YAML Metadata Warning:The pipeline tag "text2text-generation" is not in the official list: text-classification, token-classification, table-question-answering, question-answering, zero-shot-classification, translation, summarization, feature-extraction, text-generation, fill-mask, sentence-similarity, text-to-speech, text-to-audio, automatic-speech-recognition, audio-to-audio, audio-classification, audio-text-to-text, voice-activity-detection, depth-estimation, image-classification, object-detection, image-segmentation, text-to-image, image-to-text, image-to-image, image-to-video, unconditional-image-generation, video-classification, reinforcement-learning, robotics, tabular-classification, tabular-regression, tabular-to-text, table-to-text, multiple-choice, text-ranking, text-retrieval, time-series-forecasting, text-to-video, image-text-to-text, image-text-to-image, image-text-to-video, visual-question-answering, document-question-answering, zero-shot-image-classification, graph-ml, mask-generation, zero-shot-object-detection, text-to-3d, image-to-3d, image-feature-extraction, video-text-to-text, keypoint-detection, visual-document-retrieval, any-to-any, video-to-video, other

Chat T5

SourceCode

Chat T5 는 pko-flan-t5-large 를 기반으로 만들었습니다.

KoAlpaca 에서 제공하는 데이터셋과 evolve-instruct 에서 제공하는 데이터셋을 학습했습니다. 좋은 데이터를 공개해주셔서 감사합니다.

Model

Huggingface

Example

from transformers import T5TokenizerFast, T5ForConditionalGeneration
tokenizer = T5TokenizerFast.from_pretrained("paust/pko-chat-t5-large")
model = T5ForConditionalGeneration.from_pretrained("paust/pko-chat-t5-large", device_map='cuda')

prompt_tpl = "사용자가 한 말을 읽고 그에 질문에 답하거나 명령에 응답하는 비서입니다.\n\n사용자:\n{text}\n\n비서:\n"
prompt = prompt_tpl.format(text="한국의 수도는 어디인가요?")
input_ids = tokenizer(prompt, return_tensors='pt').input_ids
logits = model.generate(
    input_ids,
    max_new_tokens=1024,
    temperature=0.5,
    no_repeat_ngram_size=6,
    do_sample=True,
    num_return_sequences=1,
)
text = tokenizer.batch_decode(logits, skip_special_tokens=True)[0]
print(text)  # 한국의 수도는 서울입니다.

License

PAUST에서 만든 pko-t5는 MIT license 하에 공개되어 있습니다.

Downloads last month: 10

Safetensors

Model size

0.8B params

Tensor type

BF16