update model card README.md
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 [albert-base-v2](https://huggingface.co/albert-base-v2) on the squad dataset.
|
| 18 |
It achieves the following results on the evaluation set:
|
| 19 |
-
- Loss:
|
| 20 |
|
| 21 |
## Model description
|
| 22 |
|
|
@@ -41,18 +41,21 @@ 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 |
|
| 46 |
### Training results
|
| 47 |
|
| 48 |
-
| Training Loss | Epoch | Step
|
| 49 |
-
|
| 50 |
-
| 0.
|
|
|
|
|
|
|
|
|
|
| 51 |
|
| 52 |
|
| 53 |
### Framework versions
|
| 54 |
|
| 55 |
-
- Transformers 4.
|
| 56 |
- Pytorch 1.10.0+cu111
|
| 57 |
- Datasets 1.16.1
|
| 58 |
- Tokenizers 0.10.3
|
|
|
|
| 16 |
|
| 17 |
This model is a fine-tuned version of [albert-base-v2](https://huggingface.co/albert-base-v2) on the squad dataset.
|
| 18 |
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 1.1607
|
| 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: 4
|
| 45 |
|
| 46 |
### Training results
|
| 47 |
|
| 48 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 49 |
+
|:-------------:|:-----:|:-----:|:---------------:|
|
| 50 |
+
| 0.8695 | 1.0 | 5540 | 0.9092 |
|
| 51 |
+
| 0.6594 | 2.0 | 11080 | 0.9148 |
|
| 52 |
+
| 0.5053 | 3.0 | 16620 | 0.9641 |
|
| 53 |
+
| 0.3477 | 4.0 | 22160 | 1.1607 |
|
| 54 |
|
| 55 |
|
| 56 |
### Framework versions
|
| 57 |
|
| 58 |
+
- Transformers 4.13.0
|
| 59 |
- Pytorch 1.10.0+cu111
|
| 60 |
- Datasets 1.16.1
|
| 61 |
- Tokenizers 0.10.3
|