[CODE QUALITY][Refactor Migrations][LayoutLMDocumentQA]

docs/swarms/models/vllm.md

@@ -32,7 +32,7 @@ vLLM is designed to simplify the process of generating text using language model
 Before using vLLM, you need to install swarms. You can install vLLM using `pip`:
 
 ```bash
-pip install swarms
+pip install swarms vllm
 ```
 
 ### vLLM Class <a name="vllm-class"></a>

swarms/models/base_tts.py

@@ -26,6 +26,7 @@ class BaseTTSModel(AbstractLLM):
         save_to_file: save the speech data to a file.
 
     """
+
     def __init__(
         self,
         model_name,
@@ -37,6 +38,8 @@ class BaseTTSModel(AbstractLLM):
         self.model_name = model_name
         self.voice = voice
         self.chunk_size = chunk_size
+        self.save_to_file = save_to_file
+        self.saved_filepath = saved_filepath
 
     def save(self, filepath: Optional[str] = None):
         """Save the model to a file.
@@ -47,6 +50,11 @@ class BaseTTSModel(AbstractLLM):
         pass
 
     def load(self, filepath: Optional[str] = None):
+        """Load the model from a file.
+
+        Args:
+            filepath (Optional[str], optional): _description_. Defaults to None.
+        """
         pass
 
     @abstractmethod

swarms/models/layoutlm_document_qa.py

@@ -3,9 +3,10 @@ LayoutLMDocumentQA is a multimodal good for
 visual question answering on real world docs lik invoice, pdfs, etc
 """
 from transformers import pipeline
+from swarms.models.base_multimodal_model import BaseMultiModalModel
 
 
-class LayoutLMDocumentQA:
+class LayoutLMDocumentQA(BaseMultiModalModel):
     """
     LayoutLMDocumentQA for document question answering:
 
@@ -25,13 +26,24 @@ class LayoutLMDocumentQA:
         self,
         model_name: str = "impira/layoutlm-document-qa",
         task_type: str = "document-question-answering",
+        *args,
+        **kwargs,
     ):
+        super(LayoutLMDocumentQA, self).__init__(*args, **kwargs)
         self.model_name = model_name
         self.task_type = task_type
         self.pipeline = pipeline(task_type, model=model_name)
 
-    def __call__(self, task: str, img_path: str):
-        """Call for model"""
+    def __call__(self, task: str, img_path: str, *args, **kwargs):
+        """Call the LayoutLMDocumentQA model
+
+        Args:
+            task (str): _description_
+            img_path (str): _description_
+
+        Returns:
+            _type_: _description_
+        """
         out = self.pipeline(img_path, task)
         out = str(out)
         return out

swarms/models/multion.py

@@ -1,13 +1,14 @@
+from swarms.models.base_llm import AbstractLLM
+
+
 try:
     import multion
 
 except ImportError:
     raise ImportError(
         "Cannot import multion, please install 'pip install'"
-        
     )
 
-from swarms.models.base_llm import AbstractLLM
 
 class MultiOn(AbstractLLM):
     """
@@ -26,12 +27,34 @@ class MultiOn(AbstractLLM):
     "Order chicken tendies. https://www.google.com/"
 
     """
+
     def __init__(self, **kwargs):
         super(MultiOn, self).__init__(**kwargs)
 
     def run(self, task: str, url: str, *args, **kwargs) -> str:
+        """Run the multion model
+
+        Args:
+            task (str): _description_
+            url (str): _description_
+
+        Returns:
+            str: _description_
+        """
         response = multion.new_session({"input": task, "url": url})
         return response
-    def generate_summary(self, task: str, url: str, *args, **kwargs) -> str:
+
+    def generate_summary(
+        self, task: str, url: str, *args, **kwargs
+    ) -> str:
+        """Generate a summary from the multion model
+
+        Args:
+            task (str): _description_
+            url (str): _description_
+
+        Returns:
+            str: _description_
+        """
         response = multion.new_session({"input": task, "url": url})
         return response

swarms/models/openai_tts.py

@@ -1,9 +1,10 @@
 import os
+import subprocess
 import sys
-import openai
+
 import requests
 from dotenv import load_dotenv
-import subprocess
+
 from swarms.models.base_llm import AbstractLLM
 
 try:

swarms/structs/sequential_workflow.py

@@ -361,7 +361,7 @@ class SequentialWorkflow:
             )
 
     def workflow_bootup(self, **kwargs) -> None:
-        """Boots up the workflow."""
+        """Bootup the workflow."""
         print(
             colored(
                 """

tests/models/test_multion.py

@@ -2,40 +2,53 @@ import pytest
 from unittest.mock import Mock, patch
 from swarms.models.multion import MultiOn
 
+
 @pytest.fixture
 def multion_instance():
     return MultiOn()
 
+
 @pytest.fixture
 def mock_multion():
     return Mock()
 
+
 def test_multion_import():
     with pytest.raises(ImportError):
         import multion
 
+
 def test_multion_init():
     multion = MultiOn()
     assert isinstance(multion, MultiOn)
 
+
 def test_multion_run_with_valid_input(multion_instance, mock_multion):
     task = "Order chicken tendies"
     url = "https://www.google.com/"
-    mock_multion.new_session.return_value = "Order chicken tendies. https://www.google.com/"
+    mock_multion.new_session.return_value = (
+        "Order chicken tendies. https://www.google.com/"
+    )
 
-    with patch('swarms.models.multion.multion', mock_multion):
+    with patch("swarms.models.multion.multion", mock_multion):
         response = multion_instance.run(task, url)
 
-    assert response == "Order chicken tendies. https://www.google.com/"
+    assert (
+        response == "Order chicken tendies. https://www.google.com/"
+    )
+
 
-def test_multion_run_with_invalid_input(multion_instance, mock_multion):
+def test_multion_run_with_invalid_input(
+    multion_instance, mock_multion
+):
     task = ""
     url = "https://www.google.com/"
     mock_multion.new_session.return_value = None
 
-    with patch('swarms.models.multion.multion', mock_multion):
+    with patch("swarms.models.multion.multion", mock_multion):
         response = multion_instance.run(task, url)
 
     assert response is None
 
+
 # Add more test cases to cover different scenarios, edge cases, and error handling as needed.