类型错误:llama_tokenize() 缺少 2 个必需的位置参数:“add_bos”和“special”
问题内容
我正在运行 python 3.11 和最新版本的 llama-cpp-python 以及 一个 gguf 模型
我希望代码像聊天机器人一样正常运行,但我得到了这个错误:
traceback (most recent call last): file "d:i customi arushserver.py", line 223, in <module> init() file "d:i customi arushserver.py", line 57, in init m_eval(model, m_tokenize(model, prompt_init, true), false, "starting up...") ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ file "d:i customi arushserver.py", line 182, in m_tokenize n_tokens = llama_cpp.llama_tokenize( ^^^^^^^^^^^^^^^^^^^^^^^^^typeerror: llama_tokenize() missing 2 required positional arguments: 'add_bos' and 'special'</module>
这是我的标记化代码:
def m_tokenize(model: llama_cpp.Llama, text: bytes, add_bos=False, special=False): assert model.ctx is not None n_ctx = llama_cpp.llama_n_ctx(model.ctx) tokens = (llama_cpp.llama_token * int(n_ctx))() n_tokens = llama_cpp.llama_tokenize( model.ctx, text, tokens, n_ctx, llama_cpp.c_bool(add_bos), ) if int(n_tokens) <br><h2 class="daan">正确答案</h2><br><pre class="brush:php;toolbar:false;">typeerror: llama_tokenize() missing 2 required positional arguments: 'add_bos' and 'special'
要解决该错误,您需要将参数 add_bos 和 special 包含到 llama_tokenize() 函数中。
def m_tokenize(model: llama_cpp.llama, text: bytes, add_bos=false, special=false): assert model.ctx is not none n_ctx = llama_cpp.llama_n_ctx(model.ctx) tokens = (llama_cpp.llama_token * int(n_ctx))() # include the missing arguments in the function call n_tokens = llama_cpp.llama_tokenize( model.ctx, text, tokens, n_ctx, # you should check if llama_cpp.c_bool(add_bos) is returning a c_boo value also you have the arguments add_bos=false and special=false in this function # if i am right all you need is: add_bos # not # llama_cpp.c_bool(add_bos), # you should check if llama_cpp.c_bool(special) is returning a c_boo value # if i am right all you need is: special # include the special argument # not # llama_cpp.c_bool(special) ) if int(n_tokens) <p>来自 <a href="https://www.php.cn/link/fcbc95ccdd551da181207c0c1400c655" rel="nofollow noreferrer">llama_cpp.py (github) a>,从 1817 开始的代码行</a></p><pre class="brush:php;toolbar:false;">def llama_tokenize( model: llama_model_p, text: bytes, text_len: Union[c_int, int], tokens, # type: Array[llama_token] n_max_tokens: Union[c_int, int], add_bos: Union[c_bool, bool], special: Union[c_bool, bool],) -> int: """Convert the provided text into tokens.""" return _lib.llama_tokenize( model, text, text_len, tokens, n_max_tokens, add_bos, special )_lib.llama_tokenize.argtypes = [ llama_model_p, c_char_p, c_int32, llama_token_p, c_int32, c_bool, c_bool,]_lib.llama_tokenize.restype = c_int32