| --- |
| library_name: lucid |
| license: apache-2.0 |
| tags: |
| - base |
| - bert |
| - lucid |
| datasets: |
| - wikipedia |
| - bookcorpus |
| pipeline_tag: feature-extraction |
| --- |
| |
| # BERT-Medium |
|
|
| > https://arxiv.org/abs/1810.04805 |
|
|
| [Lucid](https://github.com/ChanLumerico/lucid) port of `transformers/google/bert_uncased_L-8_H-512_A-8`, |
| converted to Lucid-native safetensors. |
|
|
| ## Available weights |
|
|
| | Tag | Params | GFLOPs | Size | Source | |
| |---|---|---|---|---| |
| | `WIKIPEDIA_BOOKSCORPUS` *(default)* | 41.4M | — | 157.84 MB | transformers | |
|
|
| ## Usage |
|
|
| ```python |
| import lucid |
| import lucid.models as models |
| from lucid.models.weights import BertMediumWeights |
| |
| # default tag |
| model = models.bert_medium(pretrained=True) |
| |
| # explicit tag (enum or string) |
| model = models.bert_medium(weights=BertMediumWeights.WIKIPEDIA_BOOKSCORPUS) |
| model = models.bert_medium(pretrained="WIKIPEDIA_BOOKSCORPUS") |
| |
| # feed token ids (tokenize with the matching lucid.utils.tokenizer) |
| input_ids = lucid.tensor([[101, 7592, 2088, 102]], dtype=lucid.int64) |
| out = model(input_ids) |
| hidden = out.last_hidden_state # (B, T, hidden_size) |
| ``` |
|
|
| ## Conversion |
|
|
| Converted from `transformers/google/bert_uncased_L-8_H-512_A-8` via |
| `python -m tools.convert_weights bert_medium --tag WIKIPEDIA_BOOKSCORPUS`. |
| Key mapping + numerical parity verified against the source. |
|
|
| ## License |
|
|
| `apache-2.0` — inherited from the original weights. |
|
|
| ## Citation |
|
|
| ``` |
| Devlin et al., "BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding", NAACL 2019. Miniatures: Turc et al., "Well-Read Students Learn Better", 2019. |
| ``` |
|
|
