mb-rag 1.1.26__tar.gz → 1.1.29__tar.gz

{mb_rag-1.1.26 → mb_rag-1.1.29}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: mb_rag
-Version: 1.1.26
+Version: 1.1.29
 Summary: RAG function file
 Author: ['Malav Bateriwala']
 Requires-Python: >=3.8

{mb_rag-1.1.26 → mb_rag-1.1.29}/mb_rag/chatbot/basic.py

@@ -146,10 +146,11 @@ class ModelFactory:
         if not check_package("langchain_ollama"):
             raise ImportError("Langchain Community package not found. Please install it using: pip install langchain_ollama")
         
-        from langchain_ollama import ChatOllama
+        from langchain_ollama import OllamaLLM
+
         print(f"Current Ollama serve model is {os.system('ollama ps')}")
         kwargs["model"] = model_name
-        return ChatOllama(**kwargs)
+        return OllamaLLM(**kwargs)
 
     @classmethod
     def create_groq(cls, model_name: str = "llama-3.3-70b-versatile", **kwargs) -> Any:
@@ -315,21 +316,26 @@ class ModelFactory:
         str: Output from the model
     """
         base64_images = [self._image_to_base64(image) for image in images]
-        image_prompt_create = [{"type": "image_url", "image_url": {"url": f"data:image/jpeg;base64,{base64_images[i]}"}} for i in range(len(images))]
-        prompt_new = [{"type": "text", "text": prompt},
-                      *image_prompt_create,]
-        if pydantic_model is not None:
-            try:
-                self.model = self.model.with_structured_output(pydantic_model)
-            except Exception as e:
-                print(f"Error with pydantic_model: {e}")
-                print("Continuing without structured output")
-        message= HumanMessage(content=prompt_new,)
-        response = self.model.invoke([message])
-        try:
+        if self.model_name=='ollama':
+            ollama_model = self.model.bind(images=[base64_images])
+            response = ollama_model.invoke([HumanMessage(content=prompt)])
             return response.content
-        except Exception:
-            return response
+        else:
+            image_prompt_create = [{"type": "image_url", "image_url": {"url": f"data:image/jpeg;base64,{base64_images[i]}"}} for i in range(len(images))]
+            prompt_new = [{"type": "text", "text": prompt},
+                          *image_prompt_create,]
+            if pydantic_model is not None:
+                try:
+                    self.model = self.model.with_structured_output(pydantic_model)
+                except Exception as e:
+                    print(f"Error with pydantic_model: {e}")
+                    print("Continuing without structured output")
+            message= HumanMessage(content=prompt_new,)
+            response = self.model.invoke([message])
+            try:
+                return response.content
+            except Exception:
+                return response
 
 class ConversationModel:
     """

{mb_rag-1.1.26 → mb_rag-1.1.29}/mb_rag/version.py

@@ -1,5 +1,5 @@
 MAJOR_VERSION = 1
 MINOR_VERSION = 1
-PATCH_VERSION = 26
+PATCH_VERSION = 29
 version = '{}.{}.{}'.format(MAJOR_VERSION, MINOR_VERSION, PATCH_VERSION)
 __all__  = ['MAJOR_VERSION', 'MINOR_VERSION', 'PATCH_VERSION', 'version']

{mb_rag-1.1.26 → mb_rag-1.1.29}/mb_rag.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: mb_rag
-Version: 1.1.26
+Version: 1.1.29
 Summary: RAG function file
 Author: ['Malav Bateriwala']
 Requires-Python: >=3.8