update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,73 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
metrics:
|
| 6 |
+
- accuracy
|
| 7 |
+
- f1
|
| 8 |
+
model-index:
|
| 9 |
+
- name: deberta-base-CoLA
|
| 10 |
+
results: []
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 14 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 15 |
+
|
| 16 |
+
# deberta-base-CoLA
|
| 17 |
+
|
| 18 |
+
This model is a fine-tuned version of [microsoft/deberta-base](https://huggingface.co/microsoft/deberta-base) on an unknown dataset.
|
| 19 |
+
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 1.1655
|
| 21 |
+
- Accuracy: 0.8482
|
| 22 |
+
- F1: 0.8961
|
| 23 |
+
- Roc Auc: 0.8987
|
| 24 |
+
- Mcc: 0.6288
|
| 25 |
+
|
| 26 |
+
## Model description
|
| 27 |
+
|
| 28 |
+
More information needed
|
| 29 |
+
|
| 30 |
+
## Intended uses & limitations
|
| 31 |
+
|
| 32 |
+
More information needed
|
| 33 |
+
|
| 34 |
+
## Training and evaluation data
|
| 35 |
+
|
| 36 |
+
More information needed
|
| 37 |
+
|
| 38 |
+
## Training procedure
|
| 39 |
+
|
| 40 |
+
### Training hyperparameters
|
| 41 |
+
|
| 42 |
+
The following hyperparameters were used during training:
|
| 43 |
+
- learning_rate: 5e-05
|
| 44 |
+
- train_batch_size: 16
|
| 45 |
+
- eval_batch_size: 16
|
| 46 |
+
- seed: 42
|
| 47 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 48 |
+
- lr_scheduler_type: linear
|
| 49 |
+
- lr_scheduler_warmup_ratio: 0.05
|
| 50 |
+
- num_epochs: 10
|
| 51 |
+
|
| 52 |
+
### Training results
|
| 53 |
+
|
| 54 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Roc Auc | Mcc |
|
| 55 |
+
|:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:-------:|:------:|
|
| 56 |
+
| 0.5266 | 1.0 | 535 | 0.4138 | 0.8159 | 0.8698 | 0.8627 | 0.5576 |
|
| 57 |
+
| 0.3523 | 2.0 | 1070 | 0.3852 | 0.8387 | 0.8880 | 0.9041 | 0.6070 |
|
| 58 |
+
| 0.2479 | 3.0 | 1605 | 0.3981 | 0.8482 | 0.8901 | 0.9120 | 0.6447 |
|
| 59 |
+
| 0.1712 | 4.0 | 2140 | 0.4732 | 0.8558 | 0.9008 | 0.9160 | 0.6486 |
|
| 60 |
+
| 0.1354 | 5.0 | 2675 | 0.7181 | 0.8463 | 0.8938 | 0.9024 | 0.6250 |
|
| 61 |
+
| 0.0876 | 6.0 | 3210 | 0.8453 | 0.8520 | 0.8992 | 0.9123 | 0.6385 |
|
| 62 |
+
| 0.0682 | 7.0 | 3745 | 1.0282 | 0.8444 | 0.8938 | 0.9061 | 0.6189 |
|
| 63 |
+
| 0.0431 | 8.0 | 4280 | 1.1114 | 0.8463 | 0.8960 | 0.9010 | 0.6239 |
|
| 64 |
+
| 0.0323 | 9.0 | 4815 | 1.1663 | 0.8501 | 0.8970 | 0.8967 | 0.6340 |
|
| 65 |
+
| 0.0163 | 10.0 | 5350 | 1.1655 | 0.8482 | 0.8961 | 0.8987 | 0.6288 |
|
| 66 |
+
|
| 67 |
+
|
| 68 |
+
### Framework versions
|
| 69 |
+
|
| 70 |
+
- Transformers 4.11.0
|
| 71 |
+
- Pytorch 1.9.0+cu102
|
| 72 |
+
- Datasets 1.12.1
|
| 73 |
+
- Tokenizers 0.10.3
|