SimplerLLM 0.2.0__tar.gz → 0.2.3__tar.gz

simplerllm-0.2.3/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2024 Hasan Aboul Hasan
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

{SimplerLLM-0.2.0 → simplerllm-0.2.3}/PKG-INFO

@@ -1,10 +1,11 @@
 Metadata-Version: 2.1
 Name: SimplerLLM
-Version: 0.2.0
+Version: 0.2.3
 Summary: An easy-to-use Library for interacting with language models.
 Home-page: https://github.com/hassancs91/SimplerLLM
 Author: Hasan Aboul Hasan
 Author-email: hasan@learnwithhasan.com
+License: MIT
 Keywords: text generation,openai,LLM,RAG
 Classifier: Development Status :: 4 - Beta
 Classifier: Intended Audience :: Developers
@@ -15,15 +16,35 @@ Classifier: Programming Language :: Python :: 3.7
 Classifier: Programming Language :: Python :: 3.8
 Classifier: Programming Language :: Python :: 3.9
 Classifier: Programming Language :: Python :: 3.10
+Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
 Classifier: Topic :: Software Development :: Libraries :: Python Modules
 Requires-Python: >=3.6
 Description-Content-Type: text/markdown
+License-File: LICENSE
+Requires-Dist: aiohttp==3.9.4
+Requires-Dist: duckduckgo_search==5.3.0
+Requires-Dist: newspaper3k==0.2.8
+Requires-Dist: numpy==1.26.4
+Requires-Dist: openai==1.25.0
+Requires-Dist: pydantic==2.7.1
+Requires-Dist: PyPDF2==3.0.1
+Requires-Dist: python-dotenv==1.0.1
+Requires-Dist: python_docx==1.1.0
+Requires-Dist: pytube==15.0.0
+Requires-Dist: Requests==2.31.0
+Requires-Dist: youtube_transcript_api==0.6.2
 
 # ⚪ SimplerLLM (Beta)
 
 ⚡ Your Easy Pass to Advanced AI ⚡
 
+
 [![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
+[![Join the Discord chat!](https://img.shields.io/badge/Join-Discord-7289DA.svg)](https://discord.gg/HUrtZXyp3j)
+
+
+
 
 ## 🤔 What is SimplerLLM?
 
@@ -171,6 +192,10 @@ This function splits the text into chunks based on sentences.
 
 This function splits text into chunks based on paragraphs.
 
+### chunk_by_semantics
+
+This functions splits text into chunks based on semantics.
+
 Example
 
 ```python
@@ -197,3 +222,7 @@ chunks = chunker.chunk_by_max_chunk_size(text, 100, True)
 - Document Chunker
 - Advanced Document Loader
 - Integration With More Providers
+- Simple RAG With SimplerVectors
+- Integration with Vector Databases
+- Agent Builder
+- LLM Server 

{SimplerLLM-0.2.0 → simplerllm-0.2.3}/SimplerLLM/image/img_helper_funcs.py

@@ -1,33 +1,33 @@
-import base64
-import os
-
-
-def save_image_from_base64(
-    base64_str: str, file_path: str = None, image_type: str = "png"
-):
-    """
-    Saves an image from a base64 encoded string to a file.
-    If file_path is not provided, the image is saved in the current working directory.
-
-    :param base64_str: The base64 encoded string of the image.
-    :param file_path: The path (including file name) where the image will be saved.
-                      If None, saves in the current working directory with a default name.
-    :param image_type: The image type/format (e.g., 'png', 'jpg'). Default is 'png'.
-    """
-    # Decode the base64 string
-    image_data = base64.b64decode(base64_str)
-
-    # Set the default file path if not provided
-    if file_path is None:
-        file_path = os.path.join(os.getcwd(), f"default_image.{image_type}")
-
-    # Ensure the directory exists
-    os.makedirs(os.path.dirname(file_path), exist_ok=True)
-
-    # Append the file extension if not present
-    if not file_path.lower().endswith(f".{image_type.lower()}"):
-        file_path += f".{image_type}"
-
-    # Write the image data to a file
-    with open(file_path, "wb") as file:
-        file.write(image_data)
+import base64
+import os
+
+
+def save_image_from_base64(
+    base64_str: str, file_path: str = None, image_type: str = "png"
+):
+    """
+    Saves an image from a base64 encoded string to a file.
+    If file_path is not provided, the image is saved in the current working directory.
+
+    :param base64_str: The base64 encoded string of the image.
+    :param file_path: The path (including file name) where the image will be saved.
+                      If None, saves in the current working directory with a default name.
+    :param image_type: The image type/format (e.g., 'png', 'jpg'). Default is 'png'.
+    """
+    # Decode the base64 string
+    image_data = base64.b64decode(base64_str)
+
+    # Set the default file path if not provided
+    if file_path is None:
+        file_path = os.path.join(os.getcwd(), f"default_image.{image_type}")
+
+    # Ensure the directory exists
+    os.makedirs(os.path.dirname(file_path), exist_ok=True)
+
+    # Append the file extension if not present
+    if not file_path.lower().endswith(f".{image_type.lower()}"):
+        file_path += f".{image_type}"
+
+    # Write the image data to a file
+    with open(file_path, "wb") as file:
+        file.write(image_data)

{SimplerLLM-0.2.0 → simplerllm-0.2.3}/SimplerLLM/image/stability_ai.py

@@ -1,84 +1,84 @@
-from pydantic import BaseModel
-import base64
-import requests
-from typing import List
-import os
-import requests
-from dotenv import load_dotenv
-
-# Load environment variables
-load_dotenv()
-
-# Constants
-STABILITY_API_KEY = os.getenv("STABILITY_API_KEY")
-
-
-class ImageData(BaseModel):
-    base64_str: str
-    size_kb: float
-    width: int
-    height: int
-
-
-class ImageList(BaseModel):
-    images: List[ImageData]
-
-
-def generate_images(
-    model_name: str,
-    prompt: str,
-    negative_prompt: str = "",
-    width: int = 512,
-    height: int = 512,
-    samples: int = 1,
-    steps: int = 30,
-    cfg_scale: int = 7,
-    seed: int = 0,
-    style_preset: str = None,
-) -> ImageList:
-
-    engine_id = model_name
-    api_host = "https://api.stability.ai"
-    api_key = STABILITY_API_KEY
-
-    if api_key is None:
-        raise Exception("Missing Stability API key.")
-
-    response = requests.post(
-        f"{api_host}/v1/generation/{engine_id}/text-to-image",
-        headers={
-            "Content-Type": "application/json",
-            "Accept": "application/json",
-            "Authorization": f"Bearer {api_key}",
-        },
-        json={
-            "text_prompts": [{"text": prompt}],
-            "negative_prompt": negative_prompt,
-            "cfg_scale": cfg_scale,
-            "height": height,
-            "width": width,
-            "samples": samples,
-            "steps": steps,
-            "seed": seed,
-            "style_preset": style_preset,
-        },
-    )
-
-    if response.status_code != 200:
-        raise Exception("Non-200 response: " + str(response.text))
-
-    data = response.json()
-
-    images_data = []
-    for image in data["artifacts"]:
-        base64_str = image["base64"]
-        decoded_image = base64.b64decode(base64_str)
-        size_kb = len(decoded_image) / 1024
-
-        images_data.append(
-            ImageData(
-                base64_str=base64_str, size_kb=size_kb, width=width, height=height
-            )
-        )
-
-    return ImageList(images=images_data)
+from pydantic import BaseModel
+import base64
+import requests
+from typing import List
+import os
+import requests
+from dotenv import load_dotenv
+
+# Load environment variables
+load_dotenv()
+
+# Constants
+STABILITY_API_KEY = os.getenv("STABILITY_API_KEY")
+
+
+class ImageData(BaseModel):
+    base64_str: str
+    size_kb: float
+    width: int
+    height: int
+
+
+class ImageList(BaseModel):
+    images: List[ImageData]
+
+
+def generate_images(
+    model_name: str,
+    prompt: str,
+    negative_prompt: str = "",
+    width: int = 512,
+    height: int = 512,
+    samples: int = 1,
+    steps: int = 30,
+    cfg_scale: int = 7,
+    seed: int = 0,
+    style_preset: str = None,
+) -> ImageList:
+
+    engine_id = model_name
+    api_host = "https://api.stability.ai"
+    api_key = STABILITY_API_KEY
+
+    if api_key is None:
+        raise Exception("Missing Stability API key.")
+
+    response = requests.post(
+        f"{api_host}/v1/generation/{engine_id}/text-to-image",
+        headers={
+            "Content-Type": "application/json",
+            "Accept": "application/json",
+            "Authorization": f"Bearer {api_key}",
+        },
+        json={
+            "text_prompts": [{"text": prompt}],
+            "negative_prompt": negative_prompt,
+            "cfg_scale": cfg_scale,
+            "height": height,
+            "width": width,
+            "samples": samples,
+            "steps": steps,
+            "seed": seed,
+            "style_preset": style_preset,
+        },
+    )
+
+    if response.status_code != 200:
+        raise Exception("Non-200 response: " + str(response.text))
+
+    data = response.json()
+
+    images_data = []
+    for image in data["artifacts"]:
+        base64_str = image["base64"]
+        decoded_image = base64.b64decode(base64_str)
+        size_kb = len(decoded_image) / 1024
+
+        images_data.append(
+            ImageData(
+                base64_str=base64_str, size_kb=size_kb, width=width, height=height
+            )
+        )
+
+    return ImageList(images=images_data)

simplerllm-0.2.3/SimplerLLM/language/embeddings.py

@@ -0,0 +1,66 @@
+import SimplerLLM.language.llm_providers.openai_llm as openai_llm
+from enum import Enum
+
+class EmbeddingsProvider(Enum):
+    OPENAI = 1
+
+
+class LLM:
+    def __init__(
+        self, provider=EmbeddingsProvider.OPENAI, model_name="text-embedding-3-small"
+    ):
+        self.provider = provider
+        self.model_name = model_name
+
+    @staticmethod
+    def create(
+        provider=None,
+        model_name=None,
+    ):
+        if provider == EmbeddingsProvider.OPENAI:
+            return OpenAILLM(provider, model_name)
+        else:
+            return None
+
+    def set_model(self, provider):
+        if not isinstance(provider, EmbeddingsProvider):
+            raise ValueError("Provider must be an instance of EmbeddingsProvider Enum")
+        self.provider = provider
+
+
+class OpenAILLM(LLM):
+    def __init__(self, model, model_name):
+        super().__init__(model, model_name)
+
+
+    def generate_embeddings(
+        self,
+        user_input,
+        model_name=None,
+        full_response=False,
+    ):
+        # Use instance values as defaults if not provided
+        model_name = model_name if model_name is not None else self.model_name
+        
+        return openai_llm.generate_embeddings(
+            user_input=user_input,
+            model_name=model_name,
+            full_response=full_response,
+        )
+    
+    async def generate_embeddings_async(
+        self,
+        user_input,
+        model_name=None,
+        full_response=False,
+    ):
+        # Use instance values as defaults if not provided
+        model_name = model_name if model_name is not None else self.model_name
+
+        return await openai_llm.generate_embeddings_async(
+            user_input=user_input,
+            model_name=model_name,
+            full_response=full_response,
+        )
+
+

{SimplerLLM-0.2.0 → simplerllm-0.2.3}/SimplerLLM/language/llm.py

@@ -204,3 +204,5 @@ class AnthropicLLM(LLM):
             }
         )
         return await anthropic_llm.generate_response_async(**params)
+
+

simplerllm-0.2.3/SimplerLLM/language/llm_addons.py

@@ -0,0 +1,131 @@
+import time
+from typing import Type
+from pydantic import BaseModel
+from SimplerLLM.language.llm import LLM
+import asyncio
+
+from SimplerLLM.tools.json_helpers import (
+    extract_json_from_text,
+    convert_json_to_pydantic_model,
+    validate_json_with_pydantic_model,
+    generate_json_example_from_pydantic,
+)
+
+
+
+def generate_pydantic_json_model(
+    model_class: Type[BaseModel],
+    prompt: str,
+    llm_instance: LLM,
+    max_retries: int = 3,
+    initial_delay: float = 1.0,
+    custom_prompt_suffix: str = None,
+) -> BaseModel:
+    """
+    Generates a model instance based on a given prompt, retrying on validation errors.
+
+    :param model_class: The Pydantic model class to be used for validation and conversion.
+    :param prompt: The fully formatted prompt including the topic.
+    :param llm_instance: Instance of a large language model.
+    :param max_retries: Maximum number of retries on validation errors.
+    :param initial_delay: Initial delay in seconds before the first retry.
+    :param custom_prompt_suffix: Optional string to customize or override the generated prompt extension.
+
+    :return: BaseModel object if successful, otherwise error message.
+    """
+    # Concatenate prompt and JSON model outside the loop
+    json_model = generate_json_example_from_pydantic(model_class)
+    optimized_prompt = custom_prompt_suffix or (prompt + f"\n\nThe response should be in a structured JSON format that matches the following JSON: {json_model}")
+    
+    # Calculate exponential backoff before the loop
+    backoff_delays = [initial_delay * (2**attempt) for attempt in range(max_retries + 1)]
+
+    for attempt, delay in enumerate(backoff_delays):
+        try:
+            ai_response = llm_instance.generate_response(prompt=optimized_prompt)
+
+            if ai_response:
+                json_object = extract_json_from_text(ai_response)
+
+                validated, errors = validate_json_with_pydantic_model(
+                    model_class, json_object
+                )
+
+                if not errors:
+                    model_object = convert_json_to_pydantic_model(
+                        model_class, json_object[0]
+                    )
+                    return model_object
+
+        except Exception as e:  # Replace SpecificException with the appropriate exception
+            return f"Exception occurred: {e}"
+
+        # Retry logic for empty AI response or validation errors
+        if not ai_response or (errors and attempt < max_retries):
+            time.sleep(delay)
+        elif errors:
+            return f"Validation failed after {max_retries} retries: {errors}"
+
+    return "Maximum retries exceeded without successful validation."
+
+async def generate_pydantic_json_model_async(
+    model_class: Type[BaseModel],
+    prompt: str,
+    llm_instance: LLM,
+    max_retries: int = 3,
+    initial_delay: float = 1.0,
+    custom_prompt_suffix: str = None,
+) -> BaseModel:
+    """
+    Generates a model instance based on a given prompt, retrying on validation errors.
+
+    :param model_class: The Pydantic model class to be used for validation and conversion.
+    :param prompt: The fully formatted prompt including the topic.
+    :param llm_instance: Instance of a large language model.
+    :param max_retries: Maximum number of retries on validation errors.
+    :param initial_delay: Initial delay in seconds before the first retry.
+    :param custom_prompt_suffix: Optional string to customize or override the generated prompt extension.
+
+    :return: BaseModel object if successful, otherwise error message.
+    """
+    # Concatenate prompt and JSON model outside the loop
+    json_model = generate_json_example_from_pydantic(model_class)
+    optimized_prompt = custom_prompt_suffix or (prompt + f"\n\nThe response should be in a structured JSON format that matches the following JSON: {json_model}")
+    
+    # Calculate exponential backoff before the loop
+    backoff_delays = [initial_delay * (2**attempt) for attempt in range(max_retries + 1)]
+
+    for attempt, delay in enumerate(backoff_delays):
+        try:
+            ai_response = await llm_instance.generate_response_async(prompt=optimized_prompt)
+
+            if ai_response:
+                json_object = extract_json_from_text(ai_response)
+
+                validated, errors = validate_json_with_pydantic_model(
+                    model_class, json_object
+                )
+
+                if not errors:
+                    model_object = convert_json_to_pydantic_model(
+                        model_class, json_object[0]
+                    )
+                    return model_object
+
+        except Exception as e:  # Replace SpecificException with the appropriate exception
+            return f"Exception occurred: {e}"
+
+        # Retry logic for empty AI response or validation errors
+        if not ai_response or (errors and attempt < max_retries):
+            await asyncio.sleep(delay)
+        elif errors:
+            return f"Validation failed after {max_retries} retries: {errors}"
+
+    return "Maximum retries exceeded without successful validation."
+
+
+
+
+
+
+

{SimplerLLM-0.2.0 → simplerllm-0.2.3}/SimplerLLM/language/llm_providers/anthropic_llm.py

@@ -11,7 +11,7 @@ from .llm_response_models import LLMFullResponse
 load_dotenv()
 
 # Constants
-CLAUDE_API_KEY = os.getenv("CLAUDE_API_KEY", "")
+ANTHROPIC_API_KEY = os.getenv("ANTHROPIC_API_KEY", "")
 MAX_RETRIES = int(os.getenv("MAX_RETRIES", 3))
 RETRY_DELAY = int(os.getenv("RETRY_DELAY", 2))
 
@@ -46,7 +46,7 @@ def generate_response(
     # Define the URL and headers
     url = "https://api.anthropic.com/v1/messages"
     headers = {
-        "x-api-key": CLAUDE_API_KEY,
+        "x-api-key": ANTHROPIC_API_KEY,
         "anthropic-version": "2023-06-01",
         "content-type": "application/json",
     }
@@ -77,7 +77,7 @@ def generate_response(
             else:
                 return response.json()["content"][0]["text"]
 
-        except requests.RequestException as e:
+        except Exception as e:
             print(f"Attempt {attempt + 1} failed: {e}")
             time.sleep(retry_delay)
             retry_delay *= 2  # Double the delay each retry
@@ -116,7 +116,7 @@ async def generate_response_async(
     # Define the URL and headers
     url = "https://api.anthropic.com/v1/messages"
     headers = {
-        "x-api-key": CLAUDE_API_KEY,
+        "x-api-key": ANTHROPIC_API_KEY,
         "anthropic-version": "2023-06-01",
         "content-type": "application/json",
     }

{SimplerLLM-0.2.0 → simplerllm-0.2.3}/SimplerLLM/language/llm_providers/llm_response_models.py

@@ -1,9 +1,16 @@
-from pydantic import BaseModel
-from typing import Any
-
-
-class LLMFullResponse(BaseModel):
-    generated_text: str
-    model: str
-    process_time: float
-    llm_provider_response: Any
+from pydantic import BaseModel
+from typing import Any
+
+
+class LLMFullResponse(BaseModel):
+    generated_text: str
+    model: str
+    process_time: float
+    llm_provider_response: Any
+
+
+class LLMEmbeddingsResponse(BaseModel):
+    generated_embedding: Any
+    model: str
+    process_time: float
+    llm_provider_response: Any