Add support for introspecting a Jupyter notebook cell (#117)

Commit Description: Can now be run in a Jupyter notebook cell

Co-authored-by: Jack Hopkins <[email protected]>

output.spec

@@ -0,0 +1,98 @@
+Class: LLM_Finetune_API
+  finetune(self, **kwargs) -> tanuki.models.finetune_job.FinetuneJob
+  get_finetuned(self, job_id: str, **kwargs) -> tanuki.models.finetune_job.FinetuneJob
+  list_finetuned(self, limit=100, **kwargs) -> List[tanuki.models.finetune_job.FinetuneJob]
+
+Class: LLM_API
+  generate(self, model, system_message, prompt, **kwargs)
+
+Class: LLM_Finetune_API
+  finetune(self, **kwargs) -> tanuki.models.finetune_job.FinetuneJob
+  get_finetuned(self, job_id: str, **kwargs) -> tanuki.models.finetune_job.FinetuneJob
+  list_finetuned(self, limit=100, **kwargs) -> List[tanuki.models.finetune_job.FinetuneJob]
+
+Class: LLM_API
+  generate(self, model, system_message, prompt, **kwargs)
+
+Class: Embedding_API
+  embed(self, texts: List[str], model: str = None, **kwargs) -> List[tanuki.models.embedding.Embedding]
+
+Class: LLM_Finetune_API
+  Class Docstring: Helper class that provides a standard way to create an ABC using
+inheritance.
+  finetune(self, **kwargs) -> tanuki.models.finetune_job.FinetuneJob
+Docstring: Creates a fine-tuning run
+Args:
+    **kwargs: 
+
+Returns:
+  get_finetuned(self, job_id: str, **kwargs) -> tanuki.models.finetune_job.FinetuneJob
+Docstring: Gets a fine-tuning run by id
+  list_finetuned(self, limit=100, **kwargs) -> List[tanuki.models.finetune_job.FinetuneJob]
+Docstring: Gets the last N fine-tuning runs
+
+Class: Embedding_API
+  Class Docstring: Helper class that provides a standard way to create an ABC using
+inheritance.
+  embed(self, texts: List[str], model: str = None, **kwargs) -> List[tanuki.models.embedding.Embedding]
+Docstring: The main embedding function, given the model and prompt, return a vector representation
+
+Class: Embedding_API
+  Class Docstring: Helper class that provides a standard way to create an ABC using
+inheritance.
+  embed(self, texts: List[str], model: str = None, **kwargs) -> List[tanuki.models.embedding.Embedding]
+Docstring: The main embedding function, given the model and prompt, return a vector representation
+
+Class: LLM_API
+  Class Docstring: Helper class that provides a standard way to create an ABC using
+inheritance.
+  generate(self, model, system_message, prompt, **kwargs)
+Docstring: The main generation function, given the args, kwargs, function_modeler, function description and model type, generate a response and check if the datapoint can be saved to the finetune dataset
+
+Class: LLM_Finetune_API
+  Class Docstring: Helper class that provides a standard way to create an ABC using
+inheritance.
+  finetune(self, **kwargs) -> tanuki.models.finetune_job.FinetuneJob
+Docstring: Creates a fine-tuning run
+Args:
+    **kwargs: 
+
+Returns:
+  get_finetuned(self, job_id: str, **kwargs) -> tanuki.models.finetune_job.FinetuneJob
+Docstring: Gets a fine-tuning run by id
+  list_finetuned(self, limit=100, **kwargs) -> List[tanuki.models.finetune_job.FinetuneJob]
+Docstring: Gets the last N fine-tuning runs
+
+Class: LLM_API
+  Class Docstring: Helper class that provides a standard way to create an ABC using
+inheritance.
+  generate(self, model, system_message, prompt, **kwargs)
+Docstring: The main generation function, given the args, kwargs, function_modeler, function description and model type, generate a response and check if the datapoint can be saved to the finetune dataset
+
+Class: LLM_Finetune_API
+  Class Docstring: Helper class that provides a standard way to create an ABC using
+inheritance.
+  finetune(self, **kwargs) -> tanuki.models.finetune_job.FinetuneJob
+Docstring: Creates a fine-tuning run
+Args:
+    **kwargs: 
+
+Returns:
+  get_finetuned(self, job_id: str, **kwargs) -> tanuki.models.finetune_job.FinetuneJob
+Docstring: Gets a fine-tuning run by id
+  list_finetuned(self, limit=100, **kwargs) -> List[tanuki.models.finetune_job.FinetuneJob]
+Docstring: Gets the last N fine-tuning runs
+
+Class: LLM_API
+  generate(self, model, system_message, prompt, **kwargs) ->
+    """
+    The main generation function, given the args, kwargs, function_modeler, function description and model type,
+    generate a response and check if the datapoint can be saved to the finetune dataset
+    """
+
+Class: Embedding_API
+  Class Docstring: Helper class that provides a standard way to create an ABC using
+inheritance.
+  embed(self, texts: List[str], model: str = None, **kwargs) -> List[tanuki.models.embedding.Embedding]
+Docstring: The main embedding function, given the model and prompt, return a vector representation
+