End of training
Browse files- README.md +2 -2
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/Jan21_15-51-11_aabb36fdf211/events.out.tfevents.1737476616.aabb36fdf211.212.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +665 -0
README.md
CHANGED
|
@@ -18,8 +18,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [microsoft/swinv2-tiny-patch4-window8-256](https://huggingface.co/microsoft/swinv2-tiny-patch4-window8-256) on an unknown dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
-
- Loss:
|
| 22 |
-
- Accuracy: 0.
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
|
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [microsoft/swinv2-tiny-patch4-window8-256](https://huggingface.co/microsoft/swinv2-tiny-patch4-window8-256) on an unknown dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.9577
|
| 22 |
+
- Accuracy: 0.7308
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 41.93023255813954,
|
| 3 |
+
"eval_accuracy": 0.7307692307692307,
|
| 4 |
+
"eval_loss": 0.9577280879020691,
|
| 5 |
+
"eval_runtime": 1.0772,
|
| 6 |
+
"eval_samples_per_second": 48.273,
|
| 7 |
+
"eval_steps_per_second": 1.857,
|
| 8 |
+
"total_flos": 1.8345690247389512e+18,
|
| 9 |
+
"train_loss": 1.2270459805216107,
|
| 10 |
+
"train_runtime": 1356.3174,
|
| 11 |
+
"train_samples_per_second": 42.176,
|
| 12 |
+
"train_steps_per_second": 0.31
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 41.93023255813954,
|
| 3 |
+
"eval_accuracy": 0.7307692307692307,
|
| 4 |
+
"eval_loss": 0.9577280879020691,
|
| 5 |
+
"eval_runtime": 1.0772,
|
| 6 |
+
"eval_samples_per_second": 48.273,
|
| 7 |
+
"eval_steps_per_second": 1.857
|
| 8 |
+
}
|
runs/Jan21_15-51-11_aabb36fdf211/events.out.tfevents.1737476616.aabb36fdf211.212.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2db4350dd8e40a29cbb42e2f4443060de81f2261a34f39edd86eb8f9062e73f1
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 41.93023255813954,
|
| 3 |
+
"total_flos": 1.8345690247389512e+18,
|
| 4 |
+
"train_loss": 1.2270459805216107,
|
| 5 |
+
"train_runtime": 1356.3174,
|
| 6 |
+
"train_samples_per_second": 42.176,
|
| 7 |
+
"train_steps_per_second": 0.31
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,665 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.7307692307692307,
|
| 3 |
+
"best_model_checkpoint": "swinv2-tiny-patch4-window8-256-dmae-humeda-DAV21/checkpoint-90",
|
| 4 |
+
"epoch": 41.93023255813954,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 420,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.9302325581395349,
|
| 13 |
+
"eval_accuracy": 0.23076923076923078,
|
| 14 |
+
"eval_loss": 1.600624680519104,
|
| 15 |
+
"eval_runtime": 0.9466,
|
| 16 |
+
"eval_samples_per_second": 54.935,
|
| 17 |
+
"eval_steps_per_second": 2.113,
|
| 18 |
+
"step": 10
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 1.1860465116279069,
|
| 22 |
+
"grad_norm": 11.869792938232422,
|
| 23 |
+
"learning_rate": 1.4285714285714285e-05,
|
| 24 |
+
"loss": 6.7273,
|
| 25 |
+
"step": 12
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 1.9302325581395348,
|
| 29 |
+
"eval_accuracy": 0.3076923076923077,
|
| 30 |
+
"eval_loss": 1.4678974151611328,
|
| 31 |
+
"eval_runtime": 0.9407,
|
| 32 |
+
"eval_samples_per_second": 55.281,
|
| 33 |
+
"eval_steps_per_second": 2.126,
|
| 34 |
+
"step": 20
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 2.3720930232558137,
|
| 38 |
+
"grad_norm": 21.593393325805664,
|
| 39 |
+
"learning_rate": 2.857142857142857e-05,
|
| 40 |
+
"loss": 5.9474,
|
| 41 |
+
"step": 24
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 2.9302325581395348,
|
| 45 |
+
"eval_accuracy": 0.5384615384615384,
|
| 46 |
+
"eval_loss": 1.1914443969726562,
|
| 47 |
+
"eval_runtime": 0.9141,
|
| 48 |
+
"eval_samples_per_second": 56.886,
|
| 49 |
+
"eval_steps_per_second": 2.188,
|
| 50 |
+
"step": 30
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 3.558139534883721,
|
| 54 |
+
"grad_norm": 30.464523315429688,
|
| 55 |
+
"learning_rate": 4.2857142857142856e-05,
|
| 56 |
+
"loss": 4.3002,
|
| 57 |
+
"step": 36
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 3.9302325581395348,
|
| 61 |
+
"eval_accuracy": 0.46153846153846156,
|
| 62 |
+
"eval_loss": 1.1951944828033447,
|
| 63 |
+
"eval_runtime": 1.2786,
|
| 64 |
+
"eval_samples_per_second": 40.671,
|
| 65 |
+
"eval_steps_per_second": 1.564,
|
| 66 |
+
"step": 40
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 4.7441860465116275,
|
| 70 |
+
"grad_norm": 25.736072540283203,
|
| 71 |
+
"learning_rate": 4.9206349206349204e-05,
|
| 72 |
+
"loss": 2.9717,
|
| 73 |
+
"step": 48
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 4.930232558139535,
|
| 77 |
+
"eval_accuracy": 0.5769230769230769,
|
| 78 |
+
"eval_loss": 0.9782841205596924,
|
| 79 |
+
"eval_runtime": 0.9391,
|
| 80 |
+
"eval_samples_per_second": 55.375,
|
| 81 |
+
"eval_steps_per_second": 2.13,
|
| 82 |
+
"step": 50
|
| 83 |
+
},
|
| 84 |
+
{
|
| 85 |
+
"epoch": 5.930232558139535,
|
| 86 |
+
"grad_norm": 54.5507698059082,
|
| 87 |
+
"learning_rate": 4.761904761904762e-05,
|
| 88 |
+
"loss": 2.4828,
|
| 89 |
+
"step": 60
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 5.930232558139535,
|
| 93 |
+
"eval_accuracy": 0.5769230769230769,
|
| 94 |
+
"eval_loss": 0.9832176566123962,
|
| 95 |
+
"eval_runtime": 0.9619,
|
| 96 |
+
"eval_samples_per_second": 54.06,
|
| 97 |
+
"eval_steps_per_second": 2.079,
|
| 98 |
+
"step": 60
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 6.930232558139535,
|
| 102 |
+
"eval_accuracy": 0.5,
|
| 103 |
+
"eval_loss": 1.0621165037155151,
|
| 104 |
+
"eval_runtime": 1.3284,
|
| 105 |
+
"eval_samples_per_second": 39.145,
|
| 106 |
+
"eval_steps_per_second": 1.506,
|
| 107 |
+
"step": 70
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 7.186046511627907,
|
| 111 |
+
"grad_norm": 26.069013595581055,
|
| 112 |
+
"learning_rate": 4.603174603174603e-05,
|
| 113 |
+
"loss": 2.0276,
|
| 114 |
+
"step": 72
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 7.930232558139535,
|
| 118 |
+
"eval_accuracy": 0.6538461538461539,
|
| 119 |
+
"eval_loss": 0.9040729999542236,
|
| 120 |
+
"eval_runtime": 0.9561,
|
| 121 |
+
"eval_samples_per_second": 54.39,
|
| 122 |
+
"eval_steps_per_second": 2.092,
|
| 123 |
+
"step": 80
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 8.372093023255815,
|
| 127 |
+
"grad_norm": 43.163116455078125,
|
| 128 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 129 |
+
"loss": 1.7041,
|
| 130 |
+
"step": 84
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 8.930232558139535,
|
| 134 |
+
"eval_accuracy": 0.7307692307692307,
|
| 135 |
+
"eval_loss": 0.9577280879020691,
|
| 136 |
+
"eval_runtime": 0.9834,
|
| 137 |
+
"eval_samples_per_second": 52.878,
|
| 138 |
+
"eval_steps_per_second": 2.034,
|
| 139 |
+
"step": 90
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 9.55813953488372,
|
| 143 |
+
"grad_norm": 45.423728942871094,
|
| 144 |
+
"learning_rate": 4.2857142857142856e-05,
|
| 145 |
+
"loss": 1.3967,
|
| 146 |
+
"step": 96
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 9.930232558139535,
|
| 150 |
+
"eval_accuracy": 0.6730769230769231,
|
| 151 |
+
"eval_loss": 1.0290472507476807,
|
| 152 |
+
"eval_runtime": 0.9566,
|
| 153 |
+
"eval_samples_per_second": 54.362,
|
| 154 |
+
"eval_steps_per_second": 2.091,
|
| 155 |
+
"step": 100
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 10.744186046511627,
|
| 159 |
+
"grad_norm": 29.1428279876709,
|
| 160 |
+
"learning_rate": 4.126984126984127e-05,
|
| 161 |
+
"loss": 1.3079,
|
| 162 |
+
"step": 108
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 10.930232558139535,
|
| 166 |
+
"eval_accuracy": 0.7115384615384616,
|
| 167 |
+
"eval_loss": 0.9863845109939575,
|
| 168 |
+
"eval_runtime": 1.3391,
|
| 169 |
+
"eval_samples_per_second": 38.832,
|
| 170 |
+
"eval_steps_per_second": 1.494,
|
| 171 |
+
"step": 110
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 11.930232558139535,
|
| 175 |
+
"grad_norm": 25.176292419433594,
|
| 176 |
+
"learning_rate": 3.968253968253968e-05,
|
| 177 |
+
"loss": 1.0486,
|
| 178 |
+
"step": 120
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 11.930232558139535,
|
| 182 |
+
"eval_accuracy": 0.6730769230769231,
|
| 183 |
+
"eval_loss": 1.0005784034729004,
|
| 184 |
+
"eval_runtime": 0.9808,
|
| 185 |
+
"eval_samples_per_second": 53.016,
|
| 186 |
+
"eval_steps_per_second": 2.039,
|
| 187 |
+
"step": 120
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 12.930232558139535,
|
| 191 |
+
"eval_accuracy": 0.6153846153846154,
|
| 192 |
+
"eval_loss": 0.9503486156463623,
|
| 193 |
+
"eval_runtime": 0.9746,
|
| 194 |
+
"eval_samples_per_second": 53.353,
|
| 195 |
+
"eval_steps_per_second": 2.052,
|
| 196 |
+
"step": 130
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 13.186046511627907,
|
| 200 |
+
"grad_norm": 55.16592025756836,
|
| 201 |
+
"learning_rate": 3.809523809523809e-05,
|
| 202 |
+
"loss": 1.0704,
|
| 203 |
+
"step": 132
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 13.930232558139535,
|
| 207 |
+
"eval_accuracy": 0.7307692307692307,
|
| 208 |
+
"eval_loss": 0.9434700012207031,
|
| 209 |
+
"eval_runtime": 1.3209,
|
| 210 |
+
"eval_samples_per_second": 39.368,
|
| 211 |
+
"eval_steps_per_second": 1.514,
|
| 212 |
+
"step": 140
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 14.372093023255815,
|
| 216 |
+
"grad_norm": 31.888456344604492,
|
| 217 |
+
"learning_rate": 3.650793650793651e-05,
|
| 218 |
+
"loss": 0.9929,
|
| 219 |
+
"step": 144
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 14.930232558139535,
|
| 223 |
+
"eval_accuracy": 0.6346153846153846,
|
| 224 |
+
"eval_loss": 1.0601013898849487,
|
| 225 |
+
"eval_runtime": 0.9793,
|
| 226 |
+
"eval_samples_per_second": 53.098,
|
| 227 |
+
"eval_steps_per_second": 2.042,
|
| 228 |
+
"step": 150
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 15.55813953488372,
|
| 232 |
+
"grad_norm": 29.11701202392578,
|
| 233 |
+
"learning_rate": 3.492063492063492e-05,
|
| 234 |
+
"loss": 0.8911,
|
| 235 |
+
"step": 156
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 15.930232558139535,
|
| 239 |
+
"eval_accuracy": 0.6730769230769231,
|
| 240 |
+
"eval_loss": 1.0281590223312378,
|
| 241 |
+
"eval_runtime": 0.9645,
|
| 242 |
+
"eval_samples_per_second": 53.912,
|
| 243 |
+
"eval_steps_per_second": 2.074,
|
| 244 |
+
"step": 160
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 16.74418604651163,
|
| 248 |
+
"grad_norm": 32.15597152709961,
|
| 249 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 250 |
+
"loss": 0.7702,
|
| 251 |
+
"step": 168
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 16.930232558139537,
|
| 255 |
+
"eval_accuracy": 0.6923076923076923,
|
| 256 |
+
"eval_loss": 1.1857846975326538,
|
| 257 |
+
"eval_runtime": 0.9593,
|
| 258 |
+
"eval_samples_per_second": 54.207,
|
| 259 |
+
"eval_steps_per_second": 2.085,
|
| 260 |
+
"step": 170
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 17.930232558139537,
|
| 264 |
+
"grad_norm": 32.97303771972656,
|
| 265 |
+
"learning_rate": 3.1746031746031745e-05,
|
| 266 |
+
"loss": 0.718,
|
| 267 |
+
"step": 180
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 17.930232558139537,
|
| 271 |
+
"eval_accuracy": 0.6538461538461539,
|
| 272 |
+
"eval_loss": 1.2644145488739014,
|
| 273 |
+
"eval_runtime": 1.3041,
|
| 274 |
+
"eval_samples_per_second": 39.875,
|
| 275 |
+
"eval_steps_per_second": 1.534,
|
| 276 |
+
"step": 180
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 18.930232558139537,
|
| 280 |
+
"eval_accuracy": 0.6346153846153846,
|
| 281 |
+
"eval_loss": 1.1403751373291016,
|
| 282 |
+
"eval_runtime": 0.957,
|
| 283 |
+
"eval_samples_per_second": 54.339,
|
| 284 |
+
"eval_steps_per_second": 2.09,
|
| 285 |
+
"step": 190
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 19.186046511627907,
|
| 289 |
+
"grad_norm": 18.562591552734375,
|
| 290 |
+
"learning_rate": 3.0158730158730158e-05,
|
| 291 |
+
"loss": 0.6526,
|
| 292 |
+
"step": 192
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 19.930232558139537,
|
| 296 |
+
"eval_accuracy": 0.6923076923076923,
|
| 297 |
+
"eval_loss": 1.2202544212341309,
|
| 298 |
+
"eval_runtime": 0.981,
|
| 299 |
+
"eval_samples_per_second": 53.008,
|
| 300 |
+
"eval_steps_per_second": 2.039,
|
| 301 |
+
"step": 200
|
| 302 |
+
},
|
| 303 |
+
{
|
| 304 |
+
"epoch": 20.372093023255815,
|
| 305 |
+
"grad_norm": 41.22922897338867,
|
| 306 |
+
"learning_rate": 2.857142857142857e-05,
|
| 307 |
+
"loss": 0.6617,
|
| 308 |
+
"step": 204
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 20.930232558139537,
|
| 312 |
+
"eval_accuracy": 0.6730769230769231,
|
| 313 |
+
"eval_loss": 1.3465415239334106,
|
| 314 |
+
"eval_runtime": 1.3212,
|
| 315 |
+
"eval_samples_per_second": 39.359,
|
| 316 |
+
"eval_steps_per_second": 1.514,
|
| 317 |
+
"step": 210
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 21.558139534883722,
|
| 321 |
+
"grad_norm": 28.829620361328125,
|
| 322 |
+
"learning_rate": 2.6984126984126984e-05,
|
| 323 |
+
"loss": 0.5699,
|
| 324 |
+
"step": 216
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 21.930232558139537,
|
| 328 |
+
"eval_accuracy": 0.6538461538461539,
|
| 329 |
+
"eval_loss": 1.3862558603286743,
|
| 330 |
+
"eval_runtime": 0.9516,
|
| 331 |
+
"eval_samples_per_second": 54.644,
|
| 332 |
+
"eval_steps_per_second": 2.102,
|
| 333 |
+
"step": 220
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 22.74418604651163,
|
| 337 |
+
"grad_norm": 23.03827476501465,
|
| 338 |
+
"learning_rate": 2.5396825396825397e-05,
|
| 339 |
+
"loss": 0.5228,
|
| 340 |
+
"step": 228
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 22.930232558139537,
|
| 344 |
+
"eval_accuracy": 0.7115384615384616,
|
| 345 |
+
"eval_loss": 1.272707223892212,
|
| 346 |
+
"eval_runtime": 0.985,
|
| 347 |
+
"eval_samples_per_second": 52.794,
|
| 348 |
+
"eval_steps_per_second": 2.031,
|
| 349 |
+
"step": 230
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 23.930232558139537,
|
| 353 |
+
"grad_norm": 27.27239990234375,
|
| 354 |
+
"learning_rate": 2.380952380952381e-05,
|
| 355 |
+
"loss": 0.4321,
|
| 356 |
+
"step": 240
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 23.930232558139537,
|
| 360 |
+
"eval_accuracy": 0.7115384615384616,
|
| 361 |
+
"eval_loss": 1.194717288017273,
|
| 362 |
+
"eval_runtime": 0.9431,
|
| 363 |
+
"eval_samples_per_second": 55.135,
|
| 364 |
+
"eval_steps_per_second": 2.121,
|
| 365 |
+
"step": 240
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 24.930232558139537,
|
| 369 |
+
"eval_accuracy": 0.6346153846153846,
|
| 370 |
+
"eval_loss": 1.397331714630127,
|
| 371 |
+
"eval_runtime": 1.2955,
|
| 372 |
+
"eval_samples_per_second": 40.139,
|
| 373 |
+
"eval_steps_per_second": 1.544,
|
| 374 |
+
"step": 250
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 25.186046511627907,
|
| 378 |
+
"grad_norm": 24.13581085205078,
|
| 379 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 380 |
+
"loss": 0.557,
|
| 381 |
+
"step": 252
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 25.930232558139537,
|
| 385 |
+
"eval_accuracy": 0.6538461538461539,
|
| 386 |
+
"eval_loss": 1.4321506023406982,
|
| 387 |
+
"eval_runtime": 1.3216,
|
| 388 |
+
"eval_samples_per_second": 39.345,
|
| 389 |
+
"eval_steps_per_second": 1.513,
|
| 390 |
+
"step": 260
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 26.372093023255815,
|
| 394 |
+
"grad_norm": 34.4373664855957,
|
| 395 |
+
"learning_rate": 2.0634920634920636e-05,
|
| 396 |
+
"loss": 0.4569,
|
| 397 |
+
"step": 264
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 26.930232558139537,
|
| 401 |
+
"eval_accuracy": 0.6538461538461539,
|
| 402 |
+
"eval_loss": 1.3255730867385864,
|
| 403 |
+
"eval_runtime": 0.9379,
|
| 404 |
+
"eval_samples_per_second": 55.443,
|
| 405 |
+
"eval_steps_per_second": 2.132,
|
| 406 |
+
"step": 270
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 27.558139534883722,
|
| 410 |
+
"grad_norm": 19.65439224243164,
|
| 411 |
+
"learning_rate": 1.9047619047619046e-05,
|
| 412 |
+
"loss": 0.4395,
|
| 413 |
+
"step": 276
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 27.930232558139537,
|
| 417 |
+
"eval_accuracy": 0.6730769230769231,
|
| 418 |
+
"eval_loss": 1.3907530307769775,
|
| 419 |
+
"eval_runtime": 1.2894,
|
| 420 |
+
"eval_samples_per_second": 40.33,
|
| 421 |
+
"eval_steps_per_second": 1.551,
|
| 422 |
+
"step": 280
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 28.74418604651163,
|
| 426 |
+
"grad_norm": 31.272125244140625,
|
| 427 |
+
"learning_rate": 1.746031746031746e-05,
|
| 428 |
+
"loss": 0.5587,
|
| 429 |
+
"step": 288
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 28.930232558139537,
|
| 433 |
+
"eval_accuracy": 0.6923076923076923,
|
| 434 |
+
"eval_loss": 1.5055606365203857,
|
| 435 |
+
"eval_runtime": 0.9347,
|
| 436 |
+
"eval_samples_per_second": 55.632,
|
| 437 |
+
"eval_steps_per_second": 2.14,
|
| 438 |
+
"step": 290
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 29.930232558139537,
|
| 442 |
+
"grad_norm": 21.733734130859375,
|
| 443 |
+
"learning_rate": 1.5873015873015872e-05,
|
| 444 |
+
"loss": 0.439,
|
| 445 |
+
"step": 300
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 29.930232558139537,
|
| 449 |
+
"eval_accuracy": 0.6538461538461539,
|
| 450 |
+
"eval_loss": 1.3810527324676514,
|
| 451 |
+
"eval_runtime": 0.9297,
|
| 452 |
+
"eval_samples_per_second": 55.934,
|
| 453 |
+
"eval_steps_per_second": 2.151,
|
| 454 |
+
"step": 300
|
| 455 |
+
},
|
| 456 |
+
{
|
| 457 |
+
"epoch": 30.930232558139537,
|
| 458 |
+
"eval_accuracy": 0.6153846153846154,
|
| 459 |
+
"eval_loss": 1.3693809509277344,
|
| 460 |
+
"eval_runtime": 1.2587,
|
| 461 |
+
"eval_samples_per_second": 41.311,
|
| 462 |
+
"eval_steps_per_second": 1.589,
|
| 463 |
+
"step": 310
|
| 464 |
+
},
|
| 465 |
+
{
|
| 466 |
+
"epoch": 31.186046511627907,
|
| 467 |
+
"grad_norm": 24.466344833374023,
|
| 468 |
+
"learning_rate": 1.4285714285714285e-05,
|
| 469 |
+
"loss": 0.3989,
|
| 470 |
+
"step": 312
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 31.930232558139537,
|
| 474 |
+
"eval_accuracy": 0.6538461538461539,
|
| 475 |
+
"eval_loss": 1.4759361743927002,
|
| 476 |
+
"eval_runtime": 0.9328,
|
| 477 |
+
"eval_samples_per_second": 55.746,
|
| 478 |
+
"eval_steps_per_second": 2.144,
|
| 479 |
+
"step": 320
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 32.372093023255815,
|
| 483 |
+
"grad_norm": 35.87159729003906,
|
| 484 |
+
"learning_rate": 1.2698412698412699e-05,
|
| 485 |
+
"loss": 0.312,
|
| 486 |
+
"step": 324
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 32.93023255813954,
|
| 490 |
+
"eval_accuracy": 0.6346153846153846,
|
| 491 |
+
"eval_loss": 1.5513783693313599,
|
| 492 |
+
"eval_runtime": 0.9364,
|
| 493 |
+
"eval_samples_per_second": 55.53,
|
| 494 |
+
"eval_steps_per_second": 2.136,
|
| 495 |
+
"step": 330
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 33.55813953488372,
|
| 499 |
+
"grad_norm": 14.943767547607422,
|
| 500 |
+
"learning_rate": 1.1111111111111112e-05,
|
| 501 |
+
"loss": 0.3728,
|
| 502 |
+
"step": 336
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 33.93023255813954,
|
| 506 |
+
"eval_accuracy": 0.6538461538461539,
|
| 507 |
+
"eval_loss": 1.5128782987594604,
|
| 508 |
+
"eval_runtime": 1.2844,
|
| 509 |
+
"eval_samples_per_second": 40.485,
|
| 510 |
+
"eval_steps_per_second": 1.557,
|
| 511 |
+
"step": 340
|
| 512 |
+
},
|
| 513 |
+
{
|
| 514 |
+
"epoch": 34.74418604651163,
|
| 515 |
+
"grad_norm": 15.3562593460083,
|
| 516 |
+
"learning_rate": 9.523809523809523e-06,
|
| 517 |
+
"loss": 0.4031,
|
| 518 |
+
"step": 348
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 34.93023255813954,
|
| 522 |
+
"eval_accuracy": 0.6538461538461539,
|
| 523 |
+
"eval_loss": 1.4914475679397583,
|
| 524 |
+
"eval_runtime": 0.948,
|
| 525 |
+
"eval_samples_per_second": 54.852,
|
| 526 |
+
"eval_steps_per_second": 2.11,
|
| 527 |
+
"step": 350
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 35.93023255813954,
|
| 531 |
+
"grad_norm": 23.48607635498047,
|
| 532 |
+
"learning_rate": 7.936507936507936e-06,
|
| 533 |
+
"loss": 0.3523,
|
| 534 |
+
"step": 360
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 35.93023255813954,
|
| 538 |
+
"eval_accuracy": 0.6730769230769231,
|
| 539 |
+
"eval_loss": 1.5588839054107666,
|
| 540 |
+
"eval_runtime": 0.9436,
|
| 541 |
+
"eval_samples_per_second": 55.108,
|
| 542 |
+
"eval_steps_per_second": 2.12,
|
| 543 |
+
"step": 360
|
| 544 |
+
},
|
| 545 |
+
{
|
| 546 |
+
"epoch": 36.93023255813954,
|
| 547 |
+
"eval_accuracy": 0.6730769230769231,
|
| 548 |
+
"eval_loss": 1.568899154663086,
|
| 549 |
+
"eval_runtime": 1.2885,
|
| 550 |
+
"eval_samples_per_second": 40.359,
|
| 551 |
+
"eval_steps_per_second": 1.552,
|
| 552 |
+
"step": 370
|
| 553 |
+
},
|
| 554 |
+
{
|
| 555 |
+
"epoch": 37.18604651162791,
|
| 556 |
+
"grad_norm": 31.557363510131836,
|
| 557 |
+
"learning_rate": 6.349206349206349e-06,
|
| 558 |
+
"loss": 0.265,
|
| 559 |
+
"step": 372
|
| 560 |
+
},
|
| 561 |
+
{
|
| 562 |
+
"epoch": 37.93023255813954,
|
| 563 |
+
"eval_accuracy": 0.6730769230769231,
|
| 564 |
+
"eval_loss": 1.6178526878356934,
|
| 565 |
+
"eval_runtime": 0.9634,
|
| 566 |
+
"eval_samples_per_second": 53.974,
|
| 567 |
+
"eval_steps_per_second": 2.076,
|
| 568 |
+
"step": 380
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 38.372093023255815,
|
| 572 |
+
"grad_norm": 27.208581924438477,
|
| 573 |
+
"learning_rate": 4.7619047619047615e-06,
|
| 574 |
+
"loss": 0.3189,
|
| 575 |
+
"step": 384
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 38.93023255813954,
|
| 579 |
+
"eval_accuracy": 0.6538461538461539,
|
| 580 |
+
"eval_loss": 1.5962971448898315,
|
| 581 |
+
"eval_runtime": 0.9485,
|
| 582 |
+
"eval_samples_per_second": 54.826,
|
| 583 |
+
"eval_steps_per_second": 2.109,
|
| 584 |
+
"step": 390
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 39.55813953488372,
|
| 588 |
+
"grad_norm": 30.211496353149414,
|
| 589 |
+
"learning_rate": 3.1746031746031746e-06,
|
| 590 |
+
"loss": 0.2976,
|
| 591 |
+
"step": 396
|
| 592 |
+
},
|
| 593 |
+
{
|
| 594 |
+
"epoch": 39.93023255813954,
|
| 595 |
+
"eval_accuracy": 0.6346153846153846,
|
| 596 |
+
"eval_loss": 1.5738095045089722,
|
| 597 |
+
"eval_runtime": 1.1624,
|
| 598 |
+
"eval_samples_per_second": 44.733,
|
| 599 |
+
"eval_steps_per_second": 1.721,
|
| 600 |
+
"step": 400
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 40.74418604651163,
|
| 604 |
+
"grad_norm": 16.808286666870117,
|
| 605 |
+
"learning_rate": 1.5873015873015873e-06,
|
| 606 |
+
"loss": 0.3097,
|
| 607 |
+
"step": 408
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 40.93023255813954,
|
| 611 |
+
"eval_accuracy": 0.6346153846153846,
|
| 612 |
+
"eval_loss": 1.578661561012268,
|
| 613 |
+
"eval_runtime": 0.9279,
|
| 614 |
+
"eval_samples_per_second": 56.041,
|
| 615 |
+
"eval_steps_per_second": 2.155,
|
| 616 |
+
"step": 410
|
| 617 |
+
},
|
| 618 |
+
{
|
| 619 |
+
"epoch": 41.93023255813954,
|
| 620 |
+
"grad_norm": 11.15967082977295,
|
| 621 |
+
"learning_rate": 0.0,
|
| 622 |
+
"loss": 0.269,
|
| 623 |
+
"step": 420
|
| 624 |
+
},
|
| 625 |
+
{
|
| 626 |
+
"epoch": 41.93023255813954,
|
| 627 |
+
"eval_accuracy": 0.6346153846153846,
|
| 628 |
+
"eval_loss": 1.5818698406219482,
|
| 629 |
+
"eval_runtime": 1.6499,
|
| 630 |
+
"eval_samples_per_second": 31.516,
|
| 631 |
+
"eval_steps_per_second": 1.212,
|
| 632 |
+
"step": 420
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 41.93023255813954,
|
| 636 |
+
"step": 420,
|
| 637 |
+
"total_flos": 1.8345690247389512e+18,
|
| 638 |
+
"train_loss": 1.2270459805216107,
|
| 639 |
+
"train_runtime": 1356.3174,
|
| 640 |
+
"train_samples_per_second": 42.176,
|
| 641 |
+
"train_steps_per_second": 0.31
|
| 642 |
+
}
|
| 643 |
+
],
|
| 644 |
+
"logging_steps": 12,
|
| 645 |
+
"max_steps": 420,
|
| 646 |
+
"num_input_tokens_seen": 0,
|
| 647 |
+
"num_train_epochs": 42,
|
| 648 |
+
"save_steps": 500,
|
| 649 |
+
"stateful_callbacks": {
|
| 650 |
+
"TrainerControl": {
|
| 651 |
+
"args": {
|
| 652 |
+
"should_epoch_stop": false,
|
| 653 |
+
"should_evaluate": false,
|
| 654 |
+
"should_log": false,
|
| 655 |
+
"should_save": true,
|
| 656 |
+
"should_training_stop": true
|
| 657 |
+
},
|
| 658 |
+
"attributes": {}
|
| 659 |
+
}
|
| 660 |
+
},
|
| 661 |
+
"total_flos": 1.8345690247389512e+18,
|
| 662 |
+
"train_batch_size": 32,
|
| 663 |
+
"trial_name": null,
|
| 664 |
+
"trial_params": null
|
| 665 |
+
}
|