Commit
•
c3e0183
1
Parent(s):
d3b54af
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 +448 -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: 0.2660
|
| 21 |
+
- Accuracy: 0.9655
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 37.65,
|
| 3 |
+
"eval_accuracy": 0.9655172413793104,
|
| 4 |
+
"eval_loss": 0.26601171493530273,
|
| 5 |
+
"eval_runtime": 1.4626,
|
| 6 |
+
"eval_samples_per_second": 39.655,
|
| 7 |
+
"eval_steps_per_second": 1.367,
|
| 8 |
+
"total_flos": 1.5182502752352338e+18,
|
| 9 |
+
"train_loss": 0.2719826718792319,
|
| 10 |
+
"train_runtime": 1073.8442,
|
| 11 |
+
"train_samples_per_second": 19.37,
|
| 12 |
+
"train_steps_per_second": 0.149
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 37.65,
|
| 3 |
+
"eval_accuracy": 0.9655172413793104,
|
| 4 |
+
"eval_loss": 0.26601171493530273,
|
| 5 |
+
"eval_runtime": 1.4626,
|
| 6 |
+
"eval_samples_per_second": 39.655,
|
| 7 |
+
"eval_steps_per_second": 1.367
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 37.65,
|
| 3 |
+
"total_flos": 1.5182502752352338e+18,
|
| 4 |
+
"train_loss": 0.2719826718792319,
|
| 5 |
+
"train_runtime": 1073.8442,
|
| 6 |
+
"train_samples_per_second": 19.37,
|
| 7 |
+
"train_steps_per_second": 0.149
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,448 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9655172413793104,
|
| 3 |
+
"best_model_checkpoint": "vit-base-patch16-224-dmae-va-da2-40/checkpoint-46",
|
| 4 |
+
"epoch": 37.64705882352941,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 160,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.94,
|
| 13 |
+
"eval_accuracy": 0.43103448275862066,
|
| 14 |
+
"eval_loss": 1.29306960105896,
|
| 15 |
+
"eval_runtime": 1.9553,
|
| 16 |
+
"eval_samples_per_second": 29.663,
|
| 17 |
+
"eval_steps_per_second": 1.023,
|
| 18 |
+
"step": 4
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 1.88,
|
| 22 |
+
"eval_accuracy": 0.5517241379310345,
|
| 23 |
+
"eval_loss": 1.2023708820343018,
|
| 24 |
+
"eval_runtime": 1.4877,
|
| 25 |
+
"eval_samples_per_second": 38.987,
|
| 26 |
+
"eval_steps_per_second": 1.344,
|
| 27 |
+
"step": 8
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 2.82,
|
| 31 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 32 |
+
"loss": 1.2651,
|
| 33 |
+
"step": 12
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 2.82,
|
| 37 |
+
"eval_accuracy": 0.6551724137931034,
|
| 38 |
+
"eval_loss": 0.9896121621131897,
|
| 39 |
+
"eval_runtime": 1.3422,
|
| 40 |
+
"eval_samples_per_second": 43.212,
|
| 41 |
+
"eval_steps_per_second": 1.49,
|
| 42 |
+
"step": 12
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 4.0,
|
| 46 |
+
"eval_accuracy": 0.7241379310344828,
|
| 47 |
+
"eval_loss": 0.7972208261489868,
|
| 48 |
+
"eval_runtime": 1.1676,
|
| 49 |
+
"eval_samples_per_second": 49.674,
|
| 50 |
+
"eval_steps_per_second": 1.713,
|
| 51 |
+
"step": 17
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 4.94,
|
| 55 |
+
"eval_accuracy": 0.6551724137931034,
|
| 56 |
+
"eval_loss": 0.7336073517799377,
|
| 57 |
+
"eval_runtime": 1.4618,
|
| 58 |
+
"eval_samples_per_second": 39.678,
|
| 59 |
+
"eval_steps_per_second": 1.368,
|
| 60 |
+
"step": 21
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 5.65,
|
| 64 |
+
"learning_rate": 4.722222222222222e-05,
|
| 65 |
+
"loss": 0.7523,
|
| 66 |
+
"step": 24
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 5.88,
|
| 70 |
+
"eval_accuracy": 0.8103448275862069,
|
| 71 |
+
"eval_loss": 0.5780647397041321,
|
| 72 |
+
"eval_runtime": 1.4658,
|
| 73 |
+
"eval_samples_per_second": 39.569,
|
| 74 |
+
"eval_steps_per_second": 1.364,
|
| 75 |
+
"step": 25
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 6.82,
|
| 79 |
+
"eval_accuracy": 0.8793103448275862,
|
| 80 |
+
"eval_loss": 0.4911504089832306,
|
| 81 |
+
"eval_runtime": 1.1824,
|
| 82 |
+
"eval_samples_per_second": 49.051,
|
| 83 |
+
"eval_steps_per_second": 1.691,
|
| 84 |
+
"step": 29
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 8.0,
|
| 88 |
+
"eval_accuracy": 0.9137931034482759,
|
| 89 |
+
"eval_loss": 0.41120994091033936,
|
| 90 |
+
"eval_runtime": 1.1976,
|
| 91 |
+
"eval_samples_per_second": 48.428,
|
| 92 |
+
"eval_steps_per_second": 1.67,
|
| 93 |
+
"step": 34
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 8.47,
|
| 97 |
+
"learning_rate": 4.305555555555556e-05,
|
| 98 |
+
"loss": 0.4209,
|
| 99 |
+
"step": 36
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 8.94,
|
| 103 |
+
"eval_accuracy": 0.9137931034482759,
|
| 104 |
+
"eval_loss": 0.338276743888855,
|
| 105 |
+
"eval_runtime": 1.1796,
|
| 106 |
+
"eval_samples_per_second": 49.168,
|
| 107 |
+
"eval_steps_per_second": 1.695,
|
| 108 |
+
"step": 38
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 9.88,
|
| 112 |
+
"eval_accuracy": 0.9482758620689655,
|
| 113 |
+
"eval_loss": 0.3129003047943115,
|
| 114 |
+
"eval_runtime": 1.1761,
|
| 115 |
+
"eval_samples_per_second": 49.316,
|
| 116 |
+
"eval_steps_per_second": 1.701,
|
| 117 |
+
"step": 42
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 10.82,
|
| 121 |
+
"eval_accuracy": 0.9655172413793104,
|
| 122 |
+
"eval_loss": 0.26601171493530273,
|
| 123 |
+
"eval_runtime": 1.1866,
|
| 124 |
+
"eval_samples_per_second": 48.88,
|
| 125 |
+
"eval_steps_per_second": 1.686,
|
| 126 |
+
"step": 46
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 11.29,
|
| 130 |
+
"learning_rate": 3.888888888888889e-05,
|
| 131 |
+
"loss": 0.2647,
|
| 132 |
+
"step": 48
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 12.0,
|
| 136 |
+
"eval_accuracy": 0.9310344827586207,
|
| 137 |
+
"eval_loss": 0.3184232711791992,
|
| 138 |
+
"eval_runtime": 1.1991,
|
| 139 |
+
"eval_samples_per_second": 48.371,
|
| 140 |
+
"eval_steps_per_second": 1.668,
|
| 141 |
+
"step": 51
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 12.94,
|
| 145 |
+
"eval_accuracy": 0.9310344827586207,
|
| 146 |
+
"eval_loss": 0.2870759963989258,
|
| 147 |
+
"eval_runtime": 1.1806,
|
| 148 |
+
"eval_samples_per_second": 49.126,
|
| 149 |
+
"eval_steps_per_second": 1.694,
|
| 150 |
+
"step": 55
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 13.88,
|
| 154 |
+
"eval_accuracy": 0.9137931034482759,
|
| 155 |
+
"eval_loss": 0.27659285068511963,
|
| 156 |
+
"eval_runtime": 1.4456,
|
| 157 |
+
"eval_samples_per_second": 40.122,
|
| 158 |
+
"eval_steps_per_second": 1.384,
|
| 159 |
+
"step": 59
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 14.12,
|
| 163 |
+
"learning_rate": 3.472222222222222e-05,
|
| 164 |
+
"loss": 0.1743,
|
| 165 |
+
"step": 60
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 14.82,
|
| 169 |
+
"eval_accuracy": 0.896551724137931,
|
| 170 |
+
"eval_loss": 0.2726757526397705,
|
| 171 |
+
"eval_runtime": 1.1844,
|
| 172 |
+
"eval_samples_per_second": 48.97,
|
| 173 |
+
"eval_steps_per_second": 1.689,
|
| 174 |
+
"step": 63
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 16.0,
|
| 178 |
+
"eval_accuracy": 0.9310344827586207,
|
| 179 |
+
"eval_loss": 0.2282419204711914,
|
| 180 |
+
"eval_runtime": 1.198,
|
| 181 |
+
"eval_samples_per_second": 48.414,
|
| 182 |
+
"eval_steps_per_second": 1.669,
|
| 183 |
+
"step": 68
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 16.94,
|
| 187 |
+
"learning_rate": 3.055555555555556e-05,
|
| 188 |
+
"loss": 0.1511,
|
| 189 |
+
"step": 72
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 16.94,
|
| 193 |
+
"eval_accuracy": 0.896551724137931,
|
| 194 |
+
"eval_loss": 0.2891761064529419,
|
| 195 |
+
"eval_runtime": 1.1918,
|
| 196 |
+
"eval_samples_per_second": 48.667,
|
| 197 |
+
"eval_steps_per_second": 1.678,
|
| 198 |
+
"step": 72
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 17.88,
|
| 202 |
+
"eval_accuracy": 0.896551724137931,
|
| 203 |
+
"eval_loss": 0.2482166588306427,
|
| 204 |
+
"eval_runtime": 1.1922,
|
| 205 |
+
"eval_samples_per_second": 48.65,
|
| 206 |
+
"eval_steps_per_second": 1.678,
|
| 207 |
+
"step": 76
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 18.82,
|
| 211 |
+
"eval_accuracy": 0.9310344827586207,
|
| 212 |
+
"eval_loss": 0.2363000512123108,
|
| 213 |
+
"eval_runtime": 1.4565,
|
| 214 |
+
"eval_samples_per_second": 39.822,
|
| 215 |
+
"eval_steps_per_second": 1.373,
|
| 216 |
+
"step": 80
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 19.76,
|
| 220 |
+
"learning_rate": 2.6388888888888892e-05,
|
| 221 |
+
"loss": 0.1253,
|
| 222 |
+
"step": 84
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 20.0,
|
| 226 |
+
"eval_accuracy": 0.9482758620689655,
|
| 227 |
+
"eval_loss": 0.16224975883960724,
|
| 228 |
+
"eval_runtime": 1.5442,
|
| 229 |
+
"eval_samples_per_second": 37.56,
|
| 230 |
+
"eval_steps_per_second": 1.295,
|
| 231 |
+
"step": 85
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 20.94,
|
| 235 |
+
"eval_accuracy": 0.9482758620689655,
|
| 236 |
+
"eval_loss": 0.17531706392765045,
|
| 237 |
+
"eval_runtime": 1.1991,
|
| 238 |
+
"eval_samples_per_second": 48.37,
|
| 239 |
+
"eval_steps_per_second": 1.668,
|
| 240 |
+
"step": 89
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 21.88,
|
| 244 |
+
"eval_accuracy": 0.9655172413793104,
|
| 245 |
+
"eval_loss": 0.1592673361301422,
|
| 246 |
+
"eval_runtime": 1.186,
|
| 247 |
+
"eval_samples_per_second": 48.905,
|
| 248 |
+
"eval_steps_per_second": 1.686,
|
| 249 |
+
"step": 93
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 22.59,
|
| 253 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 254 |
+
"loss": 0.087,
|
| 255 |
+
"step": 96
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 22.82,
|
| 259 |
+
"eval_accuracy": 0.9482758620689655,
|
| 260 |
+
"eval_loss": 0.1334347277879715,
|
| 261 |
+
"eval_runtime": 1.1763,
|
| 262 |
+
"eval_samples_per_second": 49.307,
|
| 263 |
+
"eval_steps_per_second": 1.7,
|
| 264 |
+
"step": 97
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 24.0,
|
| 268 |
+
"eval_accuracy": 0.9482758620689655,
|
| 269 |
+
"eval_loss": 0.10879910737276077,
|
| 270 |
+
"eval_runtime": 1.2016,
|
| 271 |
+
"eval_samples_per_second": 48.269,
|
| 272 |
+
"eval_steps_per_second": 1.664,
|
| 273 |
+
"step": 102
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 24.94,
|
| 277 |
+
"eval_accuracy": 0.9482758620689655,
|
| 278 |
+
"eval_loss": 0.11301546543836594,
|
| 279 |
+
"eval_runtime": 1.2067,
|
| 280 |
+
"eval_samples_per_second": 48.064,
|
| 281 |
+
"eval_steps_per_second": 1.657,
|
| 282 |
+
"step": 106
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 25.41,
|
| 286 |
+
"learning_rate": 1.8055555555555555e-05,
|
| 287 |
+
"loss": 0.0856,
|
| 288 |
+
"step": 108
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 25.88,
|
| 292 |
+
"eval_accuracy": 0.9137931034482759,
|
| 293 |
+
"eval_loss": 0.1459188610315323,
|
| 294 |
+
"eval_runtime": 1.1981,
|
| 295 |
+
"eval_samples_per_second": 48.408,
|
| 296 |
+
"eval_steps_per_second": 1.669,
|
| 297 |
+
"step": 110
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 26.82,
|
| 301 |
+
"eval_accuracy": 0.9655172413793104,
|
| 302 |
+
"eval_loss": 0.14451515674591064,
|
| 303 |
+
"eval_runtime": 1.1942,
|
| 304 |
+
"eval_samples_per_second": 48.568,
|
| 305 |
+
"eval_steps_per_second": 1.675,
|
| 306 |
+
"step": 114
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 28.0,
|
| 310 |
+
"eval_accuracy": 0.9655172413793104,
|
| 311 |
+
"eval_loss": 0.1233600303530693,
|
| 312 |
+
"eval_runtime": 1.1858,
|
| 313 |
+
"eval_samples_per_second": 48.911,
|
| 314 |
+
"eval_steps_per_second": 1.687,
|
| 315 |
+
"step": 119
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 28.24,
|
| 319 |
+
"learning_rate": 1.388888888888889e-05,
|
| 320 |
+
"loss": 0.081,
|
| 321 |
+
"step": 120
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 28.94,
|
| 325 |
+
"eval_accuracy": 0.9482758620689655,
|
| 326 |
+
"eval_loss": 0.122359499335289,
|
| 327 |
+
"eval_runtime": 1.1811,
|
| 328 |
+
"eval_samples_per_second": 49.106,
|
| 329 |
+
"eval_steps_per_second": 1.693,
|
| 330 |
+
"step": 123
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 29.88,
|
| 334 |
+
"eval_accuracy": 0.9482758620689655,
|
| 335 |
+
"eval_loss": 0.13033324480056763,
|
| 336 |
+
"eval_runtime": 1.2613,
|
| 337 |
+
"eval_samples_per_second": 45.985,
|
| 338 |
+
"eval_steps_per_second": 1.586,
|
| 339 |
+
"step": 127
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 30.82,
|
| 343 |
+
"eval_accuracy": 0.9482758620689655,
|
| 344 |
+
"eval_loss": 0.13718828558921814,
|
| 345 |
+
"eval_runtime": 1.5095,
|
| 346 |
+
"eval_samples_per_second": 38.423,
|
| 347 |
+
"eval_steps_per_second": 1.325,
|
| 348 |
+
"step": 131
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 31.06,
|
| 352 |
+
"learning_rate": 9.722222222222223e-06,
|
| 353 |
+
"loss": 0.0554,
|
| 354 |
+
"step": 132
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 32.0,
|
| 358 |
+
"eval_accuracy": 0.9482758620689655,
|
| 359 |
+
"eval_loss": 0.14210258424282074,
|
| 360 |
+
"eval_runtime": 1.1798,
|
| 361 |
+
"eval_samples_per_second": 49.159,
|
| 362 |
+
"eval_steps_per_second": 1.695,
|
| 363 |
+
"step": 136
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 32.94,
|
| 367 |
+
"eval_accuracy": 0.9482758620689655,
|
| 368 |
+
"eval_loss": 0.13068123161792755,
|
| 369 |
+
"eval_runtime": 1.1849,
|
| 370 |
+
"eval_samples_per_second": 48.949,
|
| 371 |
+
"eval_steps_per_second": 1.688,
|
| 372 |
+
"step": 140
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 33.88,
|
| 376 |
+
"learning_rate": 5.555555555555556e-06,
|
| 377 |
+
"loss": 0.0783,
|
| 378 |
+
"step": 144
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 33.88,
|
| 382 |
+
"eval_accuracy": 0.9482758620689655,
|
| 383 |
+
"eval_loss": 0.12443631142377853,
|
| 384 |
+
"eval_runtime": 1.19,
|
| 385 |
+
"eval_samples_per_second": 48.739,
|
| 386 |
+
"eval_steps_per_second": 1.681,
|
| 387 |
+
"step": 144
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 34.82,
|
| 391 |
+
"eval_accuracy": 0.9482758620689655,
|
| 392 |
+
"eval_loss": 0.11953343451023102,
|
| 393 |
+
"eval_runtime": 1.2234,
|
| 394 |
+
"eval_samples_per_second": 47.409,
|
| 395 |
+
"eval_steps_per_second": 1.635,
|
| 396 |
+
"step": 148
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 36.0,
|
| 400 |
+
"eval_accuracy": 0.9482758620689655,
|
| 401 |
+
"eval_loss": 0.11707677692174911,
|
| 402 |
+
"eval_runtime": 1.1823,
|
| 403 |
+
"eval_samples_per_second": 49.059,
|
| 404 |
+
"eval_steps_per_second": 1.692,
|
| 405 |
+
"step": 153
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 36.71,
|
| 409 |
+
"learning_rate": 1.388888888888889e-06,
|
| 410 |
+
"loss": 0.0646,
|
| 411 |
+
"step": 156
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 36.94,
|
| 415 |
+
"eval_accuracy": 0.9482758620689655,
|
| 416 |
+
"eval_loss": 0.1165328100323677,
|
| 417 |
+
"eval_runtime": 1.1871,
|
| 418 |
+
"eval_samples_per_second": 48.857,
|
| 419 |
+
"eval_steps_per_second": 1.685,
|
| 420 |
+
"step": 157
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 37.65,
|
| 424 |
+
"eval_accuracy": 0.9482758620689655,
|
| 425 |
+
"eval_loss": 0.11630651354789734,
|
| 426 |
+
"eval_runtime": 1.1868,
|
| 427 |
+
"eval_samples_per_second": 48.872,
|
| 428 |
+
"eval_steps_per_second": 1.685,
|
| 429 |
+
"step": 160
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 37.65,
|
| 433 |
+
"step": 160,
|
| 434 |
+
"total_flos": 1.5182502752352338e+18,
|
| 435 |
+
"train_loss": 0.2719826718792319,
|
| 436 |
+
"train_runtime": 1073.8442,
|
| 437 |
+
"train_samples_per_second": 19.37,
|
| 438 |
+
"train_steps_per_second": 0.149
|
| 439 |
+
}
|
| 440 |
+
],
|
| 441 |
+
"logging_steps": 12,
|
| 442 |
+
"max_steps": 160,
|
| 443 |
+
"num_train_epochs": 40,
|
| 444 |
+
"save_steps": 500,
|
| 445 |
+
"total_flos": 1.5182502752352338e+18,
|
| 446 |
+
"trial_name": null,
|
| 447 |
+
"trial_params": null
|
| 448 |
+
}
|