from transformers import AutoModelForSeq2SeqLM, AutoTokenizer # 加载模型和分词器 model_name = "Helsinki-NLP/opus-mt-en-kk" # 英语到哈萨克语的模型 tokenizer = AutoTokenizer.from_pretrained(model_name) model = AutoModelForSeq2SeqLM.from_pretrained(model_name)

# 输入英文文本
english_text = "Hello, how are you?"

# 分词
inputs = tokenizer(english_text, return_tensors="pt")

# 生成翻译
translated_tokens = model.generate(**inputs)

# 解码翻译结果
kazakh_text = tokenizer.decode(translated_tokens[0], skip_special_tokens=True)

print(f"英文: {english_text}")
print(f"哈萨克语翻译: {kazakh_text}")

README.md

@@ -0,0 +1 @@
+Helsinki-NLP/opus-mt-en-kk