End of training
Browse files- README.md +1 -1
- all_results.json +8 -0
- runs/Jan12_12-04-40_GAN-SVR/events.out.tfevents.1736664395.GAN-SVR.3531847.1 +2 -2
- test_results.json +8 -0
- trainer_state.json +540 -0
README.md
CHANGED
|
@@ -18,7 +18,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [MCG-NJU/videomae-base](https://huggingface.co/MCG-NJU/videomae-base) on an unknown dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
-
- Loss: 3.
|
| 22 |
- Accuracy: 0.2575
|
| 23 |
|
| 24 |
## Model description
|
|
|
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [MCG-NJU/videomae-base](https://huggingface.co/MCG-NJU/videomae-base) on an unknown dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 3.0235
|
| 22 |
- Accuracy: 0.2575
|
| 23 |
|
| 24 |
## Model description
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 15.03125,
|
| 3 |
+
"eval_accuracy": 0.25752508361204013,
|
| 4 |
+
"eval_loss": 3.0235204696655273,
|
| 5 |
+
"eval_runtime": 157.8403,
|
| 6 |
+
"eval_samples_per_second": 1.894,
|
| 7 |
+
"eval_steps_per_second": 0.063
|
| 8 |
+
}
|
runs/Jan12_12-04-40_GAN-SVR/events.out.tfevents.1736664395.GAN-SVR.3531847.1
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 sha256:75698df44958717de2a419a0be7118663b127692e64a83924655f97366e8329f
|
| 3 |
+
size 734
|
test_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 15.03125,
|
| 3 |
+
"eval_accuracy": 0.25752508361204013,
|
| 4 |
+
"eval_loss": 3.0235204696655273,
|
| 5 |
+
"eval_runtime": 157.8403,
|
| 6 |
+
"eval_samples_per_second": 1.894,
|
| 7 |
+
"eval_steps_per_second": 0.063
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,540 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.25752508361204013,
|
| 3 |
+
"best_model_checkpoint": "videomae-base-finetuned-ucf101-subset/checkpoint-465",
|
| 4 |
+
"epoch": 15.03125,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 480,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.020833333333333332,
|
| 13 |
+
"grad_norm": 4.865664958953857,
|
| 14 |
+
"learning_rate": 1.0416666666666668e-05,
|
| 15 |
+
"loss": 4.2486,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.041666666666666664,
|
| 20 |
+
"grad_norm": 5.374879837036133,
|
| 21 |
+
"learning_rate": 2.0833333333333336e-05,
|
| 22 |
+
"loss": 4.2806,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.0625,
|
| 27 |
+
"grad_norm": 4.254001140594482,
|
| 28 |
+
"learning_rate": 3.125e-05,
|
| 29 |
+
"loss": 4.2682,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.06458333333333334,
|
| 34 |
+
"eval_accuracy": 0.0033444816053511705,
|
| 35 |
+
"eval_loss": 4.24931526184082,
|
| 36 |
+
"eval_runtime": 156.0528,
|
| 37 |
+
"eval_samples_per_second": 1.916,
|
| 38 |
+
"eval_steps_per_second": 0.064,
|
| 39 |
+
"step": 31
|
| 40 |
+
},
|
| 41 |
+
{
|
| 42 |
+
"epoch": 1.01875,
|
| 43 |
+
"grad_norm": 3.580606698989868,
|
| 44 |
+
"learning_rate": 4.166666666666667e-05,
|
| 45 |
+
"loss": 4.249,
|
| 46 |
+
"step": 40
|
| 47 |
+
},
|
| 48 |
+
{
|
| 49 |
+
"epoch": 1.0395833333333333,
|
| 50 |
+
"grad_norm": 3.6750593185424805,
|
| 51 |
+
"learning_rate": 4.976851851851852e-05,
|
| 52 |
+
"loss": 4.2497,
|
| 53 |
+
"step": 50
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 1.0604166666666666,
|
| 57 |
+
"grad_norm": 3.427530527114868,
|
| 58 |
+
"learning_rate": 4.8611111111111115e-05,
|
| 59 |
+
"loss": 4.2584,
|
| 60 |
+
"step": 60
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"epoch": 1.0645833333333334,
|
| 64 |
+
"eval_accuracy": 0.013377926421404682,
|
| 65 |
+
"eval_loss": 4.243381023406982,
|
| 66 |
+
"eval_runtime": 184.9,
|
| 67 |
+
"eval_samples_per_second": 1.617,
|
| 68 |
+
"eval_steps_per_second": 0.054,
|
| 69 |
+
"step": 62
|
| 70 |
+
},
|
| 71 |
+
{
|
| 72 |
+
"epoch": 2.0166666666666666,
|
| 73 |
+
"grad_norm": 3.5070836544036865,
|
| 74 |
+
"learning_rate": 4.745370370370371e-05,
|
| 75 |
+
"loss": 4.221,
|
| 76 |
+
"step": 70
|
| 77 |
+
},
|
| 78 |
+
{
|
| 79 |
+
"epoch": 2.0375,
|
| 80 |
+
"grad_norm": 3.814365863800049,
|
| 81 |
+
"learning_rate": 4.62962962962963e-05,
|
| 82 |
+
"loss": 4.249,
|
| 83 |
+
"step": 80
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 2.058333333333333,
|
| 87 |
+
"grad_norm": 3.279081344604492,
|
| 88 |
+
"learning_rate": 4.5138888888888894e-05,
|
| 89 |
+
"loss": 4.2518,
|
| 90 |
+
"step": 90
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 2.064583333333333,
|
| 94 |
+
"eval_accuracy": 0.016722408026755852,
|
| 95 |
+
"eval_loss": 4.224606990814209,
|
| 96 |
+
"eval_runtime": 182.615,
|
| 97 |
+
"eval_samples_per_second": 1.637,
|
| 98 |
+
"eval_steps_per_second": 0.055,
|
| 99 |
+
"step": 93
|
| 100 |
+
},
|
| 101 |
+
{
|
| 102 |
+
"epoch": 3.0145833333333334,
|
| 103 |
+
"grad_norm": 2.9108073711395264,
|
| 104 |
+
"learning_rate": 4.3981481481481486e-05,
|
| 105 |
+
"loss": 4.2168,
|
| 106 |
+
"step": 100
|
| 107 |
+
},
|
| 108 |
+
{
|
| 109 |
+
"epoch": 3.035416666666667,
|
| 110 |
+
"grad_norm": 3.1853749752044678,
|
| 111 |
+
"learning_rate": 4.282407407407408e-05,
|
| 112 |
+
"loss": 4.2103,
|
| 113 |
+
"step": 110
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 3.05625,
|
| 117 |
+
"grad_norm": 3.1384546756744385,
|
| 118 |
+
"learning_rate": 4.166666666666667e-05,
|
| 119 |
+
"loss": 4.2445,
|
| 120 |
+
"step": 120
|
| 121 |
+
},
|
| 122 |
+
{
|
| 123 |
+
"epoch": 3.064583333333333,
|
| 124 |
+
"eval_accuracy": 0.006688963210702341,
|
| 125 |
+
"eval_loss": 4.220835208892822,
|
| 126 |
+
"eval_runtime": 172.1522,
|
| 127 |
+
"eval_samples_per_second": 1.737,
|
| 128 |
+
"eval_steps_per_second": 0.058,
|
| 129 |
+
"step": 124
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 4.0125,
|
| 133 |
+
"grad_norm": 2.7196435928344727,
|
| 134 |
+
"learning_rate": 4.0509259259259265e-05,
|
| 135 |
+
"loss": 4.2159,
|
| 136 |
+
"step": 130
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 4.033333333333333,
|
| 140 |
+
"grad_norm": 2.6634223461151123,
|
| 141 |
+
"learning_rate": 3.935185185185186e-05,
|
| 142 |
+
"loss": 4.2063,
|
| 143 |
+
"step": 140
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 4.054166666666666,
|
| 147 |
+
"grad_norm": 2.5531411170959473,
|
| 148 |
+
"learning_rate": 3.8194444444444444e-05,
|
| 149 |
+
"loss": 4.2272,
|
| 150 |
+
"step": 150
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 4.064583333333333,
|
| 154 |
+
"eval_accuracy": 0.010033444816053512,
|
| 155 |
+
"eval_loss": 4.223015308380127,
|
| 156 |
+
"eval_runtime": 163.7094,
|
| 157 |
+
"eval_samples_per_second": 1.826,
|
| 158 |
+
"eval_steps_per_second": 0.061,
|
| 159 |
+
"step": 155
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 5.010416666666667,
|
| 163 |
+
"grad_norm": 2.824289321899414,
|
| 164 |
+
"learning_rate": 3.7037037037037037e-05,
|
| 165 |
+
"loss": 4.2151,
|
| 166 |
+
"step": 160
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 5.03125,
|
| 170 |
+
"grad_norm": 2.8366966247558594,
|
| 171 |
+
"learning_rate": 3.587962962962963e-05,
|
| 172 |
+
"loss": 4.183,
|
| 173 |
+
"step": 170
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 5.052083333333333,
|
| 177 |
+
"grad_norm": 3.340677499771118,
|
| 178 |
+
"learning_rate": 3.472222222222222e-05,
|
| 179 |
+
"loss": 4.205,
|
| 180 |
+
"step": 180
|
| 181 |
+
},
|
| 182 |
+
{
|
| 183 |
+
"epoch": 5.064583333333333,
|
| 184 |
+
"eval_accuracy": 0.023411371237458192,
|
| 185 |
+
"eval_loss": 4.211067199707031,
|
| 186 |
+
"eval_runtime": 166.937,
|
| 187 |
+
"eval_samples_per_second": 1.791,
|
| 188 |
+
"eval_steps_per_second": 0.06,
|
| 189 |
+
"step": 186
|
| 190 |
+
},
|
| 191 |
+
{
|
| 192 |
+
"epoch": 6.008333333333334,
|
| 193 |
+
"grad_norm": 2.784593105316162,
|
| 194 |
+
"learning_rate": 3.3564814814814815e-05,
|
| 195 |
+
"loss": 4.1898,
|
| 196 |
+
"step": 190
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 6.029166666666667,
|
| 200 |
+
"grad_norm": 3.389150381088257,
|
| 201 |
+
"learning_rate": 3.240740740740741e-05,
|
| 202 |
+
"loss": 4.1386,
|
| 203 |
+
"step": 200
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 6.05,
|
| 207 |
+
"grad_norm": 3.5054867267608643,
|
| 208 |
+
"learning_rate": 3.125e-05,
|
| 209 |
+
"loss": 4.1238,
|
| 210 |
+
"step": 210
|
| 211 |
+
},
|
| 212 |
+
{
|
| 213 |
+
"epoch": 6.064583333333333,
|
| 214 |
+
"eval_accuracy": 0.03678929765886288,
|
| 215 |
+
"eval_loss": 4.111179828643799,
|
| 216 |
+
"eval_runtime": 161.8854,
|
| 217 |
+
"eval_samples_per_second": 1.847,
|
| 218 |
+
"eval_steps_per_second": 0.062,
|
| 219 |
+
"step": 217
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 7.00625,
|
| 223 |
+
"grad_norm": 4.150498867034912,
|
| 224 |
+
"learning_rate": 3.0092592592592593e-05,
|
| 225 |
+
"loss": 4.0898,
|
| 226 |
+
"step": 220
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 7.027083333333334,
|
| 230 |
+
"grad_norm": 4.683104038238525,
|
| 231 |
+
"learning_rate": 2.8935185185185186e-05,
|
| 232 |
+
"loss": 4.0073,
|
| 233 |
+
"step": 230
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 7.047916666666667,
|
| 237 |
+
"grad_norm": 4.379587650299072,
|
| 238 |
+
"learning_rate": 2.777777777777778e-05,
|
| 239 |
+
"loss": 3.9136,
|
| 240 |
+
"step": 240
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 7.064583333333333,
|
| 244 |
+
"eval_accuracy": 0.07357859531772576,
|
| 245 |
+
"eval_loss": 3.8529512882232666,
|
| 246 |
+
"eval_runtime": 164.2718,
|
| 247 |
+
"eval_samples_per_second": 1.82,
|
| 248 |
+
"eval_steps_per_second": 0.061,
|
| 249 |
+
"step": 248
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 8.004166666666666,
|
| 253 |
+
"grad_norm": 5.4900078773498535,
|
| 254 |
+
"learning_rate": 2.6620370370370372e-05,
|
| 255 |
+
"loss": 3.8304,
|
| 256 |
+
"step": 250
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 8.025,
|
| 260 |
+
"grad_norm": 4.974089622497559,
|
| 261 |
+
"learning_rate": 2.5462962962962965e-05,
|
| 262 |
+
"loss": 3.7169,
|
| 263 |
+
"step": 260
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 8.045833333333333,
|
| 267 |
+
"grad_norm": 5.386253833770752,
|
| 268 |
+
"learning_rate": 2.4305555555555558e-05,
|
| 269 |
+
"loss": 3.6241,
|
| 270 |
+
"step": 270
|
| 271 |
+
},
|
| 272 |
+
{
|
| 273 |
+
"epoch": 8.064583333333333,
|
| 274 |
+
"eval_accuracy": 0.11705685618729098,
|
| 275 |
+
"eval_loss": 3.6734354496002197,
|
| 276 |
+
"eval_runtime": 159.295,
|
| 277 |
+
"eval_samples_per_second": 1.877,
|
| 278 |
+
"eval_steps_per_second": 0.063,
|
| 279 |
+
"step": 279
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 9.002083333333333,
|
| 283 |
+
"grad_norm": 5.429062843322754,
|
| 284 |
+
"learning_rate": 2.314814814814815e-05,
|
| 285 |
+
"loss": 3.4885,
|
| 286 |
+
"step": 280
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 9.022916666666667,
|
| 290 |
+
"grad_norm": 5.275113105773926,
|
| 291 |
+
"learning_rate": 2.1990740740740743e-05,
|
| 292 |
+
"loss": 3.4022,
|
| 293 |
+
"step": 290
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 9.04375,
|
| 297 |
+
"grad_norm": 6.2369704246521,
|
| 298 |
+
"learning_rate": 2.0833333333333336e-05,
|
| 299 |
+
"loss": 3.2977,
|
| 300 |
+
"step": 300
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 9.064583333333333,
|
| 304 |
+
"grad_norm": 13.635384559631348,
|
| 305 |
+
"learning_rate": 1.967592592592593e-05,
|
| 306 |
+
"loss": 3.3103,
|
| 307 |
+
"step": 310
|
| 308 |
+
},
|
| 309 |
+
{
|
| 310 |
+
"epoch": 9.064583333333333,
|
| 311 |
+
"eval_accuracy": 0.10702341137123746,
|
| 312 |
+
"eval_loss": 3.5260610580444336,
|
| 313 |
+
"eval_runtime": 170.757,
|
| 314 |
+
"eval_samples_per_second": 1.751,
|
| 315 |
+
"eval_steps_per_second": 0.059,
|
| 316 |
+
"step": 310
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 10.020833333333334,
|
| 320 |
+
"grad_norm": 6.099160194396973,
|
| 321 |
+
"learning_rate": 1.8518518518518518e-05,
|
| 322 |
+
"loss": 3.1017,
|
| 323 |
+
"step": 320
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 10.041666666666666,
|
| 327 |
+
"grad_norm": 8.271078109741211,
|
| 328 |
+
"learning_rate": 1.736111111111111e-05,
|
| 329 |
+
"loss": 3.0604,
|
| 330 |
+
"step": 330
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 10.0625,
|
| 334 |
+
"grad_norm": 6.345729827880859,
|
| 335 |
+
"learning_rate": 1.6203703703703704e-05,
|
| 336 |
+
"loss": 3.0981,
|
| 337 |
+
"step": 340
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 10.064583333333333,
|
| 341 |
+
"eval_accuracy": 0.16387959866220736,
|
| 342 |
+
"eval_loss": 3.3859572410583496,
|
| 343 |
+
"eval_runtime": 165.4502,
|
| 344 |
+
"eval_samples_per_second": 1.807,
|
| 345 |
+
"eval_steps_per_second": 0.06,
|
| 346 |
+
"step": 341
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 11.01875,
|
| 350 |
+
"grad_norm": 6.985031604766846,
|
| 351 |
+
"learning_rate": 1.5046296296296297e-05,
|
| 352 |
+
"loss": 2.9706,
|
| 353 |
+
"step": 350
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 11.039583333333333,
|
| 357 |
+
"grad_norm": 6.81059455871582,
|
| 358 |
+
"learning_rate": 1.388888888888889e-05,
|
| 359 |
+
"loss": 2.916,
|
| 360 |
+
"step": 360
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 11.060416666666667,
|
| 364 |
+
"grad_norm": 6.5202436447143555,
|
| 365 |
+
"learning_rate": 1.2731481481481482e-05,
|
| 366 |
+
"loss": 2.8216,
|
| 367 |
+
"step": 370
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 11.064583333333333,
|
| 371 |
+
"eval_accuracy": 0.2140468227424749,
|
| 372 |
+
"eval_loss": 3.179076671600342,
|
| 373 |
+
"eval_runtime": 163.8066,
|
| 374 |
+
"eval_samples_per_second": 1.825,
|
| 375 |
+
"eval_steps_per_second": 0.061,
|
| 376 |
+
"step": 372
|
| 377 |
+
},
|
| 378 |
+
{
|
| 379 |
+
"epoch": 12.016666666666667,
|
| 380 |
+
"grad_norm": 5.686159610748291,
|
| 381 |
+
"learning_rate": 1.1574074074074075e-05,
|
| 382 |
+
"loss": 2.7755,
|
| 383 |
+
"step": 380
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 12.0375,
|
| 387 |
+
"grad_norm": 6.596237659454346,
|
| 388 |
+
"learning_rate": 1.0416666666666668e-05,
|
| 389 |
+
"loss": 2.824,
|
| 390 |
+
"step": 390
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 12.058333333333334,
|
| 394 |
+
"grad_norm": 7.551661968231201,
|
| 395 |
+
"learning_rate": 9.259259259259259e-06,
|
| 396 |
+
"loss": 2.6108,
|
| 397 |
+
"step": 400
|
| 398 |
+
},
|
| 399 |
+
{
|
| 400 |
+
"epoch": 12.064583333333333,
|
| 401 |
+
"eval_accuracy": 0.24414715719063546,
|
| 402 |
+
"eval_loss": 3.1618316173553467,
|
| 403 |
+
"eval_runtime": 177.1262,
|
| 404 |
+
"eval_samples_per_second": 1.688,
|
| 405 |
+
"eval_steps_per_second": 0.056,
|
| 406 |
+
"step": 403
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 13.014583333333333,
|
| 410 |
+
"grad_norm": 6.725275039672852,
|
| 411 |
+
"learning_rate": 8.101851851851852e-06,
|
| 412 |
+
"loss": 2.7309,
|
| 413 |
+
"step": 410
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 13.035416666666666,
|
| 417 |
+
"grad_norm": 7.1535563468933105,
|
| 418 |
+
"learning_rate": 6.944444444444445e-06,
|
| 419 |
+
"loss": 2.6451,
|
| 420 |
+
"step": 420
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 13.05625,
|
| 424 |
+
"grad_norm": 7.705063819885254,
|
| 425 |
+
"learning_rate": 5.787037037037038e-06,
|
| 426 |
+
"loss": 2.598,
|
| 427 |
+
"step": 430
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 13.064583333333333,
|
| 431 |
+
"eval_accuracy": 0.23411371237458195,
|
| 432 |
+
"eval_loss": 3.0792758464813232,
|
| 433 |
+
"eval_runtime": 186.6022,
|
| 434 |
+
"eval_samples_per_second": 1.602,
|
| 435 |
+
"eval_steps_per_second": 0.054,
|
| 436 |
+
"step": 434
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 14.0125,
|
| 440 |
+
"grad_norm": 5.5132927894592285,
|
| 441 |
+
"learning_rate": 4.6296296296296296e-06,
|
| 442 |
+
"loss": 2.5532,
|
| 443 |
+
"step": 440
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 14.033333333333333,
|
| 447 |
+
"grad_norm": 6.853929042816162,
|
| 448 |
+
"learning_rate": 3.4722222222222224e-06,
|
| 449 |
+
"loss": 2.56,
|
| 450 |
+
"step": 450
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 14.054166666666667,
|
| 454 |
+
"grad_norm": 6.731771469116211,
|
| 455 |
+
"learning_rate": 2.3148148148148148e-06,
|
| 456 |
+
"loss": 2.5023,
|
| 457 |
+
"step": 460
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 14.064583333333333,
|
| 461 |
+
"eval_accuracy": 0.25752508361204013,
|
| 462 |
+
"eval_loss": 3.019373893737793,
|
| 463 |
+
"eval_runtime": 155.2839,
|
| 464 |
+
"eval_samples_per_second": 1.926,
|
| 465 |
+
"eval_steps_per_second": 0.064,
|
| 466 |
+
"step": 465
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 15.010416666666666,
|
| 470 |
+
"grad_norm": 6.570545673370361,
|
| 471 |
+
"learning_rate": 1.1574074074074074e-06,
|
| 472 |
+
"loss": 2.5538,
|
| 473 |
+
"step": 470
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 15.03125,
|
| 477 |
+
"grad_norm": 6.480160236358643,
|
| 478 |
+
"learning_rate": 0.0,
|
| 479 |
+
"loss": 2.513,
|
| 480 |
+
"step": 480
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 15.03125,
|
| 484 |
+
"eval_accuracy": 0.23745819397993312,
|
| 485 |
+
"eval_loss": 3.0668206214904785,
|
| 486 |
+
"eval_runtime": 187.3659,
|
| 487 |
+
"eval_samples_per_second": 1.596,
|
| 488 |
+
"eval_steps_per_second": 0.053,
|
| 489 |
+
"step": 480
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 15.03125,
|
| 493 |
+
"step": 480,
|
| 494 |
+
"total_flos": 1.8664399999458017e+19,
|
| 495 |
+
"train_loss": 3.5868410070737204,
|
| 496 |
+
"train_runtime": 12231.8484,
|
| 497 |
+
"train_samples_per_second": 1.256,
|
| 498 |
+
"train_steps_per_second": 0.039
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 15.03125,
|
| 502 |
+
"eval_accuracy": 0.25752508361204013,
|
| 503 |
+
"eval_loss": 3.0240976810455322,
|
| 504 |
+
"eval_runtime": 158.7962,
|
| 505 |
+
"eval_samples_per_second": 1.883,
|
| 506 |
+
"eval_steps_per_second": 0.063,
|
| 507 |
+
"step": 480
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 15.03125,
|
| 511 |
+
"eval_accuracy": 0.25752508361204013,
|
| 512 |
+
"eval_loss": 3.0235204696655273,
|
| 513 |
+
"eval_runtime": 157.8403,
|
| 514 |
+
"eval_samples_per_second": 1.894,
|
| 515 |
+
"eval_steps_per_second": 0.063,
|
| 516 |
+
"step": 480
|
| 517 |
+
}
|
| 518 |
+
],
|
| 519 |
+
"logging_steps": 10,
|
| 520 |
+
"max_steps": 480,
|
| 521 |
+
"num_input_tokens_seen": 0,
|
| 522 |
+
"num_train_epochs": 9223372036854775807,
|
| 523 |
+
"save_steps": 500,
|
| 524 |
+
"stateful_callbacks": {
|
| 525 |
+
"TrainerControl": {
|
| 526 |
+
"args": {
|
| 527 |
+
"should_epoch_stop": false,
|
| 528 |
+
"should_evaluate": false,
|
| 529 |
+
"should_log": false,
|
| 530 |
+
"should_save": true,
|
| 531 |
+
"should_training_stop": true
|
| 532 |
+
},
|
| 533 |
+
"attributes": {}
|
| 534 |
+
}
|
| 535 |
+
},
|
| 536 |
+
"total_flos": 1.8664399999458017e+19,
|
| 537 |
+
"train_batch_size": 32,
|
| 538 |
+
"trial_name": null,
|
| 539 |
+
"trial_params": null
|
| 540 |
+
}
|