End of training
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.25,
|
| 3 |
+
"eval_accuracy": 0.8451612903225807,
|
| 4 |
+
"eval_loss": 0.36612647771835327,
|
| 5 |
+
"eval_runtime": 44.7201,
|
| 6 |
+
"eval_samples_per_second": 3.466,
|
| 7 |
+
"eval_steps_per_second": 0.872
|
| 8 |
+
}
|
runs/May10_22-29-31_8d9245ac6ff8/events.out.tfevents.1715381069.8d9245ac6ff8.6532.1
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:b7f869ff6af64f3e391422b573b02a41f503d8e5d7db455b60fa517ebeb8cbb9
|
| 3 |
+
size 734
|
test_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.25,
|
| 3 |
+
"eval_accuracy": 0.8451612903225807,
|
| 4 |
+
"eval_loss": 0.36612647771835327,
|
| 5 |
+
"eval_runtime": 44.7201,
|
| 6 |
+
"eval_samples_per_second": 3.466,
|
| 7 |
+
"eval_steps_per_second": 0.872
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,294 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9285714285714286,
|
| 3 |
+
"best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-225",
|
| 4 |
+
"epoch": 3.25,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 300,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.03333333333333333,
|
| 13 |
+
"grad_norm": 10.09014892578125,
|
| 14 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 15 |
+
"loss": 2.3797,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.06666666666666667,
|
| 20 |
+
"grad_norm": 9.839759826660156,
|
| 21 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 22 |
+
"loss": 2.2833,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.1,
|
| 27 |
+
"grad_norm": 8.606062889099121,
|
| 28 |
+
"learning_rate": 5e-05,
|
| 29 |
+
"loss": 2.2797,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.13333333333333333,
|
| 34 |
+
"grad_norm": 8.876961708068848,
|
| 35 |
+
"learning_rate": 4.814814814814815e-05,
|
| 36 |
+
"loss": 2.0507,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.16666666666666666,
|
| 41 |
+
"grad_norm": 14.711962699890137,
|
| 42 |
+
"learning_rate": 4.62962962962963e-05,
|
| 43 |
+
"loss": 2.0574,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.2,
|
| 48 |
+
"grad_norm": 10.008936882019043,
|
| 49 |
+
"learning_rate": 4.4444444444444447e-05,
|
| 50 |
+
"loss": 1.8826,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.23333333333333334,
|
| 55 |
+
"grad_norm": 14.089128494262695,
|
| 56 |
+
"learning_rate": 4.259259259259259e-05,
|
| 57 |
+
"loss": 2.0257,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.25,
|
| 62 |
+
"eval_accuracy": 0.6142857142857143,
|
| 63 |
+
"eval_loss": 1.4213531017303467,
|
| 64 |
+
"eval_runtime": 21.4927,
|
| 65 |
+
"eval_samples_per_second": 3.257,
|
| 66 |
+
"eval_steps_per_second": 0.837,
|
| 67 |
+
"step": 75
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 1.0166666666666666,
|
| 71 |
+
"grad_norm": 11.685139656066895,
|
| 72 |
+
"learning_rate": 4.074074074074074e-05,
|
| 73 |
+
"loss": 1.4594,
|
| 74 |
+
"step": 80
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 1.05,
|
| 78 |
+
"grad_norm": 15.033803939819336,
|
| 79 |
+
"learning_rate": 3.888888888888889e-05,
|
| 80 |
+
"loss": 1.324,
|
| 81 |
+
"step": 90
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 1.0833333333333333,
|
| 85 |
+
"grad_norm": 10.9203462600708,
|
| 86 |
+
"learning_rate": 3.7037037037037037e-05,
|
| 87 |
+
"loss": 1.0948,
|
| 88 |
+
"step": 100
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 1.1166666666666667,
|
| 92 |
+
"grad_norm": 5.745022296905518,
|
| 93 |
+
"learning_rate": 3.518518518518519e-05,
|
| 94 |
+
"loss": 0.8842,
|
| 95 |
+
"step": 110
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 1.15,
|
| 99 |
+
"grad_norm": 9.993545532226562,
|
| 100 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 101 |
+
"loss": 1.0636,
|
| 102 |
+
"step": 120
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.1833333333333333,
|
| 106 |
+
"grad_norm": 17.138124465942383,
|
| 107 |
+
"learning_rate": 3.148148148148148e-05,
|
| 108 |
+
"loss": 0.6466,
|
| 109 |
+
"step": 130
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 1.2166666666666668,
|
| 113 |
+
"grad_norm": 5.373832702636719,
|
| 114 |
+
"learning_rate": 2.962962962962963e-05,
|
| 115 |
+
"loss": 0.6253,
|
| 116 |
+
"step": 140
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 1.25,
|
| 120 |
+
"grad_norm": 4.458014488220215,
|
| 121 |
+
"learning_rate": 2.777777777777778e-05,
|
| 122 |
+
"loss": 1.0258,
|
| 123 |
+
"step": 150
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 1.25,
|
| 127 |
+
"eval_accuracy": 0.8,
|
| 128 |
+
"eval_loss": 0.6301711797714233,
|
| 129 |
+
"eval_runtime": 20.959,
|
| 130 |
+
"eval_samples_per_second": 3.34,
|
| 131 |
+
"eval_steps_per_second": 0.859,
|
| 132 |
+
"step": 150
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 2.033333333333333,
|
| 136 |
+
"grad_norm": 10.006092071533203,
|
| 137 |
+
"learning_rate": 2.5925925925925925e-05,
|
| 138 |
+
"loss": 0.3082,
|
| 139 |
+
"step": 160
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 2.066666666666667,
|
| 143 |
+
"grad_norm": 10.732275009155273,
|
| 144 |
+
"learning_rate": 2.4074074074074074e-05,
|
| 145 |
+
"loss": 0.5016,
|
| 146 |
+
"step": 170
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 2.1,
|
| 150 |
+
"grad_norm": 6.761454105377197,
|
| 151 |
+
"learning_rate": 2.2222222222222223e-05,
|
| 152 |
+
"loss": 0.4114,
|
| 153 |
+
"step": 180
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 2.1333333333333333,
|
| 157 |
+
"grad_norm": 2.822266101837158,
|
| 158 |
+
"learning_rate": 2.037037037037037e-05,
|
| 159 |
+
"loss": 0.4457,
|
| 160 |
+
"step": 190
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 2.1666666666666665,
|
| 164 |
+
"grad_norm": 6.605894088745117,
|
| 165 |
+
"learning_rate": 1.8518518518518518e-05,
|
| 166 |
+
"loss": 0.4063,
|
| 167 |
+
"step": 200
|
| 168 |
+
},
|
| 169 |
+
{
|
| 170 |
+
"epoch": 2.2,
|
| 171 |
+
"grad_norm": 18.81662940979004,
|
| 172 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 173 |
+
"loss": 0.2629,
|
| 174 |
+
"step": 210
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 2.2333333333333334,
|
| 178 |
+
"grad_norm": 5.214437484741211,
|
| 179 |
+
"learning_rate": 1.4814814814814815e-05,
|
| 180 |
+
"loss": 0.3566,
|
| 181 |
+
"step": 220
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 2.25,
|
| 185 |
+
"eval_accuracy": 0.9285714285714286,
|
| 186 |
+
"eval_loss": 0.31571879982948303,
|
| 187 |
+
"eval_runtime": 21.9318,
|
| 188 |
+
"eval_samples_per_second": 3.192,
|
| 189 |
+
"eval_steps_per_second": 0.821,
|
| 190 |
+
"step": 225
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 3.0166666666666666,
|
| 194 |
+
"grad_norm": 2.7175469398498535,
|
| 195 |
+
"learning_rate": 1.2962962962962962e-05,
|
| 196 |
+
"loss": 0.1894,
|
| 197 |
+
"step": 230
|
| 198 |
+
},
|
| 199 |
+
{
|
| 200 |
+
"epoch": 3.05,
|
| 201 |
+
"grad_norm": 0.32148978114128113,
|
| 202 |
+
"learning_rate": 1.1111111111111112e-05,
|
| 203 |
+
"loss": 0.0987,
|
| 204 |
+
"step": 240
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 3.0833333333333335,
|
| 208 |
+
"grad_norm": 10.567233085632324,
|
| 209 |
+
"learning_rate": 9.259259259259259e-06,
|
| 210 |
+
"loss": 0.3552,
|
| 211 |
+
"step": 250
|
| 212 |
+
},
|
| 213 |
+
{
|
| 214 |
+
"epoch": 3.1166666666666667,
|
| 215 |
+
"grad_norm": 11.495248794555664,
|
| 216 |
+
"learning_rate": 7.4074074074074075e-06,
|
| 217 |
+
"loss": 0.148,
|
| 218 |
+
"step": 260
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 3.15,
|
| 222 |
+
"grad_norm": 4.4286274909973145,
|
| 223 |
+
"learning_rate": 5.555555555555556e-06,
|
| 224 |
+
"loss": 0.209,
|
| 225 |
+
"step": 270
|
| 226 |
+
},
|
| 227 |
+
{
|
| 228 |
+
"epoch": 3.183333333333333,
|
| 229 |
+
"grad_norm": 1.8519426584243774,
|
| 230 |
+
"learning_rate": 3.7037037037037037e-06,
|
| 231 |
+
"loss": 0.2113,
|
| 232 |
+
"step": 280
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 3.216666666666667,
|
| 236 |
+
"grad_norm": 3.1793460845947266,
|
| 237 |
+
"learning_rate": 1.8518518518518519e-06,
|
| 238 |
+
"loss": 0.1944,
|
| 239 |
+
"step": 290
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 3.25,
|
| 243 |
+
"grad_norm": 2.4581689834594727,
|
| 244 |
+
"learning_rate": 0.0,
|
| 245 |
+
"loss": 0.3042,
|
| 246 |
+
"step": 300
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 3.25,
|
| 250 |
+
"eval_accuracy": 0.9285714285714286,
|
| 251 |
+
"eval_loss": 0.21487949788570404,
|
| 252 |
+
"eval_runtime": 20.1773,
|
| 253 |
+
"eval_samples_per_second": 3.469,
|
| 254 |
+
"eval_steps_per_second": 0.892,
|
| 255 |
+
"step": 300
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 3.25,
|
| 259 |
+
"step": 300,
|
| 260 |
+
"total_flos": 1.495384188125184e+18,
|
| 261 |
+
"train_loss": 0.9161897947390875,
|
| 262 |
+
"train_runtime": 840.6389,
|
| 263 |
+
"train_samples_per_second": 1.427,
|
| 264 |
+
"train_steps_per_second": 0.357
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 3.25,
|
| 268 |
+
"eval_accuracy": 0.8451612903225807,
|
| 269 |
+
"eval_loss": 0.3661264479160309,
|
| 270 |
+
"eval_runtime": 46.685,
|
| 271 |
+
"eval_samples_per_second": 3.32,
|
| 272 |
+
"eval_steps_per_second": 0.835,
|
| 273 |
+
"step": 300
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 3.25,
|
| 277 |
+
"eval_accuracy": 0.8451612903225807,
|
| 278 |
+
"eval_loss": 0.36612647771835327,
|
| 279 |
+
"eval_runtime": 44.7201,
|
| 280 |
+
"eval_samples_per_second": 3.466,
|
| 281 |
+
"eval_steps_per_second": 0.872,
|
| 282 |
+
"step": 300
|
| 283 |
+
}
|
| 284 |
+
],
|
| 285 |
+
"logging_steps": 10,
|
| 286 |
+
"max_steps": 300,
|
| 287 |
+
"num_input_tokens_seen": 0,
|
| 288 |
+
"num_train_epochs": 9223372036854775807,
|
| 289 |
+
"save_steps": 500,
|
| 290 |
+
"total_flos": 1.495384188125184e+18,
|
| 291 |
+
"train_batch_size": 4,
|
| 292 |
+
"trial_name": null,
|
| 293 |
+
"trial_params": null
|
| 294 |
+
}
|