paul
commited on
Commit
•
dcc41cf
1
Parent(s):
e68b87b
End of training
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 9.99,
|
| 3 |
+
"total_flos": 4.005358668612661e+18,
|
| 4 |
+
"train_loss": 0.31261438488960264,
|
| 5 |
+
"train_runtime": 5959.0067,
|
| 6 |
+
"train_samples_per_second": 8.683,
|
| 7 |
+
"train_steps_per_second": 0.067
|
| 8 |
+
}
|
pytorch_model.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 343276209
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:31fcd36d7b965cfa08a7af2677134f3f224f4afb80676ca9545dd71ef49a79f2
|
| 3 |
size 343276209
|
runs/Sep18_12-30-38_9322f79f835b/events.out.tfevents.1663504248.9322f79f835b.71.2
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:87388a0483c92c78822e69a5d9a2e435b193f54b30a3da7a136a055b8a65cb02
|
| 3 |
+
size 13828
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 9.99,
|
| 3 |
+
"total_flos": 4.005358668612661e+18,
|
| 4 |
+
"train_loss": 0.31261438488960264,
|
| 5 |
+
"train_runtime": 5959.0067,
|
| 6 |
+
"train_samples_per_second": 8.683,
|
| 7 |
+
"train_steps_per_second": 0.067
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,355 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.8593508500772797,
|
| 3 |
+
"best_model_checkpoint": "mit-b2-finetuned-memes/checkpoint-360",
|
| 4 |
+
"epoch": 9.987654320987655,
|
| 5 |
+
"global_step": 400,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.25,
|
| 12 |
+
"learning_rate": 3e-05,
|
| 13 |
+
"loss": 1.5809,
|
| 14 |
+
"step": 10
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.49,
|
| 18 |
+
"learning_rate": 6e-05,
|
| 19 |
+
"loss": 1.3935,
|
| 20 |
+
"step": 20
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.74,
|
| 24 |
+
"learning_rate": 9e-05,
|
| 25 |
+
"loss": 1.2125,
|
| 26 |
+
"step": 30
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.99,
|
| 30 |
+
"learning_rate": 0.00012,
|
| 31 |
+
"loss": 0.9727,
|
| 32 |
+
"step": 40
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.99,
|
| 36 |
+
"eval_accuracy": 0.7333848531684699,
|
| 37 |
+
"eval_loss": 0.8400309085845947,
|
| 38 |
+
"eval_runtime": 775.4738,
|
| 39 |
+
"eval_samples_per_second": 1.669,
|
| 40 |
+
"eval_steps_per_second": 0.053,
|
| 41 |
+
"step": 40
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 1.25,
|
| 45 |
+
"learning_rate": 0.00011666666666666667,
|
| 46 |
+
"loss": 0.8348,
|
| 47 |
+
"step": 50
|
| 48 |
+
},
|
| 49 |
+
{
|
| 50 |
+
"epoch": 1.49,
|
| 51 |
+
"learning_rate": 0.00011333333333333333,
|
| 52 |
+
"loss": 0.6271,
|
| 53 |
+
"step": 60
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.74,
|
| 57 |
+
"learning_rate": 0.00011,
|
| 58 |
+
"loss": 0.6043,
|
| 59 |
+
"step": 70
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 1.99,
|
| 63 |
+
"learning_rate": 0.00010666666666666667,
|
| 64 |
+
"loss": 0.5305,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 1.99,
|
| 69 |
+
"eval_accuracy": 0.8284389489953632,
|
| 70 |
+
"eval_loss": 0.5146808624267578,
|
| 71 |
+
"eval_runtime": 23.9016,
|
| 72 |
+
"eval_samples_per_second": 54.139,
|
| 73 |
+
"eval_steps_per_second": 1.715,
|
| 74 |
+
"step": 80
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 2.25,
|
| 78 |
+
"learning_rate": 0.00010333333333333334,
|
| 79 |
+
"loss": 0.4163,
|
| 80 |
+
"step": 90
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 2.49,
|
| 84 |
+
"learning_rate": 0.0001,
|
| 85 |
+
"loss": 0.3372,
|
| 86 |
+
"step": 100
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 2.74,
|
| 90 |
+
"learning_rate": 9.666666666666667e-05,
|
| 91 |
+
"loss": 0.3527,
|
| 92 |
+
"step": 110
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 2.99,
|
| 96 |
+
"learning_rate": 9.333333333333334e-05,
|
| 97 |
+
"loss": 0.3124,
|
| 98 |
+
"step": 120
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 2.99,
|
| 102 |
+
"eval_accuracy": 0.8145285935085008,
|
| 103 |
+
"eval_loss": 0.4697549641132355,
|
| 104 |
+
"eval_runtime": 22.8386,
|
| 105 |
+
"eval_samples_per_second": 56.658,
|
| 106 |
+
"eval_steps_per_second": 1.795,
|
| 107 |
+
"step": 120
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 3.25,
|
| 111 |
+
"learning_rate": 9e-05,
|
| 112 |
+
"loss": 0.2821,
|
| 113 |
+
"step": 130
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 3.49,
|
| 117 |
+
"learning_rate": 8.666666666666667e-05,
|
| 118 |
+
"loss": 0.2588,
|
| 119 |
+
"step": 140
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 3.74,
|
| 123 |
+
"learning_rate": 8.333333333333333e-05,
|
| 124 |
+
"loss": 0.1866,
|
| 125 |
+
"step": 150
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 3.99,
|
| 129 |
+
"learning_rate": 7.999999999999999e-05,
|
| 130 |
+
"loss": 0.2263,
|
| 131 |
+
"step": 160
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 3.99,
|
| 135 |
+
"eval_accuracy": 0.8562596599690881,
|
| 136 |
+
"eval_loss": 0.3892250061035156,
|
| 137 |
+
"eval_runtime": 22.8733,
|
| 138 |
+
"eval_samples_per_second": 56.572,
|
| 139 |
+
"eval_steps_per_second": 1.792,
|
| 140 |
+
"step": 160
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 4.25,
|
| 144 |
+
"learning_rate": 7.666666666666667e-05,
|
| 145 |
+
"loss": 0.1584,
|
| 146 |
+
"step": 170
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 4.49,
|
| 150 |
+
"learning_rate": 7.333333333333334e-05,
|
| 151 |
+
"loss": 0.1588,
|
| 152 |
+
"step": 180
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 4.74,
|
| 156 |
+
"learning_rate": 7.000000000000001e-05,
|
| 157 |
+
"loss": 0.1487,
|
| 158 |
+
"step": 190
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 4.99,
|
| 162 |
+
"learning_rate": 6.666666666666667e-05,
|
| 163 |
+
"loss": 0.1453,
|
| 164 |
+
"step": 200
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 4.99,
|
| 168 |
+
"eval_accuracy": 0.857032457496136,
|
| 169 |
+
"eval_loss": 0.38735243678092957,
|
| 170 |
+
"eval_runtime": 22.8684,
|
| 171 |
+
"eval_samples_per_second": 56.585,
|
| 172 |
+
"eval_steps_per_second": 1.793,
|
| 173 |
+
"step": 200
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 5.25,
|
| 177 |
+
"learning_rate": 6.333333333333333e-05,
|
| 178 |
+
"loss": 0.1254,
|
| 179 |
+
"step": 210
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 5.49,
|
| 183 |
+
"learning_rate": 6e-05,
|
| 184 |
+
"loss": 0.1012,
|
| 185 |
+
"step": 220
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 5.74,
|
| 189 |
+
"learning_rate": 5.6666666666666664e-05,
|
| 190 |
+
"loss": 0.1078,
|
| 191 |
+
"step": 230
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 5.99,
|
| 195 |
+
"learning_rate": 5.333333333333333e-05,
|
| 196 |
+
"loss": 0.1255,
|
| 197 |
+
"step": 240
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 5.99,
|
| 201 |
+
"eval_accuracy": 0.8469860896445132,
|
| 202 |
+
"eval_loss": 0.4097033441066742,
|
| 203 |
+
"eval_runtime": 22.8903,
|
| 204 |
+
"eval_samples_per_second": 56.53,
|
| 205 |
+
"eval_steps_per_second": 1.791,
|
| 206 |
+
"step": 240
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 6.25,
|
| 210 |
+
"learning_rate": 5e-05,
|
| 211 |
+
"loss": 0.0963,
|
| 212 |
+
"step": 250
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 6.49,
|
| 216 |
+
"learning_rate": 4.666666666666667e-05,
|
| 217 |
+
"loss": 0.1155,
|
| 218 |
+
"step": 260
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 6.74,
|
| 222 |
+
"learning_rate": 4.3333333333333334e-05,
|
| 223 |
+
"loss": 0.0821,
|
| 224 |
+
"step": 270
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 6.99,
|
| 228 |
+
"learning_rate": 3.9999999999999996e-05,
|
| 229 |
+
"loss": 0.0989,
|
| 230 |
+
"step": 280
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 6.99,
|
| 234 |
+
"eval_accuracy": 0.857032457496136,
|
| 235 |
+
"eval_loss": 0.38595816493034363,
|
| 236 |
+
"eval_runtime": 23.0627,
|
| 237 |
+
"eval_samples_per_second": 56.108,
|
| 238 |
+
"eval_steps_per_second": 1.778,
|
| 239 |
+
"step": 280
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 7.25,
|
| 243 |
+
"learning_rate": 3.666666666666667e-05,
|
| 244 |
+
"loss": 0.0848,
|
| 245 |
+
"step": 290
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 7.49,
|
| 249 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 250 |
+
"loss": 0.0834,
|
| 251 |
+
"step": 300
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 7.74,
|
| 255 |
+
"learning_rate": 3e-05,
|
| 256 |
+
"loss": 0.0818,
|
| 257 |
+
"step": 310
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 7.99,
|
| 261 |
+
"learning_rate": 2.6666666666666667e-05,
|
| 262 |
+
"loss": 0.0755,
|
| 263 |
+
"step": 320
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 7.99,
|
| 267 |
+
"eval_accuracy": 0.8539412673879444,
|
| 268 |
+
"eval_loss": 0.4140683114528656,
|
| 269 |
+
"eval_runtime": 22.9765,
|
| 270 |
+
"eval_samples_per_second": 56.318,
|
| 271 |
+
"eval_steps_per_second": 1.784,
|
| 272 |
+
"step": 320
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 8.25,
|
| 276 |
+
"learning_rate": 2.3333333333333336e-05,
|
| 277 |
+
"loss": 0.0691,
|
| 278 |
+
"step": 330
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 8.49,
|
| 282 |
+
"learning_rate": 1.9999999999999998e-05,
|
| 283 |
+
"loss": 0.0721,
|
| 284 |
+
"step": 340
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 8.74,
|
| 288 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 289 |
+
"loss": 0.0804,
|
| 290 |
+
"step": 350
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 8.99,
|
| 294 |
+
"learning_rate": 1.3333333333333333e-05,
|
| 295 |
+
"loss": 0.08,
|
| 296 |
+
"step": 360
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 8.99,
|
| 300 |
+
"eval_accuracy": 0.8593508500772797,
|
| 301 |
+
"eval_loss": 0.40489473938941956,
|
| 302 |
+
"eval_runtime": 22.9961,
|
| 303 |
+
"eval_samples_per_second": 56.271,
|
| 304 |
+
"eval_steps_per_second": 1.783,
|
| 305 |
+
"step": 360
|
| 306 |
+
},
|
| 307 |
+
{
|
| 308 |
+
"epoch": 9.25,
|
| 309 |
+
"learning_rate": 9.999999999999999e-06,
|
| 310 |
+
"loss": 0.0782,
|
| 311 |
+
"step": 370
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 9.49,
|
| 315 |
+
"learning_rate": 6.666666666666667e-06,
|
| 316 |
+
"loss": 0.0811,
|
| 317 |
+
"step": 380
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 9.74,
|
| 321 |
+
"learning_rate": 3.3333333333333333e-06,
|
| 322 |
+
"loss": 0.0619,
|
| 323 |
+
"step": 390
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 9.99,
|
| 327 |
+
"learning_rate": 0.0,
|
| 328 |
+
"loss": 0.0639,
|
| 329 |
+
"step": 400
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 9.99,
|
| 333 |
+
"eval_accuracy": 0.8523956723338485,
|
| 334 |
+
"eval_loss": 0.4136844873428345,
|
| 335 |
+
"eval_runtime": 22.9197,
|
| 336 |
+
"eval_samples_per_second": 56.458,
|
| 337 |
+
"eval_steps_per_second": 1.789,
|
| 338 |
+
"step": 400
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 9.99,
|
| 342 |
+
"step": 400,
|
| 343 |
+
"total_flos": 4.005358668612661e+18,
|
| 344 |
+
"train_loss": 0.31261438488960264,
|
| 345 |
+
"train_runtime": 5959.0067,
|
| 346 |
+
"train_samples_per_second": 8.683,
|
| 347 |
+
"train_steps_per_second": 0.067
|
| 348 |
+
}
|
| 349 |
+
],
|
| 350 |
+
"max_steps": 400,
|
| 351 |
+
"num_train_epochs": 10,
|
| 352 |
+
"total_flos": 4.005358668612661e+18,
|
| 353 |
+
"trial_name": null,
|
| 354 |
+
"trial_params": null
|
| 355 |
+
}
|