End of training
Browse files
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 7.101351351351352,
|
| 3 |
+
"eval_accuracy": 0.9419354838709677,
|
| 4 |
+
"eval_loss": 0.20170918107032776,
|
| 5 |
+
"eval_runtime": 26.0431,
|
| 6 |
+
"eval_samples_per_second": 5.952,
|
| 7 |
+
"eval_steps_per_second": 0.768
|
| 8 |
+
}
|
runs/Sep21_15-38-17_b32e6c028e2e/events.out.tfevents.1726934154.b32e6c028e2e.37.1
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 sha256:9915b951a6c1f90bed5d2c374d98d928872d2f6b971905e0ca1601107c1b3f27
|
| 3 |
+
size 734
|
test_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 7.101351351351352,
|
| 3 |
+
"eval_accuracy": 0.9419354838709677,
|
| 4 |
+
"eval_loss": 0.20170918107032776,
|
| 5 |
+
"eval_runtime": 26.0431,
|
| 6 |
+
"eval_samples_per_second": 5.952,
|
| 7 |
+
"eval_steps_per_second": 0.768
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,335 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9857142857142858,
|
| 3 |
+
"best_model_checkpoint": "videomae-base-finetuned-ucf101-subset_new/checkpoint-266",
|
| 4 |
+
"epoch": 7.101351351351352,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 296,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.033783783783783786,
|
| 13 |
+
"grad_norm": 7.813199043273926,
|
| 14 |
+
"learning_rate": 6.666666666666667e-06,
|
| 15 |
+
"loss": 2.3171,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.06756756756756757,
|
| 20 |
+
"grad_norm": 7.030215740203857,
|
| 21 |
+
"learning_rate": 1.3333333333333333e-05,
|
| 22 |
+
"loss": 2.2977,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.10135135135135136,
|
| 27 |
+
"grad_norm": 6.99299430847168,
|
| 28 |
+
"learning_rate": 2e-05,
|
| 29 |
+
"loss": 2.3017,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.12837837837837837,
|
| 34 |
+
"eval_accuracy": 0.11428571428571428,
|
| 35 |
+
"eval_loss": 2.2057487964630127,
|
| 36 |
+
"eval_runtime": 11.4376,
|
| 37 |
+
"eval_samples_per_second": 6.12,
|
| 38 |
+
"eval_steps_per_second": 0.787,
|
| 39 |
+
"step": 38
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 1.0067567567567568,
|
| 43 |
+
"grad_norm": 7.603916168212891,
|
| 44 |
+
"learning_rate": 2.6666666666666667e-05,
|
| 45 |
+
"loss": 2.2659,
|
| 46 |
+
"step": 40
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 1.0405405405405406,
|
| 50 |
+
"grad_norm": 13.8406400680542,
|
| 51 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 52 |
+
"loss": 2.1036,
|
| 53 |
+
"step": 50
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.0743243243243243,
|
| 57 |
+
"grad_norm": 8.632048606872559,
|
| 58 |
+
"learning_rate": 4e-05,
|
| 59 |
+
"loss": 1.9679,
|
| 60 |
+
"step": 60
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 1.1081081081081081,
|
| 64 |
+
"grad_norm": 9.114354133605957,
|
| 65 |
+
"learning_rate": 3.8305084745762714e-05,
|
| 66 |
+
"loss": 1.7529,
|
| 67 |
+
"step": 70
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 1.1283783783783783,
|
| 71 |
+
"eval_accuracy": 0.45714285714285713,
|
| 72 |
+
"eval_loss": 1.5769890546798706,
|
| 73 |
+
"eval_runtime": 11.5101,
|
| 74 |
+
"eval_samples_per_second": 6.082,
|
| 75 |
+
"eval_steps_per_second": 0.782,
|
| 76 |
+
"step": 76
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 2.0135135135135136,
|
| 80 |
+
"grad_norm": 11.52100944519043,
|
| 81 |
+
"learning_rate": 3.6610169491525426e-05,
|
| 82 |
+
"loss": 1.5269,
|
| 83 |
+
"step": 80
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 2.0472972972972974,
|
| 87 |
+
"grad_norm": 13.187309265136719,
|
| 88 |
+
"learning_rate": 3.491525423728814e-05,
|
| 89 |
+
"loss": 1.2076,
|
| 90 |
+
"step": 90
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 2.081081081081081,
|
| 94 |
+
"grad_norm": 12.104183197021484,
|
| 95 |
+
"learning_rate": 3.322033898305085e-05,
|
| 96 |
+
"loss": 0.8936,
|
| 97 |
+
"step": 100
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 2.114864864864865,
|
| 101 |
+
"grad_norm": 7.63131856918335,
|
| 102 |
+
"learning_rate": 3.152542372881356e-05,
|
| 103 |
+
"loss": 0.738,
|
| 104 |
+
"step": 110
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 2.1283783783783785,
|
| 108 |
+
"eval_accuracy": 0.7571428571428571,
|
| 109 |
+
"eval_loss": 0.7968592643737793,
|
| 110 |
+
"eval_runtime": 11.5474,
|
| 111 |
+
"eval_samples_per_second": 6.062,
|
| 112 |
+
"eval_steps_per_second": 0.779,
|
| 113 |
+
"step": 114
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 3.02027027027027,
|
| 117 |
+
"grad_norm": 11.565006256103516,
|
| 118 |
+
"learning_rate": 2.9830508474576274e-05,
|
| 119 |
+
"loss": 0.6277,
|
| 120 |
+
"step": 120
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 3.054054054054054,
|
| 124 |
+
"grad_norm": 6.540169715881348,
|
| 125 |
+
"learning_rate": 2.8135593220338985e-05,
|
| 126 |
+
"loss": 0.4507,
|
| 127 |
+
"step": 130
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 3.0878378378378377,
|
| 131 |
+
"grad_norm": 9.234970092773438,
|
| 132 |
+
"learning_rate": 2.6440677966101696e-05,
|
| 133 |
+
"loss": 0.5129,
|
| 134 |
+
"step": 140
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 3.1216216216216215,
|
| 138 |
+
"grad_norm": 3.018888235092163,
|
| 139 |
+
"learning_rate": 2.474576271186441e-05,
|
| 140 |
+
"loss": 0.4054,
|
| 141 |
+
"step": 150
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 3.1283783783783785,
|
| 145 |
+
"eval_accuracy": 0.8142857142857143,
|
| 146 |
+
"eval_loss": 0.44068485498428345,
|
| 147 |
+
"eval_runtime": 10.5153,
|
| 148 |
+
"eval_samples_per_second": 6.657,
|
| 149 |
+
"eval_steps_per_second": 0.856,
|
| 150 |
+
"step": 152
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 4.027027027027027,
|
| 154 |
+
"grad_norm": 4.974002361297607,
|
| 155 |
+
"learning_rate": 2.3050847457627122e-05,
|
| 156 |
+
"loss": 0.3597,
|
| 157 |
+
"step": 160
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 4.0608108108108105,
|
| 161 |
+
"grad_norm": 5.819139003753662,
|
| 162 |
+
"learning_rate": 2.1355932203389833e-05,
|
| 163 |
+
"loss": 0.2099,
|
| 164 |
+
"step": 170
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 4.094594594594595,
|
| 168 |
+
"grad_norm": 12.941056251525879,
|
| 169 |
+
"learning_rate": 1.9661016949152545e-05,
|
| 170 |
+
"loss": 0.2355,
|
| 171 |
+
"step": 180
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 4.128378378378378,
|
| 175 |
+
"grad_norm": 0.7642601728439331,
|
| 176 |
+
"learning_rate": 1.7966101694915256e-05,
|
| 177 |
+
"loss": 0.1165,
|
| 178 |
+
"step": 190
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 4.128378378378378,
|
| 182 |
+
"eval_accuracy": 0.8428571428571429,
|
| 183 |
+
"eval_loss": 0.34949493408203125,
|
| 184 |
+
"eval_runtime": 10.8305,
|
| 185 |
+
"eval_samples_per_second": 6.463,
|
| 186 |
+
"eval_steps_per_second": 0.831,
|
| 187 |
+
"step": 190
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 5.033783783783784,
|
| 191 |
+
"grad_norm": 14.828243255615234,
|
| 192 |
+
"learning_rate": 1.6271186440677967e-05,
|
| 193 |
+
"loss": 0.2828,
|
| 194 |
+
"step": 200
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 5.0675675675675675,
|
| 198 |
+
"grad_norm": 0.2960444986820221,
|
| 199 |
+
"learning_rate": 1.4576271186440678e-05,
|
| 200 |
+
"loss": 0.1411,
|
| 201 |
+
"step": 210
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 5.101351351351352,
|
| 205 |
+
"grad_norm": 1.6549774408340454,
|
| 206 |
+
"learning_rate": 1.288135593220339e-05,
|
| 207 |
+
"loss": 0.1328,
|
| 208 |
+
"step": 220
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 5.128378378378378,
|
| 212 |
+
"eval_accuracy": 0.9571428571428572,
|
| 213 |
+
"eval_loss": 0.1443941295146942,
|
| 214 |
+
"eval_runtime": 10.5693,
|
| 215 |
+
"eval_samples_per_second": 6.623,
|
| 216 |
+
"eval_steps_per_second": 0.852,
|
| 217 |
+
"step": 228
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 6.006756756756757,
|
| 221 |
+
"grad_norm": 0.6144067049026489,
|
| 222 |
+
"learning_rate": 1.1186440677966102e-05,
|
| 223 |
+
"loss": 0.0745,
|
| 224 |
+
"step": 230
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 6.04054054054054,
|
| 228 |
+
"grad_norm": 1.5939418077468872,
|
| 229 |
+
"learning_rate": 9.491525423728815e-06,
|
| 230 |
+
"loss": 0.1271,
|
| 231 |
+
"step": 240
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 6.074324324324325,
|
| 235 |
+
"grad_norm": 1.3212276697158813,
|
| 236 |
+
"learning_rate": 7.796610169491526e-06,
|
| 237 |
+
"loss": 0.0804,
|
| 238 |
+
"step": 250
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 6.108108108108108,
|
| 242 |
+
"grad_norm": 0.2621954083442688,
|
| 243 |
+
"learning_rate": 6.1016949152542385e-06,
|
| 244 |
+
"loss": 0.0875,
|
| 245 |
+
"step": 260
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 6.128378378378378,
|
| 249 |
+
"eval_accuracy": 0.9857142857142858,
|
| 250 |
+
"eval_loss": 0.08640425652265549,
|
| 251 |
+
"eval_runtime": 10.4854,
|
| 252 |
+
"eval_samples_per_second": 6.676,
|
| 253 |
+
"eval_steps_per_second": 0.858,
|
| 254 |
+
"step": 266
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 7.013513513513513,
|
| 258 |
+
"grad_norm": 0.3000829517841339,
|
| 259 |
+
"learning_rate": 4.40677966101695e-06,
|
| 260 |
+
"loss": 0.0622,
|
| 261 |
+
"step": 270
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 7.047297297297297,
|
| 265 |
+
"grad_norm": 0.2126113772392273,
|
| 266 |
+
"learning_rate": 2.7118644067796613e-06,
|
| 267 |
+
"loss": 0.0534,
|
| 268 |
+
"step": 280
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 7.081081081081081,
|
| 272 |
+
"grad_norm": 0.18285629153251648,
|
| 273 |
+
"learning_rate": 1.016949152542373e-06,
|
| 274 |
+
"loss": 0.0347,
|
| 275 |
+
"step": 290
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 7.101351351351352,
|
| 279 |
+
"eval_accuracy": 0.9857142857142858,
|
| 280 |
+
"eval_loss": 0.0650578960776329,
|
| 281 |
+
"eval_runtime": 12.2768,
|
| 282 |
+
"eval_samples_per_second": 5.702,
|
| 283 |
+
"eval_steps_per_second": 0.733,
|
| 284 |
+
"step": 296
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 7.101351351351352,
|
| 288 |
+
"step": 296,
|
| 289 |
+
"total_flos": 2.915999166844109e+18,
|
| 290 |
+
"train_loss": 0.7899129955047691,
|
| 291 |
+
"train_runtime": 959.2262,
|
| 292 |
+
"train_samples_per_second": 2.469,
|
| 293 |
+
"train_steps_per_second": 0.309
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 7.101351351351352,
|
| 297 |
+
"eval_accuracy": 0.9419354838709677,
|
| 298 |
+
"eval_loss": 0.20170918107032776,
|
| 299 |
+
"eval_runtime": 25.7243,
|
| 300 |
+
"eval_samples_per_second": 6.025,
|
| 301 |
+
"eval_steps_per_second": 0.777,
|
| 302 |
+
"step": 296
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 7.101351351351352,
|
| 306 |
+
"eval_accuracy": 0.9419354838709677,
|
| 307 |
+
"eval_loss": 0.20170918107032776,
|
| 308 |
+
"eval_runtime": 26.0431,
|
| 309 |
+
"eval_samples_per_second": 5.952,
|
| 310 |
+
"eval_steps_per_second": 0.768,
|
| 311 |
+
"step": 296
|
| 312 |
+
}
|
| 313 |
+
],
|
| 314 |
+
"logging_steps": 10,
|
| 315 |
+
"max_steps": 296,
|
| 316 |
+
"num_input_tokens_seen": 0,
|
| 317 |
+
"num_train_epochs": 9223372036854775807,
|
| 318 |
+
"save_steps": 500,
|
| 319 |
+
"stateful_callbacks": {
|
| 320 |
+
"TrainerControl": {
|
| 321 |
+
"args": {
|
| 322 |
+
"should_epoch_stop": false,
|
| 323 |
+
"should_evaluate": false,
|
| 324 |
+
"should_log": false,
|
| 325 |
+
"should_save": true,
|
| 326 |
+
"should_training_stop": true
|
| 327 |
+
},
|
| 328 |
+
"attributes": {}
|
| 329 |
+
}
|
| 330 |
+
},
|
| 331 |
+
"total_flos": 2.915999166844109e+18,
|
| 332 |
+
"train_batch_size": 8,
|
| 333 |
+
"trial_name": null,
|
| 334 |
+
"trial_params": null
|
| 335 |
+
}
|