transformers/docs/source/en/model_doc/camembert.md

This model was published in HF papers on 2019-11-10 and contributed to Hugging Face Transformers on 2020-11-16.

CamemBERT

CamemBERT is a language model based on RoBERTa, but trained specifically on French text from the OSCAR dataset, making it more effective for French language tasks.

What sets CamemBERT apart is that it learned from a huge, high quality collection of French data, as opposed to mixing lots of languages. This helps it really understand French better than many multilingual models.

Common applications of CamemBERT include masked language modeling (Fill-mask prediction), text classification (sentiment analysis), token classification (entity recognition) and sentence pair classification (entailment tasks).

You can find all the original CamemBERT checkpoints under the ALMAnaCH organization.

Tip

This model was contributed by the ALMAnaCH (Inria) team.

Click on the CamemBERT models in the right sidebar for more examples of how to apply CamemBERT to different NLP tasks.

The examples below demonstrate how to predict the <mask> token with [Pipeline], [AutoModel], and from the command line.

from transformers import pipeline


pipeline = pipeline("fill-mask", model="camembert-base", device=0)
pipeline("Le camembert est un délicieux fromage <mask>.")

import torch

from transformers import AutoModelForMaskedLM, AutoTokenizer


tokenizer = AutoTokenizer.from_pretrained("camembert-base")
model = AutoModelForMaskedLM.from_pretrained("camembert-base", device_map="auto", attn_implementation="sdpa")
inputs = tokenizer("Le camembert est un délicieux fromage <mask>.", return_tensors="pt").to(model.device)

with torch.no_grad():
    outputs = model(**inputs)
    predictions = outputs.logits

masked_index = torch.where(inputs['input_ids'] == tokenizer.mask_token_id)[1]
predicted_token_id = predictions[0, masked_index].argmax(dim=-1)
predicted_token = tokenizer.decode(predicted_token_id)

print(f"The predicted token is: {predicted_token}")

Quantization reduces the memory burden of large models by representing weights in lower precision. Refer to the Quantization overview for available options.

The example below uses bitsandbytes quantization to quantize the weights to 8-bits.

import torch

from transformers import AutoModelForMaskedLM, AutoTokenizer, BitsAndBytesConfig


quant_config = BitsAndBytesConfig(load_in_8bit=True)
model = AutoModelForMaskedLM.from_pretrained(
    "almanach/camembert-large",
    quantization_config=quant_config,
    device_map="auto"
)
tokenizer = AutoTokenizer.from_pretrained("almanach/camembert-large")

inputs = tokenizer("Le camembert est un délicieux fromage <mask>.", return_tensors="pt").to(model.device)

with torch.no_grad():
    outputs = model(**inputs)
    predictions = outputs.logits

masked_index = torch.where(inputs["input_ids"] == tokenizer.mask_token_id)[1]
predicted_token_id = predictions[0, masked_index].argmax(dim=-1)
predicted_token = tokenizer.decode(predicted_token_id)

print(f"The predicted token is: {predicted_token}")

CamembertConfig

autodoc CamembertConfig

CamembertTokenizer

autodoc CamembertTokenizer - get_special_tokens_mask - save_vocabulary

CamembertTokenizerFast

autodoc CamembertTokenizerFast

CamembertModel

autodoc CamembertModel

CamembertForCausalLM

autodoc CamembertForCausalLM

CamembertForMaskedLM

autodoc CamembertForMaskedLM

CamembertForSequenceClassification

autodoc CamembertForSequenceClassification

CamembertForMultipleChoice

autodoc CamembertForMultipleChoice

CamembertForTokenClassification

autodoc CamembertForTokenClassification

CamembertForQuestionAnswering

autodoc CamembertForQuestionAnswering