README.md

---
license: mit
language:
- en
- zh
- de
- fr
- ja
- ko
- es
widget:
- text: Hi assistant How can I help you
- text: user: Python 和 C++ 哪个更好学？哪个更强大？我该怎么选择？
- 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: 
```