PiRddlmZmZmZmZmZddlmZmZddl m Z ddl m Z ddl mZmZmZgdZGddee Zd S) )AnyListMappingOptionalTuple)MessagePromptTemplate)MistralChatTemplate) Transform)ModelTokenizerSentencePieceBaseTokenizer#tokenize_messages_no_special_tokens)     c ~eZdZdZdefdedeedeefdZ e dZ e dZ e d Z e d Z dd ededededeef dZdeedefdZd ddeededeeeeeffdZ ddeeefdedeeeffdZdS)MistralTokenizera Mistral's implementation of the SentencePiece tokenizer Args: path (str): Path to pretrained tokenizer file. max_seq_len (Optional[int]): A max sequence length to truncate tokens to. Default: None prompt_template (Optional[PromptTemplate]): template used to format the messages based on their role. This is used to add structured text around the actual messages. The structured text is used in three scenarios: - Task-specific templates to gear models for a particular task that it will expect after training - Model-specific templates that are required whenever the model is prompted, such as the [INST] tags in Llama2 and in Mistral - Community standardized templates, such as :class:`~torchtune.data.ChatMLTemplate` The extra text will still get tokenized as normal text, not as special tokens. Default is :class:`~torchtune.models.mistral.MistralChatTemplate`. Examples: >>> tokenizer = MistralTokenizer("/path/to/spm_model") >>> tokenized_text = tokenizer.encode("Hello world!", add_bos=True, add_eos=True) >>> print(tokenized_text) [1, 31587, 29644, 102, 2] Npath max_seq_lenprompt_templatec|t||_d|j_|jg|_||_||_dS)Nr)r _spm_modelpad_ideos_id stop_tokensrr)selfrrrs w/home/jaya/work/projects/VOICE-AGENT/VIET/agent-env/lib/python3.11/site-packages/torchtune/models/mistral/_tokenizer.py__init__zMistralTokenizer.__init__/sC 5T::"#!K=&.c|jjSN)rrrs rrzMistralTokenizer.eos_idA %%r!c|jjSr#)rbos_idr$s rr'zMistralTokenizer.bos_idEr%r!c|jjSr#)rrr$s rrzMistralTokenizer.pad_idIr%r!c|jjSr#)r vocab_sizer$s rr*zMistralTokenizer.vocab_sizeMs ))r!TFtextadd_bosadd_eostrim_leading_whitespacereturnc>|j||||S)a4 Encode a string into a list of token IDs Args: text (str): The input text to be encoded, unbatched. add_bos (bool): Whether to prepend BOS special token (Beginning of Sentence) to the input, defaults to True. add_eos (bool): Whether to append EOS special token (End of Sentence) to the input, defaults to True. trim_leading_whitespace (bool): Whether to trim leading whitespace from underlying sentencepiece tokenization. Sentencepiece normally prepends whitespace to any tokenized text, which can cause differences where encode(s1) + encode(s2) != encode(s1 + s2) due to leading whitespace added to s2. Default: False Returns: List[int]: The encoded token IDs. )r,r-r.)rencode)rr+r,r-r.s rr1zMistralTokenizer.encodeQs/,%% $; &   r! token_idsc6|j|S)zDecode token IDs to strings. Args: token_ids (List[int]): The input token IDs to be decoded. Returns: str: The decoded text. )rdecode)rr2s rr4zMistralTokenizer.decodens%%i000r!)r-messagesc|j||n|}t|||j|r|jndS)aTokenize a list of messages one at a time then concatenate them, returning a list of tokens and a list of masks. Note: sentencepiece has problems where in general encode(s1 + s2) != encode(s1) + encode(s2) due to whitespace handling. We can get around this by prepending s2 with a known token and slicing the beginning off the tokenized s2. Example: >>> tokenizer = MistralTokenizer(tokenizer_path, max_seq_len) >>> messages = [ Message(role="system", content="system message\n", masked=True), Message(role="user", content="user prompt\n", masked=True), Message(role="assistant", content="assistant response\n"), ] >>> # tokenize_messages encodes messages separately and concats >>> tokenizer.tokenize_messages(messages)[0] [1, 1788, 2643, 13, 1792, 9508, 13, 465, 22137, 2933, 2] >>> # Same result as encoding the full string in one go >>> tokenizer.encode(''.join([message.content for message in messages])) [1, 1788, 2643, 13, 1792, 9508, 13, 465, 22137, 2933, 2] Args: messages (List[Message]): A list of messages, each containing role, content, and masked attributes. add_eos (bool): Whether to append EOS after assistant message, default to True Returns: Tuple[List[int], List[bool]]: The tokenized messages N) tokenizerr5r'r)rrr'r)rr5r-templated_messagess rtokenize_messagesz"MistralTokenizer.tokenize_messages|s^V#/   * * *  3';")34;;t     r!sample inferencect|d}||\}}||d<||d<|S)a% Apply ``tokenize_messages`` to the "messages" field in the sample. Args: sample (Mapping[str, Any]): A sample with a "messages" field containing a List[Message] to tokenize inference (bool): Whether the template is being used for inference or not. Returns: Mapping[str, Any]: The sample with added "tokens" and "mask" fields and the "messages" field removed. inference (bool): Whether the template is being used for inference or not. r5tokensmask)popr9)rr:r;r5r=r>s r__call__zMistralTokenizer.__call__sD ::j))--h77 !xv r!)TTF)F)__name__ __module__ __qualname____doc__r strrintr r propertyrr'rr*boolrr1r4rrr9rrr@r!rrrs8&*4G4G4I4I ///c]/".1 ////$&&X&&&X&&&X&**X* (-       "&  c     : 19 1  1 1 1 1$ 3 3 3 w-3  3 tCy$t*$ % 3 3 3 3 lrPs7666666666666622222222IIIIII222222 100ppppp~ypppppr!