End of training
Browse files- README.md +1 -1
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/Apr07_07-36-43_dc3c4f3a0653/events.out.tfevents.1712477196.dc3c4f3a0653.306.2 +3 -0
- train_results.json +8 -0
- trainer_state.json +495 -0
README.md
CHANGED
|
@@ -32,7 +32,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 32 |
|
| 33 |
This model is a fine-tuned version of [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) on the imagefolder dataset.
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
-
- Loss: 1.
|
| 36 |
- Accuracy: 0.9348
|
| 37 |
|
| 38 |
## Model description
|
|
|
|
| 32 |
|
| 33 |
This model is a fine-tuned version of [google/vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224) on the imagefolder dataset.
|
| 34 |
It achieves the following results on the evaluation set:
|
| 35 |
+
- Loss: 1.1736
|
| 36 |
- Accuracy: 0.9348
|
| 37 |
|
| 38 |
## Model description
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 40.0,
|
| 3 |
+
"eval_accuracy": 0.9347826086956522,
|
| 4 |
+
"eval_loss": 1.1735903024673462,
|
| 5 |
+
"eval_runtime": 9.658,
|
| 6 |
+
"eval_samples_per_second": 47.629,
|
| 7 |
+
"eval_steps_per_second": 1.553,
|
| 8 |
+
"total_flos": 1.4270027608424448e+18,
|
| 9 |
+
"train_loss": 1.9666850312550863,
|
| 10 |
+
"train_runtime": 1497.2167,
|
| 11 |
+
"train_samples_per_second": 15.362,
|
| 12 |
+
"train_steps_per_second": 0.1
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 40.0,
|
| 3 |
+
"eval_accuracy": 0.9347826086956522,
|
| 4 |
+
"eval_loss": 1.1735903024673462,
|
| 5 |
+
"eval_runtime": 9.658,
|
| 6 |
+
"eval_samples_per_second": 47.629,
|
| 7 |
+
"eval_steps_per_second": 1.553
|
| 8 |
+
}
|
runs/Apr07_07-36-43_dc3c4f3a0653/events.out.tfevents.1712477196.dc3c4f3a0653.306.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0791a0f05744fd20cd367c75d7be67b85c21a092a9a904e74341eaa820c37eb4
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 40.0,
|
| 3 |
+
"total_flos": 1.4270027608424448e+18,
|
| 4 |
+
"train_loss": 1.9666850312550863,
|
| 5 |
+
"train_runtime": 1497.2167,
|
| 6 |
+
"train_samples_per_second": 15.362,
|
| 7 |
+
"train_steps_per_second": 0.1
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,495 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9347826086956522,
|
| 3 |
+
"best_model_checkpoint": "vit-base-patch16-224-finetuned-teeth_dataset/checkpoint-138",
|
| 4 |
+
"epoch": 40.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 150,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.8,
|
| 13 |
+
"eval_accuracy": 0.008695652173913044,
|
| 14 |
+
"eval_loss": 4.653251647949219,
|
| 15 |
+
"eval_runtime": 128.9785,
|
| 16 |
+
"eval_samples_per_second": 3.566,
|
| 17 |
+
"eval_steps_per_second": 0.116,
|
| 18 |
+
"step": 3
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 1.87,
|
| 22 |
+
"eval_accuracy": 0.006521739130434782,
|
| 23 |
+
"eval_loss": 4.584750175476074,
|
| 24 |
+
"eval_runtime": 9.4797,
|
| 25 |
+
"eval_samples_per_second": 48.525,
|
| 26 |
+
"eval_steps_per_second": 1.582,
|
| 27 |
+
"step": 7
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 2.67,
|
| 31 |
+
"grad_norm": 2.967961311340332,
|
| 32 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 33 |
+
"loss": 4.6048,
|
| 34 |
+
"step": 10
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 2.93,
|
| 38 |
+
"eval_accuracy": 0.030434782608695653,
|
| 39 |
+
"eval_loss": 4.460752964019775,
|
| 40 |
+
"eval_runtime": 9.6304,
|
| 41 |
+
"eval_samples_per_second": 47.766,
|
| 42 |
+
"eval_steps_per_second": 1.558,
|
| 43 |
+
"step": 11
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 4.0,
|
| 47 |
+
"eval_accuracy": 0.08478260869565217,
|
| 48 |
+
"eval_loss": 4.285670757293701,
|
| 49 |
+
"eval_runtime": 9.7655,
|
| 50 |
+
"eval_samples_per_second": 47.104,
|
| 51 |
+
"eval_steps_per_second": 1.536,
|
| 52 |
+
"step": 15
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 4.8,
|
| 56 |
+
"eval_accuracy": 0.11521739130434783,
|
| 57 |
+
"eval_loss": 4.1469807624816895,
|
| 58 |
+
"eval_runtime": 9.6545,
|
| 59 |
+
"eval_samples_per_second": 47.646,
|
| 60 |
+
"eval_steps_per_second": 1.554,
|
| 61 |
+
"step": 18
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 5.33,
|
| 65 |
+
"grad_norm": 3.3595643043518066,
|
| 66 |
+
"learning_rate": 4.814814814814815e-05,
|
| 67 |
+
"loss": 4.2716,
|
| 68 |
+
"step": 20
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 5.87,
|
| 72 |
+
"eval_accuracy": 0.20434782608695654,
|
| 73 |
+
"eval_loss": 3.964137315750122,
|
| 74 |
+
"eval_runtime": 9.7181,
|
| 75 |
+
"eval_samples_per_second": 47.334,
|
| 76 |
+
"eval_steps_per_second": 1.544,
|
| 77 |
+
"step": 22
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 6.93,
|
| 81 |
+
"eval_accuracy": 0.31521739130434784,
|
| 82 |
+
"eval_loss": 3.7704641819000244,
|
| 83 |
+
"eval_runtime": 9.5221,
|
| 84 |
+
"eval_samples_per_second": 48.309,
|
| 85 |
+
"eval_steps_per_second": 1.575,
|
| 86 |
+
"step": 26
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 8.0,
|
| 90 |
+
"grad_norm": 4.556402683258057,
|
| 91 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 92 |
+
"loss": 3.7404,
|
| 93 |
+
"step": 30
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 8.0,
|
| 97 |
+
"eval_accuracy": 0.41956521739130437,
|
| 98 |
+
"eval_loss": 3.5808911323547363,
|
| 99 |
+
"eval_runtime": 9.3828,
|
| 100 |
+
"eval_samples_per_second": 49.026,
|
| 101 |
+
"eval_steps_per_second": 1.599,
|
| 102 |
+
"step": 30
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 8.8,
|
| 106 |
+
"eval_accuracy": 0.45217391304347826,
|
| 107 |
+
"eval_loss": 3.4765827655792236,
|
| 108 |
+
"eval_runtime": 9.5003,
|
| 109 |
+
"eval_samples_per_second": 48.419,
|
| 110 |
+
"eval_steps_per_second": 1.579,
|
| 111 |
+
"step": 33
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 9.87,
|
| 115 |
+
"eval_accuracy": 0.508695652173913,
|
| 116 |
+
"eval_loss": 3.2980847358703613,
|
| 117 |
+
"eval_runtime": 9.6725,
|
| 118 |
+
"eval_samples_per_second": 47.557,
|
| 119 |
+
"eval_steps_per_second": 1.551,
|
| 120 |
+
"step": 37
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 10.67,
|
| 124 |
+
"grad_norm": 3.8709869384765625,
|
| 125 |
+
"learning_rate": 4.074074074074074e-05,
|
| 126 |
+
"loss": 3.1589,
|
| 127 |
+
"step": 40
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 10.93,
|
| 131 |
+
"eval_accuracy": 0.6086956521739131,
|
| 132 |
+
"eval_loss": 3.1131505966186523,
|
| 133 |
+
"eval_runtime": 9.5498,
|
| 134 |
+
"eval_samples_per_second": 48.169,
|
| 135 |
+
"eval_steps_per_second": 1.571,
|
| 136 |
+
"step": 41
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 12.0,
|
| 140 |
+
"eval_accuracy": 0.6695652173913044,
|
| 141 |
+
"eval_loss": 2.949446201324463,
|
| 142 |
+
"eval_runtime": 9.5423,
|
| 143 |
+
"eval_samples_per_second": 48.206,
|
| 144 |
+
"eval_steps_per_second": 1.572,
|
| 145 |
+
"step": 45
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 12.8,
|
| 149 |
+
"eval_accuracy": 0.6782608695652174,
|
| 150 |
+
"eval_loss": 2.836071729660034,
|
| 151 |
+
"eval_runtime": 9.5423,
|
| 152 |
+
"eval_samples_per_second": 48.207,
|
| 153 |
+
"eval_steps_per_second": 1.572,
|
| 154 |
+
"step": 48
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 13.33,
|
| 158 |
+
"grad_norm": 4.395231246948242,
|
| 159 |
+
"learning_rate": 3.7037037037037037e-05,
|
| 160 |
+
"loss": 2.6384,
|
| 161 |
+
"step": 50
|
| 162 |
+
},
|
| 163 |
+
{
|
| 164 |
+
"epoch": 13.87,
|
| 165 |
+
"eval_accuracy": 0.7347826086956522,
|
| 166 |
+
"eval_loss": 2.6520774364471436,
|
| 167 |
+
"eval_runtime": 9.5638,
|
| 168 |
+
"eval_samples_per_second": 48.098,
|
| 169 |
+
"eval_steps_per_second": 1.568,
|
| 170 |
+
"step": 52
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 14.93,
|
| 174 |
+
"eval_accuracy": 0.758695652173913,
|
| 175 |
+
"eval_loss": 2.4943137168884277,
|
| 176 |
+
"eval_runtime": 9.5047,
|
| 177 |
+
"eval_samples_per_second": 48.397,
|
| 178 |
+
"eval_steps_per_second": 1.578,
|
| 179 |
+
"step": 56
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 16.0,
|
| 183 |
+
"grad_norm": 4.303729057312012,
|
| 184 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 185 |
+
"loss": 2.1342,
|
| 186 |
+
"step": 60
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 16.0,
|
| 190 |
+
"eval_accuracy": 0.7847826086956522,
|
| 191 |
+
"eval_loss": 2.3421921730041504,
|
| 192 |
+
"eval_runtime": 9.1584,
|
| 193 |
+
"eval_samples_per_second": 50.227,
|
| 194 |
+
"eval_steps_per_second": 1.638,
|
| 195 |
+
"step": 60
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 16.8,
|
| 199 |
+
"eval_accuracy": 0.8108695652173913,
|
| 200 |
+
"eval_loss": 2.2326693534851074,
|
| 201 |
+
"eval_runtime": 9.6001,
|
| 202 |
+
"eval_samples_per_second": 47.916,
|
| 203 |
+
"eval_steps_per_second": 1.562,
|
| 204 |
+
"step": 63
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 17.87,
|
| 208 |
+
"eval_accuracy": 0.8260869565217391,
|
| 209 |
+
"eval_loss": 2.083353281021118,
|
| 210 |
+
"eval_runtime": 9.5492,
|
| 211 |
+
"eval_samples_per_second": 48.172,
|
| 212 |
+
"eval_steps_per_second": 1.571,
|
| 213 |
+
"step": 67
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 18.67,
|
| 217 |
+
"grad_norm": 3.915405750274658,
|
| 218 |
+
"learning_rate": 2.962962962962963e-05,
|
| 219 |
+
"loss": 1.714,
|
| 220 |
+
"step": 70
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 18.93,
|
| 224 |
+
"eval_accuracy": 0.8565217391304348,
|
| 225 |
+
"eval_loss": 1.983383059501648,
|
| 226 |
+
"eval_runtime": 9.4581,
|
| 227 |
+
"eval_samples_per_second": 48.636,
|
| 228 |
+
"eval_steps_per_second": 1.586,
|
| 229 |
+
"step": 71
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 20.0,
|
| 233 |
+
"eval_accuracy": 0.8673913043478261,
|
| 234 |
+
"eval_loss": 1.8931976556777954,
|
| 235 |
+
"eval_runtime": 9.4385,
|
| 236 |
+
"eval_samples_per_second": 48.737,
|
| 237 |
+
"eval_steps_per_second": 1.589,
|
| 238 |
+
"step": 75
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 20.8,
|
| 242 |
+
"eval_accuracy": 0.8586956521739131,
|
| 243 |
+
"eval_loss": 1.861843228340149,
|
| 244 |
+
"eval_runtime": 9.2697,
|
| 245 |
+
"eval_samples_per_second": 49.624,
|
| 246 |
+
"eval_steps_per_second": 1.618,
|
| 247 |
+
"step": 78
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 21.33,
|
| 251 |
+
"grad_norm": 3.428410768508911,
|
| 252 |
+
"learning_rate": 2.5925925925925925e-05,
|
| 253 |
+
"loss": 1.4427,
|
| 254 |
+
"step": 80
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 21.87,
|
| 258 |
+
"eval_accuracy": 0.8891304347826087,
|
| 259 |
+
"eval_loss": 1.6974326372146606,
|
| 260 |
+
"eval_runtime": 9.5173,
|
| 261 |
+
"eval_samples_per_second": 48.333,
|
| 262 |
+
"eval_steps_per_second": 1.576,
|
| 263 |
+
"step": 82
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 22.93,
|
| 267 |
+
"eval_accuracy": 0.8891304347826087,
|
| 268 |
+
"eval_loss": 1.6662733554840088,
|
| 269 |
+
"eval_runtime": 9.3013,
|
| 270 |
+
"eval_samples_per_second": 49.456,
|
| 271 |
+
"eval_steps_per_second": 1.613,
|
| 272 |
+
"step": 86
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 24.0,
|
| 276 |
+
"grad_norm": 4.027013778686523,
|
| 277 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 278 |
+
"loss": 1.1858,
|
| 279 |
+
"step": 90
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 24.0,
|
| 283 |
+
"eval_accuracy": 0.8847826086956522,
|
| 284 |
+
"eval_loss": 1.6013683080673218,
|
| 285 |
+
"eval_runtime": 9.5313,
|
| 286 |
+
"eval_samples_per_second": 48.262,
|
| 287 |
+
"eval_steps_per_second": 1.574,
|
| 288 |
+
"step": 90
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 24.8,
|
| 292 |
+
"eval_accuracy": 0.9043478260869565,
|
| 293 |
+
"eval_loss": 1.5112110376358032,
|
| 294 |
+
"eval_runtime": 9.5654,
|
| 295 |
+
"eval_samples_per_second": 48.09,
|
| 296 |
+
"eval_steps_per_second": 1.568,
|
| 297 |
+
"step": 93
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 25.87,
|
| 301 |
+
"eval_accuracy": 0.9108695652173913,
|
| 302 |
+
"eval_loss": 1.473188042640686,
|
| 303 |
+
"eval_runtime": 9.4856,
|
| 304 |
+
"eval_samples_per_second": 48.495,
|
| 305 |
+
"eval_steps_per_second": 1.581,
|
| 306 |
+
"step": 97
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 26.67,
|
| 310 |
+
"grad_norm": 3.1277294158935547,
|
| 311 |
+
"learning_rate": 1.8518518518518518e-05,
|
| 312 |
+
"loss": 1.0222,
|
| 313 |
+
"step": 100
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 26.93,
|
| 317 |
+
"eval_accuracy": 0.9065217391304348,
|
| 318 |
+
"eval_loss": 1.430389404296875,
|
| 319 |
+
"eval_runtime": 9.3124,
|
| 320 |
+
"eval_samples_per_second": 49.396,
|
| 321 |
+
"eval_steps_per_second": 1.611,
|
| 322 |
+
"step": 101
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 28.0,
|
| 326 |
+
"eval_accuracy": 0.9130434782608695,
|
| 327 |
+
"eval_loss": 1.3915013074874878,
|
| 328 |
+
"eval_runtime": 9.4281,
|
| 329 |
+
"eval_samples_per_second": 48.79,
|
| 330 |
+
"eval_steps_per_second": 1.591,
|
| 331 |
+
"step": 105
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 28.8,
|
| 335 |
+
"eval_accuracy": 0.9217391304347826,
|
| 336 |
+
"eval_loss": 1.3509211540222168,
|
| 337 |
+
"eval_runtime": 9.5983,
|
| 338 |
+
"eval_samples_per_second": 47.925,
|
| 339 |
+
"eval_steps_per_second": 1.563,
|
| 340 |
+
"step": 108
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 29.33,
|
| 344 |
+
"grad_norm": 2.7500855922698975,
|
| 345 |
+
"learning_rate": 1.4814814814814815e-05,
|
| 346 |
+
"loss": 0.8306,
|
| 347 |
+
"step": 110
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 29.87,
|
| 351 |
+
"eval_accuracy": 0.9282608695652174,
|
| 352 |
+
"eval_loss": 1.3053604364395142,
|
| 353 |
+
"eval_runtime": 9.4585,
|
| 354 |
+
"eval_samples_per_second": 48.633,
|
| 355 |
+
"eval_steps_per_second": 1.586,
|
| 356 |
+
"step": 112
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 30.93,
|
| 360 |
+
"eval_accuracy": 0.9260869565217391,
|
| 361 |
+
"eval_loss": 1.2869884967803955,
|
| 362 |
+
"eval_runtime": 9.1724,
|
| 363 |
+
"eval_samples_per_second": 50.151,
|
| 364 |
+
"eval_steps_per_second": 1.635,
|
| 365 |
+
"step": 116
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 32.0,
|
| 369 |
+
"grad_norm": 2.7954471111297607,
|
| 370 |
+
"learning_rate": 1.1111111111111112e-05,
|
| 371 |
+
"loss": 0.7391,
|
| 372 |
+
"step": 120
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 32.0,
|
| 376 |
+
"eval_accuracy": 0.9282608695652174,
|
| 377 |
+
"eval_loss": 1.264487385749817,
|
| 378 |
+
"eval_runtime": 9.6279,
|
| 379 |
+
"eval_samples_per_second": 47.778,
|
| 380 |
+
"eval_steps_per_second": 1.558,
|
| 381 |
+
"step": 120
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 32.8,
|
| 385 |
+
"eval_accuracy": 0.9260869565217391,
|
| 386 |
+
"eval_loss": 1.2453793287277222,
|
| 387 |
+
"eval_runtime": 9.2796,
|
| 388 |
+
"eval_samples_per_second": 49.571,
|
| 389 |
+
"eval_steps_per_second": 1.616,
|
| 390 |
+
"step": 123
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 33.87,
|
| 394 |
+
"eval_accuracy": 0.9282608695652174,
|
| 395 |
+
"eval_loss": 1.2394675016403198,
|
| 396 |
+
"eval_runtime": 9.5473,
|
| 397 |
+
"eval_samples_per_second": 48.181,
|
| 398 |
+
"eval_steps_per_second": 1.571,
|
| 399 |
+
"step": 127
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 34.67,
|
| 403 |
+
"grad_norm": 2.7077043056488037,
|
| 404 |
+
"learning_rate": 7.4074074074074075e-06,
|
| 405 |
+
"loss": 0.6971,
|
| 406 |
+
"step": 130
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 34.93,
|
| 410 |
+
"eval_accuracy": 0.9304347826086956,
|
| 411 |
+
"eval_loss": 1.207598090171814,
|
| 412 |
+
"eval_runtime": 9.5844,
|
| 413 |
+
"eval_samples_per_second": 47.994,
|
| 414 |
+
"eval_steps_per_second": 1.565,
|
| 415 |
+
"step": 131
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 36.0,
|
| 419 |
+
"eval_accuracy": 0.9326086956521739,
|
| 420 |
+
"eval_loss": 1.1821191310882568,
|
| 421 |
+
"eval_runtime": 9.4964,
|
| 422 |
+
"eval_samples_per_second": 48.44,
|
| 423 |
+
"eval_steps_per_second": 1.58,
|
| 424 |
+
"step": 135
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 36.8,
|
| 428 |
+
"eval_accuracy": 0.9347826086956522,
|
| 429 |
+
"eval_loss": 1.1735903024673462,
|
| 430 |
+
"eval_runtime": 9.4168,
|
| 431 |
+
"eval_samples_per_second": 48.849,
|
| 432 |
+
"eval_steps_per_second": 1.593,
|
| 433 |
+
"step": 138
|
| 434 |
+
},
|
| 435 |
+
{
|
| 436 |
+
"epoch": 37.33,
|
| 437 |
+
"grad_norm": 2.4264731407165527,
|
| 438 |
+
"learning_rate": 3.7037037037037037e-06,
|
| 439 |
+
"loss": 0.6758,
|
| 440 |
+
"step": 140
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 37.87,
|
| 444 |
+
"eval_accuracy": 0.9326086956521739,
|
| 445 |
+
"eval_loss": 1.1671130657196045,
|
| 446 |
+
"eval_runtime": 9.6709,
|
| 447 |
+
"eval_samples_per_second": 47.565,
|
| 448 |
+
"eval_steps_per_second": 1.551,
|
| 449 |
+
"step": 142
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 38.93,
|
| 453 |
+
"eval_accuracy": 0.9347826086956522,
|
| 454 |
+
"eval_loss": 1.1656177043914795,
|
| 455 |
+
"eval_runtime": 9.2896,
|
| 456 |
+
"eval_samples_per_second": 49.518,
|
| 457 |
+
"eval_steps_per_second": 1.615,
|
| 458 |
+
"step": 146
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 40.0,
|
| 462 |
+
"grad_norm": 2.507667303085327,
|
| 463 |
+
"learning_rate": 0.0,
|
| 464 |
+
"loss": 0.6445,
|
| 465 |
+
"step": 150
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 40.0,
|
| 469 |
+
"eval_accuracy": 0.9347826086956522,
|
| 470 |
+
"eval_loss": 1.1649013757705688,
|
| 471 |
+
"eval_runtime": 9.7408,
|
| 472 |
+
"eval_samples_per_second": 47.224,
|
| 473 |
+
"eval_steps_per_second": 1.54,
|
| 474 |
+
"step": 150
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 40.0,
|
| 478 |
+
"step": 150,
|
| 479 |
+
"total_flos": 1.4270027608424448e+18,
|
| 480 |
+
"train_loss": 1.9666850312550863,
|
| 481 |
+
"train_runtime": 1497.2167,
|
| 482 |
+
"train_samples_per_second": 15.362,
|
| 483 |
+
"train_steps_per_second": 0.1
|
| 484 |
+
}
|
| 485 |
+
],
|
| 486 |
+
"logging_steps": 10,
|
| 487 |
+
"max_steps": 150,
|
| 488 |
+
"num_input_tokens_seen": 0,
|
| 489 |
+
"num_train_epochs": 50,
|
| 490 |
+
"save_steps": 500,
|
| 491 |
+
"total_flos": 1.4270027608424448e+18,
|
| 492 |
+
"train_batch_size": 32,
|
| 493 |
+
"trial_name": null,
|
| 494 |
+
"trial_params": null
|
| 495 |
+
}
|