<!-- 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. -->
combine-40-vsfc-xlm-r
This model is a fine-tuned version of xlm-roberta-base on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2145
- Precision: 0.9368
- Recall: 0.9406
- F1 Weighted: 0.9376
- F1 Macro: 0.8232
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: 64
- eval_batch_size: 128
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 100
- num_epochs: 5
Training results
Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 Weighted | F1 Macro |
---|---|---|---|---|---|---|---|
0.9949 | 0.1 | 25 | 0.8155 | 0.6773 | 0.5610 | 0.4515 | 0.3048 |
0.7063 | 0.21 | 50 | 0.4217 | 0.8504 | 0.8787 | 0.8591 | 0.5998 |
0.5545 | 0.31 | 75 | 0.4104 | 0.8674 | 0.9084 | 0.8870 | 0.6198 |
0.5285 | 0.41 | 100 | 0.3414 | 0.8648 | 0.9065 | 0.8852 | 0.6182 |
0.4649 | 0.51 | 125 | 0.2983 | 0.8755 | 0.9172 | 0.8957 | 0.6257 |
0.4305 | 0.62 | 150 | 0.3098 | 0.8763 | 0.9185 | 0.8969 | 0.6266 |
0.3806 | 0.72 | 175 | 0.2825 | 0.8780 | 0.9204 | 0.8987 | 0.6280 |
0.3606 | 0.82 | 200 | 0.2854 | 0.8804 | 0.9229 | 0.9012 | 0.6299 |
0.3633 | 0.93 | 225 | 0.2515 | 0.9321 | 0.9286 | 0.9085 | 0.6509 |
0.2473 | 1.03 | 250 | 0.2584 | 0.9277 | 0.9343 | 0.9265 | 0.7655 |
0.3906 | 1.13 | 275 | 0.2194 | 0.9264 | 0.9318 | 0.9275 | 0.7923 |
0.3153 | 1.23 | 300 | 0.2296 | 0.9427 | 0.9450 | 0.9408 | 0.8312 |
0.2881 | 1.34 | 325 | 0.2365 | 0.9276 | 0.9311 | 0.9278 | 0.8088 |
0.288 | 1.44 | 350 | 0.2472 | 0.9276 | 0.9299 | 0.9179 | 0.7305 |
0.2728 | 1.54 | 375 | 0.2390 | 0.9202 | 0.9248 | 0.9208 | 0.7915 |
0.2502 | 1.65 | 400 | 0.2188 | 0.9342 | 0.9375 | 0.9343 | 0.8172 |
0.2351 | 1.75 | 425 | 0.2145 | 0.9368 | 0.9406 | 0.9376 | 0.8232 |
Framework versions
- Transformers 4.27.1
- Pytorch 1.13.1+cu116
- Datasets 2.10.1
- Tokenizers 0.13.2