Problems with Hugging Face

[JiyangZhang]

Hi,

I noticed that you uploaded the model to hugging face transformers library which is super exciting! However, when I tried to use it through transformers with the provided guidance, I got the following error:

>>> model = AutoModelForSeq2SeqLM.from_pretrained("uclanlp/plbart-base") Traceback (most recent call last): File "<stdin>", line 1, in <module> File "/home/anaconda3/envs/mdt/lib/python3.7/site-packages/transformers/models/auto/auto_factory.py", line 397, in from_pretrained pretrained_model_name_or_path, return_unused_kwargs=True, **kwargs File "/home/anaconda3/envs/mdt/lib/python3.7/site-packages/transformers/models/auto/configuration_auto.py", line 529, in from_pretrained config_class = CONFIG_MAPPING[config_dict["model_type"]] File "/home/anaconda3/envs/mdt/lib/python3.7/site-packages/transformers/models/auto/configuration_auto.py", line 278, in __getitem__ raise KeyError(key) KeyError: 'plbart' This is the command I used: from transformers import AutoTokenizer, AutoModelForSeq2SeqLM tokenizer = AutoTokenizer.from_pretrained("uclanlp/plbart-base") model = AutoModelForSeq2SeqLM.from_pretrained("uclanlp/plbart-base")

Here is my configuration: python: 3.7.11 pytorch: 1.9.1 transformers: 4.11.3

Thanks.

[wasiahmad]

Hi, the work is still in progress. Hopefully, we will be finished by mid-November. We will make an announcement in this repository. Stay tuned!

[Tbabm]

Hi,

I also encountered this problem and would like to know whether the pre-trained model is ready for use through the Transformers library?

Thanks a lot!

[wasiahmad]

PLBart is available from transformers v4.17.0, check the release note. We will be making formal announcement soon.

Starter Code

from transformers import PLBartTokenizer, PLBartForConditionalGeneration

def load_model_and_tokenizer(model_name_or_path):
    tokenizer = PLBartTokenizer.from_pretrained(model_name_or_path, language_codes="base")
    model = PLBartForConditionalGeneration.from_pretrained(model_name_or_path)
    return model, tokenizer

def translate(
        model_name_or_path,
        input_sequences,
        src_lang=None,
        tgt_lang=None,
        max_generation_length=128,
        num_beams=10,
        num_return_sequences=1
):
    model, tokenizer = load_model_and_tokenizer(model_name_or_path)
    if src_lang:
        tokenizer.src_lang = src_lang
    decoder_start_token_id = None
    if tgt_lang:
        decoder_start_token_id = tokenizer.lang_code_to_id[tgt_lang]
    inputs = tokenizer(input_sequences, return_tensors='pt', padding=True)
    outputs = model.generate(
        **inputs,
        decoder_start_token_id=decoder_start_token_id,
        max_length=max_generation_length,
        num_beams=num_beams,
        num_return_sequences=num_return_sequences,
    )
    return tokenizer.batch_decode(outputs, skip_special_tokens=True)

src_lang, tgt_lang = "java", "en_XX"
model_name_or_path = "uclanlp/plbart-java-en_XX"
input = "static int sumDigits ( int no ) { return no == 0 ? 0 : no % 10 + sumDigits ( no / 10 ) ; }"
outputs = translate(
    model_name_or_path, [inputs], src_lang, tgt_lang
)
print("\n".join(outputs)) # ["Returns the sum of digits of the given number."]