| --- |
| tags: |
| - model_hub_mixin |
| - pytorch_model_hub_mixin |
| license: wtfpl |
| datasets: |
| - jakartaresearch/cerpen-corpus |
| language: |
| - id |
| base_model: |
| - LocalWisdom/PurpleBoW |
| pipeline_tag: feature-extraction |
| --- |
| |
| # PurpleCBoW |
|
|
| A very simple implementation of Continuous Bag-of-Words for those learning about Natural Language Processing. |
|
|
| ## What's The Difference With Skip-Gram and Word2Vec? |
|
|
| Well, Word2Vec is the concept of using a shallow neural network to turn a word into vector. |
| Meanwhile, Skip-Gram is just another variant of CBOW. Skip-Gram and CBOW *is* Word2Vec. |
|
|
| ## Usage |
|
|
| Download the `train.ipynb` file and execute every cell. |
| At the bottom of the notebook, we have the "inference" part where we can test with whatever word is in the model vocabulary. |
