sunfu-chou
commited on
Commit
•
27c6fbd
1
Parent(s):
1a6c71b
End of training
Browse files- README.md +31 -17
- model.safetensors +1 -1
README.md
CHANGED
@@ -15,8 +15,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
15 |
|
16 |
This model is a fine-tuned version of [microsoft/git-base](https://huggingface.co/microsoft/git-base) on an unknown dataset.
|
17 |
It achieves the following results on the evaluation set:
|
18 |
-
- Loss: 0.
|
19 |
-
- Wer Score:
|
20 |
|
21 |
## Model description
|
22 |
|
@@ -36,11 +36,11 @@ More information needed
|
|
36 |
|
37 |
The following hyperparameters were used during training:
|
38 |
- learning_rate: 5e-05
|
39 |
-
- train_batch_size:
|
40 |
- eval_batch_size: 2
|
41 |
- seed: 42
|
42 |
- gradient_accumulation_steps: 2
|
43 |
-
- total_train_batch_size:
|
44 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
45 |
- lr_scheduler_type: linear
|
46 |
- num_epochs: 50
|
@@ -50,19 +50,33 @@ The following hyperparameters were used during training:
|
|
50 |
|
51 |
| Training Loss | Epoch | Step | Validation Loss | Wer Score |
|
52 |
|:-------------:|:-------:|:----:|:---------------:|:---------:|
|
53 |
-
| 7.
|
54 |
-
| 2.
|
55 |
-
| 0.
|
56 |
-
| 0.
|
57 |
-
| 0.
|
58 |
-
| 0.
|
59 |
-
| 0.
|
60 |
-
| 0.
|
61 |
-
| 0.
|
62 |
-
| 0.
|
63 |
-
| 0.
|
64 |
-
| 0.
|
65 |
-
| 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
66 |
|
67 |
|
68 |
### Framework versions
|
|
|
15 |
|
16 |
This model is a fine-tuned version of [microsoft/git-base](https://huggingface.co/microsoft/git-base) on an unknown dataset.
|
17 |
It achieves the following results on the evaluation set:
|
18 |
+
- Loss: 0.0495
|
19 |
+
- Wer Score: 4.7488
|
20 |
|
21 |
## Model description
|
22 |
|
|
|
36 |
|
37 |
The following hyperparameters were used during training:
|
38 |
- learning_rate: 5e-05
|
39 |
+
- train_batch_size: 4
|
40 |
- eval_batch_size: 2
|
41 |
- seed: 42
|
42 |
- gradient_accumulation_steps: 2
|
43 |
+
- total_train_batch_size: 8
|
44 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
45 |
- lr_scheduler_type: linear
|
46 |
- num_epochs: 50
|
|
|
50 |
|
51 |
| Training Loss | Epoch | Step | Validation Loss | Wer Score |
|
52 |
|:-------------:|:-------:|:----:|:---------------:|:---------:|
|
53 |
+
| 7.2487 | 1.8182 | 50 | 4.3718 | 22.0605 |
|
54 |
+
| 2.0953 | 3.6364 | 100 | 0.2876 | 4.0186 |
|
55 |
+
| 0.0846 | 5.4545 | 150 | 0.0417 | 0.4419 |
|
56 |
+
| 0.023 | 7.2727 | 200 | 0.0380 | 0.4233 |
|
57 |
+
| 0.018 | 9.0909 | 250 | 0.0369 | 0.4186 |
|
58 |
+
| 0.0144 | 10.9091 | 300 | 0.0393 | 3.0093 |
|
59 |
+
| 0.0116 | 12.7273 | 350 | 0.0407 | 6.9628 |
|
60 |
+
| 0.0087 | 14.5455 | 400 | 0.0406 | 3.5209 |
|
61 |
+
| 0.0062 | 16.3636 | 450 | 0.0423 | 14.7023 |
|
62 |
+
| 0.0034 | 18.1818 | 500 | 0.0429 | 9.0372 |
|
63 |
+
| 0.0024 | 20.0 | 550 | 0.0471 | 8.3442 |
|
64 |
+
| 0.0013 | 21.8182 | 600 | 0.0469 | 13.5907 |
|
65 |
+
| 0.0009 | 23.6364 | 650 | 0.0464 | 14.6186 |
|
66 |
+
| 0.0005 | 25.4545 | 700 | 0.0468 | 11.1674 |
|
67 |
+
| 0.0004 | 27.2727 | 750 | 0.0476 | 7.9907 |
|
68 |
+
| 0.0003 | 29.0909 | 800 | 0.0480 | 7.3070 |
|
69 |
+
| 0.0003 | 30.9091 | 850 | 0.0480 | 7.2140 |
|
70 |
+
| 0.0003 | 32.7273 | 900 | 0.0484 | 6.9628 |
|
71 |
+
| 0.0003 | 34.5455 | 950 | 0.0487 | 6.8512 |
|
72 |
+
| 0.0003 | 36.3636 | 1000 | 0.0489 | 6.0698 |
|
73 |
+
| 0.0003 | 38.1818 | 1050 | 0.0491 | 5.4837 |
|
74 |
+
| 0.0003 | 40.0 | 1100 | 0.0492 | 4.9256 |
|
75 |
+
| 0.0002 | 41.8182 | 1150 | 0.0493 | 4.7860 |
|
76 |
+
| 0.0002 | 43.6364 | 1200 | 0.0493 | 4.8140 |
|
77 |
+
| 0.0002 | 45.4545 | 1250 | 0.0494 | 4.8 |
|
78 |
+
| 0.0002 | 47.2727 | 1300 | 0.0495 | 4.7581 |
|
79 |
+
| 0.0002 | 49.0909 | 1350 | 0.0495 | 4.7488 |
|
80 |
|
81 |
|
82 |
### Framework versions
|
model.safetensors
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 706516040
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a1b671164e2bd941a53638004974d052182327625ae9a94b38e72ee1b5fa1a7d
|
3 |
size 706516040
|