End of training
Browse files- README.md +2 -2
- all_results.json +11 -11
- eval_results.json +6 -6
- train_results.json +6 -6
- trainer_state.json +424 -37
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 the None dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss: 1.
|
| 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 the None dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 1.6126
|
| 21 |
+
- Accuracy: 0.4266
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 1.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 1.
|
| 8 |
-
"total_flos": 2.
|
| 9 |
-
"train_loss":
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second": 0.
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 27.91,
|
| 3 |
+
"eval_accuracy": 0.42657342657342656,
|
| 4 |
+
"eval_loss": 1.6125619411468506,
|
| 5 |
+
"eval_runtime": 7.9305,
|
| 6 |
+
"eval_samples_per_second": 54.095,
|
| 7 |
+
"eval_steps_per_second": 1.765,
|
| 8 |
+
"total_flos": 2.9422949721372426e+18,
|
| 9 |
+
"train_loss": 0.9840216811498006,
|
| 10 |
+
"train_runtime": 1797.847,
|
| 11 |
+
"train_samples_per_second": 22.694,
|
| 12 |
+
"train_steps_per_second": 0.167
|
| 13 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 1.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 1.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 27.91,
|
| 3 |
+
"eval_accuracy": 0.42657342657342656,
|
| 4 |
+
"eval_loss": 1.6125619411468506,
|
| 5 |
+
"eval_runtime": 7.9305,
|
| 6 |
+
"eval_samples_per_second": 54.095,
|
| 7 |
+
"eval_steps_per_second": 1.765
|
| 8 |
}
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos": 2.
|
| 4 |
-
"train_loss":
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 27.91,
|
| 3 |
+
"total_flos": 2.9422949721372426e+18,
|
| 4 |
+
"train_loss": 0.9840216811498006,
|
| 5 |
+
"train_runtime": 1797.847,
|
| 6 |
+
"train_samples_per_second": 22.694,
|
| 7 |
+
"train_steps_per_second": 0.167
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,73 +1,460 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "vit-base-patch16-224-for-pre_evaluation/checkpoint-
|
| 4 |
-
"epoch":
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.93,
|
| 13 |
-
"learning_rate":
|
| 14 |
-
"loss": 1.
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"epoch": 0.93,
|
| 19 |
-
"eval_accuracy": 0.
|
| 20 |
-
"eval_loss": 1.
|
| 21 |
-
"eval_runtime":
|
| 22 |
-
"eval_samples_per_second":
|
| 23 |
-
"eval_steps_per_second": 1.
|
| 24 |
"step": 10
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 1.86,
|
| 28 |
-
"learning_rate":
|
| 29 |
-
"loss": 1.
|
| 30 |
"step": 20
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 1.95,
|
| 34 |
-
"eval_accuracy": 0.
|
| 35 |
-
"eval_loss": 1.
|
| 36 |
-
"eval_runtime":
|
| 37 |
-
"eval_samples_per_second":
|
| 38 |
-
"eval_steps_per_second": 1.
|
| 39 |
"step": 21
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 2.79,
|
| 43 |
-
"learning_rate":
|
| 44 |
-
"loss": 1.
|
| 45 |
"step": 30
|
| 46 |
},
|
| 47 |
{
|
| 48 |
-
"epoch": 2.
|
| 49 |
-
"eval_accuracy": 0.
|
| 50 |
-
"eval_loss": 1.
|
| 51 |
-
"eval_runtime":
|
| 52 |
-
"eval_samples_per_second":
|
| 53 |
-
"eval_steps_per_second": 1.
|
| 54 |
-
"step":
|
| 55 |
},
|
| 56 |
{
|
| 57 |
-
"epoch":
|
| 58 |
-
"
|
| 59 |
-
"
|
| 60 |
-
"
|
| 61 |
-
|
| 62 |
-
|
| 63 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 64 |
}
|
| 65 |
],
|
| 66 |
"logging_steps": 10,
|
| 67 |
-
"max_steps":
|
| 68 |
-
"num_train_epochs":
|
| 69 |
"save_steps": 500,
|
| 70 |
-
"total_flos": 2.
|
| 71 |
"trial_name": null,
|
| 72 |
"trial_params": null
|
| 73 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.42657342657342656,
|
| 3 |
+
"best_model_checkpoint": "vit-base-patch16-224-for-pre_evaluation/checkpoint-236",
|
| 4 |
+
"epoch": 27.906976744186046,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 300,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.93,
|
| 13 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 14 |
+
"loss": 1.6266,
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"epoch": 0.93,
|
| 19 |
+
"eval_accuracy": 0.3006993006993007,
|
| 20 |
+
"eval_loss": 1.5307577848434448,
|
| 21 |
+
"eval_runtime": 10.578,
|
| 22 |
+
"eval_samples_per_second": 40.556,
|
| 23 |
+
"eval_steps_per_second": 1.324,
|
| 24 |
"step": 10
|
| 25 |
},
|
| 26 |
{
|
| 27 |
"epoch": 1.86,
|
| 28 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 29 |
+
"loss": 1.4973,
|
| 30 |
"step": 20
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 1.95,
|
| 34 |
+
"eval_accuracy": 0.3123543123543124,
|
| 35 |
+
"eval_loss": 1.5000418424606323,
|
| 36 |
+
"eval_runtime": 7.5932,
|
| 37 |
+
"eval_samples_per_second": 56.498,
|
| 38 |
+
"eval_steps_per_second": 1.844,
|
| 39 |
"step": 21
|
| 40 |
},
|
| 41 |
{
|
| 42 |
"epoch": 2.79,
|
| 43 |
+
"learning_rate": 5e-05,
|
| 44 |
+
"loss": 1.4636,
|
| 45 |
"step": 30
|
| 46 |
},
|
| 47 |
{
|
| 48 |
+
"epoch": 2.98,
|
| 49 |
+
"eval_accuracy": 0.34498834498834496,
|
| 50 |
+
"eval_loss": 1.472951054573059,
|
| 51 |
+
"eval_runtime": 11.6515,
|
| 52 |
+
"eval_samples_per_second": 36.819,
|
| 53 |
+
"eval_steps_per_second": 1.202,
|
| 54 |
+
"step": 32
|
| 55 |
},
|
| 56 |
{
|
| 57 |
+
"epoch": 3.72,
|
| 58 |
+
"learning_rate": 4.814814814814815e-05,
|
| 59 |
+
"loss": 1.4185,
|
| 60 |
+
"step": 40
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 4.0,
|
| 64 |
+
"eval_accuracy": 0.30536130536130535,
|
| 65 |
+
"eval_loss": 1.5051661729812622,
|
| 66 |
+
"eval_runtime": 7.0051,
|
| 67 |
+
"eval_samples_per_second": 61.241,
|
| 68 |
+
"eval_steps_per_second": 1.999,
|
| 69 |
+
"step": 43
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 4.65,
|
| 73 |
+
"learning_rate": 4.62962962962963e-05,
|
| 74 |
+
"loss": 1.4147,
|
| 75 |
+
"step": 50
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 4.93,
|
| 79 |
+
"eval_accuracy": 0.34965034965034963,
|
| 80 |
+
"eval_loss": 1.4455597400665283,
|
| 81 |
+
"eval_runtime": 6.741,
|
| 82 |
+
"eval_samples_per_second": 63.641,
|
| 83 |
+
"eval_steps_per_second": 2.077,
|
| 84 |
+
"step": 53
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 5.58,
|
| 88 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 89 |
+
"loss": 1.337,
|
| 90 |
+
"step": 60
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 5.95,
|
| 94 |
+
"eval_accuracy": 0.34265734265734266,
|
| 95 |
+
"eval_loss": 1.4735510349273682,
|
| 96 |
+
"eval_runtime": 7.0594,
|
| 97 |
+
"eval_samples_per_second": 60.77,
|
| 98 |
+
"eval_steps_per_second": 1.983,
|
| 99 |
+
"step": 64
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 6.51,
|
| 103 |
+
"learning_rate": 4.259259259259259e-05,
|
| 104 |
+
"loss": 1.2869,
|
| 105 |
+
"step": 70
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 6.98,
|
| 109 |
+
"eval_accuracy": 0.37995337995337997,
|
| 110 |
+
"eval_loss": 1.4170001745224,
|
| 111 |
+
"eval_runtime": 7.3984,
|
| 112 |
+
"eval_samples_per_second": 57.985,
|
| 113 |
+
"eval_steps_per_second": 1.892,
|
| 114 |
+
"step": 75
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 7.44,
|
| 118 |
+
"learning_rate": 4.074074074074074e-05,
|
| 119 |
+
"loss": 1.241,
|
| 120 |
+
"step": 80
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 8.0,
|
| 124 |
+
"eval_accuracy": 0.372960372960373,
|
| 125 |
+
"eval_loss": 1.5077800750732422,
|
| 126 |
+
"eval_runtime": 7.0787,
|
| 127 |
+
"eval_samples_per_second": 60.604,
|
| 128 |
+
"eval_steps_per_second": 1.978,
|
| 129 |
+
"step": 86
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 8.37,
|
| 133 |
+
"learning_rate": 3.888888888888889e-05,
|
| 134 |
+
"loss": 1.2261,
|
| 135 |
+
"step": 90
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 8.93,
|
| 139 |
+
"eval_accuracy": 0.38927738927738925,
|
| 140 |
+
"eval_loss": 1.4629043340682983,
|
| 141 |
+
"eval_runtime": 7.4136,
|
| 142 |
+
"eval_samples_per_second": 57.867,
|
| 143 |
+
"eval_steps_per_second": 1.888,
|
| 144 |
+
"step": 96
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 9.3,
|
| 148 |
+
"learning_rate": 3.7037037037037037e-05,
|
| 149 |
+
"loss": 1.1967,
|
| 150 |
+
"step": 100
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 9.95,
|
| 154 |
+
"eval_accuracy": 0.41025641025641024,
|
| 155 |
+
"eval_loss": 1.4139227867126465,
|
| 156 |
+
"eval_runtime": 7.4631,
|
| 157 |
+
"eval_samples_per_second": 57.483,
|
| 158 |
+
"eval_steps_per_second": 1.876,
|
| 159 |
+
"step": 107
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 10.23,
|
| 163 |
+
"learning_rate": 3.518518518518519e-05,
|
| 164 |
+
"loss": 1.1153,
|
| 165 |
+
"step": 110
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 10.98,
|
| 169 |
+
"eval_accuracy": 0.4195804195804196,
|
| 170 |
+
"eval_loss": 1.4337666034698486,
|
| 171 |
+
"eval_runtime": 7.0965,
|
| 172 |
+
"eval_samples_per_second": 60.452,
|
| 173 |
+
"eval_steps_per_second": 1.973,
|
| 174 |
+
"step": 118
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 11.16,
|
| 178 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 179 |
+
"loss": 1.063,
|
| 180 |
+
"step": 120
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 12.0,
|
| 184 |
+
"eval_accuracy": 0.3939393939393939,
|
| 185 |
+
"eval_loss": 1.4642902612686157,
|
| 186 |
+
"eval_runtime": 7.3068,
|
| 187 |
+
"eval_samples_per_second": 58.713,
|
| 188 |
+
"eval_steps_per_second": 1.916,
|
| 189 |
+
"step": 129
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 12.09,
|
| 193 |
+
"learning_rate": 3.148148148148148e-05,
|
| 194 |
+
"loss": 1.0434,
|
| 195 |
+
"step": 130
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 12.93,
|
| 199 |
+
"eval_accuracy": 0.40792540792540793,
|
| 200 |
+
"eval_loss": 1.4725754261016846,
|
| 201 |
+
"eval_runtime": 7.4877,
|
| 202 |
+
"eval_samples_per_second": 57.294,
|
| 203 |
+
"eval_steps_per_second": 1.87,
|
| 204 |
+
"step": 139
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 13.02,
|
| 208 |
+
"learning_rate": 2.962962962962963e-05,
|
| 209 |
+
"loss": 1.0067,
|
| 210 |
+
"step": 140
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 13.95,
|
| 214 |
+
"learning_rate": 2.777777777777778e-05,
|
| 215 |
+
"loss": 0.9849,
|
| 216 |
+
"step": 150
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 13.95,
|
| 220 |
+
"eval_accuracy": 0.4149184149184149,
|
| 221 |
+
"eval_loss": 1.528671145439148,
|
| 222 |
+
"eval_runtime": 7.2601,
|
| 223 |
+
"eval_samples_per_second": 59.09,
|
| 224 |
+
"eval_steps_per_second": 1.928,
|
| 225 |
+
"step": 150
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 14.88,
|
| 229 |
+
"learning_rate": 2.5925925925925925e-05,
|
| 230 |
+
"loss": 0.9285,
|
| 231 |
+
"step": 160
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 14.98,
|
| 235 |
+
"eval_accuracy": 0.3986013986013986,
|
| 236 |
+
"eval_loss": 1.552985429763794,
|
| 237 |
+
"eval_runtime": 7.0224,
|
| 238 |
+
"eval_samples_per_second": 61.09,
|
| 239 |
+
"eval_steps_per_second": 1.994,
|
| 240 |
+
"step": 161
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 15.81,
|
| 244 |
+
"learning_rate": 2.4074074074074074e-05,
|
| 245 |
+
"loss": 0.8724,
|
| 246 |
+
"step": 170
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 16.0,
|
| 250 |
+
"eval_accuracy": 0.3939393939393939,
|
| 251 |
+
"eval_loss": 1.5016406774520874,
|
| 252 |
+
"eval_runtime": 7.1273,
|
| 253 |
+
"eval_samples_per_second": 60.191,
|
| 254 |
+
"eval_steps_per_second": 1.964,
|
| 255 |
+
"step": 172
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 16.74,
|
| 259 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 260 |
+
"loss": 0.8063,
|
| 261 |
+
"step": 180
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 16.93,
|
| 265 |
+
"eval_accuracy": 0.40559440559440557,
|
| 266 |
+
"eval_loss": 1.5184847116470337,
|
| 267 |
+
"eval_runtime": 7.1223,
|
| 268 |
+
"eval_samples_per_second": 60.233,
|
| 269 |
+
"eval_steps_per_second": 1.966,
|
| 270 |
+
"step": 182
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 17.67,
|
| 274 |
+
"learning_rate": 2.037037037037037e-05,
|
| 275 |
+
"loss": 0.8206,
|
| 276 |
+
"step": 190
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 17.95,
|
| 280 |
+
"eval_accuracy": 0.4172494172494173,
|
| 281 |
+
"eval_loss": 1.5447591543197632,
|
| 282 |
+
"eval_runtime": 7.0975,
|
| 283 |
+
"eval_samples_per_second": 60.444,
|
| 284 |
+
"eval_steps_per_second": 1.973,
|
| 285 |
+
"step": 193
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 18.6,
|
| 289 |
+
"learning_rate": 1.8518518518518518e-05,
|
| 290 |
+
"loss": 0.7396,
|
| 291 |
+
"step": 200
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 18.98,
|
| 295 |
+
"eval_accuracy": 0.40326340326340326,
|
| 296 |
+
"eval_loss": 1.554015874862671,
|
| 297 |
+
"eval_runtime": 6.78,
|
| 298 |
+
"eval_samples_per_second": 63.275,
|
| 299 |
+
"eval_steps_per_second": 2.065,
|
| 300 |
+
"step": 204
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 19.53,
|
| 304 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 305 |
+
"loss": 0.7437,
|
| 306 |
+
"step": 210
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 20.0,
|
| 310 |
+
"eval_accuracy": 0.40093240093240096,
|
| 311 |
+
"eval_loss": 1.579628586769104,
|
| 312 |
+
"eval_runtime": 6.8036,
|
| 313 |
+
"eval_samples_per_second": 63.055,
|
| 314 |
+
"eval_steps_per_second": 2.058,
|
| 315 |
+
"step": 215
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 20.47,
|
| 319 |
+
"learning_rate": 1.4814814814814815e-05,
|
| 320 |
+
"loss": 0.7103,
|
| 321 |
+
"step": 220
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 20.93,
|
| 325 |
+
"eval_accuracy": 0.40326340326340326,
|
| 326 |
+
"eval_loss": 1.6218039989471436,
|
| 327 |
+
"eval_runtime": 7.5951,
|
| 328 |
+
"eval_samples_per_second": 56.484,
|
| 329 |
+
"eval_steps_per_second": 1.843,
|
| 330 |
+
"step": 225
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 21.4,
|
| 334 |
+
"learning_rate": 1.2962962962962962e-05,
|
| 335 |
+
"loss": 0.6861,
|
| 336 |
+
"step": 230
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 21.95,
|
| 340 |
+
"eval_accuracy": 0.42657342657342656,
|
| 341 |
+
"eval_loss": 1.6125619411468506,
|
| 342 |
+
"eval_runtime": 7.0894,
|
| 343 |
+
"eval_samples_per_second": 60.512,
|
| 344 |
+
"eval_steps_per_second": 1.975,
|
| 345 |
+
"step": 236
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 22.33,
|
| 349 |
+
"learning_rate": 1.1111111111111112e-05,
|
| 350 |
+
"loss": 0.6798,
|
| 351 |
+
"step": 240
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 22.98,
|
| 355 |
+
"eval_accuracy": 0.42657342657342656,
|
| 356 |
+
"eval_loss": 1.6051045656204224,
|
| 357 |
+
"eval_runtime": 7.0114,
|
| 358 |
+
"eval_samples_per_second": 61.186,
|
| 359 |
+
"eval_steps_per_second": 1.997,
|
| 360 |
+
"step": 247
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 23.26,
|
| 364 |
+
"learning_rate": 9.259259259259259e-06,
|
| 365 |
+
"loss": 0.6358,
|
| 366 |
+
"step": 250
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 24.0,
|
| 370 |
+
"eval_accuracy": 0.41025641025641024,
|
| 371 |
+
"eval_loss": 1.6141223907470703,
|
| 372 |
+
"eval_runtime": 6.9962,
|
| 373 |
+
"eval_samples_per_second": 61.319,
|
| 374 |
+
"eval_steps_per_second": 2.001,
|
| 375 |
+
"step": 258
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 24.19,
|
| 379 |
+
"learning_rate": 7.4074074074074075e-06,
|
| 380 |
+
"loss": 0.6189,
|
| 381 |
+
"step": 260
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 24.93,
|
| 385 |
+
"eval_accuracy": 0.40559440559440557,
|
| 386 |
+
"eval_loss": 1.6705572605133057,
|
| 387 |
+
"eval_runtime": 6.7024,
|
| 388 |
+
"eval_samples_per_second": 64.007,
|
| 389 |
+
"eval_steps_per_second": 2.089,
|
| 390 |
+
"step": 268
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 25.12,
|
| 394 |
+
"learning_rate": 5.555555555555556e-06,
|
| 395 |
+
"loss": 0.6176,
|
| 396 |
+
"step": 270
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 25.95,
|
| 400 |
+
"eval_accuracy": 0.4219114219114219,
|
| 401 |
+
"eval_loss": 1.6404389142990112,
|
| 402 |
+
"eval_runtime": 6.7355,
|
| 403 |
+
"eval_samples_per_second": 63.692,
|
| 404 |
+
"eval_steps_per_second": 2.079,
|
| 405 |
+
"step": 279
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 26.05,
|
| 409 |
+
"learning_rate": 3.7037037037037037e-06,
|
| 410 |
+
"loss": 0.575,
|
| 411 |
+
"step": 280
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 26.98,
|
| 415 |
+
"learning_rate": 1.8518518518518519e-06,
|
| 416 |
+
"loss": 0.5756,
|
| 417 |
+
"step": 290
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 26.98,
|
| 421 |
+
"eval_accuracy": 0.4172494172494173,
|
| 422 |
+
"eval_loss": 1.660282850265503,
|
| 423 |
+
"eval_runtime": 6.8437,
|
| 424 |
+
"eval_samples_per_second": 62.685,
|
| 425 |
+
"eval_steps_per_second": 2.046,
|
| 426 |
+
"step": 290
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 27.91,
|
| 430 |
+
"learning_rate": 0.0,
|
| 431 |
+
"loss": 0.5887,
|
| 432 |
+
"step": 300
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 27.91,
|
| 436 |
+
"eval_accuracy": 0.42424242424242425,
|
| 437 |
+
"eval_loss": 1.6626534461975098,
|
| 438 |
+
"eval_runtime": 6.7064,
|
| 439 |
+
"eval_samples_per_second": 63.969,
|
| 440 |
+
"eval_steps_per_second": 2.088,
|
| 441 |
+
"step": 300
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 27.91,
|
| 445 |
+
"step": 300,
|
| 446 |
+
"total_flos": 2.9422949721372426e+18,
|
| 447 |
+
"train_loss": 0.9840216811498006,
|
| 448 |
+
"train_runtime": 1797.847,
|
| 449 |
+
"train_samples_per_second": 22.694,
|
| 450 |
+
"train_steps_per_second": 0.167
|
| 451 |
}
|
| 452 |
],
|
| 453 |
"logging_steps": 10,
|
| 454 |
+
"max_steps": 300,
|
| 455 |
+
"num_train_epochs": 30,
|
| 456 |
"save_steps": 500,
|
| 457 |
+
"total_flos": 2.9422949721372426e+18,
|
| 458 |
"trial_name": null,
|
| 459 |
"trial_params": null
|
| 460 |
}
|