language: zh tags:

• exbert license: apache-2.0 widget:
• text: "女人做得越纯粹，皮肤和身材就越好"
• text: "我喜欢篮球"

liam168/c4-zh-distilbert-base-uncased

模型描述

基于["女性","体育","文学","校园"]4类数据训练的中文文本分类模型。

概览

• 语言模型: DistilBERT
• 模型参数量: 2.8亿
• 适用语言: 中文

使用示例

>>> from transformers import DistilBertForSequenceClassification , AutoTokenizer, pipeline

>>> model_name = "liam168/c4-zh-distilbert-base-uncased"
>>> class_num = 4
>>> ts_texts = ["女人做得越纯粹，皮肤和身材就越好", "我喜欢篮球"]
>>> model = DistilBertForSequenceClassification.from_pretrained(model_name, num_labels=class_num)
>>> tokenizer = AutoTokenizer.from_pretrained(model_name)
>>> classifier = pipeline('sentiment-analysis', model=model, tokenizer=tokenizer)
>>> classifier(ts_texts[0])
>>> classifier(ts_texts[1])

[{'label': '女性', 'score': 0.9137857556343079}]
[{'label': '体育', 'score': 0.8206522464752197}]