| embedding_dimension: 50 | |
| input_tokenizer: Character-level | |
| language: Darija | |
| latent_dimension: 128 | |
| model_architecture: Sequence-to-Sequence LSTM | |
| parameters_format: .json | |
| saved_model_format: .keras | |
| target_tokenizer: Character-level | |
| task: Text Normalization | |
| tokenizers_format: .json | |
| training_data: Darija Open Dataset | |