End of training
Browse files- README.md +5 -3
- all_results.json +12 -0
- eval_results.json +8 -0
- train_results.json +7 -0
- trainer_state.json +480 -0
README.md
CHANGED
|
@@ -2,6 +2,8 @@
|
|
| 2 |
license: apache-2.0
|
| 3 |
base_model: facebook/dinov2-small
|
| 4 |
tags:
|
|
|
|
|
|
|
| 5 |
- generated_from_trainer
|
| 6 |
metrics:
|
| 7 |
- accuracy
|
|
@@ -15,10 +17,10 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 15 |
|
| 16 |
# dinov2-small-types-of-film-shots-vN
|
| 17 |
|
| 18 |
-
This model is a fine-tuned version of [facebook/dinov2-small](https://huggingface.co/facebook/dinov2-small) 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: facebook/dinov2-small
|
| 4 |
tags:
|
| 5 |
+
- image-classification
|
| 6 |
+
- vision
|
| 7 |
- generated_from_trainer
|
| 8 |
metrics:
|
| 9 |
- accuracy
|
|
|
|
| 17 |
|
| 18 |
# dinov2-small-types-of-film-shots-vN
|
| 19 |
|
| 20 |
+
This model is a fine-tuned version of [facebook/dinov2-small](https://huggingface.co/facebook/dinov2-small) on the szymonrucinski/types-of-film-shots dataset.
|
| 21 |
It achieves the following results on the evaluation set:
|
| 22 |
+
- Loss: 0.9864
|
| 23 |
+
- Accuracy: 0.6259
|
| 24 |
|
| 25 |
## Model description
|
| 26 |
|
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 11.7,
|
| 3 |
+
"eval_accuracy": 0.6258992805755396,
|
| 4 |
+
"eval_loss": 0.986404299736023,
|
| 5 |
+
"eval_runtime": 0.6831,
|
| 6 |
+
"eval_samples_per_second": 203.477,
|
| 7 |
+
"eval_steps_per_second": 51.235,
|
| 8 |
+
"train_loss": 0.8871624300049411,
|
| 9 |
+
"train_runtime": 139.3419,
|
| 10 |
+
"train_samples_per_second": 67.69,
|
| 11 |
+
"train_steps_per_second": 2.067
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 11.7,
|
| 3 |
+
"eval_accuracy": 0.6258992805755396,
|
| 4 |
+
"eval_loss": 0.986404299736023,
|
| 5 |
+
"eval_runtime": 0.6831,
|
| 6 |
+
"eval_samples_per_second": 203.477,
|
| 7 |
+
"eval_steps_per_second": 51.235
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 11.7,
|
| 3 |
+
"train_loss": 0.8871624300049411,
|
| 4 |
+
"train_runtime": 139.3419,
|
| 5 |
+
"train_samples_per_second": 67.69,
|
| 6 |
+
"train_steps_per_second": 2.067
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,480 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.986404299736023,
|
| 3 |
+
"best_model_checkpoint": "./outputs/dinov2-small-types-of-film-shots-vN/checkpoint-221",
|
| 4 |
+
"epoch": 11.695431472081218,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 288,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.2,
|
| 13 |
+
"learning_rate": 6.666666666666667e-06,
|
| 14 |
+
"loss": 2.3969,
|
| 15 |
+
"step": 5
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.41,
|
| 19 |
+
"learning_rate": 1.3333333333333333e-05,
|
| 20 |
+
"loss": 2.0526,
|
| 21 |
+
"step": 10
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.61,
|
| 25 |
+
"learning_rate": 2e-05,
|
| 26 |
+
"loss": 1.8415,
|
| 27 |
+
"step": 15
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.81,
|
| 31 |
+
"learning_rate": 1.9633699633699634e-05,
|
| 32 |
+
"loss": 1.6177,
|
| 33 |
+
"step": 20
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.97,
|
| 37 |
+
"eval_accuracy": 0.41007194244604317,
|
| 38 |
+
"eval_loss": 1.5501453876495361,
|
| 39 |
+
"eval_runtime": 0.6217,
|
| 40 |
+
"eval_samples_per_second": 223.574,
|
| 41 |
+
"eval_steps_per_second": 56.296,
|
| 42 |
+
"step": 24
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 1.02,
|
| 46 |
+
"learning_rate": 1.926739926739927e-05,
|
| 47 |
+
"loss": 1.5922,
|
| 48 |
+
"step": 25
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 1.22,
|
| 52 |
+
"learning_rate": 1.8901098901098903e-05,
|
| 53 |
+
"loss": 1.3521,
|
| 54 |
+
"step": 30
|
| 55 |
+
},
|
| 56 |
+
{
|
| 57 |
+
"epoch": 1.42,
|
| 58 |
+
"learning_rate": 1.8534798534798536e-05,
|
| 59 |
+
"loss": 1.4178,
|
| 60 |
+
"step": 35
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 1.62,
|
| 64 |
+
"learning_rate": 1.816849816849817e-05,
|
| 65 |
+
"loss": 1.2448,
|
| 66 |
+
"step": 40
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 1.83,
|
| 70 |
+
"learning_rate": 1.78021978021978e-05,
|
| 71 |
+
"loss": 1.3029,
|
| 72 |
+
"step": 45
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 1.99,
|
| 76 |
+
"eval_accuracy": 0.5107913669064749,
|
| 77 |
+
"eval_loss": 1.2448219060897827,
|
| 78 |
+
"eval_runtime": 0.633,
|
| 79 |
+
"eval_samples_per_second": 219.573,
|
| 80 |
+
"eval_steps_per_second": 55.288,
|
| 81 |
+
"step": 49
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 2.03,
|
| 85 |
+
"learning_rate": 1.7435897435897438e-05,
|
| 86 |
+
"loss": 1.1814,
|
| 87 |
+
"step": 50
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 2.23,
|
| 91 |
+
"learning_rate": 1.706959706959707e-05,
|
| 92 |
+
"loss": 1.0996,
|
| 93 |
+
"step": 55
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 2.44,
|
| 97 |
+
"learning_rate": 1.6703296703296707e-05,
|
| 98 |
+
"loss": 1.098,
|
| 99 |
+
"step": 60
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 2.64,
|
| 103 |
+
"learning_rate": 1.633699633699634e-05,
|
| 104 |
+
"loss": 1.041,
|
| 105 |
+
"step": 65
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 2.84,
|
| 109 |
+
"learning_rate": 1.5970695970695972e-05,
|
| 110 |
+
"loss": 1.1785,
|
| 111 |
+
"step": 70
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 2.96,
|
| 115 |
+
"eval_accuracy": 0.5251798561151079,
|
| 116 |
+
"eval_loss": 1.0555555820465088,
|
| 117 |
+
"eval_runtime": 0.6319,
|
| 118 |
+
"eval_samples_per_second": 219.987,
|
| 119 |
+
"eval_steps_per_second": 55.392,
|
| 120 |
+
"step": 73
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 3.05,
|
| 124 |
+
"learning_rate": 1.5604395604395605e-05,
|
| 125 |
+
"loss": 1.116,
|
| 126 |
+
"step": 75
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 3.25,
|
| 130 |
+
"learning_rate": 1.523809523809524e-05,
|
| 131 |
+
"loss": 1.0266,
|
| 132 |
+
"step": 80
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 3.45,
|
| 136 |
+
"learning_rate": 1.4871794871794874e-05,
|
| 137 |
+
"loss": 0.9728,
|
| 138 |
+
"step": 85
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 3.65,
|
| 142 |
+
"learning_rate": 1.4505494505494506e-05,
|
| 143 |
+
"loss": 0.9575,
|
| 144 |
+
"step": 90
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 3.86,
|
| 148 |
+
"learning_rate": 1.4139194139194141e-05,
|
| 149 |
+
"loss": 1.2146,
|
| 150 |
+
"step": 95
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 3.98,
|
| 154 |
+
"eval_accuracy": 0.539568345323741,
|
| 155 |
+
"eval_loss": 1.2316133975982666,
|
| 156 |
+
"eval_runtime": 0.6178,
|
| 157 |
+
"eval_samples_per_second": 225.001,
|
| 158 |
+
"eval_steps_per_second": 56.655,
|
| 159 |
+
"step": 98
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 4.06,
|
| 163 |
+
"learning_rate": 1.3772893772893774e-05,
|
| 164 |
+
"loss": 0.8369,
|
| 165 |
+
"step": 100
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 4.26,
|
| 169 |
+
"learning_rate": 1.3406593406593406e-05,
|
| 170 |
+
"loss": 0.9136,
|
| 171 |
+
"step": 105
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 4.47,
|
| 175 |
+
"learning_rate": 1.3040293040293041e-05,
|
| 176 |
+
"loss": 0.9703,
|
| 177 |
+
"step": 110
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 4.67,
|
| 181 |
+
"learning_rate": 1.2673992673992674e-05,
|
| 182 |
+
"loss": 0.9575,
|
| 183 |
+
"step": 115
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 4.87,
|
| 187 |
+
"learning_rate": 1.230769230769231e-05,
|
| 188 |
+
"loss": 0.8389,
|
| 189 |
+
"step": 120
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 4.99,
|
| 193 |
+
"eval_accuracy": 0.5971223021582733,
|
| 194 |
+
"eval_loss": 1.0235121250152588,
|
| 195 |
+
"eval_runtime": 0.6239,
|
| 196 |
+
"eval_samples_per_second": 222.778,
|
| 197 |
+
"eval_steps_per_second": 56.095,
|
| 198 |
+
"step": 123
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 5.08,
|
| 202 |
+
"learning_rate": 1.1941391941391942e-05,
|
| 203 |
+
"loss": 0.8346,
|
| 204 |
+
"step": 125
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 5.28,
|
| 208 |
+
"learning_rate": 1.1575091575091577e-05,
|
| 209 |
+
"loss": 0.8608,
|
| 210 |
+
"step": 130
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 5.48,
|
| 214 |
+
"learning_rate": 1.120879120879121e-05,
|
| 215 |
+
"loss": 0.855,
|
| 216 |
+
"step": 135
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 5.69,
|
| 220 |
+
"learning_rate": 1.0842490842490842e-05,
|
| 221 |
+
"loss": 0.8467,
|
| 222 |
+
"step": 140
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 5.89,
|
| 226 |
+
"learning_rate": 1.0476190476190477e-05,
|
| 227 |
+
"loss": 0.7883,
|
| 228 |
+
"step": 145
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 5.97,
|
| 232 |
+
"eval_accuracy": 0.6258992805755396,
|
| 233 |
+
"eval_loss": 0.9959595799446106,
|
| 234 |
+
"eval_runtime": 0.6274,
|
| 235 |
+
"eval_samples_per_second": 221.561,
|
| 236 |
+
"eval_steps_per_second": 55.789,
|
| 237 |
+
"step": 147
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 6.09,
|
| 241 |
+
"learning_rate": 1.010989010989011e-05,
|
| 242 |
+
"loss": 0.7238,
|
| 243 |
+
"step": 150
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 6.29,
|
| 247 |
+
"learning_rate": 9.743589743589744e-06,
|
| 248 |
+
"loss": 0.688,
|
| 249 |
+
"step": 155
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 6.5,
|
| 253 |
+
"learning_rate": 9.377289377289379e-06,
|
| 254 |
+
"loss": 0.6678,
|
| 255 |
+
"step": 160
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 6.7,
|
| 259 |
+
"learning_rate": 9.010989010989011e-06,
|
| 260 |
+
"loss": 0.7223,
|
| 261 |
+
"step": 165
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 6.9,
|
| 265 |
+
"learning_rate": 8.644688644688646e-06,
|
| 266 |
+
"loss": 0.7899,
|
| 267 |
+
"step": 170
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 6.98,
|
| 271 |
+
"eval_accuracy": 0.5539568345323741,
|
| 272 |
+
"eval_loss": 1.1354069709777832,
|
| 273 |
+
"eval_runtime": 0.6338,
|
| 274 |
+
"eval_samples_per_second": 219.304,
|
| 275 |
+
"eval_steps_per_second": 55.22,
|
| 276 |
+
"step": 172
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 7.11,
|
| 280 |
+
"learning_rate": 8.278388278388278e-06,
|
| 281 |
+
"loss": 0.7066,
|
| 282 |
+
"step": 175
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 7.31,
|
| 286 |
+
"learning_rate": 7.912087912087913e-06,
|
| 287 |
+
"loss": 0.6458,
|
| 288 |
+
"step": 180
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 7.51,
|
| 292 |
+
"learning_rate": 7.5457875457875465e-06,
|
| 293 |
+
"loss": 0.6245,
|
| 294 |
+
"step": 185
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 7.72,
|
| 298 |
+
"learning_rate": 7.17948717948718e-06,
|
| 299 |
+
"loss": 0.6187,
|
| 300 |
+
"step": 190
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 7.92,
|
| 304 |
+
"learning_rate": 6.813186813186814e-06,
|
| 305 |
+
"loss": 0.663,
|
| 306 |
+
"step": 195
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 8.0,
|
| 310 |
+
"eval_accuracy": 0.5827338129496403,
|
| 311 |
+
"eval_loss": 1.097096562385559,
|
| 312 |
+
"eval_runtime": 0.6204,
|
| 313 |
+
"eval_samples_per_second": 224.06,
|
| 314 |
+
"eval_steps_per_second": 56.418,
|
| 315 |
+
"step": 197
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 8.12,
|
| 319 |
+
"learning_rate": 6.446886446886448e-06,
|
| 320 |
+
"loss": 0.6473,
|
| 321 |
+
"step": 200
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 8.32,
|
| 325 |
+
"learning_rate": 6.080586080586081e-06,
|
| 326 |
+
"loss": 0.5863,
|
| 327 |
+
"step": 205
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 8.53,
|
| 331 |
+
"learning_rate": 5.7142857142857145e-06,
|
| 332 |
+
"loss": 0.5137,
|
| 333 |
+
"step": 210
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 8.73,
|
| 337 |
+
"learning_rate": 5.347985347985348e-06,
|
| 338 |
+
"loss": 0.5704,
|
| 339 |
+
"step": 215
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 8.93,
|
| 343 |
+
"learning_rate": 4.9816849816849826e-06,
|
| 344 |
+
"loss": 0.6013,
|
| 345 |
+
"step": 220
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 8.97,
|
| 349 |
+
"eval_accuracy": 0.6258992805755396,
|
| 350 |
+
"eval_loss": 0.986404299736023,
|
| 351 |
+
"eval_runtime": 0.6376,
|
| 352 |
+
"eval_samples_per_second": 217.99,
|
| 353 |
+
"eval_steps_per_second": 54.89,
|
| 354 |
+
"step": 221
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 9.14,
|
| 358 |
+
"learning_rate": 4.615384615384616e-06,
|
| 359 |
+
"loss": 0.4918,
|
| 360 |
+
"step": 225
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 9.34,
|
| 364 |
+
"learning_rate": 4.249084249084249e-06,
|
| 365 |
+
"loss": 0.5953,
|
| 366 |
+
"step": 230
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 9.54,
|
| 370 |
+
"learning_rate": 3.882783882783883e-06,
|
| 371 |
+
"loss": 0.5561,
|
| 372 |
+
"step": 235
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 9.75,
|
| 376 |
+
"learning_rate": 3.516483516483517e-06,
|
| 377 |
+
"loss": 0.4991,
|
| 378 |
+
"step": 240
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 9.95,
|
| 382 |
+
"learning_rate": 3.1501831501831505e-06,
|
| 383 |
+
"loss": 0.6276,
|
| 384 |
+
"step": 245
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 9.99,
|
| 388 |
+
"eval_accuracy": 0.6115107913669064,
|
| 389 |
+
"eval_loss": 1.018210768699646,
|
| 390 |
+
"eval_runtime": 0.6196,
|
| 391 |
+
"eval_samples_per_second": 224.345,
|
| 392 |
+
"eval_steps_per_second": 56.49,
|
| 393 |
+
"step": 246
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 10.15,
|
| 397 |
+
"learning_rate": 2.783882783882784e-06,
|
| 398 |
+
"loss": 0.393,
|
| 399 |
+
"step": 250
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 10.36,
|
| 403 |
+
"learning_rate": 2.4175824175824177e-06,
|
| 404 |
+
"loss": 0.5476,
|
| 405 |
+
"step": 255
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 10.56,
|
| 409 |
+
"learning_rate": 2.0512820512820513e-06,
|
| 410 |
+
"loss": 0.457,
|
| 411 |
+
"step": 260
|
| 412 |
+
},
|
| 413 |
+
{
|
| 414 |
+
"epoch": 10.76,
|
| 415 |
+
"learning_rate": 1.6849816849816852e-06,
|
| 416 |
+
"loss": 0.4764,
|
| 417 |
+
"step": 265
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 10.96,
|
| 421 |
+
"learning_rate": 1.3186813186813187e-06,
|
| 422 |
+
"loss": 0.5196,
|
| 423 |
+
"step": 270
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 10.96,
|
| 427 |
+
"eval_accuracy": 0.6546762589928058,
|
| 428 |
+
"eval_loss": 1.0074387788772583,
|
| 429 |
+
"eval_runtime": 0.6224,
|
| 430 |
+
"eval_samples_per_second": 223.337,
|
| 431 |
+
"eval_steps_per_second": 56.236,
|
| 432 |
+
"step": 270
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 11.17,
|
| 436 |
+
"learning_rate": 9.523809523809525e-07,
|
| 437 |
+
"loss": 0.4737,
|
| 438 |
+
"step": 275
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 11.37,
|
| 442 |
+
"learning_rate": 5.860805860805862e-07,
|
| 443 |
+
"loss": 0.547,
|
| 444 |
+
"step": 280
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 11.57,
|
| 448 |
+
"learning_rate": 2.197802197802198e-07,
|
| 449 |
+
"loss": 0.4761,
|
| 450 |
+
"step": 285
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 11.7,
|
| 454 |
+
"eval_accuracy": 0.6762589928057554,
|
| 455 |
+
"eval_loss": 0.9956228137016296,
|
| 456 |
+
"eval_runtime": 0.6444,
|
| 457 |
+
"eval_samples_per_second": 215.696,
|
| 458 |
+
"eval_steps_per_second": 54.312,
|
| 459 |
+
"step": 288
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 11.7,
|
| 463 |
+
"step": 288,
|
| 464 |
+
"total_flos": 2.3928536971188634e+17,
|
| 465 |
+
"train_loss": 0.8871624300049411,
|
| 466 |
+
"train_runtime": 139.3419,
|
| 467 |
+
"train_samples_per_second": 67.69,
|
| 468 |
+
"train_steps_per_second": 2.067
|
| 469 |
+
}
|
| 470 |
+
],
|
| 471 |
+
"logging_steps": 5,
|
| 472 |
+
"max_steps": 288,
|
| 473 |
+
"num_input_tokens_seen": 0,
|
| 474 |
+
"num_train_epochs": 12,
|
| 475 |
+
"save_steps": 500,
|
| 476 |
+
"total_flos": 2.3928536971188634e+17,
|
| 477 |
+
"train_batch_size": 4,
|
| 478 |
+
"trial_name": null,
|
| 479 |
+
"trial_params": null
|
| 480 |
+
}
|