ragxo 0.1.7__tar.gz → 0.1.8__tar.gz

{ragxo-0.1.7 → ragxo-0.1.8}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: ragxo
-Version: 0.1.7
+Version: 0.1.8
 Summary: A RAG (Retrieval-Augmented Generation) toolkit with Milvus integration
 Home-page: https://github.com/yourusername/ragx
 License: MIT
@@ -67,7 +67,7 @@ def preprocess_text_remove_special_chars(text: str) -> str:
     return re.sub(r'[^a-zA-Z0-9\s]', '', text)
 
 def get_embeddings(text: str) -> list[float]:
-    return openai.embeddings.create(text=text, model="text-embedding-ada-002").data[0].embedding
+    return openai.embeddings.create(input=text, model="text-embedding-ada-002").data[0].embedding
 
 ragxo_client = Ragxo(dimension=768)
 
@@ -76,7 +76,14 @@ ragxo_client.add_preprocess(preprocess_text_remove_special_chars)
 ragxo_client.add_embedding_fn(get_embeddings)
 
 ragxo_client.add_system_prompt("You are a helpful assistant that can answer questions about the data provided.")
-ragxo_client.add_model("gpt-4o-mini")
+ragxo_client.add_model(
+    "gpt-4o-mini",
+    temperature=0.5,
+    max_tokens=1000,
+    top_p=1.0,
+    frequency_penalty=0.0,
+    presence_penalty=0.0
+)
 
 ragxo_client.index([
     Document(text="Capital of France is Paris", metadata={"source": "example"}, id=1),
@@ -94,16 +101,10 @@ ragxo_client.export("my_rag_v1.0.0")
 ```python
 loaded_ragxo_client = Ragxo.load("my_rag_v1.0.0")
 
-results = loaded_ragxo_client.query("What is the capital of France?")
+vector_search_results = loaded_ragxo_client.query("What is the capital of France?")
 
 llm_response = loaded_ragxo_client.generate_llm_response(
-    "What is the capital of France?", 
-    limit=10,
-    temperature=0.5, 
-    max_tokens=1000, 
-    top_p=1.0, 
-    frequency_penalty=0.0, 
-    presence_penalty=0.0)
+    "What is the capital of France?")
 
 ```
 

{ragxo-0.1.7 → ragxo-0.1.8}/README.md

@@ -37,7 +37,7 @@ def preprocess_text_remove_special_chars(text: str) -> str:
     return re.sub(r'[^a-zA-Z0-9\s]', '', text)
 
 def get_embeddings(text: str) -> list[float]:
-    return openai.embeddings.create(text=text, model="text-embedding-ada-002").data[0].embedding
+    return openai.embeddings.create(input=text, model="text-embedding-ada-002").data[0].embedding
 
 ragxo_client = Ragxo(dimension=768)
 
@@ -46,7 +46,14 @@ ragxo_client.add_preprocess(preprocess_text_remove_special_chars)
 ragxo_client.add_embedding_fn(get_embeddings)
 
 ragxo_client.add_system_prompt("You are a helpful assistant that can answer questions about the data provided.")
-ragxo_client.add_model("gpt-4o-mini")
+ragxo_client.add_model(
+    "gpt-4o-mini",
+    temperature=0.5,
+    max_tokens=1000,
+    top_p=1.0,
+    frequency_penalty=0.0,
+    presence_penalty=0.0
+)
 
 ragxo_client.index([
     Document(text="Capital of France is Paris", metadata={"source": "example"}, id=1),
@@ -64,16 +71,10 @@ ragxo_client.export("my_rag_v1.0.0")
 ```python
 loaded_ragxo_client = Ragxo.load("my_rag_v1.0.0")
 
-results = loaded_ragxo_client.query("What is the capital of France?")
+vector_search_results = loaded_ragxo_client.query("What is the capital of France?")
 
 llm_response = loaded_ragxo_client.generate_llm_response(
-    "What is the capital of France?", 
-    limit=10,
-    temperature=0.5, 
-    max_tokens=1000, 
-    top_p=1.0, 
-    frequency_penalty=0.0, 
-    presence_penalty=0.0)
+    "What is the capital of France?")
 
 ```
 

{ragxo-0.1.7 → ragxo-0.1.8}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "ragxo"
-version = "0.1.7"
+version = "0.1.8"
 description = "A RAG (Retrieval-Augmented Generation) toolkit with Milvus integration"
 authors = ["Mohamed Sadek <mohamedfawzydes@gmail.com>"]
 readme = "README.md"

{ragxo-0.1.7 → ragxo-0.1.8}/ragxo/client.py

@@ -20,7 +20,26 @@ class Document(BaseModel):
     id: int
 
 class Ragxo:
+    """
+    A RAG (Retrieval-Augmented Generation) system that combines vector search with LLM responses.
+    
+    Attributes:
+        dimension (int): Dimension of the embedding vectors
+        collection_name (str): Name of the Milvus collection
+        db_path (str): Path to the Milvus database
+        processing_fn (list): List of preprocessing functions
+        embedding_fn (Callable): Function to generate embeddings
+        system_prompt (str): System prompt for LLM
+        model (str): LLM model name
+    """
+    
     def __init__(self, dimension: int) -> None:
+        """
+        Initialize the Ragxo instance.
+        
+        Args:
+            dimension (int): Dimension of the embedding vectors
+        """
         self.dimension = dimension
         self.collection_name = "ragx"
         os.makedirs("ragx_artifacts", exist_ok=True)
@@ -34,20 +53,59 @@ class Ragxo:
         self.model = "gpt-4o-mini"
     
     def add_preprocess(self, fn: Callable) -> Self:
+        """
+        Add a preprocessing function to the pipeline.
+        
+        Args:
+            fn (Callable): Function that takes and returns a string
+            
+        Returns:
+            Self: The current instance for method chaining
+        """
         self.processing_fn.append(fn)
         return self
     
     def add_llm_response_fn(self, fn: Callable) -> Self:
+        """
+        Add a function to process LLM responses.
+        
+        Args:
+            fn (Callable): Function to process LLM responses
+            
+        Returns:
+            Self: The current instance for method chaining
+        """
         self.llm_response_fn = fn
         return self
     
     def add_embedding_fn(self, fn: Callable) -> Self:
+        """
+        Set the embedding function for vector generation.
+        
+        Args:
+            fn (Callable): Function that converts text to embeddings
+            
+        Returns:
+            Self: The current instance for method chaining
+            
+        Raises:
+            ValueError: If fn is None
+        """
         if not fn:
             raise ValueError("Embedding function cannot be None")
         self.embedding_fn = fn
         return self
     
     def add_system_prompt(self, prompt: str) -> Self:
+        """
+        Set the system prompt for LLM interactions.
+        
+        Args:
+            prompt (str): System prompt text
+            
+        Returns:
+            Self: The current instance for method chaining
+        """
         self.system_prompt = prompt
         return self
     
@@ -57,6 +115,21 @@ class Ragxo:
                         top_p: float = 1.0,
                         frequency_penalty: float = 0.0,
                         presence_penalty: float = 0.0) -> Self:
+        """
+        Configure the LLM model and its parameters.
+        
+        Args:
+            model (str): Name of the LLM model
+            limit (int): Maximum number of results to return from vector search
+            temperature (float): Sampling temperature
+            max_tokens (int): Maximum tokens in response
+            top_p (float): Nucleus sampling parameter
+            frequency_penalty (float): Frequency penalty parameter
+            presence_penalty (float): Presence penalty parameter
+            
+        Returns:
+            Self: The current instance for method chaining
+        """
         self.model = model
         self.limit = limit
         self.temperature = temperature
@@ -67,6 +140,18 @@ class Ragxo:
         return self
     
     def index(self, data: list[Document]) -> Self:
+        """
+        Index documents into the vector database.
+        
+        Args:
+            data (list[Document]): List of documents to index
+            
+        Returns:
+            Self: The current instance for method chaining
+            
+        Raises:
+            ValueError: If embedding function is not set
+        """
         if not self.embedding_fn:
             raise ValueError("Embedding function not set")
             
@@ -94,6 +179,20 @@ class Ragxo:
         return self
     
     def query(self, query: str, output_fields: list[str] = ['text', 'metadata'], limit: int = 10) -> list[list[dict]]:
+        """
+        Search the vector database for similar documents.
+        
+        Args:
+            query (str): Search query
+            output_fields (list[str]): Fields to return in results
+            limit (int): Maximum number of results
+            
+        Returns:
+            list[list[dict]]: Search results
+            
+        Raises:
+            ValueError: If embedding function is not set
+        """
         if not self.embedding_fn:
             raise ValueError("Embedding function not set. Please call add_embedding_fn first.")
             
@@ -253,6 +352,19 @@ class Ragxo:
     def generate_llm_response(self, 
                               query: str,
                               data: list[dict] = None) -> ChatCompletion:
+        """
+        Generate LLM response based on query and retrieved data.
+        
+        Args:
+            query (str): User query
+            data (list[dict], optional): Retrieved documents. If None, performs a new query
+            
+        Returns:
+            ChatCompletion: LLM response
+            
+        Raises:
+            ValueError: If system prompt is not set
+        """
         if data is None:
             data = self.query(query, limit=self.limit)[0]