Model save
Browse files- README.md +66 -36
- all_results.json +8 -0
- logs/events.out.tfevents.1743190156.4d701d7b0712.5422.1 +3 -0
- train_results.json +8 -0
- trainer_state.json +522 -0
- training_args.bin +1 -1
- training_metrics.xlsx +0 -0
README.md
CHANGED
|
@@ -18,8 +18,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [apple/mobilevit-small](https://huggingface.co/apple/mobilevit-small) on an unknown dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
-
- Loss: 0.
|
| 22 |
-
- Accuracy: 0.
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
@@ -45,43 +45,73 @@ The following hyperparameters were used during training:
|
|
| 45 |
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
| 46 |
- lr_scheduler_type: cosine_with_restarts
|
| 47 |
- lr_scheduler_warmup_steps: 256
|
| 48 |
-
- num_epochs:
|
| 49 |
- mixed_precision_training: Native AMP
|
| 50 |
|
| 51 |
### Training results
|
| 52 |
|
| 53 |
-
| Training Loss | Epoch | Step | Validation Loss |
|
| 54 |
-
|
| 55 |
-
| 2.0561 | 0.5 | 64 | 2.0213 |
|
| 56 |
-
| 1.9819 | 1.0 | 128 | 1.8788 |
|
| 57 |
-
| 1.771 | 1.5 | 192 | 1.5291 |
|
| 58 |
-
| 1.3911 | 2.0 | 256 | 1.0706 |
|
| 59 |
-
| 1.0026 | 2.5 | 320 | 0.
|
| 60 |
-
| 0.7657 | 3.0 | 384 | 0.
|
| 61 |
-
| 0.6082 | 3.5 | 448 | 0.
|
| 62 |
-
| 0.5313 | 4.0 | 512 | 0.
|
| 63 |
-
| 0.4623 | 4.5 | 576 | 0.
|
| 64 |
-
| 0.4212 | 5.0 | 640 | 0.
|
| 65 |
-
| 0.4025 | 5.5 | 704 | 0.
|
| 66 |
-
| 0.3892 | 6.0 | 768 | 0.
|
| 67 |
-
| 0.3959 | 6.5 | 832 | 0.
|
| 68 |
-
| 0.3279 | 7.0 | 896 | 0.
|
| 69 |
-
| 0.2793 | 7.5 | 960 | 0.
|
| 70 |
-
| 0.222 | 8.0 | 1024 | 0.
|
| 71 |
-
| 0.1862 | 8.5 | 1088 | 0.
|
| 72 |
-
| 0.1717 | 9.0 | 1152 | 0.
|
| 73 |
-
| 0.1408 | 9.5 | 1216 | 0.
|
| 74 |
-
| 0.1507 | 10.0 | 1280 | 0.
|
| 75 |
-
| 0.1369 | 10.5 | 1344 | 0.
|
| 76 |
-
| 0.1389 | 11.0 | 1408 | 0.
|
| 77 |
-
| 0.1199 | 11.5 | 1472 | 0.
|
| 78 |
-
| 0.1003 | 12.0 | 1536 | 0.
|
| 79 |
-
| 0.0756 | 12.5 | 1600 | 0.
|
| 80 |
-
| 0.0636 | 13.0 | 1664 | 0.
|
| 81 |
-
| 0.058 | 13.5 | 1728 | 0.
|
| 82 |
-
| 0.0566 | 14.0 | 1792 | 0.
|
| 83 |
-
| 0.0516 | 14.5 | 1856 | 0.
|
| 84 |
-
| 0.0426 | 15.0 | 1920 | 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 85 |
|
| 86 |
|
| 87 |
### Framework versions
|
|
@@ -89,4 +119,4 @@ The following hyperparameters were used during training:
|
|
| 89 |
- Transformers 4.48.3
|
| 90 |
- Pytorch 2.5.1+cu124
|
| 91 |
- Datasets 3.3.2
|
| 92 |
-
- Tokenizers 0.21.
|
|
|
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [apple/mobilevit-small](https://huggingface.co/apple/mobilevit-small) on an unknown dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.4089
|
| 22 |
+
- Accuracy: 0.9295
|
| 23 |
|
| 24 |
## Model description
|
| 25 |
|
|
|
|
| 45 |
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
| 46 |
- lr_scheduler_type: cosine_with_restarts
|
| 47 |
- lr_scheduler_warmup_steps: 256
|
| 48 |
+
- num_epochs: 30
|
| 49 |
- mixed_precision_training: Native AMP
|
| 50 |
|
| 51 |
### Training results
|
| 52 |
|
| 53 |
+
| Training Loss | Epoch | Step | Accuracy | Validation Loss |
|
| 54 |
+
|:-------------:|:-----:|:----:|:--------:|:---------------:|
|
| 55 |
+
| 2.0561 | 0.5 | 64 | 0.2886 | 2.0213 |
|
| 56 |
+
| 1.9819 | 1.0 | 128 | 0.5503 | 1.8788 |
|
| 57 |
+
| 1.771 | 1.5 | 192 | 0.6107 | 1.5291 |
|
| 58 |
+
| 1.3911 | 2.0 | 256 | 0.7349 | 1.0706 |
|
| 59 |
+
| 1.0026 | 2.5 | 320 | 0.8054 | 0.7560 |
|
| 60 |
+
| 0.7657 | 3.0 | 384 | 0.8356 | 0.6180 |
|
| 61 |
+
| 0.6082 | 3.5 | 448 | 0.8389 | 0.5422 |
|
| 62 |
+
| 0.5313 | 4.0 | 512 | 0.8523 | 0.4946 |
|
| 63 |
+
| 0.4623 | 4.5 | 576 | 0.8758 | 0.4512 |
|
| 64 |
+
| 0.4212 | 5.0 | 640 | 0.8792 | 0.4322 |
|
| 65 |
+
| 0.4025 | 5.5 | 704 | 0.8893 | 0.4259 |
|
| 66 |
+
| 0.3892 | 6.0 | 768 | 0.8859 | 0.4238 |
|
| 67 |
+
| 0.3959 | 6.5 | 832 | 0.8859 | 0.4083 |
|
| 68 |
+
| 0.3279 | 7.0 | 896 | 0.8826 | 0.3750 |
|
| 69 |
+
| 0.2793 | 7.5 | 960 | 0.8993 | 0.3350 |
|
| 70 |
+
| 0.222 | 8.0 | 1024 | 0.8960 | 0.3208 |
|
| 71 |
+
| 0.1862 | 8.5 | 1088 | 0.8993 | 0.3128 |
|
| 72 |
+
| 0.1717 | 9.0 | 1152 | 0.9027 | 0.3049 |
|
| 73 |
+
| 0.1408 | 9.5 | 1216 | 0.9027 | 0.3010 |
|
| 74 |
+
| 0.1507 | 10.0 | 1280 | 0.9161 | 0.3240 |
|
| 75 |
+
| 0.1369 | 10.5 | 1344 | 0.9060 | 0.3063 |
|
| 76 |
+
| 0.1389 | 11.0 | 1408 | 0.9060 | 0.3045 |
|
| 77 |
+
| 0.1199 | 11.5 | 1472 | 0.9094 | 0.3062 |
|
| 78 |
+
| 0.1003 | 12.0 | 1536 | 0.9128 | 0.3131 |
|
| 79 |
+
| 0.0756 | 12.5 | 1600 | 0.9228 | 0.3002 |
|
| 80 |
+
| 0.0636 | 13.0 | 1664 | 0.9128 | 0.3177 |
|
| 81 |
+
| 0.058 | 13.5 | 1728 | 0.9228 | 0.3143 |
|
| 82 |
+
| 0.0566 | 14.0 | 1792 | 0.9195 | 0.3136 |
|
| 83 |
+
| 0.0516 | 14.5 | 1856 | 0.9161 | 0.3447 |
|
| 84 |
+
| 0.0426 | 15.0 | 1920 | 0.9228 | 0.2911 |
|
| 85 |
+
| 0.0513 | 15.5 | 1984 | 0.3028 | 0.9228 |
|
| 86 |
+
| 0.0447 | 16.0 | 2048 | 0.3328 | 0.9195 |
|
| 87 |
+
| 0.0332 | 16.5 | 2112 | 0.3193 | 0.9262 |
|
| 88 |
+
| 0.0358 | 17.0 | 2176 | 0.3385 | 0.9161 |
|
| 89 |
+
| 0.0343 | 17.5 | 2240 | 0.3297 | 0.9295 |
|
| 90 |
+
| 0.0291 | 18.0 | 2304 | 0.3518 | 0.9161 |
|
| 91 |
+
| 0.0287 | 18.5 | 2368 | 0.3224 | 0.9195 |
|
| 92 |
+
| 0.0197 | 19.0 | 2432 | 0.3099 | 0.9228 |
|
| 93 |
+
| 0.0223 | 19.5 | 2496 | 0.3305 | 0.9295 |
|
| 94 |
+
| 0.0282 | 20.0 | 2560 | 0.3378 | 0.9161 |
|
| 95 |
+
| 0.0231 | 20.5 | 2624 | 0.3077 | 0.9228 |
|
| 96 |
+
| 0.0251 | 21.0 | 2688 | 0.3520 | 0.9161 |
|
| 97 |
+
| 0.021 | 21.5 | 2752 | 0.3506 | 0.9228 |
|
| 98 |
+
| 0.0222 | 22.0 | 2816 | 0.3561 | 0.9128 |
|
| 99 |
+
| 0.016 | 22.5 | 2880 | 0.3482 | 0.9195 |
|
| 100 |
+
| 0.0163 | 23.0 | 2944 | 0.3429 | 0.9228 |
|
| 101 |
+
| 0.0114 | 23.5 | 3008 | 0.3839 | 0.9329 |
|
| 102 |
+
| 0.0106 | 24.0 | 3072 | 0.4066 | 0.9262 |
|
| 103 |
+
| 0.0111 | 24.5 | 3136 | 0.4003 | 0.9329 |
|
| 104 |
+
| 0.009 | 25.0 | 3200 | 0.4000 | 0.9262 |
|
| 105 |
+
| 0.0088 | 25.5 | 3264 | 0.3667 | 0.9228 |
|
| 106 |
+
| 0.0057 | 26.0 | 3328 | 0.3587 | 0.9195 |
|
| 107 |
+
| 0.0073 | 26.5 | 3392 | 0.3686 | 0.9329 |
|
| 108 |
+
| 0.0085 | 27.0 | 3456 | 0.3676 | 0.9195 |
|
| 109 |
+
| 0.0087 | 27.5 | 3520 | 0.4251 | 0.9262 |
|
| 110 |
+
| 0.0061 | 28.0 | 3584 | 0.3879 | 0.9195 |
|
| 111 |
+
| 0.0062 | 28.5 | 3648 | 0.3865 | 0.9195 |
|
| 112 |
+
| 0.0068 | 29.0 | 3712 | 0.3943 | 0.9262 |
|
| 113 |
+
| 0.0092 | 29.5 | 3776 | 0.4064 | 0.9228 |
|
| 114 |
+
| 0.0078 | 30.0 | 3840 | 0.4089 | 0.9295 |
|
| 115 |
|
| 116 |
|
| 117 |
### Framework versions
|
|
|
|
| 119 |
- Transformers 4.48.3
|
| 120 |
- Pytorch 2.5.1+cu124
|
| 121 |
- Datasets 3.3.2
|
| 122 |
+
- Tokenizers 0.21.1
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 15.0,
|
| 3 |
+
"total_flos": 7.164784986292224e+17,
|
| 4 |
+
"train_loss": 0.48338687382638457,
|
| 5 |
+
"train_runtime": 5148.0361,
|
| 6 |
+
"train_samples_per_second": 23.869,
|
| 7 |
+
"train_steps_per_second": 0.373
|
| 8 |
+
}
|
logs/events.out.tfevents.1743190156.4d701d7b0712.5422.1
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:70b2c8d08479bc0fe7360f0f7efbaf4230964f41739d5218dbe876fc335fb782
|
| 3 |
+
size 22373
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 15.0,
|
| 3 |
+
"total_flos": 7.164784986292224e+17,
|
| 4 |
+
"train_loss": 0.48338687382638457,
|
| 5 |
+
"train_runtime": 5148.0361,
|
| 6 |
+
"train_samples_per_second": 23.869,
|
| 7 |
+
"train_steps_per_second": 0.373
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,522 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.2911371886730194,
|
| 3 |
+
"best_model_checkpoint": "./drive/Shareddrives/CS198-Drones/[v4] Training Output/mobilevit-small_rice-leaf-disease-augmented-v4_fft/checkpoint-1920",
|
| 4 |
+
"epoch": 15.0,
|
| 5 |
+
"eval_steps": 64,
|
| 6 |
+
"global_step": 1920,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.5,
|
| 13 |
+
"grad_norm": 0.8390832543373108,
|
| 14 |
+
"learning_rate": 7.5e-06,
|
| 15 |
+
"loss": 2.0561,
|
| 16 |
+
"step": 64
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.5,
|
| 20 |
+
"eval_accuracy": 0.28859060402684567,
|
| 21 |
+
"eval_loss": 2.0213236808776855,
|
| 22 |
+
"eval_runtime": 9.1133,
|
| 23 |
+
"eval_samples_per_second": 32.7,
|
| 24 |
+
"eval_steps_per_second": 0.549,
|
| 25 |
+
"step": 64
|
| 26 |
+
},
|
| 27 |
+
{
|
| 28 |
+
"epoch": 1.0,
|
| 29 |
+
"grad_norm": 0.9097115397453308,
|
| 30 |
+
"learning_rate": 1.5e-05,
|
| 31 |
+
"loss": 1.9819,
|
| 32 |
+
"step": 128
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 1.0,
|
| 36 |
+
"eval_accuracy": 0.5503355704697986,
|
| 37 |
+
"eval_loss": 1.8788046836853027,
|
| 38 |
+
"eval_runtime": 9.5999,
|
| 39 |
+
"eval_samples_per_second": 31.042,
|
| 40 |
+
"eval_steps_per_second": 0.521,
|
| 41 |
+
"step": 128
|
| 42 |
+
},
|
| 43 |
+
{
|
| 44 |
+
"epoch": 1.5,
|
| 45 |
+
"grad_norm": 1.2205861806869507,
|
| 46 |
+
"learning_rate": 2.25e-05,
|
| 47 |
+
"loss": 1.771,
|
| 48 |
+
"step": 192
|
| 49 |
+
},
|
| 50 |
+
{
|
| 51 |
+
"epoch": 1.5,
|
| 52 |
+
"eval_accuracy": 0.610738255033557,
|
| 53 |
+
"eval_loss": 1.5290961265563965,
|
| 54 |
+
"eval_runtime": 9.5293,
|
| 55 |
+
"eval_samples_per_second": 31.272,
|
| 56 |
+
"eval_steps_per_second": 0.525,
|
| 57 |
+
"step": 192
|
| 58 |
+
},
|
| 59 |
+
{
|
| 60 |
+
"epoch": 2.0,
|
| 61 |
+
"grad_norm": 1.496865153312683,
|
| 62 |
+
"learning_rate": 3e-05,
|
| 63 |
+
"loss": 1.3911,
|
| 64 |
+
"step": 256
|
| 65 |
+
},
|
| 66 |
+
{
|
| 67 |
+
"epoch": 2.0,
|
| 68 |
+
"eval_accuracy": 0.7348993288590604,
|
| 69 |
+
"eval_loss": 1.0706205368041992,
|
| 70 |
+
"eval_runtime": 9.6229,
|
| 71 |
+
"eval_samples_per_second": 30.968,
|
| 72 |
+
"eval_steps_per_second": 0.52,
|
| 73 |
+
"step": 256
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 2.5,
|
| 77 |
+
"grad_norm": 1.845400333404541,
|
| 78 |
+
"learning_rate": 2.9025243640281226e-05,
|
| 79 |
+
"loss": 1.0026,
|
| 80 |
+
"step": 320
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 2.5,
|
| 84 |
+
"eval_accuracy": 0.8053691275167785,
|
| 85 |
+
"eval_loss": 0.756048321723938,
|
| 86 |
+
"eval_runtime": 9.6895,
|
| 87 |
+
"eval_samples_per_second": 30.755,
|
| 88 |
+
"eval_steps_per_second": 0.516,
|
| 89 |
+
"step": 320
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 3.0,
|
| 93 |
+
"grad_norm": 2.1688945293426514,
|
| 94 |
+
"learning_rate": 2.6227661222566516e-05,
|
| 95 |
+
"loss": 0.7657,
|
| 96 |
+
"step": 384
|
| 97 |
+
},
|
| 98 |
+
{
|
| 99 |
+
"epoch": 3.0,
|
| 100 |
+
"eval_accuracy": 0.8355704697986577,
|
| 101 |
+
"eval_loss": 0.6180014610290527,
|
| 102 |
+
"eval_runtime": 9.5752,
|
| 103 |
+
"eval_samples_per_second": 31.122,
|
| 104 |
+
"eval_steps_per_second": 0.522,
|
| 105 |
+
"step": 384
|
| 106 |
+
},
|
| 107 |
+
{
|
| 108 |
+
"epoch": 3.5,
|
| 109 |
+
"grad_norm": 1.9307341575622559,
|
| 110 |
+
"learning_rate": 2.197084758065653e-05,
|
| 111 |
+
"loss": 0.6082,
|
| 112 |
+
"step": 448
|
| 113 |
+
},
|
| 114 |
+
{
|
| 115 |
+
"epoch": 3.5,
|
| 116 |
+
"eval_accuracy": 0.8389261744966443,
|
| 117 |
+
"eval_loss": 0.542212188243866,
|
| 118 |
+
"eval_runtime": 9.8282,
|
| 119 |
+
"eval_samples_per_second": 30.321,
|
| 120 |
+
"eval_steps_per_second": 0.509,
|
| 121 |
+
"step": 448
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 4.0,
|
| 125 |
+
"grad_norm": 2.4017300605773926,
|
| 126 |
+
"learning_rate": 1.6892360241408803e-05,
|
| 127 |
+
"loss": 0.5313,
|
| 128 |
+
"step": 512
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 4.0,
|
| 132 |
+
"eval_accuracy": 0.8523489932885906,
|
| 133 |
+
"eval_loss": 0.494586706161499,
|
| 134 |
+
"eval_runtime": 9.5005,
|
| 135 |
+
"eval_samples_per_second": 31.367,
|
| 136 |
+
"eval_steps_per_second": 0.526,
|
| 137 |
+
"step": 512
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 4.5,
|
| 141 |
+
"grad_norm": 1.9675699472427368,
|
| 142 |
+
"learning_rate": 1.149281235767183e-05,
|
| 143 |
+
"loss": 0.4623,
|
| 144 |
+
"step": 576
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 4.5,
|
| 148 |
+
"eval_accuracy": 0.8758389261744967,
|
| 149 |
+
"eval_loss": 0.4512217044830322,
|
| 150 |
+
"eval_runtime": 9.6531,
|
| 151 |
+
"eval_samples_per_second": 30.871,
|
| 152 |
+
"eval_steps_per_second": 0.518,
|
| 153 |
+
"step": 576
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 5.0,
|
| 157 |
+
"grad_norm": 2.900230884552002,
|
| 158 |
+
"learning_rate": 6.549084935146384e-06,
|
| 159 |
+
"loss": 0.4212,
|
| 160 |
+
"step": 640
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 5.0,
|
| 164 |
+
"eval_accuracy": 0.8791946308724832,
|
| 165 |
+
"eval_loss": 0.4321761131286621,
|
| 166 |
+
"eval_runtime": 9.537,
|
| 167 |
+
"eval_samples_per_second": 31.247,
|
| 168 |
+
"eval_steps_per_second": 0.524,
|
| 169 |
+
"step": 640
|
| 170 |
+
},
|
| 171 |
+
{
|
| 172 |
+
"epoch": 5.5,
|
| 173 |
+
"grad_norm": 2.782975196838379,
|
| 174 |
+
"learning_rate": 2.7037019399421785e-06,
|
| 175 |
+
"loss": 0.4025,
|
| 176 |
+
"step": 704
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 5.5,
|
| 180 |
+
"eval_accuracy": 0.889261744966443,
|
| 181 |
+
"eval_loss": 0.42589595913887024,
|
| 182 |
+
"eval_runtime": 9.6007,
|
| 183 |
+
"eval_samples_per_second": 31.039,
|
| 184 |
+
"eval_steps_per_second": 0.521,
|
| 185 |
+
"step": 704
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 6.0,
|
| 189 |
+
"grad_norm": 2.2286922931671143,
|
| 190 |
+
"learning_rate": 4.564382427431779e-07,
|
| 191 |
+
"loss": 0.3892,
|
| 192 |
+
"step": 768
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 6.0,
|
| 196 |
+
"eval_accuracy": 0.8859060402684564,
|
| 197 |
+
"eval_loss": 0.4237956702709198,
|
| 198 |
+
"eval_runtime": 9.608,
|
| 199 |
+
"eval_samples_per_second": 31.016,
|
| 200 |
+
"eval_steps_per_second": 0.52,
|
| 201 |
+
"step": 768
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 6.5,
|
| 205 |
+
"grad_norm": 2.1533002853393555,
|
| 206 |
+
"learning_rate": 2.9900634879009304e-05,
|
| 207 |
+
"loss": 0.3959,
|
| 208 |
+
"step": 832
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 6.5,
|
| 212 |
+
"eval_accuracy": 0.8859060402684564,
|
| 213 |
+
"eval_loss": 0.40830349922180176,
|
| 214 |
+
"eval_runtime": 9.7273,
|
| 215 |
+
"eval_samples_per_second": 30.635,
|
| 216 |
+
"eval_steps_per_second": 0.514,
|
| 217 |
+
"step": 832
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 7.0,
|
| 221 |
+
"grad_norm": 3.03064227104187,
|
| 222 |
+
"learning_rate": 2.8321109519140222e-05,
|
| 223 |
+
"loss": 0.3279,
|
| 224 |
+
"step": 896
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 7.0,
|
| 228 |
+
"eval_accuracy": 0.8825503355704698,
|
| 229 |
+
"eval_loss": 0.37496834993362427,
|
| 230 |
+
"eval_runtime": 9.1243,
|
| 231 |
+
"eval_samples_per_second": 32.66,
|
| 232 |
+
"eval_steps_per_second": 0.548,
|
| 233 |
+
"step": 896
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 7.5,
|
| 237 |
+
"grad_norm": 2.8390941619873047,
|
| 238 |
+
"learning_rate": 2.50102726629655e-05,
|
| 239 |
+
"loss": 0.2793,
|
| 240 |
+
"step": 960
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 7.5,
|
| 244 |
+
"eval_accuracy": 0.8993288590604027,
|
| 245 |
+
"eval_loss": 0.335045725107193,
|
| 246 |
+
"eval_runtime": 8.9197,
|
| 247 |
+
"eval_samples_per_second": 33.409,
|
| 248 |
+
"eval_steps_per_second": 0.561,
|
| 249 |
+
"step": 960
|
| 250 |
+
},
|
| 251 |
+
{
|
| 252 |
+
"epoch": 8.0,
|
| 253 |
+
"grad_norm": 2.106311321258545,
|
| 254 |
+
"learning_rate": 2.0398425548024824e-05,
|
| 255 |
+
"loss": 0.222,
|
| 256 |
+
"step": 1024
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 8.0,
|
| 260 |
+
"eval_accuracy": 0.8959731543624161,
|
| 261 |
+
"eval_loss": 0.3207705318927765,
|
| 262 |
+
"eval_runtime": 8.4413,
|
| 263 |
+
"eval_samples_per_second": 35.303,
|
| 264 |
+
"eval_steps_per_second": 0.592,
|
| 265 |
+
"step": 1024
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 8.5,
|
| 269 |
+
"grad_norm": 2.5288846492767334,
|
| 270 |
+
"learning_rate": 1.5084958481696745e-05,
|
| 271 |
+
"loss": 0.1862,
|
| 272 |
+
"step": 1088
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 8.5,
|
| 276 |
+
"eval_accuracy": 0.8993288590604027,
|
| 277 |
+
"eval_loss": 0.31276702880859375,
|
| 278 |
+
"eval_runtime": 8.8019,
|
| 279 |
+
"eval_samples_per_second": 33.856,
|
| 280 |
+
"eval_steps_per_second": 0.568,
|
| 281 |
+
"step": 1088
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 9.0,
|
| 285 |
+
"grad_norm": 2.468576431274414,
|
| 286 |
+
"learning_rate": 9.76044957265588e-06,
|
| 287 |
+
"loss": 0.1717,
|
| 288 |
+
"step": 1152
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 9.0,
|
| 292 |
+
"eval_accuracy": 0.9026845637583892,
|
| 293 |
+
"eval_loss": 0.3048604130744934,
|
| 294 |
+
"eval_runtime": 9.6883,
|
| 295 |
+
"eval_samples_per_second": 30.759,
|
| 296 |
+
"eval_steps_per_second": 0.516,
|
| 297 |
+
"step": 1152
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 9.5,
|
| 301 |
+
"grad_norm": 4.155700206756592,
|
| 302 |
+
"learning_rate": 5.116912010431121e-06,
|
| 303 |
+
"loss": 0.1408,
|
| 304 |
+
"step": 1216
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 9.5,
|
| 308 |
+
"eval_accuracy": 0.9026845637583892,
|
| 309 |
+
"eval_loss": 0.3009721338748932,
|
| 310 |
+
"eval_runtime": 9.5695,
|
| 311 |
+
"eval_samples_per_second": 31.141,
|
| 312 |
+
"eval_steps_per_second": 0.522,
|
| 313 |
+
"step": 1216
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 10.0,
|
| 317 |
+
"grad_norm": 2.336121082305908,
|
| 318 |
+
"learning_rate": 1.757854831072046e-06,
|
| 319 |
+
"loss": 0.1507,
|
| 320 |
+
"step": 1280
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 10.0,
|
| 324 |
+
"eval_accuracy": 0.9161073825503355,
|
| 325 |
+
"eval_loss": 0.32400238513946533,
|
| 326 |
+
"eval_runtime": 9.2278,
|
| 327 |
+
"eval_samples_per_second": 32.294,
|
| 328 |
+
"eval_steps_per_second": 0.542,
|
| 329 |
+
"step": 1280
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 10.5,
|
| 333 |
+
"grad_norm": 2.2721362113952637,
|
| 334 |
+
"learning_rate": 1.198463476772116e-07,
|
| 335 |
+
"loss": 0.1369,
|
| 336 |
+
"step": 1344
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 10.5,
|
| 340 |
+
"eval_accuracy": 0.9060402684563759,
|
| 341 |
+
"eval_loss": 0.3062884509563446,
|
| 342 |
+
"eval_runtime": 8.635,
|
| 343 |
+
"eval_samples_per_second": 34.511,
|
| 344 |
+
"eval_steps_per_second": 0.579,
|
| 345 |
+
"step": 1344
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 11.0,
|
| 349 |
+
"grad_norm": 2.3275115489959717,
|
| 350 |
+
"learning_rate": 2.9584225548225062e-05,
|
| 351 |
+
"loss": 0.1389,
|
| 352 |
+
"step": 1408
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 11.0,
|
| 356 |
+
"eval_accuracy": 0.9060402684563759,
|
| 357 |
+
"eval_loss": 0.3044714033603668,
|
| 358 |
+
"eval_runtime": 8.4938,
|
| 359 |
+
"eval_samples_per_second": 35.084,
|
| 360 |
+
"eval_steps_per_second": 0.589,
|
| 361 |
+
"step": 1408
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 11.5,
|
| 365 |
+
"grad_norm": 2.064596652984619,
|
| 366 |
+
"learning_rate": 2.7392821896833267e-05,
|
| 367 |
+
"loss": 0.1199,
|
| 368 |
+
"step": 1472
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 11.5,
|
| 372 |
+
"eval_accuracy": 0.9093959731543624,
|
| 373 |
+
"eval_loss": 0.30615922808647156,
|
| 374 |
+
"eval_runtime": 9.5329,
|
| 375 |
+
"eval_samples_per_second": 31.26,
|
| 376 |
+
"eval_steps_per_second": 0.524,
|
| 377 |
+
"step": 1472
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 12.0,
|
| 381 |
+
"grad_norm": 2.055306911468506,
|
| 382 |
+
"learning_rate": 2.359075398426809e-05,
|
| 383 |
+
"loss": 0.1003,
|
| 384 |
+
"step": 1536
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 12.0,
|
| 388 |
+
"eval_accuracy": 0.912751677852349,
|
| 389 |
+
"eval_loss": 0.3130672872066498,
|
| 390 |
+
"eval_runtime": 8.6515,
|
| 391 |
+
"eval_samples_per_second": 34.445,
|
| 392 |
+
"eval_steps_per_second": 0.578,
|
| 393 |
+
"step": 1536
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 12.5,
|
| 397 |
+
"grad_norm": 3.542067289352417,
|
| 398 |
+
"learning_rate": 1.867216712757697e-05,
|
| 399 |
+
"loss": 0.0756,
|
| 400 |
+
"step": 1600
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 12.5,
|
| 404 |
+
"eval_accuracy": 0.9228187919463087,
|
| 405 |
+
"eval_loss": 0.3002491891384125,
|
| 406 |
+
"eval_runtime": 8.8978,
|
| 407 |
+
"eval_samples_per_second": 33.491,
|
| 408 |
+
"eval_steps_per_second": 0.562,
|
| 409 |
+
"step": 1600
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 13.0,
|
| 413 |
+
"grad_norm": 1.0521228313446045,
|
| 414 |
+
"learning_rate": 1.3276317836011714e-05,
|
| 415 |
+
"loss": 0.0636,
|
| 416 |
+
"step": 1664
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 13.0,
|
| 420 |
+
"eval_accuracy": 0.912751677852349,
|
| 421 |
+
"eval_loss": 0.3176516890525818,
|
| 422 |
+
"eval_runtime": 9.2104,
|
| 423 |
+
"eval_samples_per_second": 32.355,
|
| 424 |
+
"eval_steps_per_second": 0.543,
|
| 425 |
+
"step": 1664
|
| 426 |
+
},
|
| 427 |
+
{
|
| 428 |
+
"epoch": 13.5,
|
| 429 |
+
"grad_norm": 1.8493869304656982,
|
| 430 |
+
"learning_rate": 8.104491231310648e-06,
|
| 431 |
+
"loss": 0.058,
|
| 432 |
+
"step": 1728
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 13.5,
|
| 436 |
+
"eval_accuracy": 0.9228187919463087,
|
| 437 |
+
"eval_loss": 0.3142910599708557,
|
| 438 |
+
"eval_runtime": 9.7948,
|
| 439 |
+
"eval_samples_per_second": 30.424,
|
| 440 |
+
"eval_steps_per_second": 0.51,
|
| 441 |
+
"step": 1728
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 14.0,
|
| 445 |
+
"grad_norm": 1.3355131149291992,
|
| 446 |
+
"learning_rate": 3.828856763379785e-06,
|
| 447 |
+
"loss": 0.0566,
|
| 448 |
+
"step": 1792
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 14.0,
|
| 452 |
+
"eval_accuracy": 0.9194630872483222,
|
| 453 |
+
"eval_loss": 0.3135838806629181,
|
| 454 |
+
"eval_runtime": 9.833,
|
| 455 |
+
"eval_samples_per_second": 30.306,
|
| 456 |
+
"eval_steps_per_second": 0.508,
|
| 457 |
+
"step": 1792
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 14.5,
|
| 461 |
+
"grad_norm": 2.9695112705230713,
|
| 462 |
+
"learning_rate": 1.0051080174789174e-06,
|
| 463 |
+
"loss": 0.0516,
|
| 464 |
+
"step": 1856
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 14.5,
|
| 468 |
+
"eval_accuracy": 0.9161073825503355,
|
| 469 |
+
"eval_loss": 0.34468716382980347,
|
| 470 |
+
"eval_runtime": 9.6587,
|
| 471 |
+
"eval_samples_per_second": 30.853,
|
| 472 |
+
"eval_steps_per_second": 0.518,
|
| 473 |
+
"step": 1856
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 15.0,
|
| 477 |
+
"grad_norm": 0.7186270952224731,
|
| 478 |
+
"learning_rate": 2.4060005001991235e-10,
|
| 479 |
+
"loss": 0.0426,
|
| 480 |
+
"step": 1920
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 15.0,
|
| 484 |
+
"eval_accuracy": 0.9228187919463087,
|
| 485 |
+
"eval_loss": 0.2911371886730194,
|
| 486 |
+
"eval_runtime": 9.7016,
|
| 487 |
+
"eval_samples_per_second": 30.717,
|
| 488 |
+
"eval_steps_per_second": 0.515,
|
| 489 |
+
"step": 1920
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 15.0,
|
| 493 |
+
"step": 1920,
|
| 494 |
+
"total_flos": 7.164784986292224e+17,
|
| 495 |
+
"train_loss": 0.48338687382638457,
|
| 496 |
+
"train_runtime": 5148.0361,
|
| 497 |
+
"train_samples_per_second": 23.869,
|
| 498 |
+
"train_steps_per_second": 0.373
|
| 499 |
+
}
|
| 500 |
+
],
|
| 501 |
+
"logging_steps": 64,
|
| 502 |
+
"max_steps": 1920,
|
| 503 |
+
"num_input_tokens_seen": 0,
|
| 504 |
+
"num_train_epochs": 15,
|
| 505 |
+
"save_steps": 64,
|
| 506 |
+
"stateful_callbacks": {
|
| 507 |
+
"TrainerControl": {
|
| 508 |
+
"args": {
|
| 509 |
+
"should_epoch_stop": false,
|
| 510 |
+
"should_evaluate": false,
|
| 511 |
+
"should_log": false,
|
| 512 |
+
"should_save": true,
|
| 513 |
+
"should_training_stop": true
|
| 514 |
+
},
|
| 515 |
+
"attributes": {}
|
| 516 |
+
}
|
| 517 |
+
},
|
| 518 |
+
"total_flos": 7.164784986292224e+17,
|
| 519 |
+
"train_batch_size": 64,
|
| 520 |
+
"trial_name": null,
|
| 521 |
+
"trial_params": null
|
| 522 |
+
}
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5496
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a65bff28c03c6c9e01c24d73c794599da24b606df0220c01aac2771132c075ec
|
| 3 |
size 5496
|
training_metrics.xlsx
CHANGED
|
Binary files a/training_metrics.xlsx and b/training_metrics.xlsx differ
|
|
|