SUMIN/HuggingFace_transformer
1
0
Fork 0
Code Issues Pull Requests Actions 7 Packages Projects Releases Wiki Activity

Documentation: BOS token_id deprecation change for NLLB (#32443)

Browse Source 
Update nllb.md
This commit is contained in:
Chris Toukmaji
2024-08-06 09:22:08 -07:00
committed by GitHub
parent 194cf1f392
commit 50c3ba889a
1 changed files with 4 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
docs/source/en/model_doc/nllb.md
View File

@@ -101,7 +101,7 @@ for the list of all BCP-47 in the Flores 200 dataset.
>>> inputs = tokenizer(article, return_tensors="pt")
>>> translated_tokens = model.generate(
... **inputs, forced_bos_token_id=tokenizer.lang_code_to_id["fra_Latn"], max_length=30
... **inputs, forced_bos_token_id=tokenizer.convert_tokens_to_ids("fra_Latn"), max_length=30
... )
>>> tokenizer.batch_decode(translated_tokens, skip_special_tokens=True)[0]
Le chef de l'ONU dit qu'il n'y a pas de solution militaire en Syrie
@@ -126,7 +126,7 @@ See example below for a translation from romanian to german:
>>> inputs = tokenizer(article, return_tensors="pt")
>>> translated_tokens = model.generate(
... **inputs, forced_bos_token_id=tokenizer.lang_code_to_id["deu_Latn"], max_length=30
... **inputs, forced_bos_token_id=tokenizer.convert_tokens_to_ids("deu_Latn"), max_length=30
... )
>>> tokenizer.batch_decode(translated_tokens, skip_special_tokens=True)[0]
UN-Chef sagt, es gibt keine militärische Lösung in Syrien
@@ -175,7 +175,7 @@ To load a model using Flash Attention 2, we can pass the argument `attn_implemen
>>> inputs = tokenizer(article, return_tensors="pt").to("cuda")
>>> translated_tokens = model.generate(
... **inputs, forced_bos_token_id=tokenizer.lang_code_to_id["deu_Latn"], max_length=30
... **inputs, forced_bos_token_id=tokenizer.convert_tokens_to_ids("deu_Latn"), max_length=30
... )
>>> tokenizer.batch_decode(translated_tokens, skip_special_tokens=True)[0]
"UN-Chef sagt, es gibt keine militärische Lösung in Syrien"
@@ -187,4 +187,4 @@ Below is an expected speedup diagram that compares pure inference time between t
<div style="text-align: center">
<img src="https://huggingface.co/datasets/visheratin/documentation-images/resolve/main/nllb-speedup.webp">
</div>
</div>