Commit
•
b6011ca
1
Parent(s):
b705204
End of training
Browse files- README.md +1 -1
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/Dec13_20-40-37_e60439b2b6fd/events.out.tfevents.1702500748.e60439b2b6fd.1775.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +529 -0
README.md
CHANGED
|
@@ -17,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [microsoft/swin-tiny-patch4-window7-224](https://huggingface.co/microsoft/swin-tiny-patch4-window7-224) on an unknown dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss: 0.
|
| 21 |
- Accuracy: 0.9725
|
| 22 |
|
| 23 |
## Model description
|
|
|
|
| 17 |
|
| 18 |
This model is a fine-tuned version of [microsoft/swin-tiny-patch4-window7-224](https://huggingface.co/microsoft/swin-tiny-patch4-window7-224) on an unknown dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 0.0900
|
| 21 |
- Accuracy: 0.9725
|
| 22 |
|
| 23 |
## Model description
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 36.13,
|
| 3 |
+
"eval_accuracy": 0.9724770642201835,
|
| 4 |
+
"eval_loss": 0.09002815186977386,
|
| 5 |
+
"eval_runtime": 0.8164,
|
| 6 |
+
"eval_samples_per_second": 133.521,
|
| 7 |
+
"eval_steps_per_second": 4.9,
|
| 8 |
+
"total_flos": 8.783603110659686e+17,
|
| 9 |
+
"train_loss": 0.42505432409899574,
|
| 10 |
+
"train_runtime": 631.588,
|
| 11 |
+
"train_samples_per_second": 61.939,
|
| 12 |
+
"train_steps_per_second": 0.443
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 36.13,
|
| 3 |
+
"eval_accuracy": 0.9724770642201835,
|
| 4 |
+
"eval_loss": 0.09002815186977386,
|
| 5 |
+
"eval_runtime": 0.8164,
|
| 6 |
+
"eval_samples_per_second": 133.521,
|
| 7 |
+
"eval_steps_per_second": 4.9
|
| 8 |
+
}
|
runs/Dec13_20-40-37_e60439b2b6fd/events.out.tfevents.1702500748.e60439b2b6fd.1775.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b1ab1c22b7984c08d09b884238df38552a555a5d97bfd9362463bffb4e5093c3
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 36.13,
|
| 3 |
+
"total_flos": 8.783603110659686e+17,
|
| 4 |
+
"train_loss": 0.42505432409899574,
|
| 5 |
+
"train_runtime": 631.588,
|
| 6 |
+
"train_samples_per_second": 61.939,
|
| 7 |
+
"train_steps_per_second": 0.443
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,529 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9724770642201835,
|
| 3 |
+
"best_model_checkpoint": "swin-tiny-patch4-window7-224-dmae-va-U/checkpoint-263",
|
| 4 |
+
"epoch": 36.12903225806452,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 280,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.9,
|
| 13 |
+
"eval_accuracy": 0.21100917431192662,
|
| 14 |
+
"eval_loss": 1.4643174409866333,
|
| 15 |
+
"eval_runtime": 1.0706,
|
| 16 |
+
"eval_samples_per_second": 101.808,
|
| 17 |
+
"eval_steps_per_second": 3.736,
|
| 18 |
+
"step": 7
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 1.29,
|
| 22 |
+
"learning_rate": 1.785714285714286e-05,
|
| 23 |
+
"loss": 1.4218,
|
| 24 |
+
"step": 10
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 1.94,
|
| 28 |
+
"eval_accuracy": 0.3302752293577982,
|
| 29 |
+
"eval_loss": 1.4069517850875854,
|
| 30 |
+
"eval_runtime": 0.7367,
|
| 31 |
+
"eval_samples_per_second": 147.949,
|
| 32 |
+
"eval_steps_per_second": 5.429,
|
| 33 |
+
"step": 15
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 2.58,
|
| 37 |
+
"learning_rate": 3.571428571428572e-05,
|
| 38 |
+
"loss": 1.3226,
|
| 39 |
+
"step": 20
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 2.97,
|
| 43 |
+
"eval_accuracy": 0.3486238532110092,
|
| 44 |
+
"eval_loss": 1.3453558683395386,
|
| 45 |
+
"eval_runtime": 0.7271,
|
| 46 |
+
"eval_samples_per_second": 149.91,
|
| 47 |
+
"eval_steps_per_second": 5.501,
|
| 48 |
+
"step": 23
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 3.87,
|
| 52 |
+
"learning_rate": 4.960317460317461e-05,
|
| 53 |
+
"loss": 1.1908,
|
| 54 |
+
"step": 30
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 4.0,
|
| 58 |
+
"eval_accuracy": 0.42201834862385323,
|
| 59 |
+
"eval_loss": 1.1437890529632568,
|
| 60 |
+
"eval_runtime": 0.7292,
|
| 61 |
+
"eval_samples_per_second": 149.473,
|
| 62 |
+
"eval_steps_per_second": 5.485,
|
| 63 |
+
"step": 31
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 4.9,
|
| 67 |
+
"eval_accuracy": 0.5779816513761468,
|
| 68 |
+
"eval_loss": 0.8730040788650513,
|
| 69 |
+
"eval_runtime": 0.8759,
|
| 70 |
+
"eval_samples_per_second": 124.44,
|
| 71 |
+
"eval_steps_per_second": 4.567,
|
| 72 |
+
"step": 38
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 5.16,
|
| 76 |
+
"learning_rate": 4.761904761904762e-05,
|
| 77 |
+
"loss": 0.9441,
|
| 78 |
+
"step": 40
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 5.94,
|
| 82 |
+
"eval_accuracy": 0.6422018348623854,
|
| 83 |
+
"eval_loss": 0.8099932670593262,
|
| 84 |
+
"eval_runtime": 0.7448,
|
| 85 |
+
"eval_samples_per_second": 146.348,
|
| 86 |
+
"eval_steps_per_second": 5.371,
|
| 87 |
+
"step": 46
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 6.45,
|
| 91 |
+
"learning_rate": 4.563492063492064e-05,
|
| 92 |
+
"loss": 0.7185,
|
| 93 |
+
"step": 50
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 6.97,
|
| 97 |
+
"eval_accuracy": 0.7339449541284404,
|
| 98 |
+
"eval_loss": 0.6098942160606384,
|
| 99 |
+
"eval_runtime": 0.7469,
|
| 100 |
+
"eval_samples_per_second": 145.943,
|
| 101 |
+
"eval_steps_per_second": 5.356,
|
| 102 |
+
"step": 54
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 7.74,
|
| 106 |
+
"learning_rate": 4.3650793650793655e-05,
|
| 107 |
+
"loss": 0.6526,
|
| 108 |
+
"step": 60
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 8.0,
|
| 112 |
+
"eval_accuracy": 0.7889908256880734,
|
| 113 |
+
"eval_loss": 0.5095982551574707,
|
| 114 |
+
"eval_runtime": 0.8359,
|
| 115 |
+
"eval_samples_per_second": 130.393,
|
| 116 |
+
"eval_steps_per_second": 4.785,
|
| 117 |
+
"step": 62
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 8.9,
|
| 121 |
+
"eval_accuracy": 0.8165137614678899,
|
| 122 |
+
"eval_loss": 0.4924605190753937,
|
| 123 |
+
"eval_runtime": 0.7693,
|
| 124 |
+
"eval_samples_per_second": 141.693,
|
| 125 |
+
"eval_steps_per_second": 5.2,
|
| 126 |
+
"step": 69
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 9.03,
|
| 130 |
+
"learning_rate": 4.166666666666667e-05,
|
| 131 |
+
"loss": 0.5185,
|
| 132 |
+
"step": 70
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 9.94,
|
| 136 |
+
"eval_accuracy": 0.8348623853211009,
|
| 137 |
+
"eval_loss": 0.39887699484825134,
|
| 138 |
+
"eval_runtime": 0.7535,
|
| 139 |
+
"eval_samples_per_second": 144.667,
|
| 140 |
+
"eval_steps_per_second": 5.309,
|
| 141 |
+
"step": 77
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 10.32,
|
| 145 |
+
"learning_rate": 3.968253968253968e-05,
|
| 146 |
+
"loss": 0.4946,
|
| 147 |
+
"step": 80
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 10.97,
|
| 151 |
+
"eval_accuracy": 0.8807339449541285,
|
| 152 |
+
"eval_loss": 0.3275511860847473,
|
| 153 |
+
"eval_runtime": 0.7403,
|
| 154 |
+
"eval_samples_per_second": 147.247,
|
| 155 |
+
"eval_steps_per_second": 5.404,
|
| 156 |
+
"step": 85
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 11.61,
|
| 160 |
+
"learning_rate": 3.76984126984127e-05,
|
| 161 |
+
"loss": 0.4469,
|
| 162 |
+
"step": 90
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 12.0,
|
| 166 |
+
"eval_accuracy": 0.8899082568807339,
|
| 167 |
+
"eval_loss": 0.30227306485176086,
|
| 168 |
+
"eval_runtime": 0.8725,
|
| 169 |
+
"eval_samples_per_second": 124.931,
|
| 170 |
+
"eval_steps_per_second": 4.585,
|
| 171 |
+
"step": 93
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 12.9,
|
| 175 |
+
"learning_rate": 3.571428571428572e-05,
|
| 176 |
+
"loss": 0.376,
|
| 177 |
+
"step": 100
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 12.9,
|
| 181 |
+
"eval_accuracy": 0.8990825688073395,
|
| 182 |
+
"eval_loss": 0.311217337846756,
|
| 183 |
+
"eval_runtime": 0.7759,
|
| 184 |
+
"eval_samples_per_second": 140.487,
|
| 185 |
+
"eval_steps_per_second": 5.156,
|
| 186 |
+
"step": 100
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 13.94,
|
| 190 |
+
"eval_accuracy": 0.926605504587156,
|
| 191 |
+
"eval_loss": 0.21169805526733398,
|
| 192 |
+
"eval_runtime": 0.7482,
|
| 193 |
+
"eval_samples_per_second": 145.675,
|
| 194 |
+
"eval_steps_per_second": 5.346,
|
| 195 |
+
"step": 108
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 14.19,
|
| 199 |
+
"learning_rate": 3.3730158730158734e-05,
|
| 200 |
+
"loss": 0.3156,
|
| 201 |
+
"step": 110
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 14.97,
|
| 205 |
+
"eval_accuracy": 0.9174311926605505,
|
| 206 |
+
"eval_loss": 0.202369824051857,
|
| 207 |
+
"eval_runtime": 0.797,
|
| 208 |
+
"eval_samples_per_second": 136.771,
|
| 209 |
+
"eval_steps_per_second": 5.019,
|
| 210 |
+
"step": 116
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 15.48,
|
| 214 |
+
"learning_rate": 3.1746031746031745e-05,
|
| 215 |
+
"loss": 0.366,
|
| 216 |
+
"step": 120
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 16.0,
|
| 220 |
+
"eval_accuracy": 0.944954128440367,
|
| 221 |
+
"eval_loss": 0.20653872191905975,
|
| 222 |
+
"eval_runtime": 0.7725,
|
| 223 |
+
"eval_samples_per_second": 141.108,
|
| 224 |
+
"eval_steps_per_second": 5.178,
|
| 225 |
+
"step": 124
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 16.77,
|
| 229 |
+
"learning_rate": 2.9761904761904762e-05,
|
| 230 |
+
"loss": 0.2806,
|
| 231 |
+
"step": 130
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 16.9,
|
| 235 |
+
"eval_accuracy": 0.9174311926605505,
|
| 236 |
+
"eval_loss": 0.19422601163387299,
|
| 237 |
+
"eval_runtime": 0.7563,
|
| 238 |
+
"eval_samples_per_second": 144.13,
|
| 239 |
+
"eval_steps_per_second": 5.289,
|
| 240 |
+
"step": 131
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 17.94,
|
| 244 |
+
"eval_accuracy": 0.9174311926605505,
|
| 245 |
+
"eval_loss": 0.23925282061100006,
|
| 246 |
+
"eval_runtime": 0.7499,
|
| 247 |
+
"eval_samples_per_second": 145.35,
|
| 248 |
+
"eval_steps_per_second": 5.334,
|
| 249 |
+
"step": 139
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 18.06,
|
| 253 |
+
"learning_rate": 2.777777777777778e-05,
|
| 254 |
+
"loss": 0.2695,
|
| 255 |
+
"step": 140
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 18.97,
|
| 259 |
+
"eval_accuracy": 0.9541284403669725,
|
| 260 |
+
"eval_loss": 0.1497838944196701,
|
| 261 |
+
"eval_runtime": 0.862,
|
| 262 |
+
"eval_samples_per_second": 126.456,
|
| 263 |
+
"eval_steps_per_second": 4.641,
|
| 264 |
+
"step": 147
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 19.35,
|
| 268 |
+
"learning_rate": 2.5793650793650796e-05,
|
| 269 |
+
"loss": 0.2357,
|
| 270 |
+
"step": 150
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 20.0,
|
| 274 |
+
"eval_accuracy": 0.9357798165137615,
|
| 275 |
+
"eval_loss": 0.14648140966892242,
|
| 276 |
+
"eval_runtime": 0.7642,
|
| 277 |
+
"eval_samples_per_second": 142.638,
|
| 278 |
+
"eval_steps_per_second": 5.234,
|
| 279 |
+
"step": 155
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 20.65,
|
| 283 |
+
"learning_rate": 2.380952380952381e-05,
|
| 284 |
+
"loss": 0.2345,
|
| 285 |
+
"step": 160
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 20.9,
|
| 289 |
+
"eval_accuracy": 0.963302752293578,
|
| 290 |
+
"eval_loss": 0.15224653482437134,
|
| 291 |
+
"eval_runtime": 0.7573,
|
| 292 |
+
"eval_samples_per_second": 143.934,
|
| 293 |
+
"eval_steps_per_second": 5.282,
|
| 294 |
+
"step": 162
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 21.94,
|
| 298 |
+
"learning_rate": 2.1825396825396827e-05,
|
| 299 |
+
"loss": 0.2157,
|
| 300 |
+
"step": 170
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 21.94,
|
| 304 |
+
"eval_accuracy": 0.944954128440367,
|
| 305 |
+
"eval_loss": 0.14026182889938354,
|
| 306 |
+
"eval_runtime": 0.8934,
|
| 307 |
+
"eval_samples_per_second": 122.007,
|
| 308 |
+
"eval_steps_per_second": 4.477,
|
| 309 |
+
"step": 170
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 22.97,
|
| 313 |
+
"eval_accuracy": 0.9541284403669725,
|
| 314 |
+
"eval_loss": 0.09988939017057419,
|
| 315 |
+
"eval_runtime": 0.7692,
|
| 316 |
+
"eval_samples_per_second": 141.7,
|
| 317 |
+
"eval_steps_per_second": 5.2,
|
| 318 |
+
"step": 178
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 23.23,
|
| 322 |
+
"learning_rate": 1.984126984126984e-05,
|
| 323 |
+
"loss": 0.1894,
|
| 324 |
+
"step": 180
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 24.0,
|
| 328 |
+
"eval_accuracy": 0.963302752293578,
|
| 329 |
+
"eval_loss": 0.14273326098918915,
|
| 330 |
+
"eval_runtime": 0.7562,
|
| 331 |
+
"eval_samples_per_second": 144.134,
|
| 332 |
+
"eval_steps_per_second": 5.289,
|
| 333 |
+
"step": 186
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 24.52,
|
| 337 |
+
"learning_rate": 1.785714285714286e-05,
|
| 338 |
+
"loss": 0.2195,
|
| 339 |
+
"step": 190
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 24.9,
|
| 343 |
+
"eval_accuracy": 0.963302752293578,
|
| 344 |
+
"eval_loss": 0.09490102529525757,
|
| 345 |
+
"eval_runtime": 0.9033,
|
| 346 |
+
"eval_samples_per_second": 120.668,
|
| 347 |
+
"eval_steps_per_second": 4.428,
|
| 348 |
+
"step": 193
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 25.81,
|
| 352 |
+
"learning_rate": 1.5873015873015872e-05,
|
| 353 |
+
"loss": 0.1874,
|
| 354 |
+
"step": 200
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 25.94,
|
| 358 |
+
"eval_accuracy": 0.963302752293578,
|
| 359 |
+
"eval_loss": 0.11523288488388062,
|
| 360 |
+
"eval_runtime": 0.7555,
|
| 361 |
+
"eval_samples_per_second": 144.279,
|
| 362 |
+
"eval_steps_per_second": 5.295,
|
| 363 |
+
"step": 201
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 26.97,
|
| 367 |
+
"eval_accuracy": 0.9541284403669725,
|
| 368 |
+
"eval_loss": 0.122586190700531,
|
| 369 |
+
"eval_runtime": 0.9165,
|
| 370 |
+
"eval_samples_per_second": 118.928,
|
| 371 |
+
"eval_steps_per_second": 4.364,
|
| 372 |
+
"step": 209
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 27.1,
|
| 376 |
+
"learning_rate": 1.388888888888889e-05,
|
| 377 |
+
"loss": 0.1815,
|
| 378 |
+
"step": 210
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 28.0,
|
| 382 |
+
"eval_accuracy": 0.963302752293578,
|
| 383 |
+
"eval_loss": 0.0963682234287262,
|
| 384 |
+
"eval_runtime": 0.7954,
|
| 385 |
+
"eval_samples_per_second": 137.041,
|
| 386 |
+
"eval_steps_per_second": 5.029,
|
| 387 |
+
"step": 217
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 28.39,
|
| 391 |
+
"learning_rate": 1.1904761904761905e-05,
|
| 392 |
+
"loss": 0.1619,
|
| 393 |
+
"step": 220
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 28.9,
|
| 397 |
+
"eval_accuracy": 0.963302752293578,
|
| 398 |
+
"eval_loss": 0.09118019044399261,
|
| 399 |
+
"eval_runtime": 0.761,
|
| 400 |
+
"eval_samples_per_second": 143.228,
|
| 401 |
+
"eval_steps_per_second": 5.256,
|
| 402 |
+
"step": 224
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 29.68,
|
| 406 |
+
"learning_rate": 9.92063492063492e-06,
|
| 407 |
+
"loss": 0.201,
|
| 408 |
+
"step": 230
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 29.94,
|
| 412 |
+
"eval_accuracy": 0.963302752293578,
|
| 413 |
+
"eval_loss": 0.09030430763959885,
|
| 414 |
+
"eval_runtime": 0.7673,
|
| 415 |
+
"eval_samples_per_second": 142.058,
|
| 416 |
+
"eval_steps_per_second": 5.213,
|
| 417 |
+
"step": 232
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 30.97,
|
| 421 |
+
"learning_rate": 7.936507936507936e-06,
|
| 422 |
+
"loss": 0.1659,
|
| 423 |
+
"step": 240
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 30.97,
|
| 427 |
+
"eval_accuracy": 0.963302752293578,
|
| 428 |
+
"eval_loss": 0.0744820311665535,
|
| 429 |
+
"eval_runtime": 0.7579,
|
| 430 |
+
"eval_samples_per_second": 143.812,
|
| 431 |
+
"eval_steps_per_second": 5.278,
|
| 432 |
+
"step": 240
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 32.0,
|
| 436 |
+
"eval_accuracy": 0.963302752293578,
|
| 437 |
+
"eval_loss": 0.07808323204517365,
|
| 438 |
+
"eval_runtime": 0.768,
|
| 439 |
+
"eval_samples_per_second": 141.928,
|
| 440 |
+
"eval_steps_per_second": 5.208,
|
| 441 |
+
"step": 248
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 32.26,
|
| 445 |
+
"learning_rate": 5.9523809523809525e-06,
|
| 446 |
+
"loss": 0.1459,
|
| 447 |
+
"step": 250
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 32.9,
|
| 451 |
+
"eval_accuracy": 0.963302752293578,
|
| 452 |
+
"eval_loss": 0.09302664548158646,
|
| 453 |
+
"eval_runtime": 0.763,
|
| 454 |
+
"eval_samples_per_second": 142.857,
|
| 455 |
+
"eval_steps_per_second": 5.242,
|
| 456 |
+
"step": 255
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 33.55,
|
| 460 |
+
"learning_rate": 3.968253968253968e-06,
|
| 461 |
+
"loss": 0.1459,
|
| 462 |
+
"step": 260
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 33.94,
|
| 466 |
+
"eval_accuracy": 0.9724770642201835,
|
| 467 |
+
"eval_loss": 0.09002815186977386,
|
| 468 |
+
"eval_runtime": 0.7775,
|
| 469 |
+
"eval_samples_per_second": 140.186,
|
| 470 |
+
"eval_steps_per_second": 5.144,
|
| 471 |
+
"step": 263
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 34.84,
|
| 475 |
+
"learning_rate": 1.984126984126984e-06,
|
| 476 |
+
"loss": 0.1487,
|
| 477 |
+
"step": 270
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 34.97,
|
| 481 |
+
"eval_accuracy": 0.9724770642201835,
|
| 482 |
+
"eval_loss": 0.07955732196569443,
|
| 483 |
+
"eval_runtime": 0.8544,
|
| 484 |
+
"eval_samples_per_second": 127.574,
|
| 485 |
+
"eval_steps_per_second": 4.682,
|
| 486 |
+
"step": 271
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 36.0,
|
| 490 |
+
"eval_accuracy": 0.9724770642201835,
|
| 491 |
+
"eval_loss": 0.07835839688777924,
|
| 492 |
+
"eval_runtime": 0.77,
|
| 493 |
+
"eval_samples_per_second": 141.561,
|
| 494 |
+
"eval_steps_per_second": 5.195,
|
| 495 |
+
"step": 279
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 36.13,
|
| 499 |
+
"learning_rate": 0.0,
|
| 500 |
+
"loss": 0.1504,
|
| 501 |
+
"step": 280
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 36.13,
|
| 505 |
+
"eval_accuracy": 0.9724770642201835,
|
| 506 |
+
"eval_loss": 0.07835046201944351,
|
| 507 |
+
"eval_runtime": 0.8307,
|
| 508 |
+
"eval_samples_per_second": 131.22,
|
| 509 |
+
"eval_steps_per_second": 4.815,
|
| 510 |
+
"step": 280
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 36.13,
|
| 514 |
+
"step": 280,
|
| 515 |
+
"total_flos": 8.783603110659686e+17,
|
| 516 |
+
"train_loss": 0.42505432409899574,
|
| 517 |
+
"train_runtime": 631.588,
|
| 518 |
+
"train_samples_per_second": 61.939,
|
| 519 |
+
"train_steps_per_second": 0.443
|
| 520 |
+
}
|
| 521 |
+
],
|
| 522 |
+
"logging_steps": 10,
|
| 523 |
+
"max_steps": 280,
|
| 524 |
+
"num_train_epochs": 40,
|
| 525 |
+
"save_steps": 500,
|
| 526 |
+
"total_flos": 8.783603110659686e+17,
|
| 527 |
+
"trial_name": null,
|
| 528 |
+
"trial_params": null
|
| 529 |
+
}
|