|
--- |
|
library_name: transformers |
|
license: mit |
|
base_model: microsoft/deberta-v3-small |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
- f1 |
|
- precision |
|
- recall |
|
model-index: |
|
- name: doc-topic-model_eval-00_train-01 |
|
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. --> |
|
|
|
# doc-topic-model_eval-00_train-01 |
|
|
|
This model is a fine-tuned version of [microsoft/deberta-v3-small](https://huggingface.co/microsoft/deberta-v3-small) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.0398 |
|
- Accuracy: 0.9878 |
|
- F1: 0.6321 |
|
- Precision: 0.7134 |
|
- Recall: 0.5675 |
|
|
|
## 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: 2e-05 |
|
- train_batch_size: 4 |
|
- eval_batch_size: 256 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 100 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall | |
|
|:-------------:|:------:|:-----:|:---------------:|:--------:|:------:|:---------:|:------:| |
|
| 0.0936 | 0.4931 | 1000 | 0.0882 | 0.9815 | 0.0 | 0.0 | 0.0 | |
|
| 0.0754 | 0.9862 | 2000 | 0.0682 | 0.9815 | 0.0006 | 0.5714 | 0.0003 | |
|
| 0.0614 | 1.4793 | 3000 | 0.0561 | 0.9824 | 0.1463 | 0.7360 | 0.0812 | |
|
| 0.053 | 1.9724 | 4000 | 0.0500 | 0.9842 | 0.3207 | 0.7946 | 0.2009 | |
|
| 0.0477 | 2.4655 | 5000 | 0.0463 | 0.9853 | 0.4453 | 0.7381 | 0.3189 | |
|
| 0.0445 | 2.9586 | 6000 | 0.0435 | 0.9859 | 0.4832 | 0.7548 | 0.3553 | |
|
| 0.0385 | 3.4517 | 7000 | 0.0410 | 0.9865 | 0.5406 | 0.7356 | 0.4273 | |
|
| 0.0384 | 3.9448 | 8000 | 0.0400 | 0.9867 | 0.5643 | 0.7201 | 0.4639 | |
|
| 0.0347 | 4.4379 | 9000 | 0.0386 | 0.9870 | 0.5796 | 0.7235 | 0.4834 | |
|
| 0.0336 | 4.9310 | 10000 | 0.0381 | 0.9873 | 0.5971 | 0.7223 | 0.5089 | |
|
| 0.0299 | 5.4241 | 11000 | 0.0374 | 0.9875 | 0.5941 | 0.7483 | 0.4926 | |
|
| 0.0299 | 5.9172 | 12000 | 0.0375 | 0.9874 | 0.5978 | 0.7279 | 0.5071 | |
|
| 0.0265 | 6.4103 | 13000 | 0.0377 | 0.9874 | 0.6035 | 0.7218 | 0.5185 | |
|
| 0.0271 | 6.9034 | 14000 | 0.0379 | 0.9872 | 0.6061 | 0.7061 | 0.5309 | |
|
| 0.0229 | 7.3964 | 15000 | 0.0373 | 0.9877 | 0.6254 | 0.7162 | 0.5550 | |
|
| 0.0245 | 7.8895 | 16000 | 0.0378 | 0.9879 | 0.6295 | 0.7266 | 0.5553 | |
|
| 0.0205 | 8.3826 | 17000 | 0.0376 | 0.9876 | 0.6300 | 0.7041 | 0.5701 | |
|
| 0.0213 | 8.8757 | 18000 | 0.0385 | 0.9878 | 0.6303 | 0.7156 | 0.5631 | |
|
| 0.0183 | 9.3688 | 19000 | 0.0389 | 0.9878 | 0.6300 | 0.7164 | 0.5621 | |
|
| 0.0182 | 9.8619 | 20000 | 0.0398 | 0.9878 | 0.6321 | 0.7134 | 0.5675 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.44.2 |
|
- Pytorch 2.4.1+cu121 |
|
- Datasets 2.21.0 |
|
- Tokenizers 0.19.1 |
|
|
