| --- |
| license: openrail++ |
| base_model: textdetox/bert-multilingual-toxicity-classifier |
| pipeline_tag: text-classification |
| tags: |
| - toxic |
| - onnx |
| - int8 |
| language: [en, de, fr, es, it, ru, uk, ar, hi, ja, zh, he, am, tt] |
| --- |
| |
| # Ordicio toxicity classifier (ONNX, int8) |
|
|
| An int8 ONNX export of |
| [textdetox/bert-multilingual-toxicity-classifier](https://huggingface.co/textdetox/bert-multilingual-toxicity-classifier) |
| for on-device inference in the Ordicio apps (ONNX Runtime on Android / iOS, onnxruntime-web in browsers). |
|
|
| - `manifest.json`: the current version, file hashes, tokenizer settings, labels and threshold. |
| - `v1/model.onnx`: weights quantized to int8 (dynamic quantization), inputs `input_ids`, |
| `attention_mask`, `token_type_ids` as int32 (cast to int64 inside the graph), output `logits` |
| (index 0 neutral, 1 toxic). |
| - `v1/vocab.txt`: the WordPiece vocabulary of the base model (cased, no lowercasing). |
|
|
| The int8 model agrees with the fp32 model on 98.5 % of decisions (3600 samples of |
| textdetox/multilingual_toxicity_dataset in 12 languages). For accuracy per language, see the base |
| model's card: German results are weak (F1 0.52). |
|
|
| The use restrictions of the openrail++ license of the base model apply to this export. |
|
|
