danbooru-tags-classification3
Browse files- README.md +13 -8
- all_results.json +11 -11
- eval_results.json +6 -6
- model.safetensors +1 -1
- train_results.json +6 -6
- trainer_state.json +259 -82
- training_args.bin +1 -1
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: 0.
|
| 21 |
-
- Accuracy: 0.
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
@@ -46,17 +46,22 @@ The following hyperparameters were used during training:
|
|
| 46 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
- lr_scheduler_type: linear
|
| 48 |
- lr_scheduler_warmup_ratio: 0.1
|
| 49 |
-
- num_epochs:
|
| 50 |
|
| 51 |
### Training results
|
| 52 |
|
| 53 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 54 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 55 |
-
| 0.
|
| 56 |
-
| 0.
|
| 57 |
-
| 0.
|
| 58 |
-
| 0.
|
| 59 |
-
| 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 60 |
|
| 61 |
|
| 62 |
### Framework versions
|
|
|
|
| 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: 0.3251
|
| 21 |
+
- Accuracy: 0.8761
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 46 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 47 |
- lr_scheduler_type: linear
|
| 48 |
- lr_scheduler_warmup_ratio: 0.1
|
| 49 |
+
- num_epochs: 10
|
| 50 |
|
| 51 |
### Training results
|
| 52 |
|
| 53 |
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|
| 54 |
|:-------------:|:-----:|:----:|:---------------:|:--------:|
|
| 55 |
+
| 0.6237 | 0.99 | 43 | 0.5558 | 0.7302 |
|
| 56 |
+
| 0.462 | 1.99 | 86 | 0.4139 | 0.8248 |
|
| 57 |
+
| 0.4174 | 2.98 | 129 | 0.3704 | 0.8468 |
|
| 58 |
+
| 0.4089 | 4.0 | 173 | 0.4138 | 0.8231 |
|
| 59 |
+
| 0.3656 | 4.99 | 216 | 0.3322 | 0.8688 |
|
| 60 |
+
| 0.3282 | 5.99 | 259 | 0.3251 | 0.8761 |
|
| 61 |
+
| 0.3251 | 6.98 | 302 | 0.3296 | 0.8696 |
|
| 62 |
+
| 0.3025 | 8.0 | 346 | 0.3221 | 0.8720 |
|
| 63 |
+
| 0.2937 | 8.99 | 389 | 0.3218 | 0.8729 |
|
| 64 |
+
| 0.2812 | 9.94 | 430 | 0.3219 | 0.8753 |
|
| 65 |
|
| 66 |
|
| 67 |
### Framework versions
|
all_results.json
CHANGED
|
@@ -1,13 +1,13 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 1.
|
| 8 |
-
"total_flos":
|
| 9 |
-
"train_loss":
|
| 10 |
-
"train_runtime":
|
| 11 |
-
"train_samples_per_second":
|
| 12 |
-
"train_steps_per_second": 0.
|
| 13 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 9.94,
|
| 3 |
+
"eval_accuracy": 0.876120619396903,
|
| 4 |
+
"eval_loss": 0.32511940598487854,
|
| 5 |
+
"eval_runtime": 16.2518,
|
| 6 |
+
"eval_samples_per_second": 75.499,
|
| 7 |
+
"eval_steps_per_second": 1.231,
|
| 8 |
+
"total_flos": 8.584102532322165e+18,
|
| 9 |
+
"train_loss": 0.7598524143529493,
|
| 10 |
+
"train_runtime": 2853.5799,
|
| 11 |
+
"train_samples_per_second": 38.688,
|
| 12 |
+
"train_steps_per_second": 0.151
|
| 13 |
}
|
eval_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"eval_accuracy": 0.
|
| 4 |
-
"eval_loss": 0.
|
| 5 |
-
"eval_runtime":
|
| 6 |
-
"eval_samples_per_second":
|
| 7 |
-
"eval_steps_per_second": 1.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 9.94,
|
| 3 |
+
"eval_accuracy": 0.876120619396903,
|
| 4 |
+
"eval_loss": 0.32511940598487854,
|
| 5 |
+
"eval_runtime": 16.2518,
|
| 6 |
+
"eval_samples_per_second": 75.499,
|
| 7 |
+
"eval_steps_per_second": 1.231
|
| 8 |
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 346293856
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7ba84f00089df18958c73d41ea99c1d8f356b54743e3c41b5a7a29b20b411765
|
| 3 |
size 346293856
|
train_results.json
CHANGED
|
@@ -1,8 +1,8 @@
|
|
| 1 |
{
|
| 2 |
-
"epoch":
|
| 3 |
-
"total_flos":
|
| 4 |
-
"train_loss":
|
| 5 |
-
"train_runtime":
|
| 6 |
-
"train_samples_per_second":
|
| 7 |
-
"train_steps_per_second": 0.
|
| 8 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"epoch": 9.94,
|
| 3 |
+
"total_flos": 8.584102532322165e+18,
|
| 4 |
+
"train_loss": 0.7598524143529493,
|
| 5 |
+
"train_runtime": 2853.5799,
|
| 6 |
+
"train_samples_per_second": 38.688,
|
| 7 |
+
"train_steps_per_second": 0.151
|
| 8 |
}
|
trainer_state.json
CHANGED
|
@@ -1,199 +1,376 @@
|
|
| 1 |
{
|
| 2 |
-
"best_metric": 0.
|
| 3 |
-
"best_model_checkpoint": "/kaggle/working/out/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": 0.23,
|
| 13 |
-
"learning_rate":
|
| 14 |
-
"loss":
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"epoch": 0.46,
|
| 19 |
-
"learning_rate":
|
| 20 |
-
"loss":
|
| 21 |
"step": 20
|
| 22 |
},
|
| 23 |
{
|
| 24 |
"epoch": 0.69,
|
| 25 |
-
"learning_rate":
|
| 26 |
-
"loss":
|
| 27 |
"step": 30
|
| 28 |
},
|
| 29 |
{
|
| 30 |
"epoch": 0.92,
|
| 31 |
-
"learning_rate": 4.
|
| 32 |
-
"loss": 0.
|
| 33 |
"step": 40
|
| 34 |
},
|
| 35 |
{
|
| 36 |
"epoch": 0.99,
|
| 37 |
-
"eval_accuracy": 0.
|
| 38 |
-
"eval_loss": 0.
|
| 39 |
-
"eval_runtime":
|
| 40 |
-
"eval_samples_per_second":
|
| 41 |
-
"eval_steps_per_second": 0.
|
| 42 |
"step": 43
|
| 43 |
},
|
| 44 |
{
|
| 45 |
"epoch": 1.16,
|
| 46 |
-
"learning_rate": 4.
|
| 47 |
-
"loss": 0.
|
| 48 |
"step": 50
|
| 49 |
},
|
| 50 |
{
|
| 51 |
"epoch": 1.39,
|
| 52 |
-
"learning_rate": 4.
|
| 53 |
-
"loss": 0.
|
| 54 |
"step": 60
|
| 55 |
},
|
| 56 |
{
|
| 57 |
"epoch": 1.62,
|
| 58 |
-
"learning_rate":
|
| 59 |
-
"loss": 0.
|
| 60 |
"step": 70
|
| 61 |
},
|
| 62 |
{
|
| 63 |
"epoch": 1.85,
|
| 64 |
-
"learning_rate":
|
| 65 |
-
"loss": 0.
|
| 66 |
"step": 80
|
| 67 |
},
|
| 68 |
{
|
| 69 |
"epoch": 1.99,
|
| 70 |
-
"eval_accuracy": 0.
|
| 71 |
-
"eval_loss": 0.
|
| 72 |
-
"eval_runtime": 15.
|
| 73 |
-
"eval_samples_per_second":
|
| 74 |
-
"eval_steps_per_second": 1.
|
| 75 |
"step": 86
|
| 76 |
},
|
| 77 |
{
|
| 78 |
"epoch": 2.08,
|
| 79 |
-
"learning_rate":
|
| 80 |
-
"loss": 0.
|
| 81 |
"step": 90
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"epoch": 2.31,
|
| 85 |
-
"learning_rate":
|
| 86 |
-
"loss": 0.
|
| 87 |
"step": 100
|
| 88 |
},
|
| 89 |
{
|
| 90 |
"epoch": 2.54,
|
| 91 |
-
"learning_rate":
|
| 92 |
-
"loss": 0.
|
| 93 |
"step": 110
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 2.77,
|
| 97 |
-
"learning_rate":
|
| 98 |
-
"loss": 0.
|
| 99 |
"step": 120
|
| 100 |
},
|
| 101 |
{
|
| 102 |
"epoch": 2.98,
|
| 103 |
-
"eval_accuracy": 0.
|
| 104 |
-
"eval_loss": 0.
|
| 105 |
-
"eval_runtime": 15.
|
| 106 |
-
"eval_samples_per_second":
|
| 107 |
-
"eval_steps_per_second": 1.
|
| 108 |
"step": 129
|
| 109 |
},
|
| 110 |
{
|
| 111 |
"epoch": 3.01,
|
| 112 |
-
"learning_rate":
|
| 113 |
-
"loss": 0.
|
| 114 |
"step": 130
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 3.24,
|
| 118 |
-
"learning_rate":
|
| 119 |
-
"loss": 0.
|
| 120 |
"step": 140
|
| 121 |
},
|
| 122 |
{
|
| 123 |
"epoch": 3.47,
|
| 124 |
-
"learning_rate":
|
| 125 |
-
"loss": 0.
|
| 126 |
"step": 150
|
| 127 |
},
|
| 128 |
{
|
| 129 |
"epoch": 3.7,
|
| 130 |
-
"learning_rate":
|
| 131 |
-
"loss": 0.
|
| 132 |
"step": 160
|
| 133 |
},
|
| 134 |
{
|
| 135 |
"epoch": 3.93,
|
| 136 |
-
"learning_rate":
|
| 137 |
-
"loss": 0.
|
| 138 |
"step": 170
|
| 139 |
},
|
| 140 |
{
|
| 141 |
"epoch": 4.0,
|
| 142 |
-
"eval_accuracy": 0.
|
| 143 |
-
"eval_loss": 0.
|
| 144 |
-
"eval_runtime": 15.
|
| 145 |
-
"eval_samples_per_second":
|
| 146 |
-
"eval_steps_per_second": 1.
|
| 147 |
"step": 173
|
| 148 |
},
|
| 149 |
{
|
| 150 |
"epoch": 4.16,
|
| 151 |
-
"learning_rate":
|
| 152 |
-
"loss": 0.
|
| 153 |
"step": 180
|
| 154 |
},
|
| 155 |
{
|
| 156 |
"epoch": 4.39,
|
| 157 |
-
"learning_rate":
|
| 158 |
-
"loss": 0.
|
| 159 |
"step": 190
|
| 160 |
},
|
| 161 |
{
|
| 162 |
"epoch": 4.62,
|
| 163 |
-
"learning_rate":
|
| 164 |
-
"loss": 0.
|
| 165 |
"step": 200
|
| 166 |
},
|
| 167 |
{
|
| 168 |
"epoch": 4.86,
|
| 169 |
-
"learning_rate":
|
| 170 |
-
"loss": 0.
|
| 171 |
"step": 210
|
| 172 |
},
|
| 173 |
{
|
| 174 |
-
"epoch": 4.
|
| 175 |
-
"eval_accuracy": 0.
|
| 176 |
-
"eval_loss": 0.
|
| 177 |
-
"eval_runtime": 15.
|
| 178 |
-
"eval_samples_per_second": 78.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 179 |
"eval_steps_per_second": 1.281,
|
| 180 |
-
"step":
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 181 |
},
|
| 182 |
{
|
| 183 |
-
"epoch":
|
| 184 |
-
"step":
|
| 185 |
-
"total_flos":
|
| 186 |
-
"train_loss":
|
| 187 |
-
"train_runtime":
|
| 188 |
-
"train_samples_per_second":
|
| 189 |
-
"train_steps_per_second": 0.
|
| 190 |
}
|
| 191 |
],
|
| 192 |
"logging_steps": 10,
|
| 193 |
-
"max_steps":
|
| 194 |
-
"num_train_epochs":
|
| 195 |
"save_steps": 500,
|
| 196 |
-
"total_flos":
|
| 197 |
"trial_name": null,
|
| 198 |
"trial_params": null
|
| 199 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"best_metric": 0.876120619396903,
|
| 3 |
+
"best_model_checkpoint": "/kaggle/working/out/checkpoint-259",
|
| 4 |
+
"epoch": 9.942196531791907,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 430,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.23,
|
| 13 |
+
"learning_rate": 1.1627906976744187e-05,
|
| 14 |
+
"loss": 10.3069,
|
| 15 |
"step": 10
|
| 16 |
},
|
| 17 |
{
|
| 18 |
"epoch": 0.46,
|
| 19 |
+
"learning_rate": 2.3255813953488374e-05,
|
| 20 |
+
"loss": 6.3401,
|
| 21 |
"step": 20
|
| 22 |
},
|
| 23 |
{
|
| 24 |
"epoch": 0.69,
|
| 25 |
+
"learning_rate": 3.488372093023256e-05,
|
| 26 |
+
"loss": 1.3487,
|
| 27 |
"step": 30
|
| 28 |
},
|
| 29 |
{
|
| 30 |
"epoch": 0.92,
|
| 31 |
+
"learning_rate": 4.651162790697675e-05,
|
| 32 |
+
"loss": 0.6237,
|
| 33 |
"step": 40
|
| 34 |
},
|
| 35 |
{
|
| 36 |
"epoch": 0.99,
|
| 37 |
+
"eval_accuracy": 0.7302363488182559,
|
| 38 |
+
"eval_loss": 0.55576092004776,
|
| 39 |
+
"eval_runtime": 23.793,
|
| 40 |
+
"eval_samples_per_second": 51.57,
|
| 41 |
+
"eval_steps_per_second": 0.841,
|
| 42 |
"step": 43
|
| 43 |
},
|
| 44 |
{
|
| 45 |
"epoch": 1.16,
|
| 46 |
+
"learning_rate": 4.9095607235142123e-05,
|
| 47 |
+
"loss": 0.5497,
|
| 48 |
"step": 50
|
| 49 |
},
|
| 50 |
{
|
| 51 |
"epoch": 1.39,
|
| 52 |
+
"learning_rate": 4.780361757105943e-05,
|
| 53 |
+
"loss": 0.5057,
|
| 54 |
"step": 60
|
| 55 |
},
|
| 56 |
{
|
| 57 |
"epoch": 1.62,
|
| 58 |
+
"learning_rate": 4.651162790697675e-05,
|
| 59 |
+
"loss": 0.5098,
|
| 60 |
"step": 70
|
| 61 |
},
|
| 62 |
{
|
| 63 |
"epoch": 1.85,
|
| 64 |
+
"learning_rate": 4.521963824289406e-05,
|
| 65 |
+
"loss": 0.462,
|
| 66 |
"step": 80
|
| 67 |
},
|
| 68 |
{
|
| 69 |
"epoch": 1.99,
|
| 70 |
+
"eval_accuracy": 0.8247758761206194,
|
| 71 |
+
"eval_loss": 0.4138629734516144,
|
| 72 |
+
"eval_runtime": 15.5428,
|
| 73 |
+
"eval_samples_per_second": 78.943,
|
| 74 |
+
"eval_steps_per_second": 1.287,
|
| 75 |
"step": 86
|
| 76 |
},
|
| 77 |
{
|
| 78 |
"epoch": 2.08,
|
| 79 |
+
"learning_rate": 4.392764857881137e-05,
|
| 80 |
+
"loss": 0.4337,
|
| 81 |
"step": 90
|
| 82 |
},
|
| 83 |
{
|
| 84 |
"epoch": 2.31,
|
| 85 |
+
"learning_rate": 4.263565891472868e-05,
|
| 86 |
+
"loss": 0.4433,
|
| 87 |
"step": 100
|
| 88 |
},
|
| 89 |
{
|
| 90 |
"epoch": 2.54,
|
| 91 |
+
"learning_rate": 4.1343669250646e-05,
|
| 92 |
+
"loss": 0.4448,
|
| 93 |
"step": 110
|
| 94 |
},
|
| 95 |
{
|
| 96 |
"epoch": 2.77,
|
| 97 |
+
"learning_rate": 4.005167958656331e-05,
|
| 98 |
+
"loss": 0.4174,
|
| 99 |
"step": 120
|
| 100 |
},
|
| 101 |
{
|
| 102 |
"epoch": 2.98,
|
| 103 |
+
"eval_accuracy": 0.8467807660961695,
|
| 104 |
+
"eval_loss": 0.37035566568374634,
|
| 105 |
+
"eval_runtime": 15.694,
|
| 106 |
+
"eval_samples_per_second": 78.183,
|
| 107 |
+
"eval_steps_per_second": 1.274,
|
| 108 |
"step": 129
|
| 109 |
},
|
| 110 |
{
|
| 111 |
"epoch": 3.01,
|
| 112 |
+
"learning_rate": 3.875968992248062e-05,
|
| 113 |
+
"loss": 0.4126,
|
| 114 |
"step": 130
|
| 115 |
},
|
| 116 |
{
|
| 117 |
"epoch": 3.24,
|
| 118 |
+
"learning_rate": 3.746770025839794e-05,
|
| 119 |
+
"loss": 0.3779,
|
| 120 |
"step": 140
|
| 121 |
},
|
| 122 |
{
|
| 123 |
"epoch": 3.47,
|
| 124 |
+
"learning_rate": 3.617571059431525e-05,
|
| 125 |
+
"loss": 0.4097,
|
| 126 |
"step": 150
|
| 127 |
},
|
| 128 |
{
|
| 129 |
"epoch": 3.7,
|
| 130 |
+
"learning_rate": 3.488372093023256e-05,
|
| 131 |
+
"loss": 0.3719,
|
| 132 |
"step": 160
|
| 133 |
},
|
| 134 |
{
|
| 135 |
"epoch": 3.93,
|
| 136 |
+
"learning_rate": 3.359173126614987e-05,
|
| 137 |
+
"loss": 0.4089,
|
| 138 |
"step": 170
|
| 139 |
},
|
| 140 |
{
|
| 141 |
"epoch": 4.0,
|
| 142 |
+
"eval_accuracy": 0.8231458842705787,
|
| 143 |
+
"eval_loss": 0.41381287574768066,
|
| 144 |
+
"eval_runtime": 15.3608,
|
| 145 |
+
"eval_samples_per_second": 79.878,
|
| 146 |
+
"eval_steps_per_second": 1.302,
|
| 147 |
"step": 173
|
| 148 |
},
|
| 149 |
{
|
| 150 |
"epoch": 4.16,
|
| 151 |
+
"learning_rate": 3.229974160206719e-05,
|
| 152 |
+
"loss": 0.3954,
|
| 153 |
"step": 180
|
| 154 |
},
|
| 155 |
{
|
| 156 |
"epoch": 4.39,
|
| 157 |
+
"learning_rate": 3.1007751937984497e-05,
|
| 158 |
+
"loss": 0.3791,
|
| 159 |
"step": 190
|
| 160 |
},
|
| 161 |
{
|
| 162 |
"epoch": 4.62,
|
| 163 |
+
"learning_rate": 2.971576227390181e-05,
|
| 164 |
+
"loss": 0.3473,
|
| 165 |
"step": 200
|
| 166 |
},
|
| 167 |
{
|
| 168 |
"epoch": 4.86,
|
| 169 |
+
"learning_rate": 2.842377260981912e-05,
|
| 170 |
+
"loss": 0.3656,
|
| 171 |
"step": 210
|
| 172 |
},
|
| 173 |
{
|
| 174 |
+
"epoch": 4.99,
|
| 175 |
+
"eval_accuracy": 0.8687856560717196,
|
| 176 |
+
"eval_loss": 0.3321658670902252,
|
| 177 |
+
"eval_runtime": 15.6831,
|
| 178 |
+
"eval_samples_per_second": 78.237,
|
| 179 |
+
"eval_steps_per_second": 1.275,
|
| 180 |
+
"step": 216
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 5.09,
|
| 184 |
+
"learning_rate": 2.7131782945736434e-05,
|
| 185 |
+
"loss": 0.3504,
|
| 186 |
+
"step": 220
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 5.32,
|
| 190 |
+
"learning_rate": 2.5839793281653746e-05,
|
| 191 |
+
"loss": 0.3416,
|
| 192 |
+
"step": 230
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 5.55,
|
| 196 |
+
"learning_rate": 2.4547803617571062e-05,
|
| 197 |
+
"loss": 0.3339,
|
| 198 |
+
"step": 240
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 5.78,
|
| 202 |
+
"learning_rate": 2.3255813953488374e-05,
|
| 203 |
+
"loss": 0.3282,
|
| 204 |
+
"step": 250
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 5.99,
|
| 208 |
+
"eval_accuracy": 0.876120619396903,
|
| 209 |
+
"eval_loss": 0.32511940598487854,
|
| 210 |
+
"eval_runtime": 15.6099,
|
| 211 |
+
"eval_samples_per_second": 78.604,
|
| 212 |
"eval_steps_per_second": 1.281,
|
| 213 |
+
"step": 259
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 6.01,
|
| 217 |
+
"learning_rate": 2.1963824289405686e-05,
|
| 218 |
+
"loss": 0.3505,
|
| 219 |
+
"step": 260
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 6.24,
|
| 223 |
+
"learning_rate": 2.0671834625323e-05,
|
| 224 |
+
"loss": 0.3177,
|
| 225 |
+
"step": 270
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 6.47,
|
| 229 |
+
"learning_rate": 1.937984496124031e-05,
|
| 230 |
+
"loss": 0.2742,
|
| 231 |
+
"step": 280
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 6.71,
|
| 235 |
+
"learning_rate": 1.8087855297157624e-05,
|
| 236 |
+
"loss": 0.3334,
|
| 237 |
+
"step": 290
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 6.94,
|
| 241 |
+
"learning_rate": 1.6795865633074936e-05,
|
| 242 |
+
"loss": 0.3251,
|
| 243 |
+
"step": 300
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 6.98,
|
| 247 |
+
"eval_accuracy": 0.86960065199674,
|
| 248 |
+
"eval_loss": 0.3296392858028412,
|
| 249 |
+
"eval_runtime": 15.5863,
|
| 250 |
+
"eval_samples_per_second": 78.723,
|
| 251 |
+
"eval_steps_per_second": 1.283,
|
| 252 |
+
"step": 302
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 7.17,
|
| 256 |
+
"learning_rate": 1.5503875968992248e-05,
|
| 257 |
+
"loss": 0.316,
|
| 258 |
+
"step": 310
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 7.4,
|
| 262 |
+
"learning_rate": 1.421188630490956e-05,
|
| 263 |
+
"loss": 0.3163,
|
| 264 |
+
"step": 320
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 7.63,
|
| 268 |
+
"learning_rate": 1.2919896640826873e-05,
|
| 269 |
+
"loss": 0.3176,
|
| 270 |
+
"step": 330
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 7.86,
|
| 274 |
+
"learning_rate": 1.1627906976744187e-05,
|
| 275 |
+
"loss": 0.3025,
|
| 276 |
+
"step": 340
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 8.0,
|
| 280 |
+
"eval_accuracy": 0.8720456397718012,
|
| 281 |
+
"eval_loss": 0.3221036493778229,
|
| 282 |
+
"eval_runtime": 15.5655,
|
| 283 |
+
"eval_samples_per_second": 78.828,
|
| 284 |
+
"eval_steps_per_second": 1.285,
|
| 285 |
+
"step": 346
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 8.09,
|
| 289 |
+
"learning_rate": 1.03359173126615e-05,
|
| 290 |
+
"loss": 0.3025,
|
| 291 |
+
"step": 350
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 8.32,
|
| 295 |
+
"learning_rate": 9.043927648578812e-06,
|
| 296 |
+
"loss": 0.2936,
|
| 297 |
+
"step": 360
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 8.55,
|
| 301 |
+
"learning_rate": 7.751937984496124e-06,
|
| 302 |
+
"loss": 0.2871,
|
| 303 |
+
"step": 370
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 8.79,
|
| 307 |
+
"learning_rate": 6.4599483204134365e-06,
|
| 308 |
+
"loss": 0.2937,
|
| 309 |
+
"step": 380
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 8.99,
|
| 313 |
+
"eval_accuracy": 0.8728606356968215,
|
| 314 |
+
"eval_loss": 0.32183438539505005,
|
| 315 |
+
"eval_runtime": 15.8056,
|
| 316 |
+
"eval_samples_per_second": 77.631,
|
| 317 |
+
"eval_steps_per_second": 1.265,
|
| 318 |
+
"step": 389
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 9.02,
|
| 322 |
+
"learning_rate": 5.16795865633075e-06,
|
| 323 |
+
"loss": 0.2932,
|
| 324 |
+
"step": 390
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 9.25,
|
| 328 |
+
"learning_rate": 3.875968992248062e-06,
|
| 329 |
+
"loss": 0.289,
|
| 330 |
+
"step": 400
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 9.48,
|
| 334 |
+
"learning_rate": 2.583979328165375e-06,
|
| 335 |
+
"loss": 0.2809,
|
| 336 |
+
"step": 410
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 9.71,
|
| 340 |
+
"learning_rate": 1.2919896640826874e-06,
|
| 341 |
+
"loss": 0.2906,
|
| 342 |
+
"step": 420
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 9.94,
|
| 346 |
+
"learning_rate": 0.0,
|
| 347 |
+
"loss": 0.2812,
|
| 348 |
+
"step": 430
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 9.94,
|
| 352 |
+
"eval_accuracy": 0.8753056234718827,
|
| 353 |
+
"eval_loss": 0.321910560131073,
|
| 354 |
+
"eval_runtime": 18.4587,
|
| 355 |
+
"eval_samples_per_second": 66.473,
|
| 356 |
+
"eval_steps_per_second": 1.083,
|
| 357 |
+
"step": 430
|
| 358 |
},
|
| 359 |
{
|
| 360 |
+
"epoch": 9.94,
|
| 361 |
+
"step": 430,
|
| 362 |
+
"total_flos": 8.584102532322165e+18,
|
| 363 |
+
"train_loss": 0.7598524143529493,
|
| 364 |
+
"train_runtime": 2853.5799,
|
| 365 |
+
"train_samples_per_second": 38.688,
|
| 366 |
+
"train_steps_per_second": 0.151
|
| 367 |
}
|
| 368 |
],
|
| 369 |
"logging_steps": 10,
|
| 370 |
+
"max_steps": 430,
|
| 371 |
+
"num_train_epochs": 10,
|
| 372 |
"save_steps": 500,
|
| 373 |
+
"total_flos": 8.584102532322165e+18,
|
| 374 |
"trial_name": null,
|
| 375 |
"trial_params": null
|
| 376 |
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4155
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:143845d92f8daf8b99c92077c5fb759c10e71c35de8a40a71e3c384f89ca66c4
|
| 3 |
size 4155
|