update model card README.md
Browse files
README.md
CHANGED
@@ -14,8 +14,8 @@ should probably proofread and complete it, then remove this comment. -->
|
|
14 |
|
15 |
This model is a fine-tuned version of [microsoft/deberta-v3-small](https://huggingface.co/microsoft/deberta-v3-small) on the None dataset.
|
16 |
It achieves the following results on the evaluation set:
|
17 |
-
- Loss: 0.
|
18 |
-
- Pearson: 0.
|
19 |
|
20 |
## Model description
|
21 |
|
@@ -34,35 +34,30 @@ More information needed
|
|
34 |
### Training hyperparameters
|
35 |
|
36 |
The following hyperparameters were used during training:
|
37 |
-
- learning_rate:
|
38 |
- train_batch_size: 128
|
39 |
- eval_batch_size: 256
|
40 |
- seed: 42
|
41 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
42 |
- lr_scheduler_type: cosine
|
43 |
- lr_scheduler_warmup_ratio: 0.1
|
44 |
-
- num_epochs:
|
45 |
- mixed_precision_training: Native AMP
|
46 |
|
47 |
### Training results
|
48 |
|
49 |
| Training Loss | Epoch | Step | Validation Loss | Pearson |
|
50 |
|:-------------:|:-----:|:----:|:---------------:|:-------:|
|
51 |
-
| No log | 1.0 | 12 | 0.
|
52 |
-
| No log | 2.0 | 24 | 0.
|
53 |
-
| No log | 3.0 | 36 | 0.
|
54 |
-
| No log | 4.0 | 48 | 0.
|
55 |
-
| No log | 5.0 | 60 | 0.
|
56 |
-
| No log | 6.0 | 72 | 0.
|
57 |
-
| No log | 7.0 | 84 | 0.
|
58 |
-
| No log | 8.0 | 96 | 0.
|
59 |
-
| No log | 9.0 | 108 | 0.
|
60 |
-
| No log | 10.0 | 120 | 0.
|
61 |
-
| No log | 11.0 | 132 | 0.0466 | 0.9230 |
|
62 |
-
| No log | 12.0 | 144 | 0.0491 | 0.9198 |
|
63 |
-
| No log | 13.0 | 156 | 0.0495 | 0.9196 |
|
64 |
-
| No log | 14.0 | 168 | 0.0499 | 0.9176 |
|
65 |
-
| No log | 15.0 | 180 | 0.0501 | 0.9174 |
|
66 |
|
67 |
|
68 |
### Framework versions
|
|
|
14 |
|
15 |
This model is a fine-tuned version of [microsoft/deberta-v3-small](https://huggingface.co/microsoft/deberta-v3-small) on the None dataset.
|
16 |
It achieves the following results on the evaluation set:
|
17 |
+
- Loss: 0.0475
|
18 |
+
- Pearson: 0.9261
|
19 |
|
20 |
## Model description
|
21 |
|
|
|
34 |
### Training hyperparameters
|
35 |
|
36 |
The following hyperparameters were used during training:
|
37 |
+
- learning_rate: 8e-05
|
38 |
- train_batch_size: 128
|
39 |
- eval_batch_size: 256
|
40 |
- seed: 42
|
41 |
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
42 |
- lr_scheduler_type: cosine
|
43 |
- lr_scheduler_warmup_ratio: 0.1
|
44 |
+
- num_epochs: 10
|
45 |
- mixed_precision_training: Native AMP
|
46 |
|
47 |
### Training results
|
48 |
|
49 |
| Training Loss | Epoch | Step | Validation Loss | Pearson |
|
50 |
|:-------------:|:-----:|:----:|:---------------:|:-------:|
|
51 |
+
| No log | 1.0 | 12 | 0.2617 | 0.6890 |
|
52 |
+
| No log | 2.0 | 24 | 0.2215 | 0.7383 |
|
53 |
+
| No log | 3.0 | 36 | 0.0794 | 0.8488 |
|
54 |
+
| No log | 4.0 | 48 | 0.0832 | 0.8626 |
|
55 |
+
| No log | 5.0 | 60 | 0.0753 | 0.9134 |
|
56 |
+
| No log | 6.0 | 72 | 0.0497 | 0.9271 |
|
57 |
+
| No log | 7.0 | 84 | 0.0454 | 0.9253 |
|
58 |
+
| No log | 8.0 | 96 | 0.0438 | 0.9287 |
|
59 |
+
| No log | 9.0 | 108 | 0.0496 | 0.9251 |
|
60 |
+
| No log | 10.0 | 120 | 0.0475 | 0.9261 |
|
|
|
|
|
|
|
|
|
|
|
61 |
|
62 |
|
63 |
### Framework versions
|