End of training
Browse files
README.md
CHANGED
|
@@ -16,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [naver-clova-ix/donut-base](https://huggingface.co/naver-clova-ix/donut-base) on the imagefolder dataset.
|
| 18 |
It achieves the following results on the evaluation set:
|
| 19 |
-
- Loss: 0.
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
|
@@ -41,43 +41,18 @@ The following hyperparameters were used during training:
|
|
| 41 |
- seed: 42
|
| 42 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 43 |
- lr_scheduler_type: linear
|
| 44 |
-
- num_epochs:
|
| 45 |
- mixed_precision_training: Native AMP
|
| 46 |
|
| 47 |
### Training results
|
| 48 |
|
| 49 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 50 |
|:-------------:|:-----:|:----:|:---------------:|
|
| 51 |
-
|
|
| 52 |
-
| 0.
|
| 53 |
-
| 0.
|
| 54 |
-
| 0.
|
| 55 |
-
| 0.
|
| 56 |
-
| 0.1035 | 6.0 | 282 | 0.0640 |
|
| 57 |
-
| 0.1703 | 7.0 | 329 | 0.0877 |
|
| 58 |
-
| 0.0324 | 8.0 | 376 | 0.0722 |
|
| 59 |
-
| 0.033 | 9.0 | 423 | 0.0701 |
|
| 60 |
-
| 0.0404 | 10.0 | 470 | 0.0627 |
|
| 61 |
-
| 0.0198 | 11.0 | 517 | 0.0770 |
|
| 62 |
-
| 0.007 | 12.0 | 564 | 0.0730 |
|
| 63 |
-
| 0.0207 | 13.0 | 611 | 0.0885 |
|
| 64 |
-
| 0.0528 | 14.0 | 658 | 0.0898 |
|
| 65 |
-
| 0.0189 | 15.0 | 705 | 0.0588 |
|
| 66 |
-
| 0.0977 | 16.0 | 752 | 0.0563 |
|
| 67 |
-
| 0.0542 | 17.0 | 799 | 0.0647 |
|
| 68 |
-
| 0.0114 | 18.0 | 846 | 0.0674 |
|
| 69 |
-
| 0.0056 | 19.0 | 893 | 0.0726 |
|
| 70 |
-
| 0.005 | 20.0 | 940 | 0.0797 |
|
| 71 |
-
| 0.0005 | 21.0 | 987 | 0.0704 |
|
| 72 |
-
| 0.0023 | 22.0 | 1034 | 0.0864 |
|
| 73 |
-
| 0.0105 | 23.0 | 1081 | 0.0830 |
|
| 74 |
-
| 0.0014 | 24.0 | 1128 | 0.0806 |
|
| 75 |
-
| 0.0006 | 25.0 | 1175 | 0.0826 |
|
| 76 |
-
| 0.0012 | 26.0 | 1222 | 0.0790 |
|
| 77 |
-
| 0.0006 | 27.0 | 1269 | 0.0751 |
|
| 78 |
-
| 0.0022 | 28.0 | 1316 | 0.0786 |
|
| 79 |
-
| 0.0075 | 29.0 | 1363 | 0.0816 |
|
| 80 |
-
| 0.0024 | 30.0 | 1410 | 0.0821 |
|
| 81 |
|
| 82 |
|
| 83 |
### Framework versions
|
|
|
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [naver-clova-ix/donut-base](https://huggingface.co/naver-clova-ix/donut-base) on the imagefolder dataset.
|
| 18 |
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 0.2879
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
|
|
|
| 41 |
- seed: 42
|
| 42 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 43 |
- lr_scheduler_type: linear
|
| 44 |
+
- num_epochs: 5
|
| 45 |
- mixed_precision_training: Native AMP
|
| 46 |
|
| 47 |
### Training results
|
| 48 |
|
| 49 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 50 |
|:-------------:|:-----:|:----:|:---------------:|
|
| 51 |
+
| 3.7815 | 1.0 | 47 | 2.1495 |
|
| 52 |
+
| 0.6386 | 2.0 | 94 | 0.4316 |
|
| 53 |
+
| 0.336 | 3.0 | 141 | 0.2980 |
|
| 54 |
+
| 0.2021 | 4.0 | 188 | 0.2933 |
|
| 55 |
+
| 0.2655 | 5.0 | 235 | 0.2879 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 56 |
|
| 57 |
|
| 58 |
### Framework versions
|
runs/Apr22_07-43-37_a22352671dc3/events.out.tfevents.1682149630.a22352671dc3.510.0
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 sha256:521132e022afdd16fd772d077abd9e16f9bf786f200f19ae183bb6008d31772b
|
| 3 |
+
size 13837
|