Instructions to use WindyTranslate/translate-af-fr with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use WindyTranslate/translate-af-fr with Transformers:
# Use a pipeline as a high-level helper # Warning: Pipeline type "translation" is no longer supported in transformers v5. # You must load the model directly (see below) or downgrade to v4.x with: # 'pip install "transformers<5.0.0' from transformers import pipeline pipe = pipeline("translation", model="WindyTranslate/translate-af-fr")# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("WindyTranslate/translate-af-fr", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Add tokenizer files to the lora-ct2-int8 folder so it is usable standalone
Browse filesThe folder shipped config.json + model.bin + shared_vocabulary.json only. CTranslate2 loads those, but translating needs the Marian tokenizer, which lived one folder away in lora/ - so a developer downloading just this variant got a model they could not run. Files copied from lora/ in this same repo and verified by an end-to-end translation before upload. Weights untouched. Dr. J (Opus 5), clinic log doctor-logs/2026-08-05_stt-fleet-qa.md
lora-ct2-int8/special_tokens_map.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"eos_token": "</s>",
|
| 3 |
+
"pad_token": "<pad>",
|
| 4 |
+
"unk_token": "<unk>"
|
| 5 |
+
}
|