Commit
·
ffd148f
1
Parent(s):
a13e8b7
🍻 cheers
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"eval_accuracy": 0.7166386319035604,
|
| 4 |
+
"eval_loss": 0.9879873394966125,
|
| 5 |
+
"eval_runtime": 363.7432,
|
| 6 |
+
"eval_samples_per_second": 78.451,
|
| 7 |
+
"eval_steps_per_second": 9.806,
|
| 8 |
+
"train_loss": 0.9299831045534365,
|
| 9 |
+
"train_runtime": 25938.6639,
|
| 10 |
+
"train_samples_per_second": 35.205,
|
| 11 |
+
"train_steps_per_second": 1.1
|
| 12 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"eval_accuracy": 0.7166386319035604,
|
| 4 |
+
"eval_loss": 0.9879873394966125,
|
| 5 |
+
"eval_runtime": 363.7432,
|
| 6 |
+
"eval_samples_per_second": 78.451,
|
| 7 |
+
"eval_steps_per_second": 9.806
|
| 8 |
+
}
|
runs/Jun11_14-32-56_adrian-development/events.out.tfevents.1686513103.adrian-development.17428.2
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7c81766804d59b185a8eb9af67844a622cca065008040afffb5528d9b5e9386b
|
| 3 |
+
size 369
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 4.0,
|
| 3 |
+
"train_loss": 0.9299831045534365,
|
| 4 |
+
"train_runtime": 25938.6639,
|
| 5 |
+
"train_samples_per_second": 35.205,
|
| 6 |
+
"train_steps_per_second": 1.1
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,445 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9879873394966125,
|
| 3 |
+
"best_model_checkpoint": "Transformers/vit-base-clothing-leafs-example-full-simple_highres/checkpoint-20000",
|
| 4 |
+
"epoch": 4.0,
|
| 5 |
+
"global_step": 28540,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.14,
|
| 12 |
+
"learning_rate": 2.4124036440084092e-05,
|
| 13 |
+
"loss": 2.0202,
|
| 14 |
+
"step": 1000
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.14,
|
| 18 |
+
"eval_accuracy": 0.6338309503784693,
|
| 19 |
+
"eval_loss": 1.4968905448913574,
|
| 20 |
+
"eval_runtime": 393.745,
|
| 21 |
+
"eval_samples_per_second": 72.473,
|
| 22 |
+
"eval_steps_per_second": 9.059,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.28,
|
| 27 |
+
"learning_rate": 2.32489488437281e-05,
|
| 28 |
+
"loss": 1.3694,
|
| 29 |
+
"step": 2000
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 0.28,
|
| 33 |
+
"eval_accuracy": 0.6647392767031118,
|
| 34 |
+
"eval_loss": 1.2786486148834229,
|
| 35 |
+
"eval_runtime": 367.4489,
|
| 36 |
+
"eval_samples_per_second": 77.66,
|
| 37 |
+
"eval_steps_per_second": 9.707,
|
| 38 |
+
"step": 2000
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.42,
|
| 42 |
+
"learning_rate": 2.237386124737211e-05,
|
| 43 |
+
"loss": 1.2063,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.42,
|
| 48 |
+
"eval_accuracy": 0.6793523969722456,
|
| 49 |
+
"eval_loss": 1.1788005828857422,
|
| 50 |
+
"eval_runtime": 365.4215,
|
| 51 |
+
"eval_samples_per_second": 78.091,
|
| 52 |
+
"eval_steps_per_second": 9.761,
|
| 53 |
+
"step": 3000
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.56,
|
| 57 |
+
"learning_rate": 2.1497897687456204e-05,
|
| 58 |
+
"loss": 1.1544,
|
| 59 |
+
"step": 4000
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.56,
|
| 63 |
+
"eval_accuracy": 0.6855550883095037,
|
| 64 |
+
"eval_loss": 1.1320449113845825,
|
| 65 |
+
"eval_runtime": 363.0958,
|
| 66 |
+
"eval_samples_per_second": 78.591,
|
| 67 |
+
"eval_steps_per_second": 9.824,
|
| 68 |
+
"step": 4000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.7,
|
| 72 |
+
"learning_rate": 2.0622810091100212e-05,
|
| 73 |
+
"loss": 1.1089,
|
| 74 |
+
"step": 5000
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.7,
|
| 78 |
+
"eval_accuracy": 0.6867465657415195,
|
| 79 |
+
"eval_loss": 1.1020501852035522,
|
| 80 |
+
"eval_runtime": 361.1213,
|
| 81 |
+
"eval_samples_per_second": 79.021,
|
| 82 |
+
"eval_steps_per_second": 9.878,
|
| 83 |
+
"step": 5000
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.84,
|
| 87 |
+
"learning_rate": 1.9746846531184303e-05,
|
| 88 |
+
"loss": 1.0681,
|
| 89 |
+
"step": 6000
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.84,
|
| 93 |
+
"eval_accuracy": 0.6935449957947856,
|
| 94 |
+
"eval_loss": 1.0774946212768555,
|
| 95 |
+
"eval_runtime": 364.542,
|
| 96 |
+
"eval_samples_per_second": 78.279,
|
| 97 |
+
"eval_steps_per_second": 9.785,
|
| 98 |
+
"step": 6000
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.98,
|
| 102 |
+
"learning_rate": 1.8870882971268398e-05,
|
| 103 |
+
"loss": 1.0483,
|
| 104 |
+
"step": 7000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.98,
|
| 108 |
+
"eval_accuracy": 0.7006237734791141,
|
| 109 |
+
"eval_loss": 1.0460755825042725,
|
| 110 |
+
"eval_runtime": 359.7119,
|
| 111 |
+
"eval_samples_per_second": 79.33,
|
| 112 |
+
"eval_steps_per_second": 9.916,
|
| 113 |
+
"step": 7000
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 1.12,
|
| 117 |
+
"learning_rate": 1.7995795374912406e-05,
|
| 118 |
+
"loss": 0.9591,
|
| 119 |
+
"step": 8000
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 1.12,
|
| 123 |
+
"eval_accuracy": 0.7022007289038408,
|
| 124 |
+
"eval_loss": 1.0398491621017456,
|
| 125 |
+
"eval_runtime": 363.8681,
|
| 126 |
+
"eval_samples_per_second": 78.424,
|
| 127 |
+
"eval_steps_per_second": 9.803,
|
| 128 |
+
"step": 8000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 1.26,
|
| 132 |
+
"learning_rate": 1.7119831814996497e-05,
|
| 133 |
+
"loss": 0.9541,
|
| 134 |
+
"step": 9000
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 1.26,
|
| 138 |
+
"eval_accuracy": 0.6981356882534343,
|
| 139 |
+
"eval_loss": 1.0423301458358765,
|
| 140 |
+
"eval_runtime": 365.8278,
|
| 141 |
+
"eval_samples_per_second": 78.004,
|
| 142 |
+
"eval_steps_per_second": 9.75,
|
| 143 |
+
"step": 9000
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 1.4,
|
| 147 |
+
"learning_rate": 1.624386825508059e-05,
|
| 148 |
+
"loss": 0.9382,
|
| 149 |
+
"step": 10000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 1.4,
|
| 153 |
+
"eval_accuracy": 0.7013947294645361,
|
| 154 |
+
"eval_loss": 1.0322409868240356,
|
| 155 |
+
"eval_runtime": 359.094,
|
| 156 |
+
"eval_samples_per_second": 79.467,
|
| 157 |
+
"eval_steps_per_second": 9.933,
|
| 158 |
+
"step": 10000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.54,
|
| 162 |
+
"learning_rate": 1.5369656622284514e-05,
|
| 163 |
+
"loss": 0.9363,
|
| 164 |
+
"step": 11000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 1.54,
|
| 168 |
+
"eval_accuracy": 0.7019554247266611,
|
| 169 |
+
"eval_loss": 1.0300703048706055,
|
| 170 |
+
"eval_runtime": 396.2796,
|
| 171 |
+
"eval_samples_per_second": 72.01,
|
| 172 |
+
"eval_steps_per_second": 9.001,
|
| 173 |
+
"step": 11000
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 1.68,
|
| 177 |
+
"learning_rate": 1.4493693062368607e-05,
|
| 178 |
+
"loss": 0.9199,
|
| 179 |
+
"step": 12000
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 1.68,
|
| 183 |
+
"eval_accuracy": 0.7105761143818334,
|
| 184 |
+
"eval_loss": 1.0079022645950317,
|
| 185 |
+
"eval_runtime": 360.6707,
|
| 186 |
+
"eval_samples_per_second": 79.119,
|
| 187 |
+
"eval_steps_per_second": 9.89,
|
| 188 |
+
"step": 12000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 1.82,
|
| 192 |
+
"learning_rate": 1.3617729502452698e-05,
|
| 193 |
+
"loss": 0.919,
|
| 194 |
+
"step": 13000
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 1.82,
|
| 198 |
+
"eval_accuracy": 0.7119778525371461,
|
| 199 |
+
"eval_loss": 0.9972480535507202,
|
| 200 |
+
"eval_runtime": 361.6963,
|
| 201 |
+
"eval_samples_per_second": 78.895,
|
| 202 |
+
"eval_steps_per_second": 9.862,
|
| 203 |
+
"step": 13000
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 1.96,
|
| 207 |
+
"learning_rate": 1.2741765942536791e-05,
|
| 208 |
+
"loss": 0.9203,
|
| 209 |
+
"step": 14000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 1.96,
|
| 213 |
+
"eval_accuracy": 0.7095948976731147,
|
| 214 |
+
"eval_loss": 1.0011298656463623,
|
| 215 |
+
"eval_runtime": 358.1777,
|
| 216 |
+
"eval_samples_per_second": 79.67,
|
| 217 |
+
"eval_steps_per_second": 9.959,
|
| 218 |
+
"step": 14000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 2.1,
|
| 222 |
+
"learning_rate": 1.1865802382620882e-05,
|
| 223 |
+
"loss": 0.8377,
|
| 224 |
+
"step": 15000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 2.1,
|
| 228 |
+
"eval_accuracy": 0.7146061115783572,
|
| 229 |
+
"eval_loss": 0.9911500811576843,
|
| 230 |
+
"eval_runtime": 423.0343,
|
| 231 |
+
"eval_samples_per_second": 67.456,
|
| 232 |
+
"eval_steps_per_second": 8.432,
|
| 233 |
+
"step": 15000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 2.24,
|
| 237 |
+
"learning_rate": 1.0991590749824808e-05,
|
| 238 |
+
"loss": 0.8148,
|
| 239 |
+
"step": 16000
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 2.24,
|
| 243 |
+
"eval_accuracy": 0.7121180263526773,
|
| 244 |
+
"eval_loss": 0.9991061091423035,
|
| 245 |
+
"eval_runtime": 345.3108,
|
| 246 |
+
"eval_samples_per_second": 82.639,
|
| 247 |
+
"eval_steps_per_second": 10.33,
|
| 248 |
+
"step": 16000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 2.38,
|
| 252 |
+
"learning_rate": 1.01156271899089e-05,
|
| 253 |
+
"loss": 0.8153,
|
| 254 |
+
"step": 17000
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 2.38,
|
| 258 |
+
"eval_accuracy": 0.7101906363891225,
|
| 259 |
+
"eval_loss": 1.007048487663269,
|
| 260 |
+
"eval_runtime": 342.0626,
|
| 261 |
+
"eval_samples_per_second": 83.423,
|
| 262 |
+
"eval_steps_per_second": 10.428,
|
| 263 |
+
"step": 17000
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 2.52,
|
| 267 |
+
"learning_rate": 9.239663629992992e-06,
|
| 268 |
+
"loss": 0.8004,
|
| 269 |
+
"step": 18000
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 2.52,
|
| 273 |
+
"eval_accuracy": 0.7154121110176619,
|
| 274 |
+
"eval_loss": 0.9979108572006226,
|
| 275 |
+
"eval_runtime": 365.4895,
|
| 276 |
+
"eval_samples_per_second": 78.076,
|
| 277 |
+
"eval_steps_per_second": 9.76,
|
| 278 |
+
"step": 18000
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 2.66,
|
| 282 |
+
"learning_rate": 8.363700070077085e-06,
|
| 283 |
+
"loss": 0.7937,
|
| 284 |
+
"step": 19000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 2.66,
|
| 288 |
+
"eval_accuracy": 0.7135548079618728,
|
| 289 |
+
"eval_loss": 1.0022467374801636,
|
| 290 |
+
"eval_runtime": 497.9936,
|
| 291 |
+
"eval_samples_per_second": 57.302,
|
| 292 |
+
"eval_steps_per_second": 7.163,
|
| 293 |
+
"step": 19000
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 2.8,
|
| 297 |
+
"learning_rate": 7.487736510161178e-06,
|
| 298 |
+
"loss": 0.7989,
|
| 299 |
+
"step": 20000
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 2.8,
|
| 303 |
+
"eval_accuracy": 0.7166386319035604,
|
| 304 |
+
"eval_loss": 0.9879873394966125,
|
| 305 |
+
"eval_runtime": 369.3274,
|
| 306 |
+
"eval_samples_per_second": 77.265,
|
| 307 |
+
"eval_steps_per_second": 9.658,
|
| 308 |
+
"step": 20000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 2.94,
|
| 312 |
+
"learning_rate": 6.612648913805186e-06,
|
| 313 |
+
"loss": 0.7953,
|
| 314 |
+
"step": 21000
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 2.94,
|
| 318 |
+
"eval_accuracy": 0.7175497617045136,
|
| 319 |
+
"eval_loss": 0.990685760974884,
|
| 320 |
+
"eval_runtime": 375.6866,
|
| 321 |
+
"eval_samples_per_second": 75.957,
|
| 322 |
+
"eval_steps_per_second": 9.495,
|
| 323 |
+
"step": 21000
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 3.08,
|
| 327 |
+
"learning_rate": 5.736685353889278e-06,
|
| 328 |
+
"loss": 0.7576,
|
| 329 |
+
"step": 22000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 3.08,
|
| 333 |
+
"eval_accuracy": 0.7136248948696383,
|
| 334 |
+
"eval_loss": 1.0013048648834229,
|
| 335 |
+
"eval_runtime": 367.0966,
|
| 336 |
+
"eval_samples_per_second": 77.734,
|
| 337 |
+
"eval_steps_per_second": 9.717,
|
| 338 |
+
"step": 22000
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 3.22,
|
| 342 |
+
"learning_rate": 4.861597757533287e-06,
|
| 343 |
+
"loss": 0.7018,
|
| 344 |
+
"step": 23000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 3.22,
|
| 348 |
+
"eval_accuracy": 0.7155873282870759,
|
| 349 |
+
"eval_loss": 1.0022220611572266,
|
| 350 |
+
"eval_runtime": 367.5473,
|
| 351 |
+
"eval_samples_per_second": 77.639,
|
| 352 |
+
"eval_steps_per_second": 9.705,
|
| 353 |
+
"step": 23000
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 3.36,
|
| 357 |
+
"learning_rate": 3.986510161177296e-06,
|
| 358 |
+
"loss": 0.7127,
|
| 359 |
+
"step": 24000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 3.36,
|
| 363 |
+
"eval_accuracy": 0.7150616764788338,
|
| 364 |
+
"eval_loss": 1.0079938173294067,
|
| 365 |
+
"eval_runtime": 377.1194,
|
| 366 |
+
"eval_samples_per_second": 75.668,
|
| 367 |
+
"eval_steps_per_second": 9.459,
|
| 368 |
+
"step": 24000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 3.5,
|
| 372 |
+
"learning_rate": 3.1105466012613877e-06,
|
| 373 |
+
"loss": 0.6989,
|
| 374 |
+
"step": 25000
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 3.5,
|
| 378 |
+
"eval_accuracy": 0.7159377628259042,
|
| 379 |
+
"eval_loss": 1.0025221109390259,
|
| 380 |
+
"eval_runtime": 358.2569,
|
| 381 |
+
"eval_samples_per_second": 79.652,
|
| 382 |
+
"eval_steps_per_second": 9.957,
|
| 383 |
+
"step": 25000
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 3.64,
|
| 387 |
+
"learning_rate": 2.23458304134548e-06,
|
| 388 |
+
"loss": 0.702,
|
| 389 |
+
"step": 26000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 3.64,
|
| 393 |
+
"eval_accuracy": 0.7167437622652089,
|
| 394 |
+
"eval_loss": 1.008726954460144,
|
| 395 |
+
"eval_runtime": 368.5339,
|
| 396 |
+
"eval_samples_per_second": 77.431,
|
| 397 |
+
"eval_steps_per_second": 9.679,
|
| 398 |
+
"step": 26000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 3.78,
|
| 402 |
+
"learning_rate": 1.3594954449894885e-06,
|
| 403 |
+
"loss": 0.7122,
|
| 404 |
+
"step": 27000
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 3.78,
|
| 408 |
+
"eval_accuracy": 0.7158676759181385,
|
| 409 |
+
"eval_loss": 1.0042390823364258,
|
| 410 |
+
"eval_runtime": 366.8278,
|
| 411 |
+
"eval_samples_per_second": 77.791,
|
| 412 |
+
"eval_steps_per_second": 9.724,
|
| 413 |
+
"step": 27000
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 3.92,
|
| 417 |
+
"learning_rate": 4.83531885073581e-07,
|
| 418 |
+
"loss": 0.6986,
|
| 419 |
+
"step": 28000
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 3.92,
|
| 423 |
+
"eval_accuracy": 0.7163582842724979,
|
| 424 |
+
"eval_loss": 1.0016725063323975,
|
| 425 |
+
"eval_runtime": 363.5122,
|
| 426 |
+
"eval_samples_per_second": 78.501,
|
| 427 |
+
"eval_steps_per_second": 9.813,
|
| 428 |
+
"step": 28000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 4.0,
|
| 432 |
+
"step": 28540,
|
| 433 |
+
"total_flos": 7.0790801124097475e+19,
|
| 434 |
+
"train_loss": 0.9299831045534365,
|
| 435 |
+
"train_runtime": 25938.6639,
|
| 436 |
+
"train_samples_per_second": 35.205,
|
| 437 |
+
"train_steps_per_second": 1.1
|
| 438 |
+
}
|
| 439 |
+
],
|
| 440 |
+
"max_steps": 28540,
|
| 441 |
+
"num_train_epochs": 4,
|
| 442 |
+
"total_flos": 7.0790801124097475e+19,
|
| 443 |
+
"trial_name": null,
|
| 444 |
+
"trial_params": null
|
| 445 |
+
}
|