File size: 1,872 Bytes
182734d 1f87b3c 182734d c676bd0 182734d 1f87b3c 182734d 1f87b3c 182734d 1f87b3c 182734d 1f87b3c 182734d 1f87b3c 182734d 1f87b3c 182734d 1f87b3c |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 |
---
license: apache-2.0
base_model: Twitter/twhin-bert-large
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
model-index:
- name: financial-twhin-bert-large-7labels
results: []
---
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->
# financial-twhin-bert-large-7labels
This model is a fine-tuned version of [Twitter/twhin-bert-large](https://huggingface.co/Twitter/twhin-bert-large) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.3040
- Accuracy: 0.8968
- F1: 0.8916
## Model description
More information needed
## Intended uses & limitations
More information needed
## Training and evaluation data
More information needed
## Training procedure
### Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 2.1732582582331977e-05
- train_batch_size: 16
- eval_batch_size: 8
- seed: 1203
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 2
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 |
|:-------------:|:------:|:----:|:---------------:|:--------:|:------:|
| 0.9592 | 0.3272 | 1500 | 0.6466 | 0.7665 | 0.7503 |
| 0.4705 | 0.6545 | 3000 | 0.3785 | 0.8674 | 0.8528 |
| 0.4196 | 0.9817 | 4500 | 0.5830 | 0.7892 | 0.7775 |
| 0.3403 | 1.3089 | 6000 | 0.3683 | 0.8767 | 0.8728 |
| 0.2962 | 1.6361 | 7500 | 0.3288 | 0.8889 | 0.8904 |
| 0.272 | 1.9634 | 9000 | 0.3040 | 0.8968 | 0.8916 |
### Framework versions
- Transformers 4.40.1
- Pytorch 2.2.1+cu121
- Datasets 2.19.0
- Tokenizers 0.19.1
|