File size: 3,305 Bytes
3798ae6 59bfe82 3798ae6 59bfe82 982051b 3798ae6 982051b 3798ae6 982051b 3798ae6 982051b 3798ae6 46492d4 3798ae6 59bfe82 3798ae6 982051b 3798ae6 982051b |
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 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 |
---
library_name: transformers
license: mit
base_model: microsoft/deberta-v3-large
tags:
- multi-label
- question-answering
- text-classification
- generated_from_trainer
datasets:
- saiteki-kai/BeaverTails-it
metrics:
- f1
- accuracy
- precision
- recall
language:
- it
- en
---
<!-- 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. -->
# QA-DeBERTa-v3-large
This model is a fine-tuned version of [microsoft/deberta-v3-large](https://huggingface.co/microsoft/deberta-v3-large) on the [saiteki-kai/BeaverTails-it](https://huggingface.co/datasets/saiteki-kai/BeaverTails-it) dataset.
It achieves the following results on the evaluation set:
- Loss: 0.0808
- Accuracy: 0.6938
- Macro F1: 0.6484
- Macro Precision: 0.7149
- Macro Recall: 0.6176
- Micro F1: 0.7545
- Micro Precision: 0.7874
- Micro Recall: 0.7242
- Flagged/accuracy: 0.8566
- Flagged/precision: 0.8975
- Flagged/recall: 0.8380
- Flagged/f1: 0.8667
## 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: 3.85e-06
- train_batch_size: 16
- eval_batch_size: 64
- seed: 42
- distributed_type: multi-GPU
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10
### Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Macro F1 | Macro Precision | Macro Recall | Micro F1 | Micro Precision | Micro Recall | Flagged/accuracy | Flagged/precision | Flagged/recall | Flagged/f1 |
|:-------------:|:-----:|:------:|:---------------:|:--------:|:--------:|:---------------:|:------------:|:--------:|:---------------:|:------------:|:----------------:|:-----------------:|:--------------:|:----------:|
| 0.0985 | 1.0 | 33814 | 0.0877 | 0.6750 | 0.6102 | 0.6629 | 0.5948 | 0.7406 | 0.7705 | 0.7129 | 0.8447 | 0.8701 | 0.8475 | 0.8586 |
| 0.0867 | 2.0 | 67628 | 0.0817 | 0.6910 | 0.6185 | 0.7559 | 0.5598 | 0.7446 | 0.8165 | 0.6842 | 0.8465 | 0.9093 | 0.8043 | 0.8536 |
| 0.0561 | 3.0 | 101442 | 0.0808 | 0.6938 | 0.6484 | 0.7149 | 0.6177 | 0.7545 | 0.7875 | 0.7242 | 0.8566 | 0.8975 | 0.8380 | 0.8667 |
| 0.0913 | 4.0 | 135256 | 0.0812 | 0.6877 | 0.6412 | 0.7136 | 0.6144 | 0.7516 | 0.7796 | 0.7255 | 0.8546 | 0.8902 | 0.8428 | 0.8658 |
| 0.0709 | 5.0 | 169070 | 0.0826 | 0.6911 | 0.6376 | 0.7306 | 0.5982 | 0.7500 | 0.7911 | 0.7129 | 0.8538 | 0.8936 | 0.8370 | 0.8643 |
### Framework versions
- Transformers 4.51.3
- Pytorch 2.7.0+cu118
- Datasets 3.5.1
- Tokenizers 0.21.1 |