jaclang 0.5.18__py3-none-any.whl → 0.6.0__py3-none-any.whl

jaclang/core/llms.py

@@ -1,111 +0,0 @@
-"""LLMs (Large Language Models) module for Jaclang."""
-
-
-class Anthropic:
-    """Anthropic API client for Large Language Models (LLMs)."""
-
-    MTLLM_PROMPT: str = ""
-    MTLLM_REASON_SUFFIX: str = ""
-    MTLLM_WO_REASON_SUFFIX: str = ""
-
-    def __init__(self, **kwargs: dict) -> None:
-        """Initialize the Anthropic API client."""
-        import anthropic
-
-        self.client = anthropic.Anthropic()
-        self.model_name = kwargs.get("model_name", "claude-3-sonnet-20240229")
-        self.temperature = kwargs.get("temperature", 0.7)
-        self.max_tokens = kwargs.get("max_tokens", 1024)
-
-    def __infer__(self, meaning_in: str, **kwargs: dict) -> str:
-        """Infer a response from the input meaning."""
-        messages = [{"role": "user", "content": meaning_in}]
-        output = self.client.messages.create(
-            model=kwargs.get("model_name", self.model_name),
-            temperature=kwargs.get("temperature", self.temperature),
-            max_tokens=kwargs.get("max_tokens", self.max_tokens),
-            messages=messages,
-        )
-        return output.content[0].text
-
-
-class Huggingface:
-    """Huggingface API client for Large Language Models (LLMs)."""
-
-    MTLLM_PROMPT: str = ""
-    MTLLM_REASON_SUFFIX: str = ""
-    MTLLM_WO_REASON_SUFFIX: str = ""
-
-    def __init__(self, **kwargs: dict) -> None:
-        """Initialize the Huggingface API client."""
-        import torch
-        from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline
-
-        torch.random.manual_seed(0)
-        model = AutoModelForCausalLM.from_pretrained(
-            kwargs.get("model_name", "microsoft/Phi-3-mini-128k-instruct"),
-            device_map=kwargs.get("device_map", "cuda"),
-            torch_dtype="auto",
-            trust_remote_code=True,
-        )
-        tokenizer = AutoTokenizer.from_pretrained(
-            kwargs.get("model_name", "microsoft/Phi-3-mini-128k-instruct")
-        )
-        self.pipe = pipeline("text-generation", model=model, tokenizer=tokenizer)
-        self.temperature = kwargs.get("temperature", 0.7)
-        self.max_tokens = kwargs.get("max_new_tokens", 1024)
-
-    def __infer__(self, meaning_in: str, **kwargs: dict) -> str:
-        """Infer a response from the input meaning."""
-        messages = [{"role": "user", "content": meaning_in}]
-        output = self.pipe(
-            messages,
-            temperature=kwargs.get("temperature", self.temperature),
-            max_length=kwargs.get("max_new_tokens", self.max_tokens),
-            **kwargs
-        )
-        return output[0]["generated_text"][-1]["content"]
-
-
-class Ollama:
-    """Ollama API client for Large Language Models (LLMs)."""
-
-    MTLLM_PROMPT: str = ""
-    MTLLM_REASON_SUFFIX: str = ""
-    MTLLM_WO_REASON_SUFFIX: str = ""
-
-    def __init__(self, **kwargs: dict) -> None:
-        """Initialize the Ollama API client."""
-        import ollama
-
-        self.client = ollama.Client(host=kwargs.get("host", "http://localhost:11434"))
-        self.model_name = kwargs.get("model_name", "phi3")
-        self.default_model_params = {
-            k: v for k, v in kwargs.items() if k not in ["model_name", "host"]
-        }
-
-    def __infer__(self, meaning_in: str, **kwargs: dict) -> str:
-        """Infer a response from the input meaning."""
-        model = str(kwargs.get("model_name", self.model_name))
-        if not self.check_model(model):
-            self.download_model(model)
-        model_params = {k: v for k, v in kwargs.items() if k not in ["model_name"]}
-        messages = [{"role": "user", "content": meaning_in}]
-        output = self.client.chat(
-            model=model,
-            messages=messages,
-            options={**self.default_model_params, **model_params},
-        )
-        return output["message"]["content"]
-
-    def check_model(self, model_name: str) -> bool:
-        """Check if the model is available."""
-        try:
-            self.client.show(model_name)
-            return True
-        except Exception:
-            return False
-
-    def download_model(self, model_name: str) -> None:
-        """Download the model."""
-        self.client.pull(model_name)

jaclang-0.5.18.dist-info/METADATA

@@ -1,7 +0,0 @@
-Metadata-Version: 2.1
-Name: jaclang
-Version: 0.5.18
-Home-page: https://github.com/Jaseci-Labs/jaclang
-Author: Jason Mars
-Author-email: jason@jaseci.org
-