README.md

---
license: mit
language:
- ja
pipeline_tag: text-generation
---

# japanese-gpt-1b-PII-masking

![image/png](https://cdn-uploads.huggingface.co/production/uploads/64ffe8a785a884a964b0cffe/gFQn0Oc6Nrvj8ViyTdZuM.png)

# Model Description
japanese-gpt-1b-PII-masking は、 [日本語事前学習済み1B GPTモデル](https://huggingface.co/rinna/japanese-gpt-1b)をベースとして、日本語の文章から個人情報をマスキングするように学習したモデルです。

# Usage
```python
import torch
from transformers import AutoModelForCausalLM, AutoTokenizer

input_text = ""

model_name = "cameltech/japanese-gpt-1b-PII-masking"
model = AutoModelForCausalLM.from_pretrained(best_model_path)
tokenizer = AutoTokenizer.from_pretrained(best_model_path)

if torch.cuda.is_available():
    model = model.to("cuda")

def preprocess(text):
    return text.replace("\n", "<LB>")

def postprocess(text):
    return text.replace("<LB>", "\n")
    
input_text += tokenizer.eos_token
input_text = preprocess(input_text)

with torch.no_grad():
    token_ids = tokenizer.encode(prompt, add_special_tokens=False, return_tensors="pt")

    output_ids = model.generate(
        token_ids.to(model.device),
        max_new_tokens=256,
        pad_token_id=tokenizer.pad_token_id,
        eos_token_id=tokenizer.eos_token_id,
    )
output = tokenizer.decode(output_ids.tolist()[0][token_ids.size(1) :], skip_special_tokens=True)
output = postprocess(output)

print(output)
```

# Licenese
[The MIT license](https://opensource.org/licenses/MIT)