z-dickson/CAP_coded_UK_statutory_instruments

1---2tags:3- generated_from_keras_callback4model-index:5- name: CAP_coded_UK_statutory_instruments6  results: []7  8widget:9- text: "The National Health Service (Charges for Drugs and Appliances) (Scotland) Regulations 2007"10  example_title: "example 1"11- text: "The Inshore Fishing (Prohibited Methods of Fishing) (Luce Bay) Order 2015"12  example_title: "example 2"13  14---15 16 17 18# CAP_coded_UK_statutory_instruments19 20This model predicts the CAP code of parliamentary bills/instruments (https://www.comparativeagendas.net/pages/master-codebook)21 22The model is trained on ~40k UK Parliamentary Statutory Instruments from the UK House of Commons and the Scottish Parliament. 23The model is cased (case sensitive)24 25Any questions on the model and training data feel free to message me on twitter - @sachary_26 27 28- Train Loss: 0.118829- Train Sparse Categorical Accuracy: 0.968830- Validation Loss: 0.203231- Validation Sparse Categorical Accuracy: 0.955632 33 34 35 36The following hyperparameters were used during training:37- optimizer: {'name': 'Adam', 'learning_rate': 5e-05, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e-07, 'amsgrad': False}38- training_precision: float3239 40### Training results41 42| Train Loss | Train Sparse Categorical Accuracy | Validation Loss | Validation Sparse Categorical Accuracy | Epoch |43|:----------:|:---------------------------------:|:---------------:|:--------------------------------------:|:-----:|44| 0.2167     | 0.9474                            | 0.2351          | 0.9444                                 | 0     |45| 0.1539     | 0.9592                            | 0.2076          | 0.9536                                 | 1     |46| 0.1188     | 0.9688                            | 0.2032          | 0.9556                                 | 2     |47 48 49### Framework versions50 51- Transformers 4.19.252- TensorFlow 2.8.253- Datasets 2.2.254- Tokenizers 0.12.155