Upload trained model
Browse files- README.md +1 -1
- all_results.json +8 -0
- runs/Jul06_21-04-43_0676c2d2a721/events.out.tfevents.1751837306.0676c2d2a721.299.3 +3 -0
- train_results.json +8 -0
- trainer_state.json +388 -0
README.md
CHANGED
|
@@ -21,7 +21,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 21 |
|
| 22 |
This model is a fine-tuned version of [facebook/hiera-base-224-in1k-hf](https://huggingface.co/facebook/hiera-base-224-in1k-hf) on an unknown dataset.
|
| 23 |
It achieves the following results on the evaluation set:
|
| 24 |
-
- Loss: 0.
|
| 25 |
- Accuracy: 0.9951
|
| 26 |
- F1: 0.9951
|
| 27 |
- Precision: 0.9951
|
|
|
|
| 21 |
|
| 22 |
This model is a fine-tuned version of [facebook/hiera-base-224-in1k-hf](https://huggingface.co/facebook/hiera-base-224-in1k-hf) on an unknown dataset.
|
| 23 |
It achieves the following results on the evaluation set:
|
| 24 |
+
- Loss: 0.0098
|
| 25 |
- Accuracy: 0.9951
|
| 26 |
- F1: 0.9951
|
| 27 |
- Precision: 0.9951
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 12.0,
|
| 3 |
+
"total_flos": 4.018008276329693e+18,
|
| 4 |
+
"train_loss": 0.031628868545311085,
|
| 5 |
+
"train_runtime": 1401.2705,
|
| 6 |
+
"train_samples_per_second": 62.549,
|
| 7 |
+
"train_steps_per_second": 0.985
|
| 8 |
+
}
|
runs/Jul06_21-04-43_0676c2d2a721/events.out.tfevents.1751837306.0676c2d2a721.299.3
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:21c82c3aa7ae7a7efa49247deb995a1d390f34fdba3750183657c45e790d7559
|
| 3 |
+
size 560
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 12.0,
|
| 3 |
+
"total_flos": 4.018008276329693e+18,
|
| 4 |
+
"train_loss": 0.031628868545311085,
|
| 5 |
+
"train_runtime": 1401.2705,
|
| 6 |
+
"train_samples_per_second": 62.549,
|
| 7 |
+
"train_steps_per_second": 0.985
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,388 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": 500,
|
| 3 |
+
"best_metric": 0.9950738019660021,
|
| 4 |
+
"best_model_checkpoint": "hiera-finetuned-stroke-binary-ultrasound/checkpoint-500",
|
| 5 |
+
"epoch": 12.0,
|
| 6 |
+
"eval_steps": 100,
|
| 7 |
+
"global_step": 1380,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.437636761487965,
|
| 14 |
+
"grad_norm": 4.08018684387207,
|
| 15 |
+
"learning_rate": 6.811594202898551e-06,
|
| 16 |
+
"loss": 0.0505,
|
| 17 |
+
"step": 50
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.87527352297593,
|
| 21 |
+
"grad_norm": 0.09228578954935074,
|
| 22 |
+
"learning_rate": 1.4057971014492755e-05,
|
| 23 |
+
"loss": 0.038,
|
| 24 |
+
"step": 100
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.87527352297593,
|
| 28 |
+
"eval_accuracy": 0.9938423645320197,
|
| 29 |
+
"eval_f1": 0.9938419442238937,
|
| 30 |
+
"eval_loss": 0.018425090238451958,
|
| 31 |
+
"eval_precision": 0.9939169121769106,
|
| 32 |
+
"eval_recall": 0.9938423645320197,
|
| 33 |
+
"eval_runtime": 4.7821,
|
| 34 |
+
"eval_samples_per_second": 169.799,
|
| 35 |
+
"eval_steps_per_second": 21.329,
|
| 36 |
+
"step": 100
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 1.3063457330415755,
|
| 40 |
+
"grad_norm": 13.409342765808105,
|
| 41 |
+
"learning_rate": 1.9997952651793994e-05,
|
| 42 |
+
"loss": 0.0537,
|
| 43 |
+
"step": 150
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"epoch": 1.7439824945295404,
|
| 47 |
+
"grad_norm": 4.2889485359191895,
|
| 48 |
+
"learning_rate": 1.9892575474365977e-05,
|
| 49 |
+
"loss": 0.0468,
|
| 50 |
+
"step": 200
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 1.7439824945295404,
|
| 54 |
+
"eval_accuracy": 0.9926108374384236,
|
| 55 |
+
"eval_f1": 0.9926101649260403,
|
| 56 |
+
"eval_loss": 0.02063354291021824,
|
| 57 |
+
"eval_precision": 0.9927179267509103,
|
| 58 |
+
"eval_recall": 0.9926108374384236,
|
| 59 |
+
"eval_runtime": 4.2841,
|
| 60 |
+
"eval_samples_per_second": 189.54,
|
| 61 |
+
"eval_steps_per_second": 23.809,
|
| 62 |
+
"step": 200
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 2.175054704595186,
|
| 66 |
+
"grad_norm": 27.501617431640625,
|
| 67 |
+
"learning_rate": 1.9629172873477995e-05,
|
| 68 |
+
"loss": 0.055,
|
| 69 |
+
"step": 250
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 2.612691466083151,
|
| 73 |
+
"grad_norm": 34.34234619140625,
|
| 74 |
+
"learning_rate": 1.9211952480161382e-05,
|
| 75 |
+
"loss": 0.0445,
|
| 76 |
+
"step": 300
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 2.612691466083151,
|
| 80 |
+
"eval_accuracy": 0.9901477832512315,
|
| 81 |
+
"eval_f1": 0.9901478430214063,
|
| 82 |
+
"eval_loss": 0.02250731736421585,
|
| 83 |
+
"eval_precision": 0.9901599165230894,
|
| 84 |
+
"eval_recall": 0.9901477832512315,
|
| 85 |
+
"eval_runtime": 4.3458,
|
| 86 |
+
"eval_samples_per_second": 186.849,
|
| 87 |
+
"eval_steps_per_second": 23.471,
|
| 88 |
+
"step": 300
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 3.0437636761487963,
|
| 92 |
+
"grad_norm": 1.1840153932571411,
|
| 93 |
+
"learning_rate": 1.866025403784439e-05,
|
| 94 |
+
"loss": 0.0309,
|
| 95 |
+
"step": 350
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 3.4814004376367613,
|
| 99 |
+
"grad_norm": 0.7110581994056702,
|
| 100 |
+
"learning_rate": 1.7960402776569358e-05,
|
| 101 |
+
"loss": 0.0415,
|
| 102 |
+
"step": 400
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 3.4814004376367613,
|
| 106 |
+
"eval_accuracy": 0.9889162561576355,
|
| 107 |
+
"eval_f1": 0.9889161721031502,
|
| 108 |
+
"eval_loss": 0.018740125000476837,
|
| 109 |
+
"eval_precision": 0.9889190879070968,
|
| 110 |
+
"eval_recall": 0.9889162561576355,
|
| 111 |
+
"eval_runtime": 4.4065,
|
| 112 |
+
"eval_samples_per_second": 184.274,
|
| 113 |
+
"eval_steps_per_second": 23.148,
|
| 114 |
+
"step": 400
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 3.9190371991247264,
|
| 118 |
+
"grad_norm": 44.404239654541016,
|
| 119 |
+
"learning_rate": 1.7133390896410106e-05,
|
| 120 |
+
"loss": 0.034,
|
| 121 |
+
"step": 450
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 4.350109409190372,
|
| 125 |
+
"grad_norm": 0.5970667004585266,
|
| 126 |
+
"learning_rate": 1.6192429204088022e-05,
|
| 127 |
+
"loss": 0.0465,
|
| 128 |
+
"step": 500
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 4.350109409190372,
|
| 132 |
+
"eval_accuracy": 0.9950738916256158,
|
| 133 |
+
"eval_f1": 0.9950738019660021,
|
| 134 |
+
"eval_loss": 0.009755603037774563,
|
| 135 |
+
"eval_precision": 0.9950857869728612,
|
| 136 |
+
"eval_recall": 0.9950738916256158,
|
| 137 |
+
"eval_runtime": 5.2229,
|
| 138 |
+
"eval_samples_per_second": 155.47,
|
| 139 |
+
"eval_steps_per_second": 19.53,
|
| 140 |
+
"step": 500
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 4.787746170678337,
|
| 144 |
+
"grad_norm": 0.04525822028517723,
|
| 145 |
+
"learning_rate": 1.5152548757013183e-05,
|
| 146 |
+
"loss": 0.0284,
|
| 147 |
+
"step": 550
|
| 148 |
+
},
|
| 149 |
+
{
|
| 150 |
+
"epoch": 5.218818380743983,
|
| 151 |
+
"grad_norm": 6.67152214050293,
|
| 152 |
+
"learning_rate": 1.4030360755011423e-05,
|
| 153 |
+
"loss": 0.0397,
|
| 154 |
+
"step": 600
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 5.218818380743983,
|
| 158 |
+
"eval_accuracy": 0.9901477832512315,
|
| 159 |
+
"eval_f1": 0.9901463484623846,
|
| 160 |
+
"eval_loss": 0.028631530702114105,
|
| 161 |
+
"eval_precision": 0.9903372489579386,
|
| 162 |
+
"eval_recall": 0.9901477832512315,
|
| 163 |
+
"eval_runtime": 5.1408,
|
| 164 |
+
"eval_samples_per_second": 157.953,
|
| 165 |
+
"eval_steps_per_second": 19.841,
|
| 166 |
+
"step": 600
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 5.656455142231947,
|
| 170 |
+
"grad_norm": 2.804370641708374,
|
| 171 |
+
"learning_rate": 1.284379119062912e-05,
|
| 172 |
+
"loss": 0.0334,
|
| 173 |
+
"step": 650
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 6.087527352297593,
|
| 177 |
+
"grad_norm": 0.5940968990325928,
|
| 178 |
+
"learning_rate": 1.1611794496750019e-05,
|
| 179 |
+
"loss": 0.0257,
|
| 180 |
+
"step": 700
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 6.087527352297593,
|
| 184 |
+
"eval_accuracy": 0.9926108374384236,
|
| 185 |
+
"eval_f1": 0.9926104787847378,
|
| 186 |
+
"eval_loss": 0.01877174898982048,
|
| 187 |
+
"eval_precision": 0.992658305036109,
|
| 188 |
+
"eval_recall": 0.9926108374384236,
|
| 189 |
+
"eval_runtime": 4.3633,
|
| 190 |
+
"eval_samples_per_second": 186.097,
|
| 191 |
+
"eval_steps_per_second": 23.377,
|
| 192 |
+
"step": 700
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 6.525164113785558,
|
| 196 |
+
"grad_norm": 6.283336162567139,
|
| 197 |
+
"learning_rate": 1.0354050765758148e-05,
|
| 198 |
+
"loss": 0.0226,
|
| 199 |
+
"step": 750
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 6.962800875273523,
|
| 203 |
+
"grad_norm": 0.4039391279220581,
|
| 204 |
+
"learning_rate": 9.090651376911532e-06,
|
| 205 |
+
"loss": 0.0434,
|
| 206 |
+
"step": 800
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 6.962800875273523,
|
| 210 |
+
"eval_accuracy": 0.9938423645320197,
|
| 211 |
+
"eval_f1": 0.9938419442238937,
|
| 212 |
+
"eval_loss": 0.02092530019581318,
|
| 213 |
+
"eval_precision": 0.9939169121769106,
|
| 214 |
+
"eval_recall": 0.9938423645320197,
|
| 215 |
+
"eval_runtime": 5.4149,
|
| 216 |
+
"eval_samples_per_second": 149.956,
|
| 217 |
+
"eval_steps_per_second": 18.837,
|
| 218 |
+
"step": 800
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 7.393873085339169,
|
| 222 |
+
"grad_norm": 26.558956146240234,
|
| 223 |
+
"learning_rate": 7.841778053760212e-06,
|
| 224 |
+
"loss": 0.0256,
|
| 225 |
+
"step": 850
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 7.831509846827133,
|
| 229 |
+
"grad_norm": 1.8528072834014893,
|
| 230 |
+
"learning_rate": 6.627380478391543e-06,
|
| 231 |
+
"loss": 0.0261,
|
| 232 |
+
"step": 900
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 7.831509846827133,
|
| 236 |
+
"eval_accuracy": 0.9926108374384236,
|
| 237 |
+
"eval_f1": 0.9926107029490034,
|
| 238 |
+
"eval_loss": 0.015434592962265015,
|
| 239 |
+
"eval_precision": 0.9926226132344405,
|
| 240 |
+
"eval_recall": 0.9926108374384236,
|
| 241 |
+
"eval_runtime": 4.4967,
|
| 242 |
+
"eval_samples_per_second": 180.577,
|
| 243 |
+
"eval_steps_per_second": 22.683,
|
| 244 |
+
"step": 900
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 8.262582056892779,
|
| 248 |
+
"grad_norm": 0.028279444202780724,
|
| 249 |
+
"learning_rate": 5.466857612339229e-06,
|
| 250 |
+
"loss": 0.0132,
|
| 251 |
+
"step": 950
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 8.700218818380744,
|
| 255 |
+
"grad_norm": 10.328960418701172,
|
| 256 |
+
"learning_rate": 4.378747814781629e-06,
|
| 257 |
+
"loss": 0.0198,
|
| 258 |
+
"step": 1000
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 8.700218818380744,
|
| 262 |
+
"eval_accuracy": 0.9950738916256158,
|
| 263 |
+
"eval_f1": 0.9950736525231585,
|
| 264 |
+
"eval_loss": 0.009430773556232452,
|
| 265 |
+
"eval_precision": 0.9951217179205126,
|
| 266 |
+
"eval_recall": 0.9950738916256158,
|
| 267 |
+
"eval_runtime": 5.241,
|
| 268 |
+
"eval_samples_per_second": 154.932,
|
| 269 |
+
"eval_steps_per_second": 19.462,
|
| 270 |
+
"step": 1000
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 9.13129102844639,
|
| 274 |
+
"grad_norm": 5.253521919250488,
|
| 275 |
+
"learning_rate": 3.3804327081259304e-06,
|
| 276 |
+
"loss": 0.0244,
|
| 277 |
+
"step": 1050
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 9.568927789934355,
|
| 281 |
+
"grad_norm": 0.006149813067167997,
|
| 282 |
+
"learning_rate": 2.4878595214718236e-06,
|
| 283 |
+
"loss": 0.0207,
|
| 284 |
+
"step": 1100
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 9.568927789934355,
|
| 288 |
+
"eval_accuracy": 0.9938423645320197,
|
| 289 |
+
"eval_f1": 0.9938421683977477,
|
| 290 |
+
"eval_loss": 0.012234929017722607,
|
| 291 |
+
"eval_precision": 0.9938691081078003,
|
| 292 |
+
"eval_recall": 0.9938423645320197,
|
| 293 |
+
"eval_runtime": 4.3976,
|
| 294 |
+
"eval_samples_per_second": 184.648,
|
| 295 |
+
"eval_steps_per_second": 23.195,
|
| 296 |
+
"step": 1100
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 10.0,
|
| 300 |
+
"grad_norm": 0.003324387362226844,
|
| 301 |
+
"learning_rate": 1.7152863472802195e-06,
|
| 302 |
+
"loss": 0.0212,
|
| 303 |
+
"step": 1150
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 10.437636761487965,
|
| 307 |
+
"grad_norm": 5.904305934906006,
|
| 308 |
+
"learning_rate": 1.075054380553552e-06,
|
| 309 |
+
"loss": 0.0157,
|
| 310 |
+
"step": 1200
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 10.437636761487965,
|
| 314 |
+
"eval_accuracy": 0.9950738916256158,
|
| 315 |
+
"eval_f1": 0.9950736525231585,
|
| 316 |
+
"eval_loss": 0.01014081109315157,
|
| 317 |
+
"eval_precision": 0.9951217179205126,
|
| 318 |
+
"eval_recall": 0.9950738916256158,
|
| 319 |
+
"eval_runtime": 5.4913,
|
| 320 |
+
"eval_samples_per_second": 147.869,
|
| 321 |
+
"eval_steps_per_second": 18.575,
|
| 322 |
+
"step": 1200
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 10.87527352297593,
|
| 326 |
+
"grad_norm": 0.6864893436431885,
|
| 327 |
+
"learning_rate": 5.77390778811796e-07,
|
| 328 |
+
"loss": 0.0143,
|
| 329 |
+
"step": 1250
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 11.306345733041576,
|
| 333 |
+
"grad_norm": 39.91190719604492,
|
| 334 |
+
"learning_rate": 2.3024529200728952e-07,
|
| 335 |
+
"loss": 0.0188,
|
| 336 |
+
"step": 1300
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 11.306345733041576,
|
| 340 |
+
"eval_accuracy": 0.9950738916256158,
|
| 341 |
+
"eval_f1": 0.9950736525231585,
|
| 342 |
+
"eval_loss": 0.010448944754898548,
|
| 343 |
+
"eval_precision": 0.9951217179205126,
|
| 344 |
+
"eval_recall": 0.9950738916256158,
|
| 345 |
+
"eval_runtime": 5.3991,
|
| 346 |
+
"eval_samples_per_second": 150.395,
|
| 347 |
+
"eval_steps_per_second": 18.892,
|
| 348 |
+
"step": 1300
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 11.74398249452954,
|
| 352 |
+
"grad_norm": 40.59695816040039,
|
| 353 |
+
"learning_rate": 3.916327207559967e-08,
|
| 354 |
+
"loss": 0.0198,
|
| 355 |
+
"step": 1350
|
| 356 |
+
},
|
| 357 |
+
{
|
| 358 |
+
"epoch": 12.0,
|
| 359 |
+
"step": 1380,
|
| 360 |
+
"total_flos": 4.018008276329693e+18,
|
| 361 |
+
"train_loss": 0.031628868545311085,
|
| 362 |
+
"train_runtime": 1401.2705,
|
| 363 |
+
"train_samples_per_second": 62.549,
|
| 364 |
+
"train_steps_per_second": 0.985
|
| 365 |
+
}
|
| 366 |
+
],
|
| 367 |
+
"logging_steps": 50,
|
| 368 |
+
"max_steps": 1380,
|
| 369 |
+
"num_input_tokens_seen": 0,
|
| 370 |
+
"num_train_epochs": 12,
|
| 371 |
+
"save_steps": 100,
|
| 372 |
+
"stateful_callbacks": {
|
| 373 |
+
"TrainerControl": {
|
| 374 |
+
"args": {
|
| 375 |
+
"should_epoch_stop": false,
|
| 376 |
+
"should_evaluate": false,
|
| 377 |
+
"should_log": false,
|
| 378 |
+
"should_save": true,
|
| 379 |
+
"should_training_stop": true
|
| 380 |
+
},
|
| 381 |
+
"attributes": {}
|
| 382 |
+
}
|
| 383 |
+
},
|
| 384 |
+
"total_flos": 4.018008276329693e+18,
|
| 385 |
+
"train_batch_size": 16,
|
| 386 |
+
"trial_name": null,
|
| 387 |
+
"trial_params": null
|
| 388 |
+
}
|