<!-- 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. -->
Erlangshen-Longformer-110M-finetuning-wallstreetcn-morning-news-vix-sz50-v2
This model is a fine-tuned version of IDEA-CCNL/Erlangshen-Longformer-110M on the dataset of Wallstreetcn Morning News(Sample link: https://wallstreetcn.com/articles/3691904?keyword=%E8%A7%81%E9%97%BB%E6%97%A9%E9%A4%90) to predict VIX of SZ50 up or down. It achieves the following results on the evaluation set:
- Loss: 3.5413
- Accuracy: 0.5577
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: 5e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 20
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| No log | 1.0 | 19 | 0.6847 | 0.5962 |
| No log | 2.0 | 38 | 0.9958 | 0.5385 |
| No log | 3.0 | 57 | 1.0758 | 0.5962 |
| No log | 4.0 | 76 | 1.6791 | 0.5769 |
| No log | 5.0 | 95 | 2.3408 | 0.5385 |
| No log | 6.0 | 114 | 2.4684 | 0.5385 |
| No log | 7.0 | 133 | 2.8436 | 0.5385 |
| No log | 8.0 | 152 | 2.5892 | 0.5385 |
| No log | 9.0 | 171 | 2.5652 | 0.5 |
| No log | 10.0 | 190 | 3.0316 | 0.5385 |
| No log | 11.0 | 209 | 2.9871 | 0.5577 |
| No log | 12.0 | 228 | 3.4473 | 0.5385 |
| No log | 13.0 | 247 | 3.3994 | 0.5192 |
| No log | 14.0 | 266 | 3.4179 | 0.5 |
| No log | 15.0 | 285 | 3.4592 | 0.5192 |
| No log | 16.0 | 304 | 3.4951 | 0.5192 |
| No log | 17.0 | 323 | 3.5134 | 0.5577 |
| No log | 18.0 | 342 | 3.5289 | 0.5577 |
| No log | 19.0 | 361 | 3.5380 | 0.5577 |
| No log | 20.0 | 380 | 3.5413 | 0.5577 |
Framework versions
- Transformers 4.30.2
- Pytorch 2.0.1+cu118
- Datasets 2.13.1
- Tokenizers 0.13.3
