End of training
Browse files- README.md +1 -1
- all_results.json +13 -0
- eval_results.json +8 -0
- runs/Feb10_02-57-08_e3629502e642/events.out.tfevents.1739156793.e3629502e642.69539.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +552 -0
README.md
CHANGED
|
@@ -18,7 +18,7 @@ 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: 0.
|
| 22 |
- Accuracy: 0.7308
|
| 23 |
|
| 24 |
## Model description
|
|
|
|
| 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: 0.8748
|
| 22 |
- Accuracy: 0.7308
|
| 23 |
|
| 24 |
## Model description
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 44.8421052631579,
|
| 3 |
+
"eval_accuracy": 0.7307692307692307,
|
| 4 |
+
"eval_loss": 0.8747740983963013,
|
| 5 |
+
"eval_runtime": 0.8857,
|
| 6 |
+
"eval_samples_per_second": 58.713,
|
| 7 |
+
"eval_steps_per_second": 2.258,
|
| 8 |
+
"total_flos": 8.412779636714373e+17,
|
| 9 |
+
"train_loss": 0.7686147186491225,
|
| 10 |
+
"train_runtime": 536.055,
|
| 11 |
+
"train_samples_per_second": 50.872,
|
| 12 |
+
"train_steps_per_second": 0.336
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 44.8421052631579,
|
| 3 |
+
"eval_accuracy": 0.7307692307692307,
|
| 4 |
+
"eval_loss": 0.8747740983963013,
|
| 5 |
+
"eval_runtime": 0.8857,
|
| 6 |
+
"eval_samples_per_second": 58.713,
|
| 7 |
+
"eval_steps_per_second": 2.258
|
| 8 |
+
}
|
runs/Feb10_02-57-08_e3629502e642/events.out.tfevents.1739156793.e3629502e642.69539.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:96fc582d5c53a6056be2b29c5a810778ca6e54c3c8dbda7d341213e89cfd47c8
|
| 3 |
+
size 411
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 44.8421052631579,
|
| 3 |
+
"total_flos": 8.412779636714373e+17,
|
| 4 |
+
"train_loss": 0.7686147186491225,
|
| 5 |
+
"train_runtime": 536.055,
|
| 6 |
+
"train_samples_per_second": 50.872,
|
| 7 |
+
"train_steps_per_second": 0.336
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,552 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.7307692307692307,
|
| 3 |
+
"best_model_checkpoint": "swinv2-tiny-patch4-window8-256-dmae-humeda-DAV56/checkpoint-128",
|
| 4 |
+
"epoch": 44.8421052631579,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 180,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.8421052631578947,
|
| 13 |
+
"eval_accuracy": 0.3269230769230769,
|
| 14 |
+
"eval_loss": 1.5618427991867065,
|
| 15 |
+
"eval_runtime": 0.8649,
|
| 16 |
+
"eval_samples_per_second": 60.123,
|
| 17 |
+
"eval_steps_per_second": 2.312,
|
| 18 |
+
"step": 4
|
| 19 |
+
},
|
| 20 |
+
{
|
| 21 |
+
"epoch": 1.8421052631578947,
|
| 22 |
+
"eval_accuracy": 0.4423076923076923,
|
| 23 |
+
"eval_loss": 1.5395886898040771,
|
| 24 |
+
"eval_runtime": 0.8206,
|
| 25 |
+
"eval_samples_per_second": 63.366,
|
| 26 |
+
"eval_steps_per_second": 2.437,
|
| 27 |
+
"step": 8
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 2.8421052631578947,
|
| 31 |
+
"grad_norm": 3.157193660736084,
|
| 32 |
+
"learning_rate": 1.3333333333333333e-05,
|
| 33 |
+
"loss": 1.7209,
|
| 34 |
+
"step": 12
|
| 35 |
+
},
|
| 36 |
+
{
|
| 37 |
+
"epoch": 2.8421052631578947,
|
| 38 |
+
"eval_accuracy": 0.3076923076923077,
|
| 39 |
+
"eval_loss": 1.5123951435089111,
|
| 40 |
+
"eval_runtime": 0.8398,
|
| 41 |
+
"eval_samples_per_second": 61.921,
|
| 42 |
+
"eval_steps_per_second": 2.382,
|
| 43 |
+
"step": 12
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 3.8421052631578947,
|
| 47 |
+
"eval_accuracy": 0.3269230769230769,
|
| 48 |
+
"eval_loss": 1.4679234027862549,
|
| 49 |
+
"eval_runtime": 1.2246,
|
| 50 |
+
"eval_samples_per_second": 42.462,
|
| 51 |
+
"eval_steps_per_second": 1.633,
|
| 52 |
+
"step": 16
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 4.842105263157895,
|
| 56 |
+
"eval_accuracy": 0.34615384615384615,
|
| 57 |
+
"eval_loss": 1.412973165512085,
|
| 58 |
+
"eval_runtime": 0.8428,
|
| 59 |
+
"eval_samples_per_second": 61.701,
|
| 60 |
+
"eval_steps_per_second": 2.373,
|
| 61 |
+
"step": 20
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 5.842105263157895,
|
| 65 |
+
"grad_norm": 4.583027362823486,
|
| 66 |
+
"learning_rate": 1.9932383577419432e-05,
|
| 67 |
+
"loss": 1.5864,
|
| 68 |
+
"step": 24
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 5.842105263157895,
|
| 72 |
+
"eval_accuracy": 0.5384615384615384,
|
| 73 |
+
"eval_loss": 1.31065034866333,
|
| 74 |
+
"eval_runtime": 0.8266,
|
| 75 |
+
"eval_samples_per_second": 62.908,
|
| 76 |
+
"eval_steps_per_second": 2.42,
|
| 77 |
+
"step": 24
|
| 78 |
+
},
|
| 79 |
+
{
|
| 80 |
+
"epoch": 6.842105263157895,
|
| 81 |
+
"eval_accuracy": 0.5384615384615384,
|
| 82 |
+
"eval_loss": 1.2111910581588745,
|
| 83 |
+
"eval_runtime": 0.8228,
|
| 84 |
+
"eval_samples_per_second": 63.201,
|
| 85 |
+
"eval_steps_per_second": 2.431,
|
| 86 |
+
"step": 28
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 7.842105263157895,
|
| 90 |
+
"eval_accuracy": 0.5961538461538461,
|
| 91 |
+
"eval_loss": 1.119398832321167,
|
| 92 |
+
"eval_runtime": 0.8365,
|
| 93 |
+
"eval_samples_per_second": 62.162,
|
| 94 |
+
"eval_steps_per_second": 2.391,
|
| 95 |
+
"step": 32
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 8.842105263157894,
|
| 99 |
+
"grad_norm": 5.296598434448242,
|
| 100 |
+
"learning_rate": 1.9396926207859085e-05,
|
| 101 |
+
"loss": 1.2629,
|
| 102 |
+
"step": 36
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 8.842105263157894,
|
| 106 |
+
"eval_accuracy": 0.5961538461538461,
|
| 107 |
+
"eval_loss": 1.0421706438064575,
|
| 108 |
+
"eval_runtime": 1.1444,
|
| 109 |
+
"eval_samples_per_second": 45.439,
|
| 110 |
+
"eval_steps_per_second": 1.748,
|
| 111 |
+
"step": 36
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 9.842105263157894,
|
| 115 |
+
"eval_accuracy": 0.6538461538461539,
|
| 116 |
+
"eval_loss": 0.9705998301506042,
|
| 117 |
+
"eval_runtime": 1.1966,
|
| 118 |
+
"eval_samples_per_second": 43.458,
|
| 119 |
+
"eval_steps_per_second": 1.671,
|
| 120 |
+
"step": 40
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 10.842105263157894,
|
| 124 |
+
"eval_accuracy": 0.6538461538461539,
|
| 125 |
+
"eval_loss": 0.9637848734855652,
|
| 126 |
+
"eval_runtime": 0.8287,
|
| 127 |
+
"eval_samples_per_second": 62.751,
|
| 128 |
+
"eval_steps_per_second": 2.414,
|
| 129 |
+
"step": 44
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 11.842105263157894,
|
| 133 |
+
"grad_norm": 8.424572944641113,
|
| 134 |
+
"learning_rate": 1.8354878114129368e-05,
|
| 135 |
+
"loss": 0.951,
|
| 136 |
+
"step": 48
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 11.842105263157894,
|
| 140 |
+
"eval_accuracy": 0.6153846153846154,
|
| 141 |
+
"eval_loss": 0.9906312227249146,
|
| 142 |
+
"eval_runtime": 0.8344,
|
| 143 |
+
"eval_samples_per_second": 62.318,
|
| 144 |
+
"eval_steps_per_second": 2.397,
|
| 145 |
+
"step": 48
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 12.842105263157894,
|
| 149 |
+
"eval_accuracy": 0.5961538461538461,
|
| 150 |
+
"eval_loss": 0.988998532295227,
|
| 151 |
+
"eval_runtime": 0.8289,
|
| 152 |
+
"eval_samples_per_second": 62.73,
|
| 153 |
+
"eval_steps_per_second": 2.413,
|
| 154 |
+
"step": 52
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 13.842105263157894,
|
| 158 |
+
"eval_accuracy": 0.6538461538461539,
|
| 159 |
+
"eval_loss": 0.9109674692153931,
|
| 160 |
+
"eval_runtime": 0.8337,
|
| 161 |
+
"eval_samples_per_second": 62.374,
|
| 162 |
+
"eval_steps_per_second": 2.399,
|
| 163 |
+
"step": 56
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 14.842105263157894,
|
| 167 |
+
"grad_norm": 13.505391120910645,
|
| 168 |
+
"learning_rate": 1.686241637868734e-05,
|
| 169 |
+
"loss": 0.7947,
|
| 170 |
+
"step": 60
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 14.842105263157894,
|
| 174 |
+
"eval_accuracy": 0.6730769230769231,
|
| 175 |
+
"eval_loss": 0.9282015562057495,
|
| 176 |
+
"eval_runtime": 1.167,
|
| 177 |
+
"eval_samples_per_second": 44.559,
|
| 178 |
+
"eval_steps_per_second": 1.714,
|
| 179 |
+
"step": 60
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 15.842105263157894,
|
| 183 |
+
"eval_accuracy": 0.6538461538461539,
|
| 184 |
+
"eval_loss": 0.931515634059906,
|
| 185 |
+
"eval_runtime": 0.8621,
|
| 186 |
+
"eval_samples_per_second": 60.315,
|
| 187 |
+
"eval_steps_per_second": 2.32,
|
| 188 |
+
"step": 64
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 16.842105263157894,
|
| 192 |
+
"eval_accuracy": 0.6153846153846154,
|
| 193 |
+
"eval_loss": 0.9229909181594849,
|
| 194 |
+
"eval_runtime": 0.8464,
|
| 195 |
+
"eval_samples_per_second": 61.434,
|
| 196 |
+
"eval_steps_per_second": 2.363,
|
| 197 |
+
"step": 68
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 17.842105263157894,
|
| 201 |
+
"grad_norm": 8.060040473937988,
|
| 202 |
+
"learning_rate": 1.5000000000000002e-05,
|
| 203 |
+
"loss": 0.7143,
|
| 204 |
+
"step": 72
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 17.842105263157894,
|
| 208 |
+
"eval_accuracy": 0.6538461538461539,
|
| 209 |
+
"eval_loss": 0.9067614674568176,
|
| 210 |
+
"eval_runtime": 0.8349,
|
| 211 |
+
"eval_samples_per_second": 62.281,
|
| 212 |
+
"eval_steps_per_second": 2.395,
|
| 213 |
+
"step": 72
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 18.842105263157894,
|
| 217 |
+
"eval_accuracy": 0.6153846153846154,
|
| 218 |
+
"eval_loss": 0.8996807932853699,
|
| 219 |
+
"eval_runtime": 0.8415,
|
| 220 |
+
"eval_samples_per_second": 61.795,
|
| 221 |
+
"eval_steps_per_second": 2.377,
|
| 222 |
+
"step": 76
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 19.842105263157894,
|
| 226 |
+
"eval_accuracy": 0.6923076923076923,
|
| 227 |
+
"eval_loss": 0.8648480772972107,
|
| 228 |
+
"eval_runtime": 1.1842,
|
| 229 |
+
"eval_samples_per_second": 43.912,
|
| 230 |
+
"eval_steps_per_second": 1.689,
|
| 231 |
+
"step": 80
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 20.842105263157894,
|
| 235 |
+
"grad_norm": 6.491458415985107,
|
| 236 |
+
"learning_rate": 1.2868032327110904e-05,
|
| 237 |
+
"loss": 0.6329,
|
| 238 |
+
"step": 84
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 20.842105263157894,
|
| 242 |
+
"eval_accuracy": 0.6538461538461539,
|
| 243 |
+
"eval_loss": 0.8624066710472107,
|
| 244 |
+
"eval_runtime": 1.178,
|
| 245 |
+
"eval_samples_per_second": 44.142,
|
| 246 |
+
"eval_steps_per_second": 1.698,
|
| 247 |
+
"step": 84
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 21.842105263157894,
|
| 251 |
+
"eval_accuracy": 0.6153846153846154,
|
| 252 |
+
"eval_loss": 0.8736539483070374,
|
| 253 |
+
"eval_runtime": 1.2262,
|
| 254 |
+
"eval_samples_per_second": 42.407,
|
| 255 |
+
"eval_steps_per_second": 1.631,
|
| 256 |
+
"step": 88
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 22.842105263157894,
|
| 260 |
+
"eval_accuracy": 0.6730769230769231,
|
| 261 |
+
"eval_loss": 0.8636062145233154,
|
| 262 |
+
"eval_runtime": 0.9335,
|
| 263 |
+
"eval_samples_per_second": 55.703,
|
| 264 |
+
"eval_steps_per_second": 2.142,
|
| 265 |
+
"step": 92
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 23.842105263157894,
|
| 269 |
+
"grad_norm": 13.519357681274414,
|
| 270 |
+
"learning_rate": 1.0581448289104759e-05,
|
| 271 |
+
"loss": 0.5508,
|
| 272 |
+
"step": 96
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 23.842105263157894,
|
| 276 |
+
"eval_accuracy": 0.6538461538461539,
|
| 277 |
+
"eval_loss": 0.8544560074806213,
|
| 278 |
+
"eval_runtime": 0.8147,
|
| 279 |
+
"eval_samples_per_second": 63.826,
|
| 280 |
+
"eval_steps_per_second": 2.455,
|
| 281 |
+
"step": 96
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 24.842105263157894,
|
| 285 |
+
"eval_accuracy": 0.6730769230769231,
|
| 286 |
+
"eval_loss": 0.8616629838943481,
|
| 287 |
+
"eval_runtime": 0.8014,
|
| 288 |
+
"eval_samples_per_second": 64.887,
|
| 289 |
+
"eval_steps_per_second": 2.496,
|
| 290 |
+
"step": 100
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 25.842105263157894,
|
| 294 |
+
"eval_accuracy": 0.6346153846153846,
|
| 295 |
+
"eval_loss": 0.8635188341140747,
|
| 296 |
+
"eval_runtime": 0.8235,
|
| 297 |
+
"eval_samples_per_second": 63.142,
|
| 298 |
+
"eval_steps_per_second": 2.429,
|
| 299 |
+
"step": 104
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 26.842105263157894,
|
| 303 |
+
"grad_norm": 8.143542289733887,
|
| 304 |
+
"learning_rate": 8.263518223330698e-06,
|
| 305 |
+
"loss": 0.5009,
|
| 306 |
+
"step": 108
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 26.842105263157894,
|
| 310 |
+
"eval_accuracy": 0.6346153846153846,
|
| 311 |
+
"eval_loss": 0.8650462627410889,
|
| 312 |
+
"eval_runtime": 0.8227,
|
| 313 |
+
"eval_samples_per_second": 63.207,
|
| 314 |
+
"eval_steps_per_second": 2.431,
|
| 315 |
+
"step": 108
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 27.842105263157894,
|
| 319 |
+
"eval_accuracy": 0.6538461538461539,
|
| 320 |
+
"eval_loss": 0.8637959361076355,
|
| 321 |
+
"eval_runtime": 1.2141,
|
| 322 |
+
"eval_samples_per_second": 42.831,
|
| 323 |
+
"eval_steps_per_second": 1.647,
|
| 324 |
+
"step": 112
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 28.842105263157894,
|
| 328 |
+
"eval_accuracy": 0.6538461538461539,
|
| 329 |
+
"eval_loss": 0.873045802116394,
|
| 330 |
+
"eval_runtime": 1.1581,
|
| 331 |
+
"eval_samples_per_second": 44.901,
|
| 332 |
+
"eval_steps_per_second": 1.727,
|
| 333 |
+
"step": 116
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 29.842105263157894,
|
| 337 |
+
"grad_norm": 10.985040664672852,
|
| 338 |
+
"learning_rate": 6.039202339608432e-06,
|
| 339 |
+
"loss": 0.5286,
|
| 340 |
+
"step": 120
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 29.842105263157894,
|
| 344 |
+
"eval_accuracy": 0.6346153846153846,
|
| 345 |
+
"eval_loss": 0.8886079788208008,
|
| 346 |
+
"eval_runtime": 1.168,
|
| 347 |
+
"eval_samples_per_second": 44.522,
|
| 348 |
+
"eval_steps_per_second": 1.712,
|
| 349 |
+
"step": 120
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 30.842105263157894,
|
| 353 |
+
"eval_accuracy": 0.6538461538461539,
|
| 354 |
+
"eval_loss": 0.882747232913971,
|
| 355 |
+
"eval_runtime": 1.0188,
|
| 356 |
+
"eval_samples_per_second": 51.04,
|
| 357 |
+
"eval_steps_per_second": 1.963,
|
| 358 |
+
"step": 124
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 31.842105263157894,
|
| 362 |
+
"eval_accuracy": 0.7307692307692307,
|
| 363 |
+
"eval_loss": 0.8747740983963013,
|
| 364 |
+
"eval_runtime": 0.8145,
|
| 365 |
+
"eval_samples_per_second": 63.847,
|
| 366 |
+
"eval_steps_per_second": 2.456,
|
| 367 |
+
"step": 128
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 32.8421052631579,
|
| 371 |
+
"grad_norm": 7.003128528594971,
|
| 372 |
+
"learning_rate": 4.028414082972141e-06,
|
| 373 |
+
"loss": 0.4559,
|
| 374 |
+
"step": 132
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 32.8421052631579,
|
| 378 |
+
"eval_accuracy": 0.7115384615384616,
|
| 379 |
+
"eval_loss": 0.8670610785484314,
|
| 380 |
+
"eval_runtime": 0.8405,
|
| 381 |
+
"eval_samples_per_second": 61.865,
|
| 382 |
+
"eval_steps_per_second": 2.379,
|
| 383 |
+
"step": 132
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 33.8421052631579,
|
| 387 |
+
"eval_accuracy": 0.6730769230769231,
|
| 388 |
+
"eval_loss": 0.8727457523345947,
|
| 389 |
+
"eval_runtime": 0.8367,
|
| 390 |
+
"eval_samples_per_second": 62.15,
|
| 391 |
+
"eval_steps_per_second": 2.39,
|
| 392 |
+
"step": 136
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 34.8421052631579,
|
| 396 |
+
"eval_accuracy": 0.7115384615384616,
|
| 397 |
+
"eval_loss": 0.8754667043685913,
|
| 398 |
+
"eval_runtime": 0.8489,
|
| 399 |
+
"eval_samples_per_second": 61.254,
|
| 400 |
+
"eval_steps_per_second": 2.356,
|
| 401 |
+
"step": 140
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 35.8421052631579,
|
| 405 |
+
"grad_norm": 7.350639343261719,
|
| 406 |
+
"learning_rate": 2.339555568810221e-06,
|
| 407 |
+
"loss": 0.4704,
|
| 408 |
+
"step": 144
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 35.8421052631579,
|
| 412 |
+
"eval_accuracy": 0.7307692307692307,
|
| 413 |
+
"eval_loss": 0.8759797215461731,
|
| 414 |
+
"eval_runtime": 0.845,
|
| 415 |
+
"eval_samples_per_second": 61.538,
|
| 416 |
+
"eval_steps_per_second": 2.367,
|
| 417 |
+
"step": 144
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 36.8421052631579,
|
| 421 |
+
"eval_accuracy": 0.7307692307692307,
|
| 422 |
+
"eval_loss": 0.8786163330078125,
|
| 423 |
+
"eval_runtime": 1.2012,
|
| 424 |
+
"eval_samples_per_second": 43.291,
|
| 425 |
+
"eval_steps_per_second": 1.665,
|
| 426 |
+
"step": 148
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 37.8421052631579,
|
| 430 |
+
"eval_accuracy": 0.7307692307692307,
|
| 431 |
+
"eval_loss": 0.878113329410553,
|
| 432 |
+
"eval_runtime": 0.8301,
|
| 433 |
+
"eval_samples_per_second": 62.641,
|
| 434 |
+
"eval_steps_per_second": 2.409,
|
| 435 |
+
"step": 152
|
| 436 |
+
},
|
| 437 |
+
{
|
| 438 |
+
"epoch": 38.8421052631579,
|
| 439 |
+
"grad_norm": 10.532869338989258,
|
| 440 |
+
"learning_rate": 1.0636735967658785e-06,
|
| 441 |
+
"loss": 0.4582,
|
| 442 |
+
"step": 156
|
| 443 |
+
},
|
| 444 |
+
{
|
| 445 |
+
"epoch": 38.8421052631579,
|
| 446 |
+
"eval_accuracy": 0.7307692307692307,
|
| 447 |
+
"eval_loss": 0.8771308660507202,
|
| 448 |
+
"eval_runtime": 0.807,
|
| 449 |
+
"eval_samples_per_second": 64.435,
|
| 450 |
+
"eval_steps_per_second": 2.478,
|
| 451 |
+
"step": 156
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 39.8421052631579,
|
| 455 |
+
"eval_accuracy": 0.7307692307692307,
|
| 456 |
+
"eval_loss": 0.8753777146339417,
|
| 457 |
+
"eval_runtime": 0.8106,
|
| 458 |
+
"eval_samples_per_second": 64.147,
|
| 459 |
+
"eval_steps_per_second": 2.467,
|
| 460 |
+
"step": 160
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 40.8421052631579,
|
| 464 |
+
"eval_accuracy": 0.7307692307692307,
|
| 465 |
+
"eval_loss": 0.8740986585617065,
|
| 466 |
+
"eval_runtime": 1.1627,
|
| 467 |
+
"eval_samples_per_second": 44.725,
|
| 468 |
+
"eval_steps_per_second": 1.72,
|
| 469 |
+
"step": 164
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 41.8421052631579,
|
| 473 |
+
"grad_norm": 6.582579135894775,
|
| 474 |
+
"learning_rate": 2.6955129420176193e-07,
|
| 475 |
+
"loss": 0.4538,
|
| 476 |
+
"step": 168
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 41.8421052631579,
|
| 480 |
+
"eval_accuracy": 0.7307692307692307,
|
| 481 |
+
"eval_loss": 0.8742378950119019,
|
| 482 |
+
"eval_runtime": 0.8152,
|
| 483 |
+
"eval_samples_per_second": 63.791,
|
| 484 |
+
"eval_steps_per_second": 2.453,
|
| 485 |
+
"step": 168
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 42.8421052631579,
|
| 489 |
+
"eval_accuracy": 0.7307692307692307,
|
| 490 |
+
"eval_loss": 0.873984694480896,
|
| 491 |
+
"eval_runtime": 0.8202,
|
| 492 |
+
"eval_samples_per_second": 63.397,
|
| 493 |
+
"eval_steps_per_second": 2.438,
|
| 494 |
+
"step": 172
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 43.8421052631579,
|
| 498 |
+
"eval_accuracy": 0.7307692307692307,
|
| 499 |
+
"eval_loss": 0.8740047216415405,
|
| 500 |
+
"eval_runtime": 0.8433,
|
| 501 |
+
"eval_samples_per_second": 61.665,
|
| 502 |
+
"eval_steps_per_second": 2.372,
|
| 503 |
+
"step": 176
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 44.8421052631579,
|
| 507 |
+
"grad_norm": 7.687587738037109,
|
| 508 |
+
"learning_rate": 1.8802955149865854e-09,
|
| 509 |
+
"loss": 0.4476,
|
| 510 |
+
"step": 180
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 44.8421052631579,
|
| 514 |
+
"eval_accuracy": 0.7307692307692307,
|
| 515 |
+
"eval_loss": 0.8741353750228882,
|
| 516 |
+
"eval_runtime": 1.1096,
|
| 517 |
+
"eval_samples_per_second": 46.864,
|
| 518 |
+
"eval_steps_per_second": 1.802,
|
| 519 |
+
"step": 180
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 44.8421052631579,
|
| 523 |
+
"step": 180,
|
| 524 |
+
"total_flos": 8.412779636714373e+17,
|
| 525 |
+
"train_loss": 0.7686147186491225,
|
| 526 |
+
"train_runtime": 536.055,
|
| 527 |
+
"train_samples_per_second": 50.872,
|
| 528 |
+
"train_steps_per_second": 0.336
|
| 529 |
+
}
|
| 530 |
+
],
|
| 531 |
+
"logging_steps": 12,
|
| 532 |
+
"max_steps": 180,
|
| 533 |
+
"num_input_tokens_seen": 0,
|
| 534 |
+
"num_train_epochs": 45,
|
| 535 |
+
"save_steps": 500,
|
| 536 |
+
"stateful_callbacks": {
|
| 537 |
+
"TrainerControl": {
|
| 538 |
+
"args": {
|
| 539 |
+
"should_epoch_stop": false,
|
| 540 |
+
"should_evaluate": false,
|
| 541 |
+
"should_log": false,
|
| 542 |
+
"should_save": true,
|
| 543 |
+
"should_training_stop": true
|
| 544 |
+
},
|
| 545 |
+
"attributes": {}
|
| 546 |
+
}
|
| 547 |
+
},
|
| 548 |
+
"total_flos": 8.412779636714373e+17,
|
| 549 |
+
"train_batch_size": 32,
|
| 550 |
+
"trial_name": null,
|
| 551 |
+
"trial_params": null
|
| 552 |
+
}
|