Commit
·
9839e84
1
Parent(s):
bbc9349
End of training
Browse files
README.md
CHANGED
|
@@ -21,11 +21,11 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 21 |
|
| 22 |
This model is a fine-tuned version of [bayartsogt/mongolian-roberta-base](https://huggingface.co/bayartsogt/mongolian-roberta-base) on the None dataset.
|
| 23 |
It achieves the following results on the evaluation set:
|
| 24 |
-
- Loss: 0.
|
| 25 |
-
- Precision: 0.
|
| 26 |
-
- Recall: 0.
|
| 27 |
-
- F1: 0.
|
| 28 |
-
- Accuracy: 0.
|
| 29 |
|
| 30 |
## Model description
|
| 31 |
|
|
@@ -46,17 +46,26 @@ More information needed
|
|
| 46 |
The following hyperparameters were used during training:
|
| 47 |
- learning_rate: 2e-05
|
| 48 |
- train_batch_size: 16
|
| 49 |
-
- eval_batch_size:
|
| 50 |
- seed: 42
|
| 51 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 52 |
- lr_scheduler_type: linear
|
| 53 |
-
- num_epochs:
|
| 54 |
|
| 55 |
### Training results
|
| 56 |
|
| 57 |
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|
| 58 |
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
|
| 59 |
-
| 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 60 |
|
| 61 |
|
| 62 |
### Framework versions
|
|
|
|
| 21 |
|
| 22 |
This model is a fine-tuned version of [bayartsogt/mongolian-roberta-base](https://huggingface.co/bayartsogt/mongolian-roberta-base) on the None dataset.
|
| 23 |
It achieves the following results on the evaluation set:
|
| 24 |
+
- Loss: 0.1614
|
| 25 |
+
- Precision: 0.9347
|
| 26 |
+
- Recall: 0.9439
|
| 27 |
+
- F1: 0.9393
|
| 28 |
+
- Accuracy: 0.9809
|
| 29 |
|
| 30 |
## Model description
|
| 31 |
|
|
|
|
| 46 |
The following hyperparameters were used during training:
|
| 47 |
- learning_rate: 2e-05
|
| 48 |
- train_batch_size: 16
|
| 49 |
+
- eval_batch_size: 16
|
| 50 |
- seed: 42
|
| 51 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 52 |
- lr_scheduler_type: linear
|
| 53 |
+
- num_epochs: 10
|
| 54 |
|
| 55 |
### Training results
|
| 56 |
|
| 57 |
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|
| 58 |
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
|
| 59 |
+
| 0.003 | 1.0 | 477 | 0.1548 | 0.9130 | 0.9269 | 0.9199 | 0.9762 |
|
| 60 |
+
| 0.0048 | 2.0 | 954 | 0.1259 | 0.9345 | 0.9434 | 0.9389 | 0.9806 |
|
| 61 |
+
| 0.0024 | 3.0 | 1431 | 0.1324 | 0.9291 | 0.9432 | 0.9361 | 0.9806 |
|
| 62 |
+
| 0.0023 | 4.0 | 1908 | 0.1416 | 0.9315 | 0.9431 | 0.9372 | 0.9802 |
|
| 63 |
+
| 0.0012 | 5.0 | 2385 | 0.1466 | 0.9329 | 0.9427 | 0.9378 | 0.9808 |
|
| 64 |
+
| 0.001 | 6.0 | 2862 | 0.1507 | 0.9335 | 0.9428 | 0.9381 | 0.9804 |
|
| 65 |
+
| 0.0007 | 7.0 | 3339 | 0.1558 | 0.9350 | 0.9438 | 0.9394 | 0.9805 |
|
| 66 |
+
| 0.0007 | 8.0 | 3816 | 0.1588 | 0.9355 | 0.9453 | 0.9404 | 0.9804 |
|
| 67 |
+
| 0.0003 | 9.0 | 4293 | 0.1605 | 0.9338 | 0.9429 | 0.9383 | 0.9807 |
|
| 68 |
+
| 0.0003 | 10.0 | 4770 | 0.1614 | 0.9347 | 0.9439 | 0.9393 | 0.9809 |
|
| 69 |
|
| 70 |
|
| 71 |
### Framework versions
|