🍻 cheers
Browse files- README.md +4 -3
- all_results.json +13 -0
- eval_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +661 -0
README.md
CHANGED
|
@@ -2,6 +2,7 @@
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: microsoft/swin-tiny-patch4-window7-224
|
| 4 |
tags:
|
|
|
|
| 5 |
- generated_from_trainer
|
| 6 |
metrics:
|
| 7 |
- accuracy
|
|
@@ -15,10 +16,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# swin-tiny-patch4-window7-224-finetuned_ASL_Isolated_Swin_dataset2
|
| 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
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
-
- Loss: 0.
|
| 21 |
-
- Accuracy: 0.
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: microsoft/swin-tiny-patch4-window7-224
|
| 4 |
tags:
|
| 5 |
+
- image-classification
|
| 6 |
- generated_from_trainer
|
| 7 |
metrics:
|
| 8 |
- accuracy
|
|
|
|
| 16 |
|
| 17 |
# swin-tiny-patch4-window7-224-finetuned_ASL_Isolated_Swin_dataset2
|
| 18 |
|
| 19 |
+
This model is a fine-tuned version of [microsoft/swin-tiny-patch4-window7-224](https://huggingface.co/microsoft/swin-tiny-patch4-window7-224) on the ASL_Isolated_Swin_dataset dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.0640
|
| 22 |
+
- Accuracy: 0.9846
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.9846153846153847,
|
| 4 |
+
"eval_loss": 0.06404715776443481,
|
| 5 |
+
"eval_runtime": 2.1321,
|
| 6 |
+
"eval_samples_per_second": 121.943,
|
| 7 |
+
"eval_steps_per_second": 15.477,
|
| 8 |
+
"total_flos": 3.651306173035315e+17,
|
| 9 |
+
"train_loss": 0.6966753153697304,
|
| 10 |
+
"train_runtime": 217.2288,
|
| 11 |
+
"train_samples_per_second": 67.579,
|
| 12 |
+
"train_steps_per_second": 4.235
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.9846153846153847,
|
| 4 |
+
"eval_loss": 0.06404715776443481,
|
| 5 |
+
"eval_runtime": 2.1321,
|
| 6 |
+
"eval_samples_per_second": 121.943,
|
| 7 |
+
"eval_steps_per_second": 15.477
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 3.651306173035315e+17,
|
| 4 |
+
"train_loss": 0.6966753153697304,
|
| 5 |
+
"train_runtime": 217.2288,
|
| 6 |
+
"train_samples_per_second": 67.579,
|
| 7 |
+
"train_steps_per_second": 4.235
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,661 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9846153846153847,
|
| 3 |
+
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned_ASL_Isolated_Swin_dataset2/checkpoint-800",
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 920,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.11,
|
| 13 |
+
"learning_rate": 0.00019782608695652175,
|
| 14 |
+
"loss": 3.3609,
|
| 15 |
+
"step": 10
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.22,
|
| 19 |
+
"learning_rate": 0.0001956521739130435,
|
| 20 |
+
"loss": 3.1386,
|
| 21 |
+
"step": 20
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.33,
|
| 25 |
+
"learning_rate": 0.00019347826086956523,
|
| 26 |
+
"loss": 2.9847,
|
| 27 |
+
"step": 30
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.43,
|
| 31 |
+
"learning_rate": 0.00019130434782608697,
|
| 32 |
+
"loss": 2.7624,
|
| 33 |
+
"step": 40
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.54,
|
| 37 |
+
"learning_rate": 0.0001891304347826087,
|
| 38 |
+
"loss": 2.2708,
|
| 39 |
+
"step": 50
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.65,
|
| 43 |
+
"learning_rate": 0.00018717391304347826,
|
| 44 |
+
"loss": 2.1984,
|
| 45 |
+
"step": 60
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.76,
|
| 49 |
+
"learning_rate": 0.00018500000000000002,
|
| 50 |
+
"loss": 1.8882,
|
| 51 |
+
"step": 70
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.87,
|
| 55 |
+
"learning_rate": 0.00018282608695652174,
|
| 56 |
+
"loss": 1.6971,
|
| 57 |
+
"step": 80
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.98,
|
| 61 |
+
"learning_rate": 0.00018065217391304348,
|
| 62 |
+
"loss": 1.4803,
|
| 63 |
+
"step": 90
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 1.09,
|
| 67 |
+
"learning_rate": 0.00017847826086956522,
|
| 68 |
+
"loss": 1.39,
|
| 69 |
+
"step": 100
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 1.09,
|
| 73 |
+
"eval_accuracy": 0.6346153846153846,
|
| 74 |
+
"eval_loss": 1.182677984237671,
|
| 75 |
+
"eval_runtime": 2.4526,
|
| 76 |
+
"eval_samples_per_second": 106.009,
|
| 77 |
+
"eval_steps_per_second": 13.455,
|
| 78 |
+
"step": 100
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 1.2,
|
| 82 |
+
"learning_rate": 0.00017630434782608695,
|
| 83 |
+
"loss": 1.2313,
|
| 84 |
+
"step": 110
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 1.3,
|
| 88 |
+
"learning_rate": 0.00017413043478260872,
|
| 89 |
+
"loss": 1.2511,
|
| 90 |
+
"step": 120
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 1.41,
|
| 94 |
+
"learning_rate": 0.00017217391304347827,
|
| 95 |
+
"loss": 1.1545,
|
| 96 |
+
"step": 130
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 1.52,
|
| 100 |
+
"learning_rate": 0.00017,
|
| 101 |
+
"loss": 1.0298,
|
| 102 |
+
"step": 140
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.63,
|
| 106 |
+
"learning_rate": 0.00016782608695652175,
|
| 107 |
+
"loss": 1.0783,
|
| 108 |
+
"step": 150
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 1.74,
|
| 112 |
+
"learning_rate": 0.0001656521739130435,
|
| 113 |
+
"loss": 1.0336,
|
| 114 |
+
"step": 160
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 1.85,
|
| 118 |
+
"learning_rate": 0.00016347826086956523,
|
| 119 |
+
"loss": 1.0377,
|
| 120 |
+
"step": 170
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 1.96,
|
| 124 |
+
"learning_rate": 0.00016130434782608697,
|
| 125 |
+
"loss": 1.1622,
|
| 126 |
+
"step": 180
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 2.07,
|
| 130 |
+
"learning_rate": 0.00015913043478260868,
|
| 131 |
+
"loss": 0.9864,
|
| 132 |
+
"step": 190
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 2.17,
|
| 136 |
+
"learning_rate": 0.00015695652173913045,
|
| 137 |
+
"loss": 0.8972,
|
| 138 |
+
"step": 200
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 2.17,
|
| 142 |
+
"eval_accuracy": 0.7807692307692308,
|
| 143 |
+
"eval_loss": 0.6287302374839783,
|
| 144 |
+
"eval_runtime": 2.6916,
|
| 145 |
+
"eval_samples_per_second": 96.596,
|
| 146 |
+
"eval_steps_per_second": 12.26,
|
| 147 |
+
"step": 200
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 2.28,
|
| 151 |
+
"learning_rate": 0.0001547826086956522,
|
| 152 |
+
"loss": 0.7377,
|
| 153 |
+
"step": 210
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 2.39,
|
| 157 |
+
"learning_rate": 0.00015260869565217393,
|
| 158 |
+
"loss": 0.8899,
|
| 159 |
+
"step": 220
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 2.5,
|
| 163 |
+
"learning_rate": 0.00015043478260869567,
|
| 164 |
+
"loss": 0.6429,
|
| 165 |
+
"step": 230
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 2.61,
|
| 169 |
+
"learning_rate": 0.0001482608695652174,
|
| 170 |
+
"loss": 0.8942,
|
| 171 |
+
"step": 240
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 2.72,
|
| 175 |
+
"learning_rate": 0.00014608695652173914,
|
| 176 |
+
"loss": 0.8716,
|
| 177 |
+
"step": 250
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 2.83,
|
| 181 |
+
"learning_rate": 0.00014391304347826086,
|
| 182 |
+
"loss": 0.8024,
|
| 183 |
+
"step": 260
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 2.93,
|
| 187 |
+
"learning_rate": 0.00014173913043478262,
|
| 188 |
+
"loss": 0.5412,
|
| 189 |
+
"step": 270
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 3.04,
|
| 193 |
+
"learning_rate": 0.00013956521739130434,
|
| 194 |
+
"loss": 0.7387,
|
| 195 |
+
"step": 280
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 3.15,
|
| 199 |
+
"learning_rate": 0.0001373913043478261,
|
| 200 |
+
"loss": 0.5791,
|
| 201 |
+
"step": 290
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 3.26,
|
| 205 |
+
"learning_rate": 0.00013521739130434784,
|
| 206 |
+
"loss": 0.4884,
|
| 207 |
+
"step": 300
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 3.26,
|
| 211 |
+
"eval_accuracy": 0.8961538461538462,
|
| 212 |
+
"eval_loss": 0.29269811511039734,
|
| 213 |
+
"eval_runtime": 1.9876,
|
| 214 |
+
"eval_samples_per_second": 130.808,
|
| 215 |
+
"eval_steps_per_second": 16.603,
|
| 216 |
+
"step": 300
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 3.37,
|
| 220 |
+
"learning_rate": 0.00013304347826086958,
|
| 221 |
+
"loss": 0.6485,
|
| 222 |
+
"step": 310
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 3.48,
|
| 226 |
+
"learning_rate": 0.00013086956521739132,
|
| 227 |
+
"loss": 0.7623,
|
| 228 |
+
"step": 320
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 3.59,
|
| 232 |
+
"learning_rate": 0.00012869565217391303,
|
| 233 |
+
"loss": 0.6887,
|
| 234 |
+
"step": 330
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 3.7,
|
| 238 |
+
"learning_rate": 0.0001265217391304348,
|
| 239 |
+
"loss": 0.5794,
|
| 240 |
+
"step": 340
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 3.8,
|
| 244 |
+
"learning_rate": 0.0001243478260869565,
|
| 245 |
+
"loss": 0.6654,
|
| 246 |
+
"step": 350
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 3.91,
|
| 250 |
+
"learning_rate": 0.00012217391304347828,
|
| 251 |
+
"loss": 0.5027,
|
| 252 |
+
"step": 360
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 4.02,
|
| 256 |
+
"learning_rate": 0.00012,
|
| 257 |
+
"loss": 0.4336,
|
| 258 |
+
"step": 370
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 4.13,
|
| 262 |
+
"learning_rate": 0.00011782608695652176,
|
| 263 |
+
"loss": 0.5295,
|
| 264 |
+
"step": 380
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 4.24,
|
| 268 |
+
"learning_rate": 0.00011565217391304348,
|
| 269 |
+
"loss": 0.6389,
|
| 270 |
+
"step": 390
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 4.35,
|
| 274 |
+
"learning_rate": 0.00011347826086956523,
|
| 275 |
+
"loss": 0.4179,
|
| 276 |
+
"step": 400
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 4.35,
|
| 280 |
+
"eval_accuracy": 0.9423076923076923,
|
| 281 |
+
"eval_loss": 0.1795085072517395,
|
| 282 |
+
"eval_runtime": 2.6974,
|
| 283 |
+
"eval_samples_per_second": 96.391,
|
| 284 |
+
"eval_steps_per_second": 12.234,
|
| 285 |
+
"step": 400
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 4.46,
|
| 289 |
+
"learning_rate": 0.00011152173913043477,
|
| 290 |
+
"loss": 0.5516,
|
| 291 |
+
"step": 410
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 4.57,
|
| 295 |
+
"learning_rate": 0.00010934782608695653,
|
| 296 |
+
"loss": 0.4264,
|
| 297 |
+
"step": 420
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 4.67,
|
| 301 |
+
"learning_rate": 0.00010717391304347825,
|
| 302 |
+
"loss": 0.4827,
|
| 303 |
+
"step": 430
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 4.78,
|
| 307 |
+
"learning_rate": 0.000105,
|
| 308 |
+
"loss": 0.4434,
|
| 309 |
+
"step": 440
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 4.89,
|
| 313 |
+
"learning_rate": 0.00010282608695652174,
|
| 314 |
+
"loss": 0.4812,
|
| 315 |
+
"step": 450
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 5.0,
|
| 319 |
+
"learning_rate": 0.0001006521739130435,
|
| 320 |
+
"loss": 0.5592,
|
| 321 |
+
"step": 460
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 5.11,
|
| 325 |
+
"learning_rate": 9.847826086956522e-05,
|
| 326 |
+
"loss": 0.4754,
|
| 327 |
+
"step": 470
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 5.22,
|
| 331 |
+
"learning_rate": 9.630434782608696e-05,
|
| 332 |
+
"loss": 0.3776,
|
| 333 |
+
"step": 480
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 5.33,
|
| 337 |
+
"learning_rate": 9.41304347826087e-05,
|
| 338 |
+
"loss": 0.4084,
|
| 339 |
+
"step": 490
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 5.43,
|
| 343 |
+
"learning_rate": 9.195652173913044e-05,
|
| 344 |
+
"loss": 0.4169,
|
| 345 |
+
"step": 500
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 5.43,
|
| 349 |
+
"eval_accuracy": 0.95,
|
| 350 |
+
"eval_loss": 0.1563507616519928,
|
| 351 |
+
"eval_runtime": 1.8046,
|
| 352 |
+
"eval_samples_per_second": 144.078,
|
| 353 |
+
"eval_steps_per_second": 18.287,
|
| 354 |
+
"step": 500
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 5.54,
|
| 358 |
+
"learning_rate": 8.978260869565218e-05,
|
| 359 |
+
"loss": 0.371,
|
| 360 |
+
"step": 510
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 5.65,
|
| 364 |
+
"learning_rate": 8.760869565217392e-05,
|
| 365 |
+
"loss": 0.4745,
|
| 366 |
+
"step": 520
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 5.76,
|
| 370 |
+
"learning_rate": 8.543478260869566e-05,
|
| 371 |
+
"loss": 0.4209,
|
| 372 |
+
"step": 530
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 5.87,
|
| 376 |
+
"learning_rate": 8.32608695652174e-05,
|
| 377 |
+
"loss": 0.4645,
|
| 378 |
+
"step": 540
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 5.98,
|
| 382 |
+
"learning_rate": 8.108695652173914e-05,
|
| 383 |
+
"loss": 0.4322,
|
| 384 |
+
"step": 550
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 6.09,
|
| 388 |
+
"learning_rate": 7.891304347826088e-05,
|
| 389 |
+
"loss": 0.489,
|
| 390 |
+
"step": 560
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 6.2,
|
| 394 |
+
"learning_rate": 7.673913043478262e-05,
|
| 395 |
+
"loss": 0.4531,
|
| 396 |
+
"step": 570
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 6.3,
|
| 400 |
+
"learning_rate": 7.456521739130435e-05,
|
| 401 |
+
"loss": 0.2837,
|
| 402 |
+
"step": 580
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 6.41,
|
| 406 |
+
"learning_rate": 7.23913043478261e-05,
|
| 407 |
+
"loss": 0.37,
|
| 408 |
+
"step": 590
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 6.52,
|
| 412 |
+
"learning_rate": 7.021739130434783e-05,
|
| 413 |
+
"loss": 0.3427,
|
| 414 |
+
"step": 600
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 6.52,
|
| 418 |
+
"eval_accuracy": 0.95,
|
| 419 |
+
"eval_loss": 0.14258505403995514,
|
| 420 |
+
"eval_runtime": 1.9248,
|
| 421 |
+
"eval_samples_per_second": 135.078,
|
| 422 |
+
"eval_steps_per_second": 17.145,
|
| 423 |
+
"step": 600
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 6.63,
|
| 427 |
+
"learning_rate": 6.804347826086956e-05,
|
| 428 |
+
"loss": 0.3713,
|
| 429 |
+
"step": 610
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 6.74,
|
| 433 |
+
"learning_rate": 6.58695652173913e-05,
|
| 434 |
+
"loss": 0.2211,
|
| 435 |
+
"step": 620
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 6.85,
|
| 439 |
+
"learning_rate": 6.369565217391305e-05,
|
| 440 |
+
"loss": 0.2769,
|
| 441 |
+
"step": 630
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 6.96,
|
| 445 |
+
"learning_rate": 6.152173913043479e-05,
|
| 446 |
+
"loss": 0.2778,
|
| 447 |
+
"step": 640
|
| 448 |
+
},
|
| 449 |
+
{
|
| 450 |
+
"epoch": 7.07,
|
| 451 |
+
"learning_rate": 5.934782608695652e-05,
|
| 452 |
+
"loss": 0.3071,
|
| 453 |
+
"step": 650
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 7.17,
|
| 457 |
+
"learning_rate": 5.717391304347827e-05,
|
| 458 |
+
"loss": 0.2988,
|
| 459 |
+
"step": 660
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 7.28,
|
| 463 |
+
"learning_rate": 5.500000000000001e-05,
|
| 464 |
+
"loss": 0.2773,
|
| 465 |
+
"step": 670
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 7.39,
|
| 469 |
+
"learning_rate": 5.2826086956521735e-05,
|
| 470 |
+
"loss": 0.2628,
|
| 471 |
+
"step": 680
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 7.5,
|
| 475 |
+
"learning_rate": 5.0652173913043474e-05,
|
| 476 |
+
"loss": 0.4372,
|
| 477 |
+
"step": 690
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 7.61,
|
| 481 |
+
"learning_rate": 4.847826086956522e-05,
|
| 482 |
+
"loss": 0.2939,
|
| 483 |
+
"step": 700
|
| 484 |
+
},
|
| 485 |
+
{
|
| 486 |
+
"epoch": 7.61,
|
| 487 |
+
"eval_accuracy": 0.9730769230769231,
|
| 488 |
+
"eval_loss": 0.11737523972988129,
|
| 489 |
+
"eval_runtime": 1.9862,
|
| 490 |
+
"eval_samples_per_second": 130.905,
|
| 491 |
+
"eval_steps_per_second": 16.615,
|
| 492 |
+
"step": 700
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 7.72,
|
| 496 |
+
"learning_rate": 4.630434782608696e-05,
|
| 497 |
+
"loss": 0.3101,
|
| 498 |
+
"step": 710
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 7.83,
|
| 502 |
+
"learning_rate": 4.41304347826087e-05,
|
| 503 |
+
"loss": 0.1584,
|
| 504 |
+
"step": 720
|
| 505 |
+
},
|
| 506 |
+
{
|
| 507 |
+
"epoch": 7.93,
|
| 508 |
+
"learning_rate": 4.195652173913044e-05,
|
| 509 |
+
"loss": 0.3143,
|
| 510 |
+
"step": 730
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 8.04,
|
| 514 |
+
"learning_rate": 3.978260869565217e-05,
|
| 515 |
+
"loss": 0.3705,
|
| 516 |
+
"step": 740
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 8.15,
|
| 520 |
+
"learning_rate": 3.7608695652173917e-05,
|
| 521 |
+
"loss": 0.3003,
|
| 522 |
+
"step": 750
|
| 523 |
+
},
|
| 524 |
+
{
|
| 525 |
+
"epoch": 8.26,
|
| 526 |
+
"learning_rate": 3.5434782608695656e-05,
|
| 527 |
+
"loss": 0.3864,
|
| 528 |
+
"step": 760
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 8.37,
|
| 532 |
+
"learning_rate": 3.3260869565217395e-05,
|
| 533 |
+
"loss": 0.1961,
|
| 534 |
+
"step": 770
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 8.48,
|
| 538 |
+
"learning_rate": 3.108695652173913e-05,
|
| 539 |
+
"loss": 0.2859,
|
| 540 |
+
"step": 780
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 8.59,
|
| 544 |
+
"learning_rate": 2.891304347826087e-05,
|
| 545 |
+
"loss": 0.3521,
|
| 546 |
+
"step": 790
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 8.7,
|
| 550 |
+
"learning_rate": 2.673913043478261e-05,
|
| 551 |
+
"loss": 0.1605,
|
| 552 |
+
"step": 800
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 8.7,
|
| 556 |
+
"eval_accuracy": 0.9846153846153847,
|
| 557 |
+
"eval_loss": 0.06404715776443481,
|
| 558 |
+
"eval_runtime": 1.8828,
|
| 559 |
+
"eval_samples_per_second": 138.091,
|
| 560 |
+
"eval_steps_per_second": 17.527,
|
| 561 |
+
"step": 800
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 8.8,
|
| 565 |
+
"learning_rate": 2.456521739130435e-05,
|
| 566 |
+
"loss": 0.1828,
|
| 567 |
+
"step": 810
|
| 568 |
+
},
|
| 569 |
+
{
|
| 570 |
+
"epoch": 8.91,
|
| 571 |
+
"learning_rate": 2.239130434782609e-05,
|
| 572 |
+
"loss": 0.1448,
|
| 573 |
+
"step": 820
|
| 574 |
+
},
|
| 575 |
+
{
|
| 576 |
+
"epoch": 9.02,
|
| 577 |
+
"learning_rate": 2.0217391304347828e-05,
|
| 578 |
+
"loss": 0.2442,
|
| 579 |
+
"step": 830
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 9.13,
|
| 583 |
+
"learning_rate": 1.8043478260869567e-05,
|
| 584 |
+
"loss": 0.2182,
|
| 585 |
+
"step": 840
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 9.24,
|
| 589 |
+
"learning_rate": 1.5869565217391306e-05,
|
| 590 |
+
"loss": 0.1857,
|
| 591 |
+
"step": 850
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 9.35,
|
| 595 |
+
"learning_rate": 1.3695652173913042e-05,
|
| 596 |
+
"loss": 0.123,
|
| 597 |
+
"step": 860
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 9.46,
|
| 601 |
+
"learning_rate": 1.1521739130434783e-05,
|
| 602 |
+
"loss": 0.1286,
|
| 603 |
+
"step": 870
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 9.57,
|
| 607 |
+
"learning_rate": 9.347826086956523e-06,
|
| 608 |
+
"loss": 0.3302,
|
| 609 |
+
"step": 880
|
| 610 |
+
},
|
| 611 |
+
{
|
| 612 |
+
"epoch": 9.67,
|
| 613 |
+
"learning_rate": 7.173913043478261e-06,
|
| 614 |
+
"loss": 0.3513,
|
| 615 |
+
"step": 890
|
| 616 |
+
},
|
| 617 |
+
{
|
| 618 |
+
"epoch": 9.78,
|
| 619 |
+
"learning_rate": 5e-06,
|
| 620 |
+
"loss": 0.1865,
|
| 621 |
+
"step": 900
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 9.78,
|
| 625 |
+
"eval_accuracy": 0.9807692307692307,
|
| 626 |
+
"eval_loss": 0.07018034905195236,
|
| 627 |
+
"eval_runtime": 2.3303,
|
| 628 |
+
"eval_samples_per_second": 111.572,
|
| 629 |
+
"eval_steps_per_second": 14.161,
|
| 630 |
+
"step": 900
|
| 631 |
+
},
|
| 632 |
+
{
|
| 633 |
+
"epoch": 9.89,
|
| 634 |
+
"learning_rate": 2.8260869565217393e-06,
|
| 635 |
+
"loss": 0.1316,
|
| 636 |
+
"step": 910
|
| 637 |
+
},
|
| 638 |
+
{
|
| 639 |
+
"epoch": 10.0,
|
| 640 |
+
"learning_rate": 6.521739130434782e-07,
|
| 641 |
+
"loss": 0.1979,
|
| 642 |
+
"step": 920
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 10.0,
|
| 646 |
+
"step": 920,
|
| 647 |
+
"total_flos": 3.651306173035315e+17,
|
| 648 |
+
"train_loss": 0.6966753153697304,
|
| 649 |
+
"train_runtime": 217.2288,
|
| 650 |
+
"train_samples_per_second": 67.579,
|
| 651 |
+
"train_steps_per_second": 4.235
|
| 652 |
+
}
|
| 653 |
+
],
|
| 654 |
+
"logging_steps": 10,
|
| 655 |
+
"max_steps": 920,
|
| 656 |
+
"num_train_epochs": 10,
|
| 657 |
+
"save_steps": 100,
|
| 658 |
+
"total_flos": 3.651306173035315e+17,
|
| 659 |
+
"trial_name": null,
|
| 660 |
+
"trial_params": null
|
| 661 |
+
}
|