classifier = pipeline('sentiment-analysis') 这个内部封装的模型是只针对英文的吗?我用中文发现完全不对,比如:words = '这是一个不错的服务' Judge1 = classifier(words) Words = '这是一个很好的服务' Judge2 = classifier(words) print(judge1,judge2)
结果展示: [{'label': 'NEGATIVE', 'score': 0.9500061273574829}] [{'label': 'NEGATIVE', 'score': 0.92015540599823}]