Commit
•
5dd6591
1
Parent(s):
4d54b02
End of training
Browse files- README.md +2 -2
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/Dec13_16-29-59_88adfa788d35/events.out.tfevents.1702486638.88adfa788d35.980.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +499 -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: 0.0534
|
| 21 |
+
- Accuracy: 0.9908
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 36.13,
|
| 3 |
+
"eval_accuracy": 0.9908256880733946,
|
| 4 |
+
"eval_loss": 0.053366996347904205,
|
| 5 |
+
"eval_runtime": 1.5189,
|
| 6 |
+
"eval_samples_per_second": 71.765,
|
| 7 |
+
"eval_steps_per_second": 2.634,
|
| 8 |
+
"total_flos": 2.738306029282984e+18,
|
| 9 |
+
"train_loss": 0.23923614642449786,
|
| 10 |
+
"train_runtime": 1562.4778,
|
| 11 |
+
"train_samples_per_second": 25.037,
|
| 12 |
+
"train_steps_per_second": 0.179
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 36.13,
|
| 3 |
+
"eval_accuracy": 0.9908256880733946,
|
| 4 |
+
"eval_loss": 0.053366996347904205,
|
| 5 |
+
"eval_runtime": 1.5189,
|
| 6 |
+
"eval_samples_per_second": 71.765,
|
| 7 |
+
"eval_steps_per_second": 2.634
|
| 8 |
+
}
|
runs/Dec13_16-29-59_88adfa788d35/events.out.tfevents.1702486638.88adfa788d35.980.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e5008772dff5b92c633de137b4206336389e499f6dbc9ee47b9b90ad8bade301
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 36.13,
|
| 3 |
+
"total_flos": 2.738306029282984e+18,
|
| 4 |
+
"train_loss": 0.23923614642449786,
|
| 5 |
+
"train_runtime": 1562.4778,
|
| 6 |
+
"train_samples_per_second": 25.037,
|
| 7 |
+
"train_steps_per_second": 0.179
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,499 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9908256880733946,
|
| 3 |
+
"best_model_checkpoint": "vit-base-patch16-224-dmae-va-U/checkpoint-217",
|
| 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": 0.25688073394495414,
|
| 14 |
+
"eval_loss": 1.4318687915802002,
|
| 15 |
+
"eval_runtime": 1.4321,
|
| 16 |
+
"eval_samples_per_second": 76.112,
|
| 17 |
+
"eval_steps_per_second": 2.793,
|
| 18 |
+
"step": 7
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 1.55,
|
| 22 |
+
"learning_rate": 2.1428571428571428e-05,
|
| 23 |
+
"loss": 1.3911,
|
| 24 |
+
"step": 12
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 1.94,
|
| 28 |
+
"eval_accuracy": 0.47706422018348627,
|
| 29 |
+
"eval_loss": 1.213340163230896,
|
| 30 |
+
"eval_runtime": 1.4871,
|
| 31 |
+
"eval_samples_per_second": 73.295,
|
| 32 |
+
"eval_steps_per_second": 2.69,
|
| 33 |
+
"step": 15
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 2.97,
|
| 37 |
+
"eval_accuracy": 0.6055045871559633,
|
| 38 |
+
"eval_loss": 0.9487143158912659,
|
| 39 |
+
"eval_runtime": 1.5892,
|
| 40 |
+
"eval_samples_per_second": 68.589,
|
| 41 |
+
"eval_steps_per_second": 2.517,
|
| 42 |
+
"step": 23
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 3.1,
|
| 46 |
+
"learning_rate": 4.2857142857142856e-05,
|
| 47 |
+
"loss": 1.0766,
|
| 48 |
+
"step": 24
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 4.0,
|
| 52 |
+
"eval_accuracy": 0.7155963302752294,
|
| 53 |
+
"eval_loss": 0.6542130708694458,
|
| 54 |
+
"eval_runtime": 1.5217,
|
| 55 |
+
"eval_samples_per_second": 71.63,
|
| 56 |
+
"eval_steps_per_second": 2.629,
|
| 57 |
+
"step": 31
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 4.65,
|
| 61 |
+
"learning_rate": 4.841269841269841e-05,
|
| 62 |
+
"loss": 0.6974,
|
| 63 |
+
"step": 36
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 4.9,
|
| 67 |
+
"eval_accuracy": 0.8715596330275229,
|
| 68 |
+
"eval_loss": 0.4643765389919281,
|
| 69 |
+
"eval_runtime": 1.561,
|
| 70 |
+
"eval_samples_per_second": 69.827,
|
| 71 |
+
"eval_steps_per_second": 2.562,
|
| 72 |
+
"step": 38
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 5.94,
|
| 76 |
+
"eval_accuracy": 0.8715596330275229,
|
| 77 |
+
"eval_loss": 0.39188772439956665,
|
| 78 |
+
"eval_runtime": 1.5611,
|
| 79 |
+
"eval_samples_per_second": 69.823,
|
| 80 |
+
"eval_steps_per_second": 2.562,
|
| 81 |
+
"step": 46
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 6.19,
|
| 85 |
+
"learning_rate": 4.603174603174603e-05,
|
| 86 |
+
"loss": 0.421,
|
| 87 |
+
"step": 48
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 6.97,
|
| 91 |
+
"eval_accuracy": 0.8715596330275229,
|
| 92 |
+
"eval_loss": 0.30943310260772705,
|
| 93 |
+
"eval_runtime": 1.6488,
|
| 94 |
+
"eval_samples_per_second": 66.109,
|
| 95 |
+
"eval_steps_per_second": 2.426,
|
| 96 |
+
"step": 54
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 7.74,
|
| 100 |
+
"learning_rate": 4.3650793650793655e-05,
|
| 101 |
+
"loss": 0.2513,
|
| 102 |
+
"step": 60
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 8.0,
|
| 106 |
+
"eval_accuracy": 0.8990825688073395,
|
| 107 |
+
"eval_loss": 0.2334030568599701,
|
| 108 |
+
"eval_runtime": 1.6748,
|
| 109 |
+
"eval_samples_per_second": 65.082,
|
| 110 |
+
"eval_steps_per_second": 2.388,
|
| 111 |
+
"step": 62
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 8.9,
|
| 115 |
+
"eval_accuracy": 0.9174311926605505,
|
| 116 |
+
"eval_loss": 0.19148482382297516,
|
| 117 |
+
"eval_runtime": 1.68,
|
| 118 |
+
"eval_samples_per_second": 64.88,
|
| 119 |
+
"eval_steps_per_second": 2.381,
|
| 120 |
+
"step": 69
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 9.29,
|
| 124 |
+
"learning_rate": 4.126984126984127e-05,
|
| 125 |
+
"loss": 0.1931,
|
| 126 |
+
"step": 72
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 9.94,
|
| 130 |
+
"eval_accuracy": 0.8807339449541285,
|
| 131 |
+
"eval_loss": 0.24314457178115845,
|
| 132 |
+
"eval_runtime": 1.7464,
|
| 133 |
+
"eval_samples_per_second": 62.412,
|
| 134 |
+
"eval_steps_per_second": 2.29,
|
| 135 |
+
"step": 77
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 10.84,
|
| 139 |
+
"learning_rate": 3.888888888888889e-05,
|
| 140 |
+
"loss": 0.1757,
|
| 141 |
+
"step": 84
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 10.97,
|
| 145 |
+
"eval_accuracy": 0.944954128440367,
|
| 146 |
+
"eval_loss": 0.16078265011310577,
|
| 147 |
+
"eval_runtime": 1.6471,
|
| 148 |
+
"eval_samples_per_second": 66.177,
|
| 149 |
+
"eval_steps_per_second": 2.428,
|
| 150 |
+
"step": 85
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 12.0,
|
| 154 |
+
"eval_accuracy": 0.926605504587156,
|
| 155 |
+
"eval_loss": 0.1423795074224472,
|
| 156 |
+
"eval_runtime": 1.7337,
|
| 157 |
+
"eval_samples_per_second": 62.87,
|
| 158 |
+
"eval_steps_per_second": 2.307,
|
| 159 |
+
"step": 93
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 12.39,
|
| 163 |
+
"learning_rate": 3.650793650793651e-05,
|
| 164 |
+
"loss": 0.1442,
|
| 165 |
+
"step": 96
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 12.9,
|
| 169 |
+
"eval_accuracy": 0.944954128440367,
|
| 170 |
+
"eval_loss": 0.1280096173286438,
|
| 171 |
+
"eval_runtime": 1.6549,
|
| 172 |
+
"eval_samples_per_second": 65.864,
|
| 173 |
+
"eval_steps_per_second": 2.417,
|
| 174 |
+
"step": 100
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 13.94,
|
| 178 |
+
"learning_rate": 3.412698412698413e-05,
|
| 179 |
+
"loss": 0.1085,
|
| 180 |
+
"step": 108
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 13.94,
|
| 184 |
+
"eval_accuracy": 0.9541284403669725,
|
| 185 |
+
"eval_loss": 0.10545489937067032,
|
| 186 |
+
"eval_runtime": 1.6964,
|
| 187 |
+
"eval_samples_per_second": 64.255,
|
| 188 |
+
"eval_steps_per_second": 2.358,
|
| 189 |
+
"step": 108
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 14.97,
|
| 193 |
+
"eval_accuracy": 0.9541284403669725,
|
| 194 |
+
"eval_loss": 0.1080455482006073,
|
| 195 |
+
"eval_runtime": 1.5479,
|
| 196 |
+
"eval_samples_per_second": 70.417,
|
| 197 |
+
"eval_steps_per_second": 2.584,
|
| 198 |
+
"step": 116
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 15.48,
|
| 202 |
+
"learning_rate": 3.1746031746031745e-05,
|
| 203 |
+
"loss": 0.1056,
|
| 204 |
+
"step": 120
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 16.0,
|
| 208 |
+
"eval_accuracy": 0.963302752293578,
|
| 209 |
+
"eval_loss": 0.09967872500419617,
|
| 210 |
+
"eval_runtime": 1.7093,
|
| 211 |
+
"eval_samples_per_second": 63.77,
|
| 212 |
+
"eval_steps_per_second": 2.34,
|
| 213 |
+
"step": 124
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 16.9,
|
| 217 |
+
"eval_accuracy": 0.963302752293578,
|
| 218 |
+
"eval_loss": 0.11852575093507767,
|
| 219 |
+
"eval_runtime": 1.5272,
|
| 220 |
+
"eval_samples_per_second": 71.372,
|
| 221 |
+
"eval_steps_per_second": 2.619,
|
| 222 |
+
"step": 131
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 17.03,
|
| 226 |
+
"learning_rate": 2.9365079365079366e-05,
|
| 227 |
+
"loss": 0.0926,
|
| 228 |
+
"step": 132
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 17.94,
|
| 232 |
+
"eval_accuracy": 0.963302752293578,
|
| 233 |
+
"eval_loss": 0.07731892913579941,
|
| 234 |
+
"eval_runtime": 1.5604,
|
| 235 |
+
"eval_samples_per_second": 69.856,
|
| 236 |
+
"eval_steps_per_second": 2.564,
|
| 237 |
+
"step": 139
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 18.58,
|
| 241 |
+
"learning_rate": 2.6984126984126984e-05,
|
| 242 |
+
"loss": 0.103,
|
| 243 |
+
"step": 144
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 18.97,
|
| 247 |
+
"eval_accuracy": 0.963302752293578,
|
| 248 |
+
"eval_loss": 0.12785662710666656,
|
| 249 |
+
"eval_runtime": 1.527,
|
| 250 |
+
"eval_samples_per_second": 71.384,
|
| 251 |
+
"eval_steps_per_second": 2.62,
|
| 252 |
+
"step": 147
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 20.0,
|
| 256 |
+
"eval_accuracy": 0.963302752293578,
|
| 257 |
+
"eval_loss": 0.10433010756969452,
|
| 258 |
+
"eval_runtime": 1.5327,
|
| 259 |
+
"eval_samples_per_second": 71.115,
|
| 260 |
+
"eval_steps_per_second": 2.61,
|
| 261 |
+
"step": 155
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 20.13,
|
| 265 |
+
"learning_rate": 2.4603174603174602e-05,
|
| 266 |
+
"loss": 0.0938,
|
| 267 |
+
"step": 156
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 20.9,
|
| 271 |
+
"eval_accuracy": 0.981651376146789,
|
| 272 |
+
"eval_loss": 0.08241702616214752,
|
| 273 |
+
"eval_runtime": 1.5631,
|
| 274 |
+
"eval_samples_per_second": 69.735,
|
| 275 |
+
"eval_steps_per_second": 2.559,
|
| 276 |
+
"step": 162
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 21.68,
|
| 280 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 281 |
+
"loss": 0.0891,
|
| 282 |
+
"step": 168
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 21.94,
|
| 286 |
+
"eval_accuracy": 0.9541284403669725,
|
| 287 |
+
"eval_loss": 0.14486828446388245,
|
| 288 |
+
"eval_runtime": 1.5236,
|
| 289 |
+
"eval_samples_per_second": 71.54,
|
| 290 |
+
"eval_steps_per_second": 2.625,
|
| 291 |
+
"step": 170
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 22.97,
|
| 295 |
+
"eval_accuracy": 0.963302752293578,
|
| 296 |
+
"eval_loss": 0.13658782839775085,
|
| 297 |
+
"eval_runtime": 1.7185,
|
| 298 |
+
"eval_samples_per_second": 63.429,
|
| 299 |
+
"eval_steps_per_second": 2.328,
|
| 300 |
+
"step": 178
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 23.23,
|
| 304 |
+
"learning_rate": 1.984126984126984e-05,
|
| 305 |
+
"loss": 0.0754,
|
| 306 |
+
"step": 180
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 24.0,
|
| 310 |
+
"eval_accuracy": 0.9357798165137615,
|
| 311 |
+
"eval_loss": 0.11480199545621872,
|
| 312 |
+
"eval_runtime": 1.6552,
|
| 313 |
+
"eval_samples_per_second": 65.853,
|
| 314 |
+
"eval_steps_per_second": 2.417,
|
| 315 |
+
"step": 186
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 24.77,
|
| 319 |
+
"learning_rate": 1.746031746031746e-05,
|
| 320 |
+
"loss": 0.0882,
|
| 321 |
+
"step": 192
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 24.9,
|
| 325 |
+
"eval_accuracy": 0.9357798165137615,
|
| 326 |
+
"eval_loss": 0.19915136694908142,
|
| 327 |
+
"eval_runtime": 1.6537,
|
| 328 |
+
"eval_samples_per_second": 65.913,
|
| 329 |
+
"eval_steps_per_second": 2.419,
|
| 330 |
+
"step": 193
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 25.94,
|
| 334 |
+
"eval_accuracy": 0.981651376146789,
|
| 335 |
+
"eval_loss": 0.07425253838300705,
|
| 336 |
+
"eval_runtime": 1.4966,
|
| 337 |
+
"eval_samples_per_second": 72.83,
|
| 338 |
+
"eval_steps_per_second": 2.673,
|
| 339 |
+
"step": 201
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 26.32,
|
| 343 |
+
"learning_rate": 1.5079365079365079e-05,
|
| 344 |
+
"loss": 0.078,
|
| 345 |
+
"step": 204
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 26.97,
|
| 349 |
+
"eval_accuracy": 0.9724770642201835,
|
| 350 |
+
"eval_loss": 0.06684111058712006,
|
| 351 |
+
"eval_runtime": 1.5659,
|
| 352 |
+
"eval_samples_per_second": 69.61,
|
| 353 |
+
"eval_steps_per_second": 2.554,
|
| 354 |
+
"step": 209
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 27.87,
|
| 358 |
+
"learning_rate": 1.2698412698412699e-05,
|
| 359 |
+
"loss": 0.0666,
|
| 360 |
+
"step": 216
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 28.0,
|
| 364 |
+
"eval_accuracy": 0.9908256880733946,
|
| 365 |
+
"eval_loss": 0.053366996347904205,
|
| 366 |
+
"eval_runtime": 1.5382,
|
| 367 |
+
"eval_samples_per_second": 70.862,
|
| 368 |
+
"eval_steps_per_second": 2.6,
|
| 369 |
+
"step": 217
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 28.9,
|
| 373 |
+
"eval_accuracy": 0.9908256880733946,
|
| 374 |
+
"eval_loss": 0.049850545823574066,
|
| 375 |
+
"eval_runtime": 1.5346,
|
| 376 |
+
"eval_samples_per_second": 71.027,
|
| 377 |
+
"eval_steps_per_second": 2.607,
|
| 378 |
+
"step": 224
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 29.42,
|
| 382 |
+
"learning_rate": 1.0317460317460318e-05,
|
| 383 |
+
"loss": 0.0514,
|
| 384 |
+
"step": 228
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 29.94,
|
| 388 |
+
"eval_accuracy": 0.9724770642201835,
|
| 389 |
+
"eval_loss": 0.043298669159412384,
|
| 390 |
+
"eval_runtime": 1.5329,
|
| 391 |
+
"eval_samples_per_second": 71.108,
|
| 392 |
+
"eval_steps_per_second": 2.609,
|
| 393 |
+
"step": 232
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 30.97,
|
| 397 |
+
"learning_rate": 7.936507936507936e-06,
|
| 398 |
+
"loss": 0.062,
|
| 399 |
+
"step": 240
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 30.97,
|
| 403 |
+
"eval_accuracy": 0.963302752293578,
|
| 404 |
+
"eval_loss": 0.08402539044618607,
|
| 405 |
+
"eval_runtime": 1.7473,
|
| 406 |
+
"eval_samples_per_second": 62.382,
|
| 407 |
+
"eval_steps_per_second": 2.289,
|
| 408 |
+
"step": 240
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 32.0,
|
| 412 |
+
"eval_accuracy": 0.9724770642201835,
|
| 413 |
+
"eval_loss": 0.05133233219385147,
|
| 414 |
+
"eval_runtime": 1.6441,
|
| 415 |
+
"eval_samples_per_second": 66.297,
|
| 416 |
+
"eval_steps_per_second": 2.433,
|
| 417 |
+
"step": 248
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 32.52,
|
| 421 |
+
"learning_rate": 5.555555555555556e-06,
|
| 422 |
+
"loss": 0.0712,
|
| 423 |
+
"step": 252
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 32.9,
|
| 427 |
+
"eval_accuracy": 0.981651376146789,
|
| 428 |
+
"eval_loss": 0.04824218526482582,
|
| 429 |
+
"eval_runtime": 1.6796,
|
| 430 |
+
"eval_samples_per_second": 64.896,
|
| 431 |
+
"eval_steps_per_second": 2.382,
|
| 432 |
+
"step": 255
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 33.94,
|
| 436 |
+
"eval_accuracy": 0.981651376146789,
|
| 437 |
+
"eval_loss": 0.055325187742710114,
|
| 438 |
+
"eval_runtime": 1.5821,
|
| 439 |
+
"eval_samples_per_second": 68.894,
|
| 440 |
+
"eval_steps_per_second": 2.528,
|
| 441 |
+
"step": 263
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 34.06,
|
| 445 |
+
"learning_rate": 3.1746031746031746e-06,
|
| 446 |
+
"loss": 0.0703,
|
| 447 |
+
"step": 264
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 34.97,
|
| 451 |
+
"eval_accuracy": 0.9724770642201835,
|
| 452 |
+
"eval_loss": 0.060162752866744995,
|
| 453 |
+
"eval_runtime": 1.5164,
|
| 454 |
+
"eval_samples_per_second": 71.882,
|
| 455 |
+
"eval_steps_per_second": 2.638,
|
| 456 |
+
"step": 271
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 35.61,
|
| 460 |
+
"learning_rate": 7.936507936507937e-07,
|
| 461 |
+
"loss": 0.0553,
|
| 462 |
+
"step": 276
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 36.0,
|
| 466 |
+
"eval_accuracy": 0.9724770642201835,
|
| 467 |
+
"eval_loss": 0.05953967571258545,
|
| 468 |
+
"eval_runtime": 1.5574,
|
| 469 |
+
"eval_samples_per_second": 69.99,
|
| 470 |
+
"eval_steps_per_second": 2.568,
|
| 471 |
+
"step": 279
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 36.13,
|
| 475 |
+
"eval_accuracy": 0.9724770642201835,
|
| 476 |
+
"eval_loss": 0.05953451991081238,
|
| 477 |
+
"eval_runtime": 1.6854,
|
| 478 |
+
"eval_samples_per_second": 64.671,
|
| 479 |
+
"eval_steps_per_second": 2.373,
|
| 480 |
+
"step": 280
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 36.13,
|
| 484 |
+
"step": 280,
|
| 485 |
+
"total_flos": 2.738306029282984e+18,
|
| 486 |
+
"train_loss": 0.23923614642449786,
|
| 487 |
+
"train_runtime": 1562.4778,
|
| 488 |
+
"train_samples_per_second": 25.037,
|
| 489 |
+
"train_steps_per_second": 0.179
|
| 490 |
+
}
|
| 491 |
+
],
|
| 492 |
+
"logging_steps": 12,
|
| 493 |
+
"max_steps": 280,
|
| 494 |
+
"num_train_epochs": 40,
|
| 495 |
+
"save_steps": 500,
|
| 496 |
+
"total_flos": 2.738306029282984e+18,
|
| 497 |
+
"trial_name": null,
|
| 498 |
+
"trial_params": null
|
| 499 |
+
}
|