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 +454 -0
README.md
CHANGED
|
@@ -33,7 +33,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 33 |
|
| 34 |
This model is a fine-tuned version of [microsoft/swin-tiny-patch4-window7-224](https://huggingface.co/microsoft/swin-tiny-patch4-window7-224) on the imagefolder dataset.
|
| 35 |
It achieves the following results on the evaluation set:
|
| 36 |
-
- Loss: 0.
|
| 37 |
- Accuracy: 0.9969
|
| 38 |
|
| 39 |
## Model description
|
|
|
|
| 33 |
|
| 34 |
This model is a fine-tuned version of [microsoft/swin-tiny-patch4-window7-224](https://huggingface.co/microsoft/swin-tiny-patch4-window7-224) on the imagefolder dataset.
|
| 35 |
It achieves the following results on the evaluation set:
|
| 36 |
+
- Loss: 0.0326
|
| 37 |
- Accuracy: 0.9969
|
| 38 |
|
| 39 |
## Model description
|
all_results.json
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.9969230769230769,
|
| 4 |
+
"eval_loss": 0.0326334573328495,
|
| 5 |
+
"eval_runtime": 0.6271,
|
| 6 |
+
"eval_samples_per_second": 518.272,
|
| 7 |
+
"eval_steps_per_second": 17.542,
|
| 8 |
+
"total_flos": 7.265405580565709e+17,
|
| 9 |
+
"train_loss": 0.07611252427425073,
|
| 10 |
+
"train_runtime": 145.9787,
|
| 11 |
+
"train_samples_per_second": 200.235,
|
| 12 |
+
"train_steps_per_second": 1.576
|
| 13 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_accuracy": 0.9969230769230769,
|
| 4 |
+
"eval_loss": 0.0326334573328495,
|
| 5 |
+
"eval_runtime": 0.6271,
|
| 6 |
+
"eval_samples_per_second": 518.272,
|
| 7 |
+
"eval_steps_per_second": 17.542
|
| 8 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"total_flos": 7.265405580565709e+17,
|
| 4 |
+
"train_loss": 0.07611252427425073,
|
| 5 |
+
"train_runtime": 145.9787,
|
| 6 |
+
"train_samples_per_second": 200.235,
|
| 7 |
+
"train_steps_per_second": 1.576
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,454 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.9969230769230769,
|
| 3 |
+
"best_model_checkpoint": "bridalMakeupClassifier/checkpoint-69",
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 230,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.21739130434782608,
|
| 13 |
+
"grad_norm": 5.045365810394287,
|
| 14 |
+
"learning_rate": 1.0869565217391305e-05,
|
| 15 |
+
"loss": 0.5777,
|
| 16 |
+
"step": 5
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.43478260869565216,
|
| 20 |
+
"grad_norm": 2.5848591327667236,
|
| 21 |
+
"learning_rate": 2.173913043478261e-05,
|
| 22 |
+
"loss": 0.4351,
|
| 23 |
+
"step": 10
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.6521739130434783,
|
| 27 |
+
"grad_norm": 1.8966680765151978,
|
| 28 |
+
"learning_rate": 3.260869565217392e-05,
|
| 29 |
+
"loss": 0.2392,
|
| 30 |
+
"step": 15
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.8695652173913043,
|
| 34 |
+
"grad_norm": 1.950931429862976,
|
| 35 |
+
"learning_rate": 4.347826086956522e-05,
|
| 36 |
+
"loss": 0.1604,
|
| 37 |
+
"step": 20
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 1.0,
|
| 41 |
+
"eval_accuracy": 0.9846153846153847,
|
| 42 |
+
"eval_loss": 0.050897907465696335,
|
| 43 |
+
"eval_runtime": 0.6159,
|
| 44 |
+
"eval_samples_per_second": 527.667,
|
| 45 |
+
"eval_steps_per_second": 17.86,
|
| 46 |
+
"step": 23
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 1.0869565217391304,
|
| 50 |
+
"grad_norm": 3.094477891921997,
|
| 51 |
+
"learning_rate": 4.9516908212560386e-05,
|
| 52 |
+
"loss": 0.1457,
|
| 53 |
+
"step": 25
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.3043478260869565,
|
| 57 |
+
"grad_norm": 3.4824132919311523,
|
| 58 |
+
"learning_rate": 4.830917874396135e-05,
|
| 59 |
+
"loss": 0.0999,
|
| 60 |
+
"step": 30
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 1.5217391304347827,
|
| 64 |
+
"grad_norm": 1.870570182800293,
|
| 65 |
+
"learning_rate": 4.710144927536232e-05,
|
| 66 |
+
"loss": 0.0597,
|
| 67 |
+
"step": 35
|
| 68 |
+
},
|
| 69 |
+
{
|
| 70 |
+
"epoch": 1.7391304347826086,
|
| 71 |
+
"grad_norm": 2.119906187057495,
|
| 72 |
+
"learning_rate": 4.589371980676328e-05,
|
| 73 |
+
"loss": 0.076,
|
| 74 |
+
"step": 40
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 1.9565217391304348,
|
| 78 |
+
"grad_norm": 1.0238789319992065,
|
| 79 |
+
"learning_rate": 4.4685990338164255e-05,
|
| 80 |
+
"loss": 0.0837,
|
| 81 |
+
"step": 45
|
| 82 |
+
},
|
| 83 |
+
{
|
| 84 |
+
"epoch": 2.0,
|
| 85 |
+
"eval_accuracy": 0.9876923076923076,
|
| 86 |
+
"eval_loss": 0.035339485853910446,
|
| 87 |
+
"eval_runtime": 0.6194,
|
| 88 |
+
"eval_samples_per_second": 524.743,
|
| 89 |
+
"eval_steps_per_second": 17.761,
|
| 90 |
+
"step": 46
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 2.1739130434782608,
|
| 94 |
+
"grad_norm": 1.9959893226623535,
|
| 95 |
+
"learning_rate": 4.347826086956522e-05,
|
| 96 |
+
"loss": 0.0421,
|
| 97 |
+
"step": 50
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 2.391304347826087,
|
| 101 |
+
"grad_norm": 1.87074875831604,
|
| 102 |
+
"learning_rate": 4.2270531400966186e-05,
|
| 103 |
+
"loss": 0.0812,
|
| 104 |
+
"step": 55
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 2.608695652173913,
|
| 108 |
+
"grad_norm": 2.448767900466919,
|
| 109 |
+
"learning_rate": 4.106280193236715e-05,
|
| 110 |
+
"loss": 0.0712,
|
| 111 |
+
"step": 60
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 2.8260869565217392,
|
| 115 |
+
"grad_norm": 0.7633253931999207,
|
| 116 |
+
"learning_rate": 3.985507246376812e-05,
|
| 117 |
+
"loss": 0.0588,
|
| 118 |
+
"step": 65
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 3.0,
|
| 122 |
+
"eval_accuracy": 0.9969230769230769,
|
| 123 |
+
"eval_loss": 0.0326334573328495,
|
| 124 |
+
"eval_runtime": 0.6252,
|
| 125 |
+
"eval_samples_per_second": 519.854,
|
| 126 |
+
"eval_steps_per_second": 17.595,
|
| 127 |
+
"step": 69
|
| 128 |
+
},
|
| 129 |
+
{
|
| 130 |
+
"epoch": 3.0434782608695654,
|
| 131 |
+
"grad_norm": 1.0322834253311157,
|
| 132 |
+
"learning_rate": 3.864734299516908e-05,
|
| 133 |
+
"loss": 0.0537,
|
| 134 |
+
"step": 70
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 3.260869565217391,
|
| 138 |
+
"grad_norm": 2.0880322456359863,
|
| 139 |
+
"learning_rate": 3.743961352657005e-05,
|
| 140 |
+
"loss": 0.0633,
|
| 141 |
+
"step": 75
|
| 142 |
+
},
|
| 143 |
+
{
|
| 144 |
+
"epoch": 3.4782608695652173,
|
| 145 |
+
"grad_norm": 0.9349754452705383,
|
| 146 |
+
"learning_rate": 3.6231884057971014e-05,
|
| 147 |
+
"loss": 0.0647,
|
| 148 |
+
"step": 80
|
| 149 |
+
},
|
| 150 |
+
{
|
| 151 |
+
"epoch": 3.6956521739130435,
|
| 152 |
+
"grad_norm": 3.263050079345703,
|
| 153 |
+
"learning_rate": 3.502415458937198e-05,
|
| 154 |
+
"loss": 0.0699,
|
| 155 |
+
"step": 85
|
| 156 |
+
},
|
| 157 |
+
{
|
| 158 |
+
"epoch": 3.9130434782608696,
|
| 159 |
+
"grad_norm": 2.5234551429748535,
|
| 160 |
+
"learning_rate": 3.381642512077295e-05,
|
| 161 |
+
"loss": 0.05,
|
| 162 |
+
"step": 90
|
| 163 |
+
},
|
| 164 |
+
{
|
| 165 |
+
"epoch": 4.0,
|
| 166 |
+
"eval_accuracy": 0.9969230769230769,
|
| 167 |
+
"eval_loss": 0.030185265466570854,
|
| 168 |
+
"eval_runtime": 0.6333,
|
| 169 |
+
"eval_samples_per_second": 513.207,
|
| 170 |
+
"eval_steps_per_second": 17.37,
|
| 171 |
+
"step": 92
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 4.130434782608695,
|
| 175 |
+
"grad_norm": 2.066303014755249,
|
| 176 |
+
"learning_rate": 3.260869565217392e-05,
|
| 177 |
+
"loss": 0.058,
|
| 178 |
+
"step": 95
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 4.3478260869565215,
|
| 182 |
+
"grad_norm": 3.373596429824829,
|
| 183 |
+
"learning_rate": 3.140096618357488e-05,
|
| 184 |
+
"loss": 0.0826,
|
| 185 |
+
"step": 100
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 4.565217391304348,
|
| 189 |
+
"grad_norm": 0.9573812484741211,
|
| 190 |
+
"learning_rate": 3.0193236714975848e-05,
|
| 191 |
+
"loss": 0.0314,
|
| 192 |
+
"step": 105
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 4.782608695652174,
|
| 196 |
+
"grad_norm": 1.094375729560852,
|
| 197 |
+
"learning_rate": 2.8985507246376814e-05,
|
| 198 |
+
"loss": 0.0412,
|
| 199 |
+
"step": 110
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 5.0,
|
| 203 |
+
"grad_norm": 1.668631672859192,
|
| 204 |
+
"learning_rate": 2.777777777777778e-05,
|
| 205 |
+
"loss": 0.0284,
|
| 206 |
+
"step": 115
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 5.0,
|
| 210 |
+
"eval_accuracy": 0.9938461538461538,
|
| 211 |
+
"eval_loss": 0.031306326389312744,
|
| 212 |
+
"eval_runtime": 0.6258,
|
| 213 |
+
"eval_samples_per_second": 519.339,
|
| 214 |
+
"eval_steps_per_second": 17.578,
|
| 215 |
+
"step": 115
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 5.217391304347826,
|
| 219 |
+
"grad_norm": 2.7304937839508057,
|
| 220 |
+
"learning_rate": 2.6570048309178748e-05,
|
| 221 |
+
"loss": 0.039,
|
| 222 |
+
"step": 120
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 5.434782608695652,
|
| 226 |
+
"grad_norm": 2.803903341293335,
|
| 227 |
+
"learning_rate": 2.5362318840579714e-05,
|
| 228 |
+
"loss": 0.0409,
|
| 229 |
+
"step": 125
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 5.6521739130434785,
|
| 233 |
+
"grad_norm": 2.4052670001983643,
|
| 234 |
+
"learning_rate": 2.4154589371980676e-05,
|
| 235 |
+
"loss": 0.0489,
|
| 236 |
+
"step": 130
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 5.869565217391305,
|
| 240 |
+
"grad_norm": 0.9198804497718811,
|
| 241 |
+
"learning_rate": 2.294685990338164e-05,
|
| 242 |
+
"loss": 0.0372,
|
| 243 |
+
"step": 135
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 6.0,
|
| 247 |
+
"eval_accuracy": 0.9938461538461538,
|
| 248 |
+
"eval_loss": 0.02726001664996147,
|
| 249 |
+
"eval_runtime": 0.6257,
|
| 250 |
+
"eval_samples_per_second": 519.426,
|
| 251 |
+
"eval_steps_per_second": 17.581,
|
| 252 |
+
"step": 138
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 6.086956521739131,
|
| 256 |
+
"grad_norm": 1.1955087184906006,
|
| 257 |
+
"learning_rate": 2.173913043478261e-05,
|
| 258 |
+
"loss": 0.0498,
|
| 259 |
+
"step": 140
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 6.304347826086957,
|
| 263 |
+
"grad_norm": 1.0752404928207397,
|
| 264 |
+
"learning_rate": 2.0531400966183576e-05,
|
| 265 |
+
"loss": 0.0372,
|
| 266 |
+
"step": 145
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 6.521739130434782,
|
| 270 |
+
"grad_norm": 0.4848777651786804,
|
| 271 |
+
"learning_rate": 1.932367149758454e-05,
|
| 272 |
+
"loss": 0.0229,
|
| 273 |
+
"step": 150
|
| 274 |
+
},
|
| 275 |
+
{
|
| 276 |
+
"epoch": 6.739130434782608,
|
| 277 |
+
"grad_norm": 2.153843641281128,
|
| 278 |
+
"learning_rate": 1.8115942028985507e-05,
|
| 279 |
+
"loss": 0.0242,
|
| 280 |
+
"step": 155
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 6.956521739130435,
|
| 284 |
+
"grad_norm": 0.9689924120903015,
|
| 285 |
+
"learning_rate": 1.6908212560386476e-05,
|
| 286 |
+
"loss": 0.0461,
|
| 287 |
+
"step": 160
|
| 288 |
+
},
|
| 289 |
+
{
|
| 290 |
+
"epoch": 7.0,
|
| 291 |
+
"eval_accuracy": 0.9969230769230769,
|
| 292 |
+
"eval_loss": 0.026803696528077126,
|
| 293 |
+
"eval_runtime": 0.6324,
|
| 294 |
+
"eval_samples_per_second": 513.955,
|
| 295 |
+
"eval_steps_per_second": 17.395,
|
| 296 |
+
"step": 161
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 7.173913043478261,
|
| 300 |
+
"grad_norm": 3.230715751647949,
|
| 301 |
+
"learning_rate": 1.570048309178744e-05,
|
| 302 |
+
"loss": 0.0415,
|
| 303 |
+
"step": 165
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 7.391304347826087,
|
| 307 |
+
"grad_norm": 1.7875250577926636,
|
| 308 |
+
"learning_rate": 1.4492753623188407e-05,
|
| 309 |
+
"loss": 0.0427,
|
| 310 |
+
"step": 170
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 7.608695652173913,
|
| 314 |
+
"grad_norm": 0.38386890292167664,
|
| 315 |
+
"learning_rate": 1.3285024154589374e-05,
|
| 316 |
+
"loss": 0.0542,
|
| 317 |
+
"step": 175
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 7.826086956521739,
|
| 321 |
+
"grad_norm": 2.5545527935028076,
|
| 322 |
+
"learning_rate": 1.2077294685990338e-05,
|
| 323 |
+
"loss": 0.0338,
|
| 324 |
+
"step": 180
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 8.0,
|
| 328 |
+
"eval_accuracy": 0.9969230769230769,
|
| 329 |
+
"eval_loss": 0.02585289627313614,
|
| 330 |
+
"eval_runtime": 0.6295,
|
| 331 |
+
"eval_samples_per_second": 516.298,
|
| 332 |
+
"eval_steps_per_second": 17.475,
|
| 333 |
+
"step": 184
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 8.043478260869565,
|
| 337 |
+
"grad_norm": 1.940019965171814,
|
| 338 |
+
"learning_rate": 1.0869565217391305e-05,
|
| 339 |
+
"loss": 0.0395,
|
| 340 |
+
"step": 185
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 8.26086956521739,
|
| 344 |
+
"grad_norm": 1.6294324398040771,
|
| 345 |
+
"learning_rate": 9.66183574879227e-06,
|
| 346 |
+
"loss": 0.0309,
|
| 347 |
+
"step": 190
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 8.478260869565217,
|
| 351 |
+
"grad_norm": 1.2202316522598267,
|
| 352 |
+
"learning_rate": 8.454106280193238e-06,
|
| 353 |
+
"loss": 0.0169,
|
| 354 |
+
"step": 195
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 8.695652173913043,
|
| 358 |
+
"grad_norm": 0.7180597186088562,
|
| 359 |
+
"learning_rate": 7.246376811594203e-06,
|
| 360 |
+
"loss": 0.0098,
|
| 361 |
+
"step": 200
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 8.91304347826087,
|
| 365 |
+
"grad_norm": 2.1890971660614014,
|
| 366 |
+
"learning_rate": 6.038647342995169e-06,
|
| 367 |
+
"loss": 0.0253,
|
| 368 |
+
"step": 205
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 9.0,
|
| 372 |
+
"eval_accuracy": 0.9938461538461538,
|
| 373 |
+
"eval_loss": 0.02562897279858589,
|
| 374 |
+
"eval_runtime": 0.6368,
|
| 375 |
+
"eval_samples_per_second": 510.384,
|
| 376 |
+
"eval_steps_per_second": 17.275,
|
| 377 |
+
"step": 207
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 9.130434782608695,
|
| 381 |
+
"grad_norm": 1.944470763206482,
|
| 382 |
+
"learning_rate": 4.830917874396135e-06,
|
| 383 |
+
"loss": 0.0497,
|
| 384 |
+
"step": 210
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 9.347826086956522,
|
| 388 |
+
"grad_norm": 1.3670103549957275,
|
| 389 |
+
"learning_rate": 3.6231884057971017e-06,
|
| 390 |
+
"loss": 0.0237,
|
| 391 |
+
"step": 215
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 9.565217391304348,
|
| 395 |
+
"grad_norm": 1.2348929643630981,
|
| 396 |
+
"learning_rate": 2.4154589371980677e-06,
|
| 397 |
+
"loss": 0.0266,
|
| 398 |
+
"step": 220
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 9.782608695652174,
|
| 402 |
+
"grad_norm": 2.2366175651550293,
|
| 403 |
+
"learning_rate": 1.2077294685990338e-06,
|
| 404 |
+
"loss": 0.054,
|
| 405 |
+
"step": 225
|
| 406 |
+
},
|
| 407 |
+
{
|
| 408 |
+
"epoch": 10.0,
|
| 409 |
+
"grad_norm": 0.43929433822631836,
|
| 410 |
+
"learning_rate": 0.0,
|
| 411 |
+
"loss": 0.0326,
|
| 412 |
+
"step": 230
|
| 413 |
+
},
|
| 414 |
+
{
|
| 415 |
+
"epoch": 10.0,
|
| 416 |
+
"eval_accuracy": 0.9969230769230769,
|
| 417 |
+
"eval_loss": 0.02662992849946022,
|
| 418 |
+
"eval_runtime": 0.6361,
|
| 419 |
+
"eval_samples_per_second": 510.892,
|
| 420 |
+
"eval_steps_per_second": 17.292,
|
| 421 |
+
"step": 230
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 10.0,
|
| 425 |
+
"step": 230,
|
| 426 |
+
"total_flos": 7.265405580565709e+17,
|
| 427 |
+
"train_loss": 0.07611252427425073,
|
| 428 |
+
"train_runtime": 145.9787,
|
| 429 |
+
"train_samples_per_second": 200.235,
|
| 430 |
+
"train_steps_per_second": 1.576
|
| 431 |
+
}
|
| 432 |
+
],
|
| 433 |
+
"logging_steps": 5,
|
| 434 |
+
"max_steps": 230,
|
| 435 |
+
"num_input_tokens_seen": 0,
|
| 436 |
+
"num_train_epochs": 10,
|
| 437 |
+
"save_steps": 500,
|
| 438 |
+
"stateful_callbacks": {
|
| 439 |
+
"TrainerControl": {
|
| 440 |
+
"args": {
|
| 441 |
+
"should_epoch_stop": false,
|
| 442 |
+
"should_evaluate": false,
|
| 443 |
+
"should_log": false,
|
| 444 |
+
"should_save": true,
|
| 445 |
+
"should_training_stop": true
|
| 446 |
+
},
|
| 447 |
+
"attributes": {}
|
| 448 |
+
}
|
| 449 |
+
},
|
| 450 |
+
"total_flos": 7.265405580565709e+17,
|
| 451 |
+
"train_batch_size": 32,
|
| 452 |
+
"trial_name": null,
|
| 453 |
+
"trial_params": null
|
| 454 |
+
}
|