anas-awadalla
commited on
Commit
•
fa060e4
1
Parent(s):
c1063d5
update model card README.md
Browse files
README.md
CHANGED
@@ -15,8 +15,6 @@ should probably proofread and complete it, then remove this comment. -->
|
|
15 |
# bert-tiny-finetuned-squad
|
16 |
|
17 |
This model is a fine-tuned version of [prajjwal1/bert-tiny](https://huggingface.co/prajjwal1/bert-tiny) on the squad dataset.
|
18 |
-
It achieves the following results on the evaluation set:
|
19 |
-
- Loss: 2.3478
|
20 |
|
21 |
## Model description
|
22 |
|
@@ -35,43 +33,22 @@ More information needed
|
|
35 |
### Training hyperparameters
|
36 |
|
37 |
The following hyperparameters were used during training:
|
38 |
-
- learning_rate:
|
39 |
-
- train_batch_size:
|
40 |
-
- eval_batch_size:
|
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 |
|
46 |
### Training results
|
47 |
|
48 |
-
| Training Loss | Epoch | Step | Validation Loss |
|
49 |
-
|:-------------:|:-----:|:-----:|:---------------:|
|
50 |
-
| 2.7198 | 1.0 | 2767 | 2.5916 |
|
51 |
-
| 2.6502 | 2.0 | 5534 | 2.5309 |
|
52 |
-
| 2.5589 | 3.0 | 8301 | 2.4849 |
|
53 |
-
| 2.5128 | 4.0 | 11068 | 2.4474 |
|
54 |
-
| 2.4631 | 5.0 | 13835 | 2.4226 |
|
55 |
-
| 2.4165 | 6.0 | 16602 | 2.3987 |
|
56 |
-
| 2.3634 | 7.0 | 19369 | 2.4002 |
|
57 |
-
| 2.3243 | 8.0 | 22136 | 2.3855 |
|
58 |
-
| 2.2799 | 9.0 | 24903 | 2.3827 |
|
59 |
-
| 2.2596 | 10.0 | 27670 | 2.3641 |
|
60 |
-
| 2.2133 | 11.0 | 30437 | 2.3464 |
|
61 |
-
| 2.1911 | 12.0 | 33204 | 2.3551 |
|
62 |
-
| 2.1621 | 13.0 | 35971 | 2.3553 |
|
63 |
-
| 2.1421 | 14.0 | 38738 | 2.3584 |
|
64 |
-
| 2.1202 | 15.0 | 41505 | 2.3501 |
|
65 |
-
| 2.1154 | 16.0 | 44272 | 2.3548 |
|
66 |
-
| 2.1237 | 17.0 | 47039 | 2.3456 |
|
67 |
-
| 2.0952 | 18.0 | 49806 | 2.3507 |
|
68 |
-
| 2.0886 | 19.0 | 52573 | 2.3500 |
|
69 |
-
| 2.1062 | 20.0 | 55340 | 2.3478 |
|
70 |
|
71 |
|
72 |
### Framework versions
|
73 |
|
74 |
-
- Transformers 4.
|
75 |
-
- Pytorch 1.
|
76 |
-
- Datasets
|
77 |
-
- Tokenizers 0.
|
|
|
15 |
# bert-tiny-finetuned-squad
|
16 |
|
17 |
This model is a fine-tuned version of [prajjwal1/bert-tiny](https://huggingface.co/prajjwal1/bert-tiny) on the squad dataset.
|
|
|
|
|
18 |
|
19 |
## Model description
|
20 |
|
|
|
33 |
### Training hyperparameters
|
34 |
|
35 |
The following hyperparameters were used during training:
|
36 |
+
- learning_rate: 3e-05
|
37 |
+
- train_batch_size: 64
|
38 |
+
- eval_batch_size: 8
|
39 |
- seed: 42
|
40 |
+
- distributed_type: multi-GPU
|
41 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
42 |
- lr_scheduler_type: linear
|
43 |
+
- num_epochs: 2.0
|
44 |
|
45 |
### Training results
|
46 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
47 |
|
48 |
|
49 |
### Framework versions
|
50 |
|
51 |
+
- Transformers 4.17.0
|
52 |
+
- Pytorch 1.11.0+cu113
|
53 |
+
- Datasets 2.0.0
|
54 |
+
- Tokenizers 0.11.6
|