409546,
|
| 267 |
+
"eval_runtime": 0.6137,
|
| 268 |
+
"eval_samples_per_second": 70.068,
|
| 269 |
+
"eval_steps_per_second": 3.259,
|
| 270 |
+
"step": 81
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 25.85,
|
| 274 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 275 |
+
"loss": 0.1485,
|
| 276 |
+
"step": 84
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 25.85,
|
| 280 |
+
"eval_accuracy": 0.813953488372093,
|
| 281 |
+
"eval_loss": 0.4280536472797394,
|
| 282 |
+
"eval_runtime": 0.6787,
|
| 283 |
+
"eval_samples_per_second": 63.361,
|
| 284 |
+
"eval_steps_per_second": 2.947,
|
| 285 |
+
"step": 84
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 26.77,
|
| 289 |
+
"eval_accuracy": 0.8837209302325582,
|
| 290 |
+
"eval_loss": 0.3776519298553467,
|
| 291 |
+
"eval_runtime": 0.6159,
|
| 292 |
+
"eval_samples_per_second": 69.822,
|
| 293 |
+
"eval_steps_per_second": 3.248,
|
| 294 |
+
"step": 87
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 28.0,
|
| 298 |
+
"eval_accuracy": 0.9302325581395349,
|
| 299 |
+
"eval_loss": 0.340448260307312,
|
| 300 |
+
"eval_runtime": 0.6251,
|
| 301 |
+
"eval_samples_per_second": 68.79,
|
| 302 |
+
"eval_steps_per_second": 3.2,
|
| 303 |
+
"step": 91
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 28.92,
|
| 307 |
+
"eval_accuracy": 0.9069767441860465,
|
| 308 |
+
"eval_loss": 0.3580568730831146,
|
| 309 |
+
"eval_runtime": 0.6149,
|
| 310 |
+
"eval_samples_per_second": 69.927,
|
| 311 |
+
"eval_steps_per_second": 3.252,
|
| 312 |
+
"step": 94
|
| 313 |
+
},
|
| 314 |
+
{
|
| 315 |
+
"epoch": 29.54,
|
| 316 |
+
"learning_rate": 1.1111111111111112e-05,
|
| 317 |
+
"loss": 0.1001,
|
| 318 |
+
"step": 96
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 29.85,
|
| 322 |
+
"eval_accuracy": 0.8604651162790697,
|
| 323 |
+
"eval_loss": 0.38068118691444397,
|
| 324 |
+
"eval_runtime": 0.6756,
|
| 325 |
+
"eval_samples_per_second": 63.646,
|
| 326 |
+
"eval_steps_per_second": 2.96,
|
| 327 |
+
"step": 97
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 30.77,
|
| 331 |
+
"eval_accuracy": 0.8837209302325582,
|
| 332 |
+
"eval_loss": 0.3700192868709564,
|
| 333 |
+
"eval_runtime": 0.6122,
|
| 334 |
+
"eval_samples_per_second": 70.236,
|
| 335 |
+
"eval_steps_per_second": 3.267,
|
| 336 |
+
"step": 100
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 32.0,
|
| 340 |
+
"eval_accuracy": 0.8837209302325582,
|
| 341 |
+
"eval_loss": 0.37302762269973755,
|
| 342 |
+
"eval_runtime": 0.6248,
|
| 343 |
+
"eval_samples_per_second": 68.822,
|
| 344 |
+
"eval_steps_per_second": 3.201,
|
| 345 |
+
"step": 104
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 32.92,
|
| 349 |
+
"eval_accuracy": 0.8837209302325582,
|
| 350 |
+
"eval_loss": 0.38681066036224365,
|
| 351 |
+
"eval_runtime": 0.6675,
|
| 352 |
+
"eval_samples_per_second": 64.42,
|
| 353 |
+
"eval_steps_per_second": 2.996,
|
| 354 |
+
"step": 107
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 33.23,
|
| 358 |
+
"learning_rate": 5.555555555555556e-06,
|
| 359 |
+
"loss": 0.0797,
|
| 360 |
+
"step": 108
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 33.85,
|
| 364 |
+
"eval_accuracy": 0.8604651162790697,
|
| 365 |
+
"eval_loss": 0.38833436369895935,
|
| 366 |
+
"eval_runtime": 0.6127,
|
| 367 |
+
"eval_samples_per_second": 70.184,
|
| 368 |
+
"eval_steps_per_second": 3.264,
|
| 369 |
+
"step": 110
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 34.77,
|
| 373 |
+
"eval_accuracy": 0.8372093023255814,
|
| 374 |
+
"eval_loss": 0.39326179027557373,
|
| 375 |
+
"eval_runtime": 0.6145,
|
| 376 |
+
"eval_samples_per_second": 69.979,
|
| 377 |
+
"eval_steps_per_second": 3.255,
|
| 378 |
+
"step": 113
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 36.0,
|
| 382 |
+
"eval_accuracy": 0.8372093023255814,
|
| 383 |
+
"eval_loss": 0.39980578422546387,
|
| 384 |
+
"eval_runtime": 0.6129,
|
| 385 |
+
"eval_samples_per_second": 70.154,
|
| 386 |
+
"eval_steps_per_second": 3.263,
|
| 387 |
+
"step": 117
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 36.92,
|
| 391 |
+
"learning_rate": 0.0,
|
| 392 |
+
"loss": 0.0991,
|
| 393 |
+
"step": 120
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 36.92,
|
| 397 |
+
"eval_accuracy": 0.8372093023255814,
|
| 398 |
+
"eval_loss": 0.40144917368888855,
|
| 399 |
+
"eval_runtime": 0.6828,
|
| 400 |
+
"eval_samples_per_second": 62.975,
|
| 401 |
+
"eval_steps_per_second": 2.929,
|
| 402 |
+
"step": 120
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 36.92,
|
| 406 |
+
"step": 120,
|
| 407 |
+
"total_flos": 1.1038156860172861e+18,
|
| 408 |
+
"train_loss": 0.3578883737325668,
|
| 409 |
+
"train_runtime": 762.5104,
|
| 410 |
+
"train_samples_per_second": 20.196,
|
| 411 |
+
"train_steps_per_second": 0.157
|
| 412 |
+
}
|
| 413 |
+
],
|
| 414 |
+
"logging_steps": 12,
|
| 415 |
+
"max_steps": 120,
|
| 416 |
+
"num_train_epochs": 40,
|
| 417 |
+
"save_steps": 500,
|
| 418 |
+
"total_flos": 1.1038156860172861e+18,
|
| 419 |
+
"trial_name": null,
|
| 420 |
+
"trial_params": null
|
| 421 |
+
}
|