|
--- |
|
license: mit |
|
language: |
|
- en |
|
- zh |
|
- de |
|
- fr |
|
- ja |
|
- ko |
|
- es |
|
widget: |
|
- text: Hi assistant How can I help you |
|
- text: Guten Morgen! Wie kann ich Ihnen helfen? |
|
- text: どうすれば運動を続けられますか？ 運動を続けることは、健康的な生活を維持する上で非常に重要ですが、モチベーションを維持することが難しい場合があります。以下にいくつかの方法を紹介します。 |
|
- text: 세계 1차 대전은 1914년부터 1918년까지 전 세계적으로 벌어진 대규모 전쟁입니다. 주요한 참전국으로는 독일, 오스트리아-헝가리 제국, 영국, 프랑스, 러시아, 이탈리아, 미국 등이 있었습니다. |
|
- text: こんにちは！お元気ですか？何かお手伝いできることがありますか？ |
|
- text: user: Python 和 C++ 哪个更好学？哪个更强大？我该怎么选择？ |
|
- text: 在大熱天裡，墨鏡的銷售與冰淇淋的銷售有著高度相關性。當天氣很熱的時候，兩個都十分熱賣，而天氣轉涼以後兩者的銷售就底落谷底。當有一天，墨鏡批發商車輛在上班途中拋錨，因此無法開業，導致墨鏡銷售變 0 。請問當天冰淇淋的銷售如何？ |
|
- text: >- |
|
user: Good morning\n assistant: Good morning! How can I assist you |
|
today? |
|
pipeline_tag: text2text-generation |
|
tags: |
|
- text-generation-inference |
|
--- |
|
|
|
# Generate title for conversation |
|
|
|
## How to use |
|
|
|
```python |
|
model_name = "theblackcat102/alpaca-title-generator-mt0-large" |
|
tokenizer = AutoTokenizer.from_pretrained(model_name) |
|
model = AutoModelForSeq2SeqLM.from_pretrained(model_name) |
|
question = 'Hi\nHow can I help you?' |
|
encodes = tokenizer(question, return_tensors='pt') |
|
outputs = model.generate(encodes.input_ids, |
|
max_length=512, |
|
do_sample=True, |
|
repetition_penalty=1.2, |
|
top_k=50, |
|
num_return_sequences=1, |
|
early_stopping=True |
|
) |
|
for i, beam_output in enumerate(outputs): |
|
print('-----') |
|
print("{}".format(tokenizer.decode(beam_output, skip_special_tokens=True))) |
|
# > Help requested. |
|
``` |
|
|
|
## Generate title data |
|
|
|
data was generated using response pair from `yahma/alpaca-cleaned` and use openai turbo model for title. |
|
|
|
``` |
|
"" |
|
user: {} |
|
assistant: {} |
|
"" |
|
|
|
Generate a very short title within 5 words of the conversation above, title must be as relevant as possible. Title language must be same as the context |
|
|
|
TITLE: |
|
``` |
