End of training
Browse files- README.md +3 -1
- all_results.json +18 -0
- eval_results.json +14 -0
- train_results.json +7 -0
- trainer_state.json +403 -0
README.md
CHANGED
|
@@ -2,6 +2,8 @@
|
|
| 2 |
license: other
|
| 3 |
base_model: nvidia/mit-b5
|
| 4 |
tags:
|
|
|
|
|
|
|
| 5 |
- generated_from_trainer
|
| 6 |
model-index:
|
| 7 |
- name: ecc_segformerv3
|
|
@@ -13,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 13 |
|
| 14 |
# ecc_segformerv3
|
| 15 |
|
| 16 |
-
This model is a fine-tuned version of [nvidia/mit-b5](https://huggingface.co/nvidia/mit-b5) on
|
| 17 |
It achieves the following results on the evaluation set:
|
| 18 |
- Loss: 0.1344
|
| 19 |
- Mean Iou: 0.0005
|
|
|
|
| 2 |
license: other
|
| 3 |
base_model: nvidia/mit-b5
|
| 4 |
tags:
|
| 5 |
+
- image-segmentation
|
| 6 |
+
- vision
|
| 7 |
- generated_from_trainer
|
| 8 |
model-index:
|
| 9 |
- name: ecc_segformerv3
|
|
|
|
| 15 |
|
| 16 |
# ecc_segformerv3
|
| 17 |
|
| 18 |
+
This model is a fine-tuned version of [nvidia/mit-b5](https://huggingface.co/nvidia/mit-b5) on the rishitunu/ecc_crackdetector_dataset dataset.
|
| 19 |
It achieves the following results on the evaluation set:
|
| 20 |
- Loss: 0.1344
|
| 21 |
- Mean Iou: 0.0005
|
all_results.json
ADDED
|
@@ -0,0 +1,18 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_accuracy_background": NaN,
|
| 4 |
+
"eval_accuracy_crack": 0.0009813165224114941,
|
| 5 |
+
"eval_iou_background": 0.0,
|
| 6 |
+
"eval_iou_crack": 0.0009813165224114941,
|
| 7 |
+
"eval_loss": 0.1343701034784317,
|
| 8 |
+
"eval_mean_accuracy": 0.0009813165224114941,
|
| 9 |
+
"eval_mean_iou": 0.0004906582612057471,
|
| 10 |
+
"eval_overall_accuracy": 0.0009813165224114941,
|
| 11 |
+
"eval_runtime": 532.3161,
|
| 12 |
+
"eval_samples_per_second": 0.333,
|
| 13 |
+
"eval_steps_per_second": 0.333,
|
| 14 |
+
"train_loss": 0.09988212461471557,
|
| 15 |
+
"train_runtime": 34878.9562,
|
| 16 |
+
"train_samples_per_second": 0.143,
|
| 17 |
+
"train_steps_per_second": 0.143
|
| 18 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"eval_accuracy_background": NaN,
|
| 4 |
+
"eval_accuracy_crack": 0.0009813165224114941,
|
| 5 |
+
"eval_iou_background": 0.0,
|
| 6 |
+
"eval_iou_crack": 0.0009813165224114941,
|
| 7 |
+
"eval_loss": 0.1343701034784317,
|
| 8 |
+
"eval_mean_accuracy": 0.0009813165224114941,
|
| 9 |
+
"eval_mean_iou": 0.0004906582612057471,
|
| 10 |
+
"eval_overall_accuracy": 0.0009813165224114941,
|
| 11 |
+
"eval_runtime": 532.3161,
|
| 12 |
+
"eval_samples_per_second": 0.333,
|
| 13 |
+
"eval_steps_per_second": 0.333
|
| 14 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 5.0,
|
| 3 |
+
"train_loss": 0.09988212461471557,
|
| 4 |
+
"train_runtime": 34878.9562,
|
| 5 |
+
"train_samples_per_second": 0.143,
|
| 6 |
+
"train_steps_per_second": 0.143
|
| 7 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,403 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 4.995004995004995,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 5000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.1,
|
| 13 |
+
"learning_rate": 0.000588,
|
| 14 |
+
"loss": 0.1326,
|
| 15 |
+
"step": 100
|
| 16 |
+
},
|
| 17 |
+
{
|
| 18 |
+
"epoch": 0.2,
|
| 19 |
+
"learning_rate": 0.0005759999999999999,
|
| 20 |
+
"loss": 0.1179,
|
| 21 |
+
"step": 200
|
| 22 |
+
},
|
| 23 |
+
{
|
| 24 |
+
"epoch": 0.3,
|
| 25 |
+
"learning_rate": 0.0005639999999999999,
|
| 26 |
+
"loss": 0.1292,
|
| 27 |
+
"step": 300
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"epoch": 0.4,
|
| 31 |
+
"learning_rate": 0.000552,
|
| 32 |
+
"loss": 0.1292,
|
| 33 |
+
"step": 400
|
| 34 |
+
},
|
| 35 |
+
{
|
| 36 |
+
"epoch": 0.5,
|
| 37 |
+
"learning_rate": 0.00054,
|
| 38 |
+
"loss": 0.1189,
|
| 39 |
+
"step": 500
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 0.6,
|
| 43 |
+
"learning_rate": 0.0005279999999999999,
|
| 44 |
+
"loss": 0.0963,
|
| 45 |
+
"step": 600
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.7,
|
| 49 |
+
"learning_rate": 0.000516,
|
| 50 |
+
"loss": 0.0949,
|
| 51 |
+
"step": 700
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.8,
|
| 55 |
+
"learning_rate": 0.0005039999999999999,
|
| 56 |
+
"loss": 0.1309,
|
| 57 |
+
"step": 800
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 0.9,
|
| 61 |
+
"learning_rate": 0.0004919999999999999,
|
| 62 |
+
"loss": 0.1353,
|
| 63 |
+
"step": 900
|
| 64 |
+
},
|
| 65 |
+
{
|
| 66 |
+
"epoch": 1.0,
|
| 67 |
+
"learning_rate": 0.00047999999999999996,
|
| 68 |
+
"loss": 0.1306,
|
| 69 |
+
"step": 1000
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 1.0,
|
| 73 |
+
"eval_accuracy_background": NaN,
|
| 74 |
+
"eval_accuracy_crack": 0.0,
|
| 75 |
+
"eval_iou_background": 0.0,
|
| 76 |
+
"eval_iou_crack": 0.0,
|
| 77 |
+
"eval_loss": 0.11144199222326279,
|
| 78 |
+
"eval_mean_accuracy": 0.0,
|
| 79 |
+
"eval_mean_iou": 0.0,
|
| 80 |
+
"eval_overall_accuracy": 0.0,
|
| 81 |
+
"eval_runtime": 523.6071,
|
| 82 |
+
"eval_samples_per_second": 0.338,
|
| 83 |
+
"eval_steps_per_second": 0.338,
|
| 84 |
+
"step": 1001
|
| 85 |
+
},
|
| 86 |
+
{
|
| 87 |
+
"epoch": 1.1,
|
| 88 |
+
"learning_rate": 0.000468,
|
| 89 |
+
"loss": 0.113,
|
| 90 |
+
"step": 1100
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 1.2,
|
| 94 |
+
"learning_rate": 0.00045599999999999997,
|
| 95 |
+
"loss": 0.1031,
|
| 96 |
+
"step": 1200
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 1.3,
|
| 100 |
+
"learning_rate": 0.00044399999999999995,
|
| 101 |
+
"loss": 0.1021,
|
| 102 |
+
"step": 1300
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.4,
|
| 106 |
+
"learning_rate": 0.00043199999999999993,
|
| 107 |
+
"loss": 0.1107,
|
| 108 |
+
"step": 1400
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 1.5,
|
| 112 |
+
"learning_rate": 0.00041999999999999996,
|
| 113 |
+
"loss": 0.0944,
|
| 114 |
+
"step": 1500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 1.6,
|
| 118 |
+
"learning_rate": 0.000408,
|
| 119 |
+
"loss": 0.1082,
|
| 120 |
+
"step": 1600
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 1.7,
|
| 124 |
+
"learning_rate": 0.000396,
|
| 125 |
+
"loss": 0.0868,
|
| 126 |
+
"step": 1700
|
| 127 |
+
},
|
| 128 |
+
{
|
| 129 |
+
"epoch": 1.8,
|
| 130 |
+
"learning_rate": 0.00038399999999999996,
|
| 131 |
+
"loss": 0.1218,
|
| 132 |
+
"step": 1800
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 1.9,
|
| 136 |
+
"learning_rate": 0.000372,
|
| 137 |
+
"loss": 0.1062,
|
| 138 |
+
"step": 1900
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 2.0,
|
| 142 |
+
"learning_rate": 0.00035999999999999997,
|
| 143 |
+
"loss": 0.107,
|
| 144 |
+
"step": 2000
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 2.0,
|
| 148 |
+
"eval_accuracy_background": NaN,
|
| 149 |
+
"eval_accuracy_crack": 5.917486567305492e-06,
|
| 150 |
+
"eval_iou_background": 0.0,
|
| 151 |
+
"eval_iou_crack": 5.917486567305492e-06,
|
| 152 |
+
"eval_loss": 0.12381087243556976,
|
| 153 |
+
"eval_mean_accuracy": 5.917486567305492e-06,
|
| 154 |
+
"eval_mean_iou": 2.958743283652746e-06,
|
| 155 |
+
"eval_overall_accuracy": 5.917486567305492e-06,
|
| 156 |
+
"eval_runtime": 527.1442,
|
| 157 |
+
"eval_samples_per_second": 0.336,
|
| 158 |
+
"eval_steps_per_second": 0.336,
|
| 159 |
+
"step": 2002
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 2.1,
|
| 163 |
+
"learning_rate": 0.00034799999999999995,
|
| 164 |
+
"loss": 0.0988,
|
| 165 |
+
"step": 2100
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 2.2,
|
| 169 |
+
"learning_rate": 0.000336,
|
| 170 |
+
"loss": 0.0966,
|
| 171 |
+
"step": 2200
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 2.3,
|
| 175 |
+
"learning_rate": 0.000324,
|
| 176 |
+
"loss": 0.1167,
|
| 177 |
+
"step": 2300
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 2.4,
|
| 181 |
+
"learning_rate": 0.000312,
|
| 182 |
+
"loss": 0.088,
|
| 183 |
+
"step": 2400
|
| 184 |
+
},
|
| 185 |
+
{
|
| 186 |
+
"epoch": 2.5,
|
| 187 |
+
"learning_rate": 0.0003,
|
| 188 |
+
"loss": 0.0806,
|
| 189 |
+
"step": 2500
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 2.6,
|
| 193 |
+
"learning_rate": 0.00028799999999999995,
|
| 194 |
+
"loss": 0.0864,
|
| 195 |
+
"step": 2600
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 2.7,
|
| 199 |
+
"learning_rate": 0.000276,
|
| 200 |
+
"loss": 0.102,
|
| 201 |
+
"step": 2700
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 2.8,
|
| 205 |
+
"learning_rate": 0.00026399999999999997,
|
| 206 |
+
"loss": 0.0957,
|
| 207 |
+
"step": 2800
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 2.9,
|
| 211 |
+
"learning_rate": 0.00025199999999999995,
|
| 212 |
+
"loss": 0.111,
|
| 213 |
+
"step": 2900
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 3.0,
|
| 217 |
+
"learning_rate": 0.00023999999999999998,
|
| 218 |
+
"loss": 0.1285,
|
| 219 |
+
"step": 3000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 3.0,
|
| 223 |
+
"eval_accuracy_background": NaN,
|
| 224 |
+
"eval_accuracy_crack": 0.004886857656833119,
|
| 225 |
+
"eval_iou_background": 0.0,
|
| 226 |
+
"eval_iou_crack": 0.004827905390592649,
|
| 227 |
+
"eval_loss": 0.1630975902080536,
|
| 228 |
+
"eval_mean_accuracy": 0.004886857656833119,
|
| 229 |
+
"eval_mean_iou": 0.0024139526952963244,
|
| 230 |
+
"eval_overall_accuracy": 0.004886857656833119,
|
| 231 |
+
"eval_runtime": 529.3979,
|
| 232 |
+
"eval_samples_per_second": 0.334,
|
| 233 |
+
"eval_steps_per_second": 0.334,
|
| 234 |
+
"step": 3003
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 3.1,
|
| 238 |
+
"learning_rate": 0.00022799999999999999,
|
| 239 |
+
"loss": 0.0961,
|
| 240 |
+
"step": 3100
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 3.2,
|
| 244 |
+
"learning_rate": 0.00021599999999999996,
|
| 245 |
+
"loss": 0.0937,
|
| 246 |
+
"step": 3200
|
| 247 |
+
},
|
| 248 |
+
{
|
| 249 |
+
"epoch": 3.3,
|
| 250 |
+
"learning_rate": 0.000204,
|
| 251 |
+
"loss": 0.1059,
|
| 252 |
+
"step": 3300
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 3.4,
|
| 256 |
+
"learning_rate": 0.00019199999999999998,
|
| 257 |
+
"loss": 0.1069,
|
| 258 |
+
"step": 3400
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 3.5,
|
| 262 |
+
"learning_rate": 0.00017999999999999998,
|
| 263 |
+
"loss": 0.0571,
|
| 264 |
+
"step": 3500
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 3.6,
|
| 268 |
+
"learning_rate": 0.000168,
|
| 269 |
+
"loss": 0.0958,
|
| 270 |
+
"step": 3600
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 3.7,
|
| 274 |
+
"learning_rate": 0.000156,
|
| 275 |
+
"loss": 0.0836,
|
| 276 |
+
"step": 3700
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 3.8,
|
| 280 |
+
"learning_rate": 0.00014399999999999998,
|
| 281 |
+
"loss": 0.0988,
|
| 282 |
+
"step": 3800
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 3.9,
|
| 286 |
+
"learning_rate": 0.00013199999999999998,
|
| 287 |
+
"loss": 0.0675,
|
| 288 |
+
"step": 3900
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 4.0,
|
| 292 |
+
"learning_rate": 0.00011999999999999999,
|
| 293 |
+
"loss": 0.0887,
|
| 294 |
+
"step": 4000
|
| 295 |
+
},
|
| 296 |
+
{
|
| 297 |
+
"epoch": 4.0,
|
| 298 |
+
"eval_accuracy_background": NaN,
|
| 299 |
+
"eval_accuracy_crack": 0.0003008055671713625,
|
| 300 |
+
"eval_iou_background": 0.0,
|
| 301 |
+
"eval_iou_crack": 0.0003008055671713625,
|
| 302 |
+
"eval_loss": 0.10831139981746674,
|
| 303 |
+
"eval_mean_accuracy": 0.0003008055671713625,
|
| 304 |
+
"eval_mean_iou": 0.00015040278358568126,
|
| 305 |
+
"eval_overall_accuracy": 0.0003008055671713625,
|
| 306 |
+
"eval_runtime": 530.7687,
|
| 307 |
+
"eval_samples_per_second": 0.333,
|
| 308 |
+
"eval_steps_per_second": 0.333,
|
| 309 |
+
"step": 4004
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 4.1,
|
| 313 |
+
"learning_rate": 0.00010799999999999998,
|
| 314 |
+
"loss": 0.075,
|
| 315 |
+
"step": 4100
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 4.2,
|
| 319 |
+
"learning_rate": 9.599999999999999e-05,
|
| 320 |
+
"loss": 0.0891,
|
| 321 |
+
"step": 4200
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 4.3,
|
| 325 |
+
"learning_rate": 8.4e-05,
|
| 326 |
+
"loss": 0.0745,
|
| 327 |
+
"step": 4300
|
| 328 |
+
},
|
| 329 |
+
{
|
| 330 |
+
"epoch": 4.4,
|
| 331 |
+
"learning_rate": 7.199999999999999e-05,
|
| 332 |
+
"loss": 0.092,
|
| 333 |
+
"step": 4400
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 4.5,
|
| 337 |
+
"learning_rate": 5.9999999999999995e-05,
|
| 338 |
+
"loss": 0.0706,
|
| 339 |
+
"step": 4500
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 4.6,
|
| 343 |
+
"learning_rate": 4.7999999999999994e-05,
|
| 344 |
+
"loss": 0.0957,
|
| 345 |
+
"step": 4600
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 4.7,
|
| 349 |
+
"learning_rate": 3.5999999999999994e-05,
|
| 350 |
+
"loss": 0.0776,
|
| 351 |
+
"step": 4700
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 4.8,
|
| 355 |
+
"learning_rate": 2.3999999999999997e-05,
|
| 356 |
+
"loss": 0.0895,
|
| 357 |
+
"step": 4800
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 4.9,
|
| 361 |
+
"learning_rate": 1.1999999999999999e-05,
|
| 362 |
+
"loss": 0.0795,
|
| 363 |
+
"step": 4900
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 5.0,
|
| 367 |
+
"learning_rate": 0.0,
|
| 368 |
+
"loss": 0.0828,
|
| 369 |
+
"step": 5000
|
| 370 |
+
},
|
| 371 |
+
{
|
| 372 |
+
"epoch": 5.0,
|
| 373 |
+
"eval_accuracy_background": NaN,
|
| 374 |
+
"eval_accuracy_crack": 0.0009813165224114941,
|
| 375 |
+
"eval_iou_background": 0.0,
|
| 376 |
+
"eval_iou_crack": 0.0009813165224114941,
|
| 377 |
+
"eval_loss": 0.1343701034784317,
|
| 378 |
+
"eval_mean_accuracy": 0.0009813165224114941,
|
| 379 |
+
"eval_mean_iou": 0.0004906582612057471,
|
| 380 |
+
"eval_overall_accuracy": 0.0009813165224114941,
|
| 381 |
+
"eval_runtime": 531.91,
|
| 382 |
+
"eval_samples_per_second": 0.333,
|
| 383 |
+
"eval_steps_per_second": 0.333,
|
| 384 |
+
"step": 5000
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 5.0,
|
| 388 |
+
"step": 5000,
|
| 389 |
+
"total_flos": 1.99584366723072e+18,
|
| 390 |
+
"train_loss": 0.09988212461471557,
|
| 391 |
+
"train_runtime": 34878.9562,
|
| 392 |
+
"train_samples_per_second": 0.143,
|
| 393 |
+
"train_steps_per_second": 0.143
|
| 394 |
+
}
|
| 395 |
+
],
|
| 396 |
+
"logging_steps": 100,
|
| 397 |
+
"max_steps": 5000,
|
| 398 |
+
"num_train_epochs": 5,
|
| 399 |
+
"save_steps": 500,
|
| 400 |
+
"total_flos": 1.99584366723072e+18,
|
| 401 |
+
"trial_name": null,
|
| 402 |
+
"trial_params": null
|
| 403 |
+
}
|