End of training
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.496124031007753,
|
| 3 |
+
"eval_accuracy": 0.9488919041157847,
|
| 4 |
+
"eval_f1": 0.9483693001512903,
|
| 5 |
+
"eval_loss": 0.15268893539905548,
|
| 6 |
+
"eval_precision": 0.9504887785210339,
|
| 7 |
+
"eval_recall": 0.9488919041157847,
|
| 8 |
+
"eval_runtime": 8.6365,
|
| 9 |
+
"eval_samples_per_second": 256.007,
|
| 10 |
+
"eval_steps_per_second": 32.073,
|
| 11 |
+
"total_flos": 5.998049534242161e+18,
|
| 12 |
+
"train_loss": 0.14669023766237146,
|
| 13 |
+
"train_runtime": 774.0157,
|
| 14 |
+
"train_samples_per_second": 319.869,
|
| 15 |
+
"train_steps_per_second": 9.984
|
| 16 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.496124031007753,
|
| 3 |
+
"eval_accuracy": 0.9488919041157847,
|
| 4 |
+
"eval_f1": 0.9483693001512903,
|
| 5 |
+
"eval_loss": 0.15268893539905548,
|
| 6 |
+
"eval_precision": 0.9504887785210339,
|
| 7 |
+
"eval_recall": 0.9488919041157847,
|
| 8 |
+
"eval_runtime": 8.6365,
|
| 9 |
+
"eval_samples_per_second": 256.007,
|
| 10 |
+
"eval_steps_per_second": 32.073
|
| 11 |
+
}
|
runs/Mar16_19-32-23_774ccf98b3fe/events.out.tfevents.1742158382.774ccf98b3fe.5066.29
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:89e7f51520db145abdf9c1a6d5fc28a1eda7f709213a10bf83aaac42a08112e8
|
| 3 |
+
size 560
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.496124031007753,
|
| 3 |
+
"total_flos": 5.998049534242161e+18,
|
| 4 |
+
"train_loss": 0.14669023766237146,
|
| 5 |
+
"train_runtime": 774.0157,
|
| 6 |
+
"train_samples_per_second": 319.869,
|
| 7 |
+
"train_steps_per_second": 9.984
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,484 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9483693001512903,
|
| 3 |
+
"best_model_checkpoint": "deit-base-patch16-224-finetuned-stroke-binary/checkpoint-1700",
|
| 4 |
+
"epoch": 10.496124031007753,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 1700,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.31007751937984496,
|
| 13 |
+
"grad_norm": 1.1434299945831299,
|
| 14 |
+
"learning_rate": 1.29366106080207e-06,
|
| 15 |
+
"loss": 0.1628,
|
| 16 |
+
"step": 50
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.6201550387596899,
|
| 20 |
+
"grad_norm": 0.9698119163513184,
|
| 21 |
+
"learning_rate": 2.58732212160414e-06,
|
| 22 |
+
"loss": 0.1646,
|
| 23 |
+
"step": 100
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.6201550387596899,
|
| 27 |
+
"eval_accuracy": 0.9430122116689281,
|
| 28 |
+
"eval_f1": 0.9424717047287801,
|
| 29 |
+
"eval_loss": 0.15877728164196014,
|
| 30 |
+
"eval_precision": 0.9442262324920166,
|
| 31 |
+
"eval_recall": 0.9430122116689281,
|
| 32 |
+
"eval_runtime": 9.3645,
|
| 33 |
+
"eval_samples_per_second": 236.103,
|
| 34 |
+
"eval_steps_per_second": 29.58,
|
| 35 |
+
"step": 100
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"epoch": 0.9302325581395349,
|
| 39 |
+
"grad_norm": 1.4945017099380493,
|
| 40 |
+
"learning_rate": 3.88098318240621e-06,
|
| 41 |
+
"loss": 0.1355,
|
| 42 |
+
"step": 150
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 1.235658914728682,
|
| 46 |
+
"grad_norm": 1.1095311641693115,
|
| 47 |
+
"learning_rate": 5.17464424320828e-06,
|
| 48 |
+
"loss": 0.1417,
|
| 49 |
+
"step": 200
|
| 50 |
+
},
|
| 51 |
+
{
|
| 52 |
+
"epoch": 1.235658914728682,
|
| 53 |
+
"eval_accuracy": 0.9439167797376753,
|
| 54 |
+
"eval_f1": 0.9432839027033096,
|
| 55 |
+
"eval_loss": 0.16397197544574738,
|
| 56 |
+
"eval_precision": 0.9457751540323792,
|
| 57 |
+
"eval_recall": 0.9439167797376753,
|
| 58 |
+
"eval_runtime": 9.5146,
|
| 59 |
+
"eval_samples_per_second": 232.38,
|
| 60 |
+
"eval_steps_per_second": 29.113,
|
| 61 |
+
"step": 200
|
| 62 |
+
},
|
| 63 |
+
{
|
| 64 |
+
"epoch": 1.5457364341085271,
|
| 65 |
+
"grad_norm": 2.9033849239349365,
|
| 66 |
+
"learning_rate": 6.468305304010349e-06,
|
| 67 |
+
"loss": 0.1576,
|
| 68 |
+
"step": 250
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 1.8558139534883722,
|
| 72 |
+
"grad_norm": 0.8375345468521118,
|
| 73 |
+
"learning_rate": 7.76196636481242e-06,
|
| 74 |
+
"loss": 0.1681,
|
| 75 |
+
"step": 300
|
| 76 |
+
},
|
| 77 |
+
{
|
| 78 |
+
"epoch": 1.8558139534883722,
|
| 79 |
+
"eval_accuracy": 0.945273631840796,
|
| 80 |
+
"eval_f1": 0.9446810798244,
|
| 81 |
+
"eval_loss": 0.16215302050113678,
|
| 82 |
+
"eval_precision": 0.9469984434420068,
|
| 83 |
+
"eval_recall": 0.945273631840796,
|
| 84 |
+
"eval_runtime": 9.4184,
|
| 85 |
+
"eval_samples_per_second": 234.752,
|
| 86 |
+
"eval_steps_per_second": 29.41,
|
| 87 |
+
"step": 300
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 2.1612403100775195,
|
| 91 |
+
"grad_norm": 1.2844864130020142,
|
| 92 |
+
"learning_rate": 9.055627425614489e-06,
|
| 93 |
+
"loss": 0.1487,
|
| 94 |
+
"step": 350
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 2.471317829457364,
|
| 98 |
+
"grad_norm": 2.3872156143188477,
|
| 99 |
+
"learning_rate": 1.034928848641656e-05,
|
| 100 |
+
"loss": 0.1512,
|
| 101 |
+
"step": 400
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 2.471317829457364,
|
| 105 |
+
"eval_accuracy": 0.9434644957033017,
|
| 106 |
+
"eval_f1": 0.9430495459530136,
|
| 107 |
+
"eval_loss": 0.15099050104618073,
|
| 108 |
+
"eval_precision": 0.9440827208056073,
|
| 109 |
+
"eval_recall": 0.9434644957033017,
|
| 110 |
+
"eval_runtime": 9.4764,
|
| 111 |
+
"eval_samples_per_second": 233.316,
|
| 112 |
+
"eval_steps_per_second": 29.23,
|
| 113 |
+
"step": 400
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 2.781395348837209,
|
| 117 |
+
"grad_norm": 1.7295209169387817,
|
| 118 |
+
"learning_rate": 1.164294954721863e-05,
|
| 119 |
+
"loss": 0.161,
|
| 120 |
+
"step": 450
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 3.0868217054263565,
|
| 124 |
+
"grad_norm": 3.4684560298919678,
|
| 125 |
+
"learning_rate": 1.2936610608020698e-05,
|
| 126 |
+
"loss": 0.1506,
|
| 127 |
+
"step": 500
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 3.0868217054263565,
|
| 131 |
+
"eval_accuracy": 0.9339665309814563,
|
| 132 |
+
"eval_f1": 0.9326951378602786,
|
| 133 |
+
"eval_loss": 0.19129638373851776,
|
| 134 |
+
"eval_precision": 0.9391386500820929,
|
| 135 |
+
"eval_recall": 0.9339665309814563,
|
| 136 |
+
"eval_runtime": 9.3995,
|
| 137 |
+
"eval_samples_per_second": 235.226,
|
| 138 |
+
"eval_steps_per_second": 29.47,
|
| 139 |
+
"step": 500
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 3.3968992248062015,
|
| 143 |
+
"grad_norm": 2.042165517807007,
|
| 144 |
+
"learning_rate": 1.423027166882277e-05,
|
| 145 |
+
"loss": 0.1263,
|
| 146 |
+
"step": 550
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 3.7069767441860466,
|
| 150 |
+
"grad_norm": 0.8381805419921875,
|
| 151 |
+
"learning_rate": 1.552393272962484e-05,
|
| 152 |
+
"loss": 0.1654,
|
| 153 |
+
"step": 600
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 3.7069767441860466,
|
| 157 |
+
"eval_accuracy": 0.9425599276345545,
|
| 158 |
+
"eval_f1": 0.9419379928735436,
|
| 159 |
+
"eval_loss": 0.1679351031780243,
|
| 160 |
+
"eval_precision": 0.9442161451256601,
|
| 161 |
+
"eval_recall": 0.9425599276345545,
|
| 162 |
+
"eval_runtime": 9.4346,
|
| 163 |
+
"eval_samples_per_second": 234.351,
|
| 164 |
+
"eval_steps_per_second": 29.36,
|
| 165 |
+
"step": 600
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 4.0124031007751935,
|
| 169 |
+
"grad_norm": 2.1265010833740234,
|
| 170 |
+
"learning_rate": 1.6817593790426908e-05,
|
| 171 |
+
"loss": 0.1507,
|
| 172 |
+
"step": 650
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 4.322480620155039,
|
| 176 |
+
"grad_norm": 2.001044511795044,
|
| 177 |
+
"learning_rate": 1.8111254851228977e-05,
|
| 178 |
+
"loss": 0.1482,
|
| 179 |
+
"step": 700
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 4.322480620155039,
|
| 183 |
+
"eval_accuracy": 0.9402985074626866,
|
| 184 |
+
"eval_f1": 0.9402037869673187,
|
| 185 |
+
"eval_loss": 0.1550845354795456,
|
| 186 |
+
"eval_precision": 0.9401885824772098,
|
| 187 |
+
"eval_recall": 0.9402985074626866,
|
| 188 |
+
"eval_runtime": 9.4957,
|
| 189 |
+
"eval_samples_per_second": 232.841,
|
| 190 |
+
"eval_steps_per_second": 29.171,
|
| 191 |
+
"step": 700
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 4.632558139534884,
|
| 195 |
+
"grad_norm": 0.40356162190437317,
|
| 196 |
+
"learning_rate": 1.940491591203105e-05,
|
| 197 |
+
"loss": 0.1451,
|
| 198 |
+
"step": 750
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 4.942635658914728,
|
| 202 |
+
"grad_norm": 0.9926322102546692,
|
| 203 |
+
"learning_rate": 1.999925630026586e-05,
|
| 204 |
+
"loss": 0.1599,
|
| 205 |
+
"step": 800
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 4.942635658914728,
|
| 209 |
+
"eval_accuracy": 0.9461781999095432,
|
| 210 |
+
"eval_f1": 0.9457377367508388,
|
| 211 |
+
"eval_loss": 0.14886853098869324,
|
| 212 |
+
"eval_precision": 0.9470663350522793,
|
| 213 |
+
"eval_recall": 0.9461781999095432,
|
| 214 |
+
"eval_runtime": 9.4883,
|
| 215 |
+
"eval_samples_per_second": 233.025,
|
| 216 |
+
"eval_steps_per_second": 29.194,
|
| 217 |
+
"step": 800
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 5.248062015503876,
|
| 221 |
+
"grad_norm": 1.0929352045059204,
|
| 222 |
+
"learning_rate": 1.9993951980962474e-05,
|
| 223 |
+
"loss": 0.1598,
|
| 224 |
+
"step": 850
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 5.558139534883721,
|
| 228 |
+
"grad_norm": 1.2597020864486694,
|
| 229 |
+
"learning_rate": 1.9983550078926357e-05,
|
| 230 |
+
"loss": 0.1477,
|
| 231 |
+
"step": 900
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 5.558139534883721,
|
| 235 |
+
"eval_accuracy": 0.9425599276345545,
|
| 236 |
+
"eval_f1": 0.9423650667942064,
|
| 237 |
+
"eval_loss": 0.14367012679576874,
|
| 238 |
+
"eval_precision": 0.942514608438234,
|
| 239 |
+
"eval_recall": 0.9425599276345545,
|
| 240 |
+
"eval_runtime": 9.4876,
|
| 241 |
+
"eval_samples_per_second": 233.041,
|
| 242 |
+
"eval_steps_per_second": 29.196,
|
| 243 |
+
"step": 900
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 5.868217054263566,
|
| 247 |
+
"grad_norm": 1.0700503587722778,
|
| 248 |
+
"learning_rate": 1.9968055899822005e-05,
|
| 249 |
+
"loss": 0.1432,
|
| 250 |
+
"step": 950
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 6.173643410852713,
|
| 254 |
+
"grad_norm": 2.013948440551758,
|
| 255 |
+
"learning_rate": 1.9947477346715192e-05,
|
| 256 |
+
"loss": 0.1308,
|
| 257 |
+
"step": 1000
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 6.173643410852713,
|
| 261 |
+
"eval_accuracy": 0.9416553595658074,
|
| 262 |
+
"eval_f1": 0.9414428935954616,
|
| 263 |
+
"eval_loss": 0.15271881222724915,
|
| 264 |
+
"eval_precision": 0.9416239036557209,
|
| 265 |
+
"eval_recall": 0.9416553595658074,
|
| 266 |
+
"eval_runtime": 9.4259,
|
| 267 |
+
"eval_samples_per_second": 234.566,
|
| 268 |
+
"eval_steps_per_second": 29.387,
|
| 269 |
+
"step": 1000
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 6.4837209302325585,
|
| 273 |
+
"grad_norm": 2.9245336055755615,
|
| 274 |
+
"learning_rate": 1.9921824916041882e-05,
|
| 275 |
+
"loss": 0.1355,
|
| 276 |
+
"step": 1050
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 6.793798449612403,
|
| 280 |
+
"grad_norm": 2.073958396911621,
|
| 281 |
+
"learning_rate": 1.9891111692254346e-05,
|
| 282 |
+
"loss": 0.1362,
|
| 283 |
+
"step": 1100
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 6.793798449612403,
|
| 287 |
+
"eval_accuracy": 0.9425599276345545,
|
| 288 |
+
"eval_f1": 0.9421222945979488,
|
| 289 |
+
"eval_loss": 0.16084641218185425,
|
| 290 |
+
"eval_precision": 0.943232512538474,
|
| 291 |
+
"eval_recall": 0.9425599276345545,
|
| 292 |
+
"eval_runtime": 9.4085,
|
| 293 |
+
"eval_samples_per_second": 235.0,
|
| 294 |
+
"eval_steps_per_second": 29.441,
|
| 295 |
+
"step": 1100
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 7.09922480620155,
|
| 299 |
+
"grad_norm": 3.3903515338897705,
|
| 300 |
+
"learning_rate": 1.98553533411472e-05,
|
| 301 |
+
"loss": 0.1437,
|
| 302 |
+
"step": 1150
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 7.409302325581395,
|
| 306 |
+
"grad_norm": 0.6468881964683533,
|
| 307 |
+
"learning_rate": 1.9814568101866843e-05,
|
| 308 |
+
"loss": 0.1494,
|
| 309 |
+
"step": 1200
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 7.409302325581395,
|
| 313 |
+
"eval_accuracy": 0.9434644957033017,
|
| 314 |
+
"eval_f1": 0.9428523551904958,
|
| 315 |
+
"eval_loss": 0.16010308265686035,
|
| 316 |
+
"eval_precision": 0.9451435778977758,
|
| 317 |
+
"eval_recall": 0.9434644957033017,
|
| 318 |
+
"eval_runtime": 9.4239,
|
| 319 |
+
"eval_samples_per_second": 234.617,
|
| 320 |
+
"eval_steps_per_second": 29.394,
|
| 321 |
+
"step": 1200
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 7.7193798449612405,
|
| 325 |
+
"grad_norm": 1.8461512327194214,
|
| 326 |
+
"learning_rate": 1.9768776777608227e-05,
|
| 327 |
+
"loss": 0.1437,
|
| 328 |
+
"step": 1250
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 8.024806201550387,
|
| 332 |
+
"grad_norm": 1.100502371788025,
|
| 333 |
+
"learning_rate": 1.971800272500388e-05,
|
| 334 |
+
"loss": 0.1592,
|
| 335 |
+
"step": 1300
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 8.024806201550387,
|
| 339 |
+
"eval_accuracy": 0.9430122116689281,
|
| 340 |
+
"eval_f1": 0.9428538886036008,
|
| 341 |
+
"eval_loss": 0.1429978460073471,
|
| 342 |
+
"eval_precision": 0.9429310401027086,
|
| 343 |
+
"eval_recall": 0.9430122116689281,
|
| 344 |
+
"eval_runtime": 9.449,
|
| 345 |
+
"eval_samples_per_second": 233.992,
|
| 346 |
+
"eval_steps_per_second": 29.315,
|
| 347 |
+
"step": 1300
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 8.334883720930232,
|
| 351 |
+
"grad_norm": 4.64835786819458,
|
| 352 |
+
"learning_rate": 1.9662271842210433e-05,
|
| 353 |
+
"loss": 0.131,
|
| 354 |
+
"step": 1350
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 8.644961240310078,
|
| 358 |
+
"grad_norm": 1.8132234811782837,
|
| 359 |
+
"learning_rate": 1.960161255569886e-05,
|
| 360 |
+
"loss": 0.16,
|
| 361 |
+
"step": 1400
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 8.644961240310078,
|
| 365 |
+
"eval_accuracy": 0.9457259158751696,
|
| 366 |
+
"eval_f1": 0.9451300205802559,
|
| 367 |
+
"eval_loss": 0.1504276692867279,
|
| 368 |
+
"eval_precision": 0.9475179562129199,
|
| 369 |
+
"eval_recall": 0.9457259158751696,
|
| 370 |
+
"eval_runtime": 9.4602,
|
| 371 |
+
"eval_samples_per_second": 233.715,
|
| 372 |
+
"eval_steps_per_second": 29.28,
|
| 373 |
+
"step": 1400
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 8.955038759689922,
|
| 377 |
+
"grad_norm": 1.1923645734786987,
|
| 378 |
+
"learning_rate": 1.9536055805755044e-05,
|
| 379 |
+
"loss": 0.1105,
|
| 380 |
+
"step": 1450
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 9.26046511627907,
|
| 384 |
+
"grad_norm": 2.665161609649658,
|
| 385 |
+
"learning_rate": 1.9465635030698203e-05,
|
| 386 |
+
"loss": 0.1245,
|
| 387 |
+
"step": 1500
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 9.26046511627907,
|
| 391 |
+
"eval_accuracy": 0.9461781999095432,
|
| 392 |
+
"eval_f1": 0.9457529907245916,
|
| 393 |
+
"eval_loss": 0.1505969762802124,
|
| 394 |
+
"eval_precision": 0.9469878940244384,
|
| 395 |
+
"eval_recall": 0.9461781999095432,
|
| 396 |
+
"eval_runtime": 9.4828,
|
| 397 |
+
"eval_samples_per_second": 233.159,
|
| 398 |
+
"eval_steps_per_second": 29.211,
|
| 399 |
+
"step": 1500
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 9.570542635658915,
|
| 403 |
+
"grad_norm": 0.45807984471321106,
|
| 404 |
+
"learning_rate": 1.939038614982509e-05,
|
| 405 |
+
"loss": 0.1342,
|
| 406 |
+
"step": 1550
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 9.88062015503876,
|
| 410 |
+
"grad_norm": 1.0274362564086914,
|
| 411 |
+
"learning_rate": 1.9310347545088764e-05,
|
| 412 |
+
"loss": 0.1397,
|
| 413 |
+
"step": 1600
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 9.88062015503876,
|
| 417 |
+
"eval_accuracy": 0.9312528267752148,
|
| 418 |
+
"eval_f1": 0.9299781453592552,
|
| 419 |
+
"eval_loss": 0.19713716208934784,
|
| 420 |
+
"eval_precision": 0.9359197905804808,
|
| 421 |
+
"eval_recall": 0.9312528267752148,
|
| 422 |
+
"eval_runtime": 9.4563,
|
| 423 |
+
"eval_samples_per_second": 233.811,
|
| 424 |
+
"eval_steps_per_second": 29.293,
|
| 425 |
+
"step": 1600
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 10.186046511627907,
|
| 429 |
+
"grad_norm": 2.8293967247009277,
|
| 430 |
+
"learning_rate": 1.9225560041521225e-05,
|
| 431 |
+
"loss": 0.1615,
|
| 432 |
+
"step": 1650
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 10.496124031007753,
|
| 436 |
+
"grad_norm": 0.916153609752655,
|
| 437 |
+
"learning_rate": 1.913606688640993e-05,
|
| 438 |
+
"loss": 0.1396,
|
| 439 |
+
"step": 1700
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 10.496124031007753,
|
| 443 |
+
"eval_accuracy": 0.9488919041157847,
|
| 444 |
+
"eval_f1": 0.9483693001512903,
|
| 445 |
+
"eval_loss": 0.15268893539905548,
|
| 446 |
+
"eval_precision": 0.9504887785210339,
|
| 447 |
+
"eval_recall": 0.9488919041157847,
|
| 448 |
+
"eval_runtime": 9.4488,
|
| 449 |
+
"eval_samples_per_second": 233.998,
|
| 450 |
+
"eval_steps_per_second": 29.316,
|
| 451 |
+
"step": 1700
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 10.496124031007753,
|
| 455 |
+
"step": 1700,
|
| 456 |
+
"total_flos": 5.998049534242161e+18,
|
| 457 |
+
"train_loss": 0.14669023766237146,
|
| 458 |
+
"train_runtime": 774.0157,
|
| 459 |
+
"train_samples_per_second": 319.869,
|
| 460 |
+
"train_steps_per_second": 9.984
|
| 461 |
+
}
|
| 462 |
+
],
|
| 463 |
+
"logging_steps": 50,
|
| 464 |
+
"max_steps": 7728,
|
| 465 |
+
"num_input_tokens_seen": 0,
|
| 466 |
+
"num_train_epochs": 48,
|
| 467 |
+
"save_steps": 100,
|
| 468 |
+
"stateful_callbacks": {
|
| 469 |
+
"TrainerControl": {
|
| 470 |
+
"args": {
|
| 471 |
+
"should_epoch_stop": false,
|
| 472 |
+
"should_evaluate": false,
|
| 473 |
+
"should_log": false,
|
| 474 |
+
"should_save": true,
|
| 475 |
+
"should_training_stop": true
|
| 476 |
+
},
|
| 477 |
+
"attributes": {}
|
| 478 |
+
}
|
| 479 |
+
},
|
| 480 |
+
"total_flos": 5.998049534242161e+18,
|
| 481 |
+
"train_batch_size": 8,
|
| 482 |
+
"trial_name": null,
|
| 483 |
+
"trial_params": null
|
| 484 |
+
}
|