|
--- |
|
license: mit |
|
base_model: xlnet/xlnet-base-cased |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- f1 |
|
model-index: |
|
- name: xlnet-base-cased-airlines-news-multi-label-7-actions |
|
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. --> |
|
|
|
# xlnet-base-cased-airlines-news-multi-label-7-actions |
|
|
|
This model is a fine-tuned version of [xlnet/xlnet-base-cased](https://huggingface.co./xlnet/xlnet-base-cased) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.2151 |
|
- F1: 0.9188 |
|
- Jaccard: 0.6316 |
|
|
|
## 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: 0.0001 |
|
- train_batch_size: 32 |
|
- eval_batch_size: 32 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_steps: 150 |
|
- num_epochs: 45 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | F1 | Jaccard | |
|
|:-------------:|:-----:|:----:|:---------------:|:------:|:-------:| |
|
| No log | 1.0 | 57 | 0.4002 | 0.8330 | 0.4071 | |
|
| No log | 2.0 | 114 | 0.3221 | 0.8443 | 0.4115 | |
|
| No log | 3.0 | 171 | 0.2836 | 0.8703 | 0.4565 | |
|
| No log | 4.0 | 228 | 0.2654 | 0.8731 | 0.4786 | |
|
| No log | 5.0 | 285 | 0.2478 | 0.8825 | 0.5288 | |
|
| No log | 6.0 | 342 | 0.2410 | 0.8925 | 0.5395 | |
|
| No log | 7.0 | 399 | 0.2337 | 0.8990 | 0.5830 | |
|
| No log | 8.0 | 456 | 0.2313 | 0.8997 | 0.5586 | |
|
| 0.3113 | 9.0 | 513 | 0.2287 | 0.8974 | 0.5579 | |
|
| 0.3113 | 10.0 | 570 | 0.2242 | 0.9037 | 0.6080 | |
|
| 0.3113 | 11.0 | 627 | 0.2215 | 0.9036 | 0.5911 | |
|
| 0.3113 | 12.0 | 684 | 0.2209 | 0.9072 | 0.5977 | |
|
| 0.3113 | 13.0 | 741 | 0.2192 | 0.9093 | 0.6169 | |
|
| 0.3113 | 14.0 | 798 | 0.2193 | 0.9076 | 0.6125 | |
|
| 0.3113 | 15.0 | 855 | 0.2206 | 0.9067 | 0.6257 | |
|
| 0.3113 | 16.0 | 912 | 0.2166 | 0.9121 | 0.6265 | |
|
| 0.3113 | 17.0 | 969 | 0.2151 | 0.9188 | 0.6316 | |
|
| 0.2282 | 18.0 | 1026 | 0.2143 | 0.9107 | 0.6279 | |
|
| 0.2282 | 19.0 | 1083 | 0.2132 | 0.9105 | 0.6346 | |
|
| 0.2282 | 20.0 | 1140 | 0.2132 | 0.9106 | 0.6279 | |
|
| 0.2282 | 21.0 | 1197 | 0.2174 | 0.9122 | 0.6243 | |
|
| 0.2282 | 22.0 | 1254 | 0.2161 | 0.9139 | 0.6221 | |
|
| 0.2282 | 23.0 | 1311 | 0.2127 | 0.9119 | 0.6287 | |
|
| 0.2282 | 24.0 | 1368 | 0.2129 | 0.9119 | 0.6353 | |
|
| 0.2282 | 25.0 | 1425 | 0.2112 | 0.9147 | 0.6324 | |
|
| 0.2282 | 26.0 | 1482 | 0.2122 | 0.9129 | 0.6386 | |
|
| 0.2142 | 27.0 | 1539 | 0.2126 | 0.9146 | 0.6272 | |
|
| 0.2142 | 28.0 | 1596 | 0.2114 | 0.9144 | 0.6331 | |
|
| 0.2142 | 29.0 | 1653 | 0.2109 | 0.9159 | 0.6434 | |
|
| 0.2142 | 30.0 | 1710 | 0.2102 | 0.9165 | 0.6479 | |
|
| 0.2142 | 31.0 | 1767 | 0.2120 | 0.9114 | 0.6397 | |
|
| 0.2142 | 32.0 | 1824 | 0.2110 | 0.9125 | 0.6405 | |
|
| 0.2142 | 33.0 | 1881 | 0.2111 | 0.9093 | 0.625 | |
|
| 0.2142 | 34.0 | 1938 | 0.2110 | 0.9109 | 0.6243 | |
|
| 0.2142 | 35.0 | 1995 | 0.2111 | 0.9110 | 0.6324 | |
|
| 0.2084 | 36.0 | 2052 | 0.2127 | 0.9141 | 0.6394 | |
|
| 0.2084 | 37.0 | 2109 | 0.2112 | 0.9098 | 0.6265 | |
|
| 0.2084 | 38.0 | 2166 | 0.2109 | 0.9112 | 0.6309 | |
|
| 0.2084 | 39.0 | 2223 | 0.2109 | 0.9137 | 0.6405 | |
|
| 0.2084 | 40.0 | 2280 | 0.2111 | 0.9122 | 0.6346 | |
|
| 0.2084 | 41.0 | 2337 | 0.2111 | 0.9111 | 0.6368 | |
|
| 0.2084 | 42.0 | 2394 | 0.2113 | 0.9099 | 0.6279 | |
|
| 0.2084 | 43.0 | 2451 | 0.2112 | 0.9105 | 0.6324 | |
|
| 0.2054 | 44.0 | 2508 | 0.2111 | 0.9105 | 0.6287 | |
|
| 0.2054 | 45.0 | 2565 | 0.2110 | 0.9089 | 0.6221 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.41.1 |
|
- Pytorch 2.3.0+cu121 |
|
- Datasets 2.19.1 |
|
- Tokenizers 0.19.1 |
|
|
