---
tags:
- generated_from_keras_callback
model-index:
- name: CAP_coded_UK_statutory_instruments
  results: []
  
widget:
- text: "The National Health Service (Charges for Drugs and Appliances) (Scotland) Regulations 2007"
  example_title: "'label': 'health', 'score': 0.9882584810256958"
- text: "The Licensing of Relevant Permanent Sites (Scotland) Regulations 2016"
  example_title: "'label': 'housing', 'score': 0.6133455038070679"
  
---



# CAP_coded_UK_statutory_instruments

This model predicts the CAP code of parliamentary bills/instruments (https://www.comparativeagendas.net/pages/master-codebook)

The model is trained on ~40k UK Parliamentary Statutory Instruments from the UK House of Commons and the Scottish Parliament. 
The model is cased (case sensitive)

See the README for more information on the model and training data or message me on twitter - @sachary_


- Train Loss: 0.1188
- Train Sparse Categorical Accuracy: 0.9688
- Validation Loss: 0.2032
- Validation Sparse Categorical Accuracy: 0.9556
- Epoch: 2



The following hyperparameters were used during training:
- optimizer: {'name': 'Adam', 'learning_rate': 5e-05, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False}
- training_precision: float32

### Training results

| Train Loss | Train Sparse Categorical Accuracy | Validation Loss | Validation Sparse Categorical Accuracy | Epoch |
|:----------:|:---------------------------------:|:---------------:|:--------------------------------------:|:-----:|
| 0.2167     | 0.9474                            | 0.2351          | 0.9444                                 | 0     |
| 0.1539     | 0.9592                            | 0.2076          | 0.9536                                 | 1     |
| 0.1188     | 0.9688                            | 0.2032          | 0.9556                                 | 2     |


### Framework versions

- Transformers 4.19.2
- TensorFlow 2.8.2
- Datasets 2.2.2
- Tokenizers 0.12.1