Commit
•
cf7506a
1
Parent(s):
0605b70
End of training
Browse files- README.md +1 -1
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/Feb01_15-59-58_56aa26a8524f/events.out.tfevents.1706803463.56aa26a8524f.1432.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +529 -0
README.md
CHANGED
|
@@ -16,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [MBZUAI/swiftformer-xs](https://huggingface.co/MBZUAI/swiftformer-xs) on an unknown dataset.
|
| 18 |
It achieves the following results on the evaluation set:
|
| 19 |
-
- Loss: 0.
|
| 20 |
- Accuracy: 0.7523
|
| 21 |
|
| 22 |
## Model description
|
|
|
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [MBZUAI/swiftformer-xs](https://huggingface.co/MBZUAI/swiftformer-xs) on an unknown dataset.
|
| 18 |
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 0.7007
|
| 20 |
- Accuracy: 0.7523
|
| 21 |
|
| 22 |
## Model description
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 36.13,
|
| 3 |
+
"eval_accuracy": 0.7522935779816514,
|
| 4 |
+
"eval_loss": 0.7007379531860352,
|
| 5 |
+
"eval_runtime": 0.4794,
|
| 6 |
+
"eval_samples_per_second": 227.386,
|
| 7 |
+
"eval_steps_per_second": 8.344,
|
| 8 |
+
"total_flos": 9.686412043576934e+16,
|
| 9 |
+
"train_loss": 0.878055340903146,
|
| 10 |
+
"train_runtime": 251.6613,
|
| 11 |
+
"train_samples_per_second": 155.447,
|
| 12 |
+
"train_steps_per_second": 1.113
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 36.13,
|
| 3 |
+
"eval_accuracy": 0.7522935779816514,
|
| 4 |
+
"eval_loss": 0.7007379531860352,
|
| 5 |
+
"eval_runtime": 0.4794,
|
| 6 |
+
"eval_samples_per_second": 227.386,
|
| 7 |
+
"eval_steps_per_second": 8.344
|
| 8 |
+
}
|
runs/Feb01_15-59-58_56aa26a8524f/events.out.tfevents.1706803463.56aa26a8524f.1432.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9791ea8015279dfc50ca507b8a6956f014088eb63f1d09d21ca529bd782e201e
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 36.13,
|
| 3 |
+
"total_flos": 9.686412043576934e+16,
|
| 4 |
+
"train_loss": 0.878055340903146,
|
| 5 |
+
"train_runtime": 251.6613,
|
| 6 |
+
"train_samples_per_second": 155.447,
|
| 7 |
+
"train_steps_per_second": 1.113
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,529 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.7522935779816514,
|
| 3 |
+
"best_model_checkpoint": "swiftformer-xs-dmae-va-U-40/checkpoint-224",
|
| 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": 0.29357798165137616,
|
| 14 |
+
"eval_loss": 1.3578405380249023,
|
| 15 |
+
"eval_runtime": 0.4941,
|
| 16 |
+
"eval_samples_per_second": 220.605,
|
| 17 |
+
"eval_steps_per_second": 8.096,
|
| 18 |
+
"step": 7
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 1.29,
|
| 22 |
+
"learning_rate": 1.785714285714286e-05,
|
| 23 |
+
"loss": 1.3702,
|
| 24 |
+
"step": 10
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 1.94,
|
| 28 |
+
"eval_accuracy": 0.29357798165137616,
|
| 29 |
+
"eval_loss": 1.3703261613845825,
|
| 30 |
+
"eval_runtime": 0.5833,
|
| 31 |
+
"eval_samples_per_second": 186.883,
|
| 32 |
+
"eval_steps_per_second": 6.858,
|
| 33 |
+
"step": 15
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 2.58,
|
| 37 |
+
"learning_rate": 3.571428571428572e-05,
|
| 38 |
+
"loss": 1.3497,
|
| 39 |
+
"step": 20
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 2.97,
|
| 43 |
+
"eval_accuracy": 0.3394495412844037,
|
| 44 |
+
"eval_loss": 1.3361120223999023,
|
| 45 |
+
"eval_runtime": 0.4226,
|
| 46 |
+
"eval_samples_per_second": 257.899,
|
| 47 |
+
"eval_steps_per_second": 9.464,
|
| 48 |
+
"step": 23
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 3.87,
|
| 52 |
+
"learning_rate": 4.960317460317461e-05,
|
| 53 |
+
"loss": 1.3004,
|
| 54 |
+
"step": 30
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 4.0,
|
| 58 |
+
"eval_accuracy": 0.3669724770642202,
|
| 59 |
+
"eval_loss": 1.2852015495300293,
|
| 60 |
+
"eval_runtime": 0.4154,
|
| 61 |
+
"eval_samples_per_second": 262.426,
|
| 62 |
+
"eval_steps_per_second": 9.63,
|
| 63 |
+
"step": 31
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 4.9,
|
| 67 |
+
"eval_accuracy": 0.43119266055045874,
|
| 68 |
+
"eval_loss": 1.231681227684021,
|
| 69 |
+
"eval_runtime": 0.4157,
|
| 70 |
+
"eval_samples_per_second": 262.204,
|
| 71 |
+
"eval_steps_per_second": 9.622,
|
| 72 |
+
"step": 38
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 5.16,
|
| 76 |
+
"learning_rate": 4.761904761904762e-05,
|
| 77 |
+
"loss": 1.2248,
|
| 78 |
+
"step": 40
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 5.94,
|
| 82 |
+
"eval_accuracy": 0.45871559633027525,
|
| 83 |
+
"eval_loss": 1.1785550117492676,
|
| 84 |
+
"eval_runtime": 0.4211,
|
| 85 |
+
"eval_samples_per_second": 258.867,
|
| 86 |
+
"eval_steps_per_second": 9.5,
|
| 87 |
+
"step": 46
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 6.45,
|
| 91 |
+
"learning_rate": 4.563492063492064e-05,
|
| 92 |
+
"loss": 1.1485,
|
| 93 |
+
"step": 50
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 6.97,
|
| 97 |
+
"eval_accuracy": 0.5045871559633027,
|
| 98 |
+
"eval_loss": 1.123984694480896,
|
| 99 |
+
"eval_runtime": 0.4143,
|
| 100 |
+
"eval_samples_per_second": 263.124,
|
| 101 |
+
"eval_steps_per_second": 9.656,
|
| 102 |
+
"step": 54
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 7.74,
|
| 106 |
+
"learning_rate": 4.3650793650793655e-05,
|
| 107 |
+
"loss": 1.0759,
|
| 108 |
+
"step": 60
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 8.0,
|
| 112 |
+
"eval_accuracy": 0.5504587155963303,
|
| 113 |
+
"eval_loss": 1.0727368593215942,
|
| 114 |
+
"eval_runtime": 0.4109,
|
| 115 |
+
"eval_samples_per_second": 265.292,
|
| 116 |
+
"eval_steps_per_second": 9.735,
|
| 117 |
+
"step": 62
|
| 118 |
+
},
|
| 119 |
+
{
|
| 120 |
+
"epoch": 8.9,
|
| 121 |
+
"eval_accuracy": 0.5596330275229358,
|
| 122 |
+
"eval_loss": 1.0403714179992676,
|
| 123 |
+
"eval_runtime": 0.5431,
|
| 124 |
+
"eval_samples_per_second": 200.69,
|
| 125 |
+
"eval_steps_per_second": 7.365,
|
| 126 |
+
"step": 69
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 9.03,
|
| 130 |
+
"learning_rate": 4.166666666666667e-05,
|
| 131 |
+
"loss": 1.0244,
|
| 132 |
+
"step": 70
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 9.94,
|
| 136 |
+
"eval_accuracy": 0.6238532110091743,
|
| 137 |
+
"eval_loss": 0.974201500415802,
|
| 138 |
+
"eval_runtime": 0.4104,
|
| 139 |
+
"eval_samples_per_second": 265.623,
|
| 140 |
+
"eval_steps_per_second": 9.748,
|
| 141 |
+
"step": 77
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 10.32,
|
| 145 |
+
"learning_rate": 3.968253968253968e-05,
|
| 146 |
+
"loss": 0.9782,
|
| 147 |
+
"step": 80
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 10.97,
|
| 151 |
+
"eval_accuracy": 0.6422018348623854,
|
| 152 |
+
"eval_loss": 0.9374117851257324,
|
| 153 |
+
"eval_runtime": 0.6713,
|
| 154 |
+
"eval_samples_per_second": 162.361,
|
| 155 |
+
"eval_steps_per_second": 5.958,
|
| 156 |
+
"step": 85
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 11.61,
|
| 160 |
+
"learning_rate": 3.76984126984127e-05,
|
| 161 |
+
"loss": 0.9359,
|
| 162 |
+
"step": 90
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 12.0,
|
| 166 |
+
"eval_accuracy": 0.6788990825688074,
|
| 167 |
+
"eval_loss": 0.9196614027023315,
|
| 168 |
+
"eval_runtime": 0.4175,
|
| 169 |
+
"eval_samples_per_second": 261.061,
|
| 170 |
+
"eval_steps_per_second": 9.58,
|
| 171 |
+
"step": 93
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 12.9,
|
| 175 |
+
"learning_rate": 3.571428571428572e-05,
|
| 176 |
+
"loss": 0.9051,
|
| 177 |
+
"step": 100
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 12.9,
|
| 181 |
+
"eval_accuracy": 0.6880733944954128,
|
| 182 |
+
"eval_loss": 0.8753331899642944,
|
| 183 |
+
"eval_runtime": 0.41,
|
| 184 |
+
"eval_samples_per_second": 265.867,
|
| 185 |
+
"eval_steps_per_second": 9.757,
|
| 186 |
+
"step": 100
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 13.94,
|
| 190 |
+
"eval_accuracy": 0.6972477064220184,
|
| 191 |
+
"eval_loss": 0.8679403066635132,
|
| 192 |
+
"eval_runtime": 0.4077,
|
| 193 |
+
"eval_samples_per_second": 267.322,
|
| 194 |
+
"eval_steps_per_second": 9.81,
|
| 195 |
+
"step": 108
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 14.19,
|
| 199 |
+
"learning_rate": 3.3730158730158734e-05,
|
| 200 |
+
"loss": 0.8652,
|
| 201 |
+
"step": 110
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 14.97,
|
| 205 |
+
"eval_accuracy": 0.7155963302752294,
|
| 206 |
+
"eval_loss": 0.8316473364830017,
|
| 207 |
+
"eval_runtime": 0.4256,
|
| 208 |
+
"eval_samples_per_second": 256.117,
|
| 209 |
+
"eval_steps_per_second": 9.399,
|
| 210 |
+
"step": 116
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 15.48,
|
| 214 |
+
"learning_rate": 3.1746031746031745e-05,
|
| 215 |
+
"loss": 0.8336,
|
| 216 |
+
"step": 120
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 16.0,
|
| 220 |
+
"eval_accuracy": 0.6972477064220184,
|
| 221 |
+
"eval_loss": 0.8222222924232483,
|
| 222 |
+
"eval_runtime": 0.4301,
|
| 223 |
+
"eval_samples_per_second": 253.434,
|
| 224 |
+
"eval_steps_per_second": 9.3,
|
| 225 |
+
"step": 124
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 16.77,
|
| 229 |
+
"learning_rate": 2.9761904761904762e-05,
|
| 230 |
+
"loss": 0.8177,
|
| 231 |
+
"step": 130
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 16.9,
|
| 235 |
+
"eval_accuracy": 0.6972477064220184,
|
| 236 |
+
"eval_loss": 0.8177938461303711,
|
| 237 |
+
"eval_runtime": 0.4098,
|
| 238 |
+
"eval_samples_per_second": 266.011,
|
| 239 |
+
"eval_steps_per_second": 9.762,
|
| 240 |
+
"step": 131
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 17.94,
|
| 244 |
+
"eval_accuracy": 0.7339449541284404,
|
| 245 |
+
"eval_loss": 0.7817714810371399,
|
| 246 |
+
"eval_runtime": 0.5687,
|
| 247 |
+
"eval_samples_per_second": 191.682,
|
| 248 |
+
"eval_steps_per_second": 7.034,
|
| 249 |
+
"step": 139
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 18.06,
|
| 253 |
+
"learning_rate": 2.777777777777778e-05,
|
| 254 |
+
"loss": 0.8077,
|
| 255 |
+
"step": 140
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 18.97,
|
| 259 |
+
"eval_accuracy": 0.7339449541284404,
|
| 260 |
+
"eval_loss": 0.7627159357070923,
|
| 261 |
+
"eval_runtime": 0.4134,
|
| 262 |
+
"eval_samples_per_second": 263.676,
|
| 263 |
+
"eval_steps_per_second": 9.676,
|
| 264 |
+
"step": 147
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 19.35,
|
| 268 |
+
"learning_rate": 2.5793650793650796e-05,
|
| 269 |
+
"loss": 0.7796,
|
| 270 |
+
"step": 150
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 20.0,
|
| 274 |
+
"eval_accuracy": 0.7339449541284404,
|
| 275 |
+
"eval_loss": 0.7478492259979248,
|
| 276 |
+
"eval_runtime": 0.4144,
|
| 277 |
+
"eval_samples_per_second": 263.034,
|
| 278 |
+
"eval_steps_per_second": 9.653,
|
| 279 |
+
"step": 155
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 20.65,
|
| 283 |
+
"learning_rate": 2.380952380952381e-05,
|
| 284 |
+
"loss": 0.7673,
|
| 285 |
+
"step": 160
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 20.9,
|
| 289 |
+
"eval_accuracy": 0.7431192660550459,
|
| 290 |
+
"eval_loss": 0.7414626479148865,
|
| 291 |
+
"eval_runtime": 0.408,
|
| 292 |
+
"eval_samples_per_second": 267.182,
|
| 293 |
+
"eval_steps_per_second": 9.805,
|
| 294 |
+
"step": 162
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 21.94,
|
| 298 |
+
"learning_rate": 2.1825396825396827e-05,
|
| 299 |
+
"loss": 0.7445,
|
| 300 |
+
"step": 170
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 21.94,
|
| 304 |
+
"eval_accuracy": 0.7155963302752294,
|
| 305 |
+
"eval_loss": 0.7413556575775146,
|
| 306 |
+
"eval_runtime": 0.4043,
|
| 307 |
+
"eval_samples_per_second": 269.576,
|
| 308 |
+
"eval_steps_per_second": 9.893,
|
| 309 |
+
"step": 170
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 22.97,
|
| 313 |
+
"eval_accuracy": 0.7155963302752294,
|
| 314 |
+
"eval_loss": 0.7375438809394836,
|
| 315 |
+
"eval_runtime": 0.4062,
|
| 316 |
+
"eval_samples_per_second": 268.371,
|
| 317 |
+
"eval_steps_per_second": 9.848,
|
| 318 |
+
"step": 178
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 23.23,
|
| 322 |
+
"learning_rate": 1.984126984126984e-05,
|
| 323 |
+
"loss": 0.7413,
|
| 324 |
+
"step": 180
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 24.0,
|
| 328 |
+
"eval_accuracy": 0.7155963302752294,
|
| 329 |
+
"eval_loss": 0.7353999018669128,
|
| 330 |
+
"eval_runtime": 0.4171,
|
| 331 |
+
"eval_samples_per_second": 261.311,
|
| 332 |
+
"eval_steps_per_second": 9.589,
|
| 333 |
+
"step": 186
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 24.52,
|
| 337 |
+
"learning_rate": 1.785714285714286e-05,
|
| 338 |
+
"loss": 0.739,
|
| 339 |
+
"step": 190
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 24.9,
|
| 343 |
+
"eval_accuracy": 0.7431192660550459,
|
| 344 |
+
"eval_loss": 0.71100252866745,
|
| 345 |
+
"eval_runtime": 0.4123,
|
| 346 |
+
"eval_samples_per_second": 264.367,
|
| 347 |
+
"eval_steps_per_second": 9.702,
|
| 348 |
+
"step": 193
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 25.81,
|
| 352 |
+
"learning_rate": 1.5873015873015872e-05,
|
| 353 |
+
"loss": 0.6992,
|
| 354 |
+
"step": 200
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 25.94,
|
| 358 |
+
"eval_accuracy": 0.7339449541284404,
|
| 359 |
+
"eval_loss": 0.7120506763458252,
|
| 360 |
+
"eval_runtime": 0.4281,
|
| 361 |
+
"eval_samples_per_second": 254.6,
|
| 362 |
+
"eval_steps_per_second": 9.343,
|
| 363 |
+
"step": 201
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 26.97,
|
| 367 |
+
"eval_accuracy": 0.7431192660550459,
|
| 368 |
+
"eval_loss": 0.7044178247451782,
|
| 369 |
+
"eval_runtime": 0.5666,
|
| 370 |
+
"eval_samples_per_second": 192.368,
|
| 371 |
+
"eval_steps_per_second": 7.059,
|
| 372 |
+
"step": 209
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 27.1,
|
| 376 |
+
"learning_rate": 1.388888888888889e-05,
|
| 377 |
+
"loss": 0.7111,
|
| 378 |
+
"step": 210
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 28.0,
|
| 382 |
+
"eval_accuracy": 0.7339449541284404,
|
| 383 |
+
"eval_loss": 0.6947001218795776,
|
| 384 |
+
"eval_runtime": 0.4105,
|
| 385 |
+
"eval_samples_per_second": 265.532,
|
| 386 |
+
"eval_steps_per_second": 9.744,
|
| 387 |
+
"step": 217
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 28.39,
|
| 391 |
+
"learning_rate": 1.1904761904761905e-05,
|
| 392 |
+
"loss": 0.7013,
|
| 393 |
+
"step": 220
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 28.9,
|
| 397 |
+
"eval_accuracy": 0.7522935779816514,
|
| 398 |
+
"eval_loss": 0.7007379531860352,
|
| 399 |
+
"eval_runtime": 0.4106,
|
| 400 |
+
"eval_samples_per_second": 265.441,
|
| 401 |
+
"eval_steps_per_second": 9.741,
|
| 402 |
+
"step": 224
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 29.68,
|
| 406 |
+
"learning_rate": 9.92063492063492e-06,
|
| 407 |
+
"loss": 0.712,
|
| 408 |
+
"step": 230
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 29.94,
|
| 412 |
+
"eval_accuracy": 0.7431192660550459,
|
| 413 |
+
"eval_loss": 0.6792589426040649,
|
| 414 |
+
"eval_runtime": 0.4139,
|
| 415 |
+
"eval_samples_per_second": 263.354,
|
| 416 |
+
"eval_steps_per_second": 9.664,
|
| 417 |
+
"step": 232
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 30.97,
|
| 421 |
+
"learning_rate": 7.936507936507936e-06,
|
| 422 |
+
"loss": 0.671,
|
| 423 |
+
"step": 240
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 30.97,
|
| 427 |
+
"eval_accuracy": 0.7431192660550459,
|
| 428 |
+
"eval_loss": 0.6808269619941711,
|
| 429 |
+
"eval_runtime": 0.408,
|
| 430 |
+
"eval_samples_per_second": 267.189,
|
| 431 |
+
"eval_steps_per_second": 9.805,
|
| 432 |
+
"step": 240
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 32.0,
|
| 436 |
+
"eval_accuracy": 0.7339449541284404,
|
| 437 |
+
"eval_loss": 0.6820599436759949,
|
| 438 |
+
"eval_runtime": 0.4244,
|
| 439 |
+
"eval_samples_per_second": 256.807,
|
| 440 |
+
"eval_steps_per_second": 9.424,
|
| 441 |
+
"step": 248
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 32.26,
|
| 445 |
+
"learning_rate": 5.9523809523809525e-06,
|
| 446 |
+
"loss": 0.6862,
|
| 447 |
+
"step": 250
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 32.9,
|
| 451 |
+
"eval_accuracy": 0.7339449541284404,
|
| 452 |
+
"eval_loss": 0.6705361008644104,
|
| 453 |
+
"eval_runtime": 0.4159,
|
| 454 |
+
"eval_samples_per_second": 262.074,
|
| 455 |
+
"eval_steps_per_second": 9.617,
|
| 456 |
+
"step": 255
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 33.55,
|
| 460 |
+
"learning_rate": 3.968253968253968e-06,
|
| 461 |
+
"loss": 0.6606,
|
| 462 |
+
"step": 260
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 33.94,
|
| 466 |
+
"eval_accuracy": 0.7431192660550459,
|
| 467 |
+
"eval_loss": 0.6783888339996338,
|
| 468 |
+
"eval_runtime": 0.4197,
|
| 469 |
+
"eval_samples_per_second": 259.739,
|
| 470 |
+
"eval_steps_per_second": 9.532,
|
| 471 |
+
"step": 263
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 34.84,
|
| 475 |
+
"learning_rate": 1.984126984126984e-06,
|
| 476 |
+
"loss": 0.6667,
|
| 477 |
+
"step": 270
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 34.97,
|
| 481 |
+
"eval_accuracy": 0.7522935779816514,
|
| 482 |
+
"eval_loss": 0.6764441132545471,
|
| 483 |
+
"eval_runtime": 0.4166,
|
| 484 |
+
"eval_samples_per_second": 261.642,
|
| 485 |
+
"eval_steps_per_second": 9.602,
|
| 486 |
+
"step": 271
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 36.0,
|
| 490 |
+
"eval_accuracy": 0.7522935779816514,
|
| 491 |
+
"eval_loss": 0.6716886758804321,
|
| 492 |
+
"eval_runtime": 0.5848,
|
| 493 |
+
"eval_samples_per_second": 186.395,
|
| 494 |
+
"eval_steps_per_second": 6.84,
|
| 495 |
+
"step": 279
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 36.13,
|
| 499 |
+
"learning_rate": 0.0,
|
| 500 |
+
"loss": 0.6687,
|
| 501 |
+
"step": 280
|
| 502 |
+
},
|
| 503 |
+
{
|
| 504 |
+
"epoch": 36.13,
|
| 505 |
+
"eval_accuracy": 0.7522935779816514,
|
| 506 |
+
"eval_loss": 0.6736045479774475,
|
| 507 |
+
"eval_runtime": 0.4181,
|
| 508 |
+
"eval_samples_per_second": 260.729,
|
| 509 |
+
"eval_steps_per_second": 9.568,
|
| 510 |
+
"step": 280
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 36.13,
|
| 514 |
+
"step": 280,
|
| 515 |
+
"total_flos": 9.686412043576934e+16,
|
| 516 |
+
"train_loss": 0.878055340903146,
|
| 517 |
+
"train_runtime": 251.6613,
|
| 518 |
+
"train_samples_per_second": 155.447,
|
| 519 |
+
"train_steps_per_second": 1.113
|
| 520 |
+
}
|
| 521 |
+
],
|
| 522 |
+
"logging_steps": 10,
|
| 523 |
+
"max_steps": 280,
|
| 524 |
+
"num_train_epochs": 40,
|
| 525 |
+
"save_steps": 500,
|
| 526 |
+
"total_flos": 9.686412043576934e+16,
|
| 527 |
+
"trial_name": null,
|
| 528 |
+
"trial_params": null
|
| 529 |
+
}
|