---
tags:
- generated_from_trainer
metrics:
- precision
- recall
- f1
- accuracy
model-index:
- name: RoBERTa-base-PM-M3-Voc-distill-align-hf-finetuned-ner
  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. -->

# RoBERTa-base-PM-M3-Voc-distill-align-hf-finetuned-ner

This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.5981
- Precision: 0.5371
- Recall: 0.7479
- F1: 0.6252
- Accuracy: 0.8321

## 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: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- num_epochs: 20

### Training results

| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1     | Accuracy |
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
| No log        | 1.0   | 23   | 1.5355          | 0.0       | 0.0    | 0.0    | 0.6968   |
| No log        | 2.0   | 46   | 1.2530          | 0.2813    | 0.1177 | 0.1659 | 0.7177   |
| No log        | 3.0   | 69   | 0.9868          | 0.3861    | 0.3212 | 0.3507 | 0.7462   |
| No log        | 4.0   | 92   | 0.8347          | 0.4115    | 0.4235 | 0.4174 | 0.7629   |
| No log        | 5.0   | 115  | 0.7452          | 0.4892    | 0.4197 | 0.4518 | 0.7884   |
| No log        | 6.0   | 138  | 0.6956          | 0.4935    | 0.5810 | 0.5337 | 0.7952   |
| No log        | 7.0   | 161  | 0.7015          | 0.4673    | 0.6515 | 0.5443 | 0.7932   |
| No log        | 8.0   | 184  | 0.6316          | 0.5150    | 0.6704 | 0.5825 | 0.8137   |
| No log        | 9.0   | 207  | 0.6475          | 0.4944    | 0.6976 | 0.5787 | 0.8085   |
| No log        | 10.0  | 230  | 0.6144          | 0.5157    | 0.7105 | 0.5977 | 0.8175   |
| No log        | 11.0  | 253  | 0.6339          | 0.5075    | 0.7161 | 0.5941 | 0.8146   |
| No log        | 12.0  | 276  | 0.6114          | 0.5250    | 0.7235 | 0.6084 | 0.8221   |
| No log        | 13.0  | 299  | 0.6389          | 0.5127    | 0.7339 | 0.6037 | 0.8160   |
| No log        | 14.0  | 322  | 0.6256          | 0.5222    | 0.7350 | 0.6106 | 0.8228   |
| No log        | 15.0  | 345  | 0.6445          | 0.5161    | 0.7455 | 0.6099 | 0.8188   |
| No log        | 16.0  | 368  | 0.5887          | 0.5411    | 0.7416 | 0.6257 | 0.8323   |
| No log        | 17.0  | 391  | 0.5906          | 0.5387    | 0.7455 | 0.6255 | 0.8332   |
| No log        | 18.0  | 414  | 0.5858          | 0.5400    | 0.7490 | 0.6276 | 0.8353   |
| No log        | 19.0  | 437  | 0.6005          | 0.5369    | 0.7497 | 0.6257 | 0.8316   |
| No log        | 20.0  | 460  | 0.5981          | 0.5371    | 0.7479 | 0.6252 | 0.8321   |


### Framework versions

- Transformers 4.28.1
- Pytorch 2.0.0+cu118
- Datasets 2.11.0
- Tokenizers 0.13.3