Commit
•
5df4032
1
Parent(s):
82143c6
End of training
Browse files- README.md +1 -1
- all_results.json +13 -0
- eval_results.json +8 -0
- train_results.json +8 -0
- trainer_state.json +377 -0
README.md
CHANGED
|
@@ -30,7 +30,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 30 |
|
| 31 |
This model was trained from scratch on the imagefolder dataset.
|
| 32 |
It achieves the following results on the evaluation set:
|
| 33 |
-
- Loss: 0.
|
| 34 |
- Accuracy: 0.6984
|
| 35 |
|
| 36 |
## Model description
|
|
|
|
| 30 |
|
| 31 |
This model was trained from scratch on the imagefolder dataset.
|
| 32 |
It achieves the following results on the evaluation set:
|
| 33 |
+
- Loss: 0.6463
|
| 34 |
- Accuracy: 0.6984
|
| 35 |
|
| 36 |
## Model description
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.6984126984126984,
|
| 4 |
+
"eval_loss": 0.6463457942008972,
|
| 5 |
+
"eval_runtime": 10.8458,
|
| 6 |
+
"eval_samples_per_second": 5.809,
|
| 7 |
+
"eval_steps_per_second": 1.475,
|
| 8 |
+
"total_flos": 4.3240530204942336e+17,
|
| 9 |
+
"train_loss": 0.6562235273633684,
|
| 10 |
+
"train_runtime": 1176.6645,
|
| 11 |
+
"train_samples_per_second": 4.742,
|
| 12 |
+
"train_steps_per_second": 0.297
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.6984126984126984,
|
| 4 |
+
"eval_loss": 0.6463457942008972,
|
| 5 |
+
"eval_runtime": 10.8458,
|
| 6 |
+
"eval_samples_per_second": 5.809,
|
| 7 |
+
"eval_steps_per_second": 1.475
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 4.3240530204942336e+17,
|
| 4 |
+
"train_loss": 0.6562235273633684,
|
| 5 |
+
"train_runtime": 1176.6645,
|
| 6 |
+
"train_samples_per_second": 4.742,
|
| 7 |
+
"train_steps_per_second": 0.297
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,377 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.6984126984126984,
|
| 3 |
+
"best_model_checkpoint": "vit-base-patch16-224/checkpoint-245",
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 350,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.2857142857142857,
|
| 13 |
+
"grad_norm": 6.601813793182373,
|
| 14 |
+
"learning_rate": 1.4285714285714286e-06,
|
| 15 |
+
"loss": 0.785,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.5714285714285714,
|
| 20 |
+
"grad_norm": 5.436784267425537,
|
| 21 |
+
"learning_rate": 2.8571428571428573e-06,
|
| 22 |
+
"loss": 0.7224,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.8571428571428571,
|
| 27 |
+
"grad_norm": 5.551754474639893,
|
| 28 |
+
"learning_rate": 4.2857142857142855e-06,
|
| 29 |
+
"loss": 0.6893,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 1.0,
|
| 34 |
+
"eval_accuracy": 0.4126984126984127,
|
| 35 |
+
"eval_loss": 0.7319461703300476,
|
| 36 |
+
"eval_runtime": 12.0068,
|
| 37 |
+
"eval_samples_per_second": 5.247,
|
| 38 |
+
"eval_steps_per_second": 1.333,
|
| 39 |
+
"step": 35
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 1.1428571428571428,
|
| 43 |
+
"grad_norm": 7.985645294189453,
|
| 44 |
+
"learning_rate": 4.920634920634921e-06,
|
| 45 |
+
"loss": 0.7348,
|
| 46 |
+
"step": 40
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 1.4285714285714286,
|
| 50 |
+
"grad_norm": 4.910062313079834,
|
| 51 |
+
"learning_rate": 4.761904761904762e-06,
|
| 52 |
+
"loss": 0.7049,
|
| 53 |
+
"step": 50
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.7142857142857144,
|
| 57 |
+
"grad_norm": 4.96162223815918,
|
| 58 |
+
"learning_rate": 4.603174603174604e-06,
|
| 59 |
+
"loss": 0.7185,
|
| 60 |
+
"step": 60
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 2.0,
|
| 64 |
+
"grad_norm": 7.957433223724365,
|
| 65 |
+
"learning_rate": 4.444444444444444e-06,
|
| 66 |
+
"loss": 0.702,
|
| 67 |
+
"step": 70
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 2.0,
|
| 71 |
+
"eval_accuracy": 0.5238095238095238,
|
| 72 |
+
"eval_loss": 0.686264157295227,
|
| 73 |
+
"eval_runtime": 10.8952,
|
| 74 |
+
"eval_samples_per_second": 5.782,
|
| 75 |
+
"eval_steps_per_second": 1.469,
|
| 76 |
+
"step": 70
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 2.2857142857142856,
|
| 80 |
+
"grad_norm": 5.780089378356934,
|
| 81 |
+
"learning_rate": 4.2857142857142855e-06,
|
| 82 |
+
"loss": 0.6473,
|
| 83 |
+
"step": 80
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 2.571428571428571,
|
| 87 |
+
"grad_norm": 3.8612518310546875,
|
| 88 |
+
"learning_rate": 4.126984126984127e-06,
|
| 89 |
+
"loss": 0.6843,
|
| 90 |
+
"step": 90
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 2.857142857142857,
|
| 94 |
+
"grad_norm": 5.266788482666016,
|
| 95 |
+
"learning_rate": 3.968253968253968e-06,
|
| 96 |
+
"loss": 0.6644,
|
| 97 |
+
"step": 100
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 3.0,
|
| 101 |
+
"eval_accuracy": 0.5873015873015873,
|
| 102 |
+
"eval_loss": 0.6796107888221741,
|
| 103 |
+
"eval_runtime": 10.7178,
|
| 104 |
+
"eval_samples_per_second": 5.878,
|
| 105 |
+
"eval_steps_per_second": 1.493,
|
| 106 |
+
"step": 105
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 3.142857142857143,
|
| 110 |
+
"grad_norm": 4.334442615509033,
|
| 111 |
+
"learning_rate": 3.80952380952381e-06,
|
| 112 |
+
"loss": 0.6628,
|
| 113 |
+
"step": 110
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 3.4285714285714284,
|
| 117 |
+
"grad_norm": 4.810235500335693,
|
| 118 |
+
"learning_rate": 3.6507936507936507e-06,
|
| 119 |
+
"loss": 0.6903,
|
| 120 |
+
"step": 120
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 3.7142857142857144,
|
| 124 |
+
"grad_norm": 7.596153736114502,
|
| 125 |
+
"learning_rate": 3.492063492063492e-06,
|
| 126 |
+
"loss": 0.6859,
|
| 127 |
+
"step": 130
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 4.0,
|
| 131 |
+
"grad_norm": 8.868383407592773,
|
| 132 |
+
"learning_rate": 3.3333333333333333e-06,
|
| 133 |
+
"loss": 0.645,
|
| 134 |
+
"step": 140
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 4.0,
|
| 138 |
+
"eval_accuracy": 0.5714285714285714,
|
| 139 |
+
"eval_loss": 0.6722006797790527,
|
| 140 |
+
"eval_runtime": 10.7432,
|
| 141 |
+
"eval_samples_per_second": 5.864,
|
| 142 |
+
"eval_steps_per_second": 1.489,
|
| 143 |
+
"step": 140
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 4.285714285714286,
|
| 147 |
+
"grad_norm": 4.674743175506592,
|
| 148 |
+
"learning_rate": 3.1746031746031746e-06,
|
| 149 |
+
"loss": 0.6336,
|
| 150 |
+
"step": 150
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 4.571428571428571,
|
| 154 |
+
"grad_norm": 5.1306023597717285,
|
| 155 |
+
"learning_rate": 3.015873015873016e-06,
|
| 156 |
+
"loss": 0.6607,
|
| 157 |
+
"step": 160
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 4.857142857142857,
|
| 161 |
+
"grad_norm": 5.9358015060424805,
|
| 162 |
+
"learning_rate": 2.8571428571428573e-06,
|
| 163 |
+
"loss": 0.6455,
|
| 164 |
+
"step": 170
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 5.0,
|
| 168 |
+
"eval_accuracy": 0.6507936507936508,
|
| 169 |
+
"eval_loss": 0.6544514298439026,
|
| 170 |
+
"eval_runtime": 10.8969,
|
| 171 |
+
"eval_samples_per_second": 5.781,
|
| 172 |
+
"eval_steps_per_second": 1.468,
|
| 173 |
+
"step": 175
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 5.142857142857143,
|
| 177 |
+
"grad_norm": 4.706002712249756,
|
| 178 |
+
"learning_rate": 2.6984126984126986e-06,
|
| 179 |
+
"loss": 0.6039,
|
| 180 |
+
"step": 180
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 5.428571428571429,
|
| 184 |
+
"grad_norm": 5.049504280090332,
|
| 185 |
+
"learning_rate": 2.53968253968254e-06,
|
| 186 |
+
"loss": 0.6881,
|
| 187 |
+
"step": 190
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 5.714285714285714,
|
| 191 |
+
"grad_norm": 4.565422534942627,
|
| 192 |
+
"learning_rate": 2.380952380952381e-06,
|
| 193 |
+
"loss": 0.5841,
|
| 194 |
+
"step": 200
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 6.0,
|
| 198 |
+
"grad_norm": 5.591720104217529,
|
| 199 |
+
"learning_rate": 2.222222222222222e-06,
|
| 200 |
+
"loss": 0.6456,
|
| 201 |
+
"step": 210
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 6.0,
|
| 205 |
+
"eval_accuracy": 0.6507936507936508,
|
| 206 |
+
"eval_loss": 0.6535508036613464,
|
| 207 |
+
"eval_runtime": 10.8606,
|
| 208 |
+
"eval_samples_per_second": 5.801,
|
| 209 |
+
"eval_steps_per_second": 1.473,
|
| 210 |
+
"step": 210
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 6.285714285714286,
|
| 214 |
+
"grad_norm": 5.09126091003418,
|
| 215 |
+
"learning_rate": 2.0634920634920634e-06,
|
| 216 |
+
"loss": 0.6349,
|
| 217 |
+
"step": 220
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 6.571428571428571,
|
| 221 |
+
"grad_norm": 4.8962812423706055,
|
| 222 |
+
"learning_rate": 1.904761904761905e-06,
|
| 223 |
+
"loss": 0.5735,
|
| 224 |
+
"step": 230
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 6.857142857142857,
|
| 228 |
+
"grad_norm": 4.864034175872803,
|
| 229 |
+
"learning_rate": 1.746031746031746e-06,
|
| 230 |
+
"loss": 0.6745,
|
| 231 |
+
"step": 240
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 7.0,
|
| 235 |
+
"eval_accuracy": 0.6984126984126984,
|
| 236 |
+
"eval_loss": 0.6463457942008972,
|
| 237 |
+
"eval_runtime": 10.8225,
|
| 238 |
+
"eval_samples_per_second": 5.821,
|
| 239 |
+
"eval_steps_per_second": 1.478,
|
| 240 |
+
"step": 245
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 7.142857142857143,
|
| 244 |
+
"grad_norm": 7.017258167266846,
|
| 245 |
+
"learning_rate": 1.5873015873015873e-06,
|
| 246 |
+
"loss": 0.6431,
|
| 247 |
+
"step": 250
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 7.428571428571429,
|
| 251 |
+
"grad_norm": 4.79843282699585,
|
| 252 |
+
"learning_rate": 1.4285714285714286e-06,
|
| 253 |
+
"loss": 0.6164,
|
| 254 |
+
"step": 260
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 7.714285714285714,
|
| 258 |
+
"grad_norm": 6.843749523162842,
|
| 259 |
+
"learning_rate": 1.26984126984127e-06,
|
| 260 |
+
"loss": 0.644,
|
| 261 |
+
"step": 270
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 8.0,
|
| 265 |
+
"grad_norm": 5.495963096618652,
|
| 266 |
+
"learning_rate": 1.111111111111111e-06,
|
| 267 |
+
"loss": 0.6369,
|
| 268 |
+
"step": 280
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 8.0,
|
| 272 |
+
"eval_accuracy": 0.6666666666666666,
|
| 273 |
+
"eval_loss": 0.6524515748023987,
|
| 274 |
+
"eval_runtime": 11.0752,
|
| 275 |
+
"eval_samples_per_second": 5.688,
|
| 276 |
+
"eval_steps_per_second": 1.445,
|
| 277 |
+
"step": 280
|
| 278 |
+
},
|
| 279 |
+
{
|
| 280 |
+
"epoch": 8.285714285714286,
|
| 281 |
+
"grad_norm": 7.264893531799316,
|
| 282 |
+
"learning_rate": 9.523809523809525e-07,
|
| 283 |
+
"loss": 0.6661,
|
| 284 |
+
"step": 290
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 8.571428571428571,
|
| 288 |
+
"grad_norm": 3.773362636566162,
|
| 289 |
+
"learning_rate": 7.936507936507937e-07,
|
| 290 |
+
"loss": 0.607,
|
| 291 |
+
"step": 300
|
| 292 |
+
},
|
| 293 |
+
{
|
| 294 |
+
"epoch": 8.857142857142858,
|
| 295 |
+
"grad_norm": 6.338254451751709,
|
| 296 |
+
"learning_rate": 6.34920634920635e-07,
|
| 297 |
+
"loss": 0.6012,
|
| 298 |
+
"step": 310
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 9.0,
|
| 302 |
+
"eval_accuracy": 0.6984126984126984,
|
| 303 |
+
"eval_loss": 0.6485763788223267,
|
| 304 |
+
"eval_runtime": 10.9179,
|
| 305 |
+
"eval_samples_per_second": 5.77,
|
| 306 |
+
"eval_steps_per_second": 1.465,
|
| 307 |
+
"step": 315
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 9.142857142857142,
|
| 311 |
+
"grad_norm": 5.557713031768799,
|
| 312 |
+
"learning_rate": 4.7619047619047623e-07,
|
| 313 |
+
"loss": 0.625,
|
| 314 |
+
"step": 320
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 9.428571428571429,
|
| 318 |
+
"grad_norm": 5.070873737335205,
|
| 319 |
+
"learning_rate": 3.174603174603175e-07,
|
| 320 |
+
"loss": 0.6012,
|
| 321 |
+
"step": 330
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 9.714285714285714,
|
| 325 |
+
"grad_norm": 6.369267463684082,
|
| 326 |
+
"learning_rate": 1.5873015873015874e-07,
|
| 327 |
+
"loss": 0.6237,
|
| 328 |
+
"step": 340
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 10.0,
|
| 332 |
+
"grad_norm": 8.102092742919922,
|
| 333 |
+
"learning_rate": 0.0,
|
| 334 |
+
"loss": 0.6219,
|
| 335 |
+
"step": 350
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 10.0,
|
| 339 |
+
"eval_accuracy": 0.6984126984126984,
|
| 340 |
+
"eval_loss": 0.646605372428894,
|
| 341 |
+
"eval_runtime": 11.2095,
|
| 342 |
+
"eval_samples_per_second": 5.62,
|
| 343 |
+
"eval_steps_per_second": 1.427,
|
| 344 |
+
"step": 350
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 10.0,
|
| 348 |
+
"step": 350,
|
| 349 |
+
"total_flos": 4.3240530204942336e+17,
|
| 350 |
+
"train_loss": 0.6562235273633684,
|
| 351 |
+
"train_runtime": 1176.6645,
|
| 352 |
+
"train_samples_per_second": 4.742,
|
| 353 |
+
"train_steps_per_second": 0.297
|
| 354 |
+
}
|
| 355 |
+
],
|
| 356 |
+
"logging_steps": 10,
|
| 357 |
+
"max_steps": 350,
|
| 358 |
+
"num_input_tokens_seen": 0,
|
| 359 |
+
"num_train_epochs": 10,
|
| 360 |
+
"save_steps": 500,
|
| 361 |
+
"stateful_callbacks": {
|
| 362 |
+
"TrainerControl": {
|
| 363 |
+
"args": {
|
| 364 |
+
"should_epoch_stop": false,
|
| 365 |
+
"should_evaluate": false,
|
| 366 |
+
"should_log": false,
|
| 367 |
+
"should_save": true,
|
| 368 |
+
"should_training_stop": true
|
| 369 |
+
},
|
| 370 |
+
"attributes": {}
|
| 371 |
+
}
|
| 372 |
+
},
|
| 373 |
+
"total_flos": 4.3240530204942336e+17,
|
| 374 |
+
"train_batch_size": 4,
|
| 375 |
+
"trial_name": null,
|
| 376 |
+
"trial_params": null
|
| 377 |
+
}
|