Model save
Browse files- all_results.json +11 -0
- eval_results.json +7 -0
- train_results.json +7 -0
- trainer_state.json +221 -0
- training_args.bin +1 -1
all_results.json
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.04,
|
| 3 |
+
"eval_loss": 0.07713703066110611,
|
| 4 |
+
"eval_runtime": 1039.9476,
|
| 5 |
+
"eval_samples_per_second": 37.261,
|
| 6 |
+
"eval_steps_per_second": 4.658,
|
| 7 |
+
"train_loss": 0.0033295607640092873,
|
| 8 |
+
"train_runtime": 264.6407,
|
| 9 |
+
"train_samples_per_second": 442.695,
|
| 10 |
+
"train_steps_per_second": 55.339
|
| 11 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.04,
|
| 3 |
+
"eval_loss": 0.07713703066110611,
|
| 4 |
+
"eval_runtime": 1039.9476,
|
| 5 |
+
"eval_samples_per_second": 37.261,
|
| 6 |
+
"eval_steps_per_second": 4.658
|
| 7 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 0.04,
|
| 3 |
+
"train_loss": 0.0033295607640092873,
|
| 4 |
+
"train_runtime": 264.6407,
|
| 5 |
+
"train_samples_per_second": 442.695,
|
| 6 |
+
"train_steps_per_second": 55.339
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,221 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.07713755965232849,
|
| 3 |
+
"best_model_checkpoint": "/nasty/data/tpid/vizwiz/ViTGPT2_VW/checkpoint-14000",
|
| 4 |
+
"epoch": 0.04404233526800956,
|
| 5 |
+
"global_step": 14645,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.03,
|
| 12 |
+
"learning_rate": 1.9863839666769097e-05,
|
| 13 |
+
"loss": 0.1256,
|
| 14 |
+
"step": 1000
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.03,
|
| 18 |
+
"eval_loss": 0.09279213845729828,
|
| 19 |
+
"eval_runtime": 1047.3083,
|
| 20 |
+
"eval_samples_per_second": 37.0,
|
| 21 |
+
"eval_steps_per_second": 9.25,
|
| 22 |
+
"step": 1000
|
| 23 |
+
},
|
| 24 |
+
{
|
| 25 |
+
"epoch": 0.07,
|
| 26 |
+
"learning_rate": 1.9727269623408106e-05,
|
| 27 |
+
"loss": 0.0947,
|
| 28 |
+
"step": 2000
|
| 29 |
+
},
|
| 30 |
+
{
|
| 31 |
+
"epoch": 0.07,
|
| 32 |
+
"eval_loss": 0.0897236242890358,
|
| 33 |
+
"eval_runtime": 864.4123,
|
| 34 |
+
"eval_samples_per_second": 44.828,
|
| 35 |
+
"eval_steps_per_second": 11.208,
|
| 36 |
+
"step": 2000
|
| 37 |
+
},
|
| 38 |
+
{
|
| 39 |
+
"epoch": 0.1,
|
| 40 |
+
"learning_rate": 1.9590699580047118e-05,
|
| 41 |
+
"loss": 0.0889,
|
| 42 |
+
"step": 3000
|
| 43 |
+
},
|
| 44 |
+
{
|
| 45 |
+
"epoch": 0.1,
|
| 46 |
+
"eval_loss": 0.08590871840715408,
|
| 47 |
+
"eval_runtime": 857.9482,
|
| 48 |
+
"eval_samples_per_second": 45.166,
|
| 49 |
+
"eval_steps_per_second": 11.292,
|
| 50 |
+
"step": 3000
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.14,
|
| 54 |
+
"learning_rate": 1.945412953668613e-05,
|
| 55 |
+
"loss": 0.0888,
|
| 56 |
+
"step": 4000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.14,
|
| 60 |
+
"eval_loss": 0.08419814705848694,
|
| 61 |
+
"eval_runtime": 866.4231,
|
| 62 |
+
"eval_samples_per_second": 44.724,
|
| 63 |
+
"eval_steps_per_second": 11.182,
|
| 64 |
+
"step": 4000
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 0.17,
|
| 68 |
+
"learning_rate": 1.931755949332514e-05,
|
| 69 |
+
"loss": 0.0866,
|
| 70 |
+
"step": 5000
|
| 71 |
+
},
|
| 72 |
+
{
|
| 73 |
+
"epoch": 0.17,
|
| 74 |
+
"eval_loss": 0.08309577405452728,
|
| 75 |
+
"eval_runtime": 864.2485,
|
| 76 |
+
"eval_samples_per_second": 44.837,
|
| 77 |
+
"eval_steps_per_second": 11.21,
|
| 78 |
+
"step": 5000
|
| 79 |
+
},
|
| 80 |
+
{
|
| 81 |
+
"epoch": 0.2,
|
| 82 |
+
"learning_rate": 1.918098944996415e-05,
|
| 83 |
+
"loss": 0.0852,
|
| 84 |
+
"step": 6000
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 0.2,
|
| 88 |
+
"eval_loss": 0.08189540356397629,
|
| 89 |
+
"eval_runtime": 865.5776,
|
| 90 |
+
"eval_samples_per_second": 44.768,
|
| 91 |
+
"eval_steps_per_second": 11.193,
|
| 92 |
+
"step": 6000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.24,
|
| 96 |
+
"learning_rate": 1.9044419406603162e-05,
|
| 97 |
+
"loss": 0.0833,
|
| 98 |
+
"step": 7000
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.24,
|
| 102 |
+
"eval_loss": 0.08095835894346237,
|
| 103 |
+
"eval_runtime": 864.7144,
|
| 104 |
+
"eval_samples_per_second": 44.812,
|
| 105 |
+
"eval_steps_per_second": 11.204,
|
| 106 |
+
"step": 7000
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 0.27,
|
| 110 |
+
"learning_rate": 1.8907849363242174e-05,
|
| 111 |
+
"loss": 0.0835,
|
| 112 |
+
"step": 8000
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 0.27,
|
| 116 |
+
"eval_loss": 0.08023638278245926,
|
| 117 |
+
"eval_runtime": 866.728,
|
| 118 |
+
"eval_samples_per_second": 44.708,
|
| 119 |
+
"eval_steps_per_second": 11.178,
|
| 120 |
+
"step": 8000
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 0.31,
|
| 124 |
+
"learning_rate": 1.8771279319881186e-05,
|
| 125 |
+
"loss": 0.081,
|
| 126 |
+
"step": 9000
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 0.31,
|
| 130 |
+
"eval_loss": 0.07963699847459793,
|
| 131 |
+
"eval_runtime": 865.0337,
|
| 132 |
+
"eval_samples_per_second": 44.796,
|
| 133 |
+
"eval_steps_per_second": 11.2,
|
| 134 |
+
"step": 9000
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.34,
|
| 138 |
+
"learning_rate": 1.86347092765202e-05,
|
| 139 |
+
"loss": 0.0803,
|
| 140 |
+
"step": 10000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.34,
|
| 144 |
+
"eval_loss": 0.07886938005685806,
|
| 145 |
+
"eval_runtime": 864.2901,
|
| 146 |
+
"eval_samples_per_second": 44.834,
|
| 147 |
+
"eval_steps_per_second": 11.209,
|
| 148 |
+
"step": 10000
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 0.38,
|
| 152 |
+
"learning_rate": 1.849813923315921e-05,
|
| 153 |
+
"loss": 0.0814,
|
| 154 |
+
"step": 11000
|
| 155 |
+
},
|
| 156 |
+
{
|
| 157 |
+
"epoch": 0.38,
|
| 158 |
+
"eval_loss": 0.07851768285036087,
|
| 159 |
+
"eval_runtime": 863.0619,
|
| 160 |
+
"eval_samples_per_second": 44.898,
|
| 161 |
+
"eval_steps_per_second": 11.225,
|
| 162 |
+
"step": 11000
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 0.41,
|
| 166 |
+
"learning_rate": 1.836156918979822e-05,
|
| 167 |
+
"loss": 0.0799,
|
| 168 |
+
"step": 12000
|
| 169 |
+
},
|
| 170 |
+
{
|
| 171 |
+
"epoch": 0.41,
|
| 172 |
+
"eval_loss": 0.0779803916811943,
|
| 173 |
+
"eval_runtime": 862.8886,
|
| 174 |
+
"eval_samples_per_second": 44.907,
|
| 175 |
+
"eval_steps_per_second": 11.227,
|
| 176 |
+
"step": 12000
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.44,
|
| 180 |
+
"learning_rate": 1.822513571648059e-05,
|
| 181 |
+
"loss": 0.0786,
|
| 182 |
+
"step": 13000
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.44,
|
| 186 |
+
"eval_loss": 0.07762513309717178,
|
| 187 |
+
"eval_runtime": 861.7094,
|
| 188 |
+
"eval_samples_per_second": 44.969,
|
| 189 |
+
"eval_steps_per_second": 11.243,
|
| 190 |
+
"step": 13000
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 0.48,
|
| 194 |
+
"learning_rate": 1.8088565673119603e-05,
|
| 195 |
+
"loss": 0.0796,
|
| 196 |
+
"step": 14000
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 0.48,
|
| 200 |
+
"eval_loss": 0.07713755965232849,
|
| 201 |
+
"eval_runtime": 866.4253,
|
| 202 |
+
"eval_samples_per_second": 44.724,
|
| 203 |
+
"eval_steps_per_second": 11.182,
|
| 204 |
+
"step": 14000
|
| 205 |
+
},
|
| 206 |
+
{
|
| 207 |
+
"epoch": 0.04,
|
| 208 |
+
"step": 14645,
|
| 209 |
+
"total_flos": 7.904602466056929e+18,
|
| 210 |
+
"train_loss": 0.0033295607640092873,
|
| 211 |
+
"train_runtime": 264.6407,
|
| 212 |
+
"train_samples_per_second": 442.695,
|
| 213 |
+
"train_steps_per_second": 55.339
|
| 214 |
+
}
|
| 215 |
+
],
|
| 216 |
+
"max_steps": 14645,
|
| 217 |
+
"num_train_epochs": 1,
|
| 218 |
+
"total_flos": 7.904602466056929e+18,
|
| 219 |
+
"trial_name": null,
|
| 220 |
+
"trial_params": null
|
| 221 |
+
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 3119
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2e84ce83a470576c9eb64aa02b3093a5b38f263e31818dbce9bc12767b9c6053
|
| 3 |
size 3119
|