update model card README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,79 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- generated_from_trainer
|
| 5 |
+
metrics:
|
| 6 |
+
- accuracy
|
| 7 |
+
- f1
|
| 8 |
+
- recall
|
| 9 |
+
- precision
|
| 10 |
+
model-index:
|
| 11 |
+
- name: distilbert-undersampled
|
| 12 |
+
results: []
|
| 13 |
+
---
|
| 14 |
+
|
| 15 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 16 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 17 |
+
|
| 18 |
+
# distilbert-undersampled
|
| 19 |
+
|
| 20 |
+
This model is a fine-tuned version of [distilbert-base-uncased](https://huggingface.co/distilbert-base-uncased) on the None dataset.
|
| 21 |
+
It achieves the following results on the evaluation set:
|
| 22 |
+
- Loss: 0.0826
|
| 23 |
+
- Accuracy: 0.9811
|
| 24 |
+
- F1: 0.9810
|
| 25 |
+
- Recall: 0.9811
|
| 26 |
+
- Precision: 0.9812
|
| 27 |
+
|
| 28 |
+
## Model description
|
| 29 |
+
|
| 30 |
+
More information needed
|
| 31 |
+
|
| 32 |
+
## Intended uses & limitations
|
| 33 |
+
|
| 34 |
+
More information needed
|
| 35 |
+
|
| 36 |
+
## Training and evaluation data
|
| 37 |
+
|
| 38 |
+
More information needed
|
| 39 |
+
|
| 40 |
+
## Training procedure
|
| 41 |
+
|
| 42 |
+
### Training hyperparameters
|
| 43 |
+
|
| 44 |
+
The following hyperparameters were used during training:
|
| 45 |
+
- learning_rate: 3e-05
|
| 46 |
+
- train_batch_size: 16
|
| 47 |
+
- eval_batch_size: 16
|
| 48 |
+
- seed: 33
|
| 49 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 50 |
+
- lr_scheduler_type: linear
|
| 51 |
+
- lr_scheduler_warmup_steps: 100
|
| 52 |
+
- num_epochs: 5
|
| 53 |
+
- mixed_precision_training: Native AMP
|
| 54 |
+
|
| 55 |
+
### Training results
|
| 56 |
+
|
| 57 |
+
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Recall | Precision |
|
| 58 |
+
|:-------------:|:-----:|:-----:|:---------------:|:--------:|:------:|:------:|:---------:|
|
| 59 |
+
| 0.0959 | 0.2 | 2000 | 0.0999 | 0.9651 | 0.9628 | 0.9651 | 0.9655 |
|
| 60 |
+
| 0.0618 | 0.41 | 4000 | 0.0886 | 0.9717 | 0.9717 | 0.9717 | 0.9731 |
|
| 61 |
+
| 0.159 | 0.61 | 6000 | 0.0884 | 0.9719 | 0.9720 | 0.9719 | 0.9728 |
|
| 62 |
+
| 0.0513 | 0.81 | 8000 | 0.0785 | 0.9782 | 0.9782 | 0.9782 | 0.9788 |
|
| 63 |
+
| 0.0219 | 1.01 | 10000 | 0.0680 | 0.9779 | 0.9779 | 0.9779 | 0.9783 |
|
| 64 |
+
| 0.036 | 1.22 | 12000 | 0.0745 | 0.9787 | 0.9787 | 0.9787 | 0.9792 |
|
| 65 |
+
| 0.0892 | 1.42 | 14000 | 0.0675 | 0.9786 | 0.9786 | 0.9786 | 0.9789 |
|
| 66 |
+
| 0.0214 | 1.62 | 16000 | 0.0760 | 0.9799 | 0.9798 | 0.9799 | 0.9801 |
|
| 67 |
+
| 0.0882 | 1.83 | 18000 | 0.0800 | 0.9800 | 0.9800 | 0.9800 | 0.9802 |
|
| 68 |
+
| 0.0234 | 2.03 | 20000 | 0.0720 | 0.9813 | 0.9813 | 0.9813 | 0.9815 |
|
| 69 |
+
| 0.0132 | 2.23 | 22000 | 0.0738 | 0.9803 | 0.9803 | 0.9803 | 0.9805 |
|
| 70 |
+
| 0.0136 | 2.43 | 24000 | 0.0847 | 0.9804 | 0.9804 | 0.9804 | 0.9806 |
|
| 71 |
+
| 0.0119 | 2.64 | 26000 | 0.0826 | 0.9811 | 0.9810 | 0.9811 | 0.9812 |
|
| 72 |
+
|
| 73 |
+
|
| 74 |
+
### Framework versions
|
| 75 |
+
|
| 76 |
+
- Transformers 4.16.2
|
| 77 |
+
- Pytorch 1.10.0+cu111
|
| 78 |
+
- Datasets 1.18.3
|
| 79 |
+
- Tokenizers 0.11.0
|