Migrating from llm-wrapper to allms. All code refactored, docs and RE…

…ADME.md updated

.github/workflows/release.yml

@@ -20,9 +20,9 @@ jobs:
         run: make install-poetry
       - name: Install dependencies
         run: make install-env
-      - name: Build llm-wrapper package
+      - name: Build allms package
         run: make build
-      - name: Publish llm-wrapper package to PyPI
+      - name: Publish allms package to PyPI
         env:
           TWINE_PASSWORD: ${{ secrets.PYPI_API_TOKEN }}
         run: |

Makefile

@@ -12,7 +12,7 @@ build::
 	poetry run python -m build --sdist --wheel .
 
 linter::
-	poetry run pylint llm_wrapper --reports=no --output-format=colorized --fail-under=8.0
+	poetry run pylint allms --reports=no --output-format=colorized --fail-under=8.0
 
 tests::
 	poetry run python -m pytest -s --verbose

README.md

@@ -1,16 +1,16 @@
-# llm-wrapper
+# allms
 
 ___
 ## About
 
-llm-wrapper is a versatile and powerful library designed to streamline the process of querying Large Language Models
+allms is a versatile and powerful library designed to streamline the process of querying Large Language Models
 (LLMs) 🤖💬
 
-Developed by the Allegro engineers, llm-wrapper is based on popular libraries like transformers, pydantic, and langchain. It takes care 
+Developed by the Allegro engineers, allms is based on popular libraries like transformers, pydantic, and langchain. It takes care 
 of the boring boiler-plate code you write around your LLM applications, quickly enabling you to prototype ideas, and eventually helping you to scale up 
 for production use-cases!
 
-Among the llm-wrapper most notable features, you will find:
+Among the allms most notable features, you will find:
 
 * **😊 Simple and User-Friendly Interface**: The module offers an intuitive and easy-to-use interface, making it straightforward to work with the model.
 
@@ -28,7 +28,7 @@ ___
 
 Full documentation available at **[llm-wrapper.allegro.tech](https://llm-wrapper.allegro.tech/)**
 
-Get familiar with llm-wrapper 🚀: [introductory jupyter notebook](https://github.com/allegro/llm-wrapper/blob/main/examples/introduction.ipynb)
+Get familiar with allms 🚀: [introductory jupyter notebook](https://github.com/allegro/allms/blob/main/examples/introduction.ipynb)
 
 ___
 
@@ -39,23 +39,23 @@ ___
 Install the package via pip:
 
 ```
-pip install llm-wrapper
+pip install allms
 ```
 
 ### Basic Usage ⭐
 
 Configure endpoint credentials and start querying the model with any prompt:
 
 ```python
-from llm_wrapper.models import AzureOpenAIModel
-from llm_wrapper.domain.configuration import AzureOpenAIConfiguration
+from allms.models import AzureOpenAIModel
+from allms.domain.configuration import AzureOpenAIConfiguration
 
 configuration = AzureOpenAIConfiguration(
-    api_key="your-secret-api-key",
-    base_url="https://endpoint.openai.azure.com/",
-    api_version="2023-03-15-preview",
-    deployment="gpt-35-turbo",
-    model_name="gpt-3.5-turbo"
+   api_key="your-secret-api-key",
+   base_url="https://endpoint.openai.azure.com/",
+   api_version="2023-03-15-preview",
+   deployment="gpt-35-turbo",
+   model_name="gpt-3.5-turbo"
 )
 
 gpt_model = AzureOpenAIModel(config=configuration)
@@ -102,7 +102,7 @@ responses = model.generate(prompt=prompt, input_data=input_data)
 
 ### Forcing Structured Output Format
 
-Through pydantic integration, in llm-wrapper you can pass an output dataclass and force the LLM to provide
+Through pydantic integration, in allms you can pass an output dataclass and force the LLM to provide
 you the response in a structured way.
 
 ```python
@@ -151,7 +151,7 @@ ___
 
 We assume that you have python `3.10.*` installed on your machine. 
 You can set it up using [pyenv](https://github.com/pyenv/pyenv#installationbrew) 
-([How to install pyenv on MacOS](https://jordanthomasg.medium.com/python-development-on-macos-with-pyenv-2509c694a808)). To install llm-wrapper env locally:
+([How to install pyenv on MacOS](https://jordanthomasg.medium.com/python-development-on-macos-with-pyenv-2509c694a808)). To install allms env locally:
 
 * Activate your pyenv;
 * Install Poetry via:
@@ -160,7 +160,7 @@ You can set it up using [pyenv](https://github.com/pyenv/pyenv#installationbrew)
 make install-poetry
 ```
 
-* Install llm-wrapper dependencies with the command:
+* Install allms dependencies with the command:
 
 ```bash
 make install-env
@@ -189,7 +189,7 @@ via the github action `.github/workflows/docs.yml`
 
 ### Make a new release
 
-When a new version of llm-wrapper is ready to be released, do the following operations:
+When a new version of allms is ready to be released, do the following operations:
 
 1. **Merge to master** the dev branch in which the new version has been specified:
     1. In this branch, `version` under `[tool.poetry]` section in `pyproject.toml` should be updated, e.g `0.1.0`;
@@ -207,5 +207,5 @@ When a new version of llm-wrapper is ready to be released, do the following oper
     1. Go to _Releases_ → _Draft a new release_;
     2. Select the recently created tag in _Choose a tag_ window;
     3. Copy/paste all the content present in the CHANGELOG under the version you are about to release;
-    4. Upload `llm_wrapper-<NEW-VERSION>.whl` and `llm_wrapper-<NEW-VERSION>.tar.gz` as assets;
+    4. Upload `allms-<NEW-VERSION>.whl` and `allms-<NEW-VERSION>.tar.gz` as assets;
     5. Click `Publish release`.

llm_wrapper/chains/long_text_processing_chain.py → allms/chains/long_text_processing_chain.py

@@ -7,10 +7,10 @@
 from langchain.chains.combine_documents.base import BaseCombineDocumentsChain
 from langchain.schema import Document
 
-from llm_wrapper.domain.enumerables import AggregationLogicForLongInputData, LanguageModelTask
-from llm_wrapper.domain.input_data import InputData
-from llm_wrapper.domain.prompt_dto import AggregateOutputClass, KeywordsOutputClass, SummaryOutputClass
-from llm_wrapper.utils.long_text_processing_utils import split_text_to_max_size
+from allms.domain.enumerables import AggregationLogicForLongInputData, LanguageModelTask
+from allms.domain.input_data import InputData
+from allms.domain.prompt_dto import AggregateOutputClass, KeywordsOutputClass, SummaryOutputClass
+from allms.utils.long_text_processing_utils import split_text_to_max_size
 
 
 class LongTextProcessingChain(BaseCombineDocumentsChain):

llm_wrapper/domain/response.py → allms/domain/response.py

@@ -2,7 +2,7 @@
 
 from pydantic import BaseModel
 
-from llm_wrapper.domain.input_data import InputData
+from allms.domain.input_data import InputData
 
 
 class ResponseData(BaseModel):

allms/models/__init__.py

@@ -0,0 +1,7 @@
+from allms.models.azure_llama2 import AzureLlama2Model
+from allms.models.azure_mistral import AzureMistralModel
+from allms.models.azure_openai import AzureOpenAIModel
+from allms.models.vertexai_gemini import VertexAIGeminiModel
+from allms.models.vertexai_palm import VertexAIPalmModel
+
+__all__ = ["AzureOpenAIModel", "AzureLlama2Model", "AzureMistralModel", "VertexAIPalmModel", "VertexAIGeminiModel"]

llm_wrapper/models/abstract.py → allms/models/abstract.py

@@ -17,21 +17,21 @@
 from langchain.schema import OutputParserException
 from pydantic import BaseModel
 
-from llm_wrapper.chains.long_text_processing_chain import (
+from allms.chains.long_text_processing_chain import (
     LongTextProcessingChain,
     load_long_text_processing_chain
 )
-from llm_wrapper.constants.input_data import IODataConstants
-from llm_wrapper.constants.prompt import PromptConstants
-from llm_wrapper.defaults.general_defaults import GeneralDefaults
-from llm_wrapper.domain.enumerables import AggregationLogicForLongInputData, LanguageModelTask
-
-from llm_wrapper.defaults.long_text_chain import LongTextChainDefaults
-from llm_wrapper.domain.input_data import InputData
-from llm_wrapper.domain.prompt_dto import SummaryOutputClass, KeywordsOutputClass
-from llm_wrapper.domain.response import ResponseData
-import llm_wrapper.exceptions.validation_input_data_exceptions as input_exception_message
-from llm_wrapper.utils.long_text_processing_utils import get_max_allowed_number_of_tokens
+from allms.constants.input_data import IODataConstants
+from allms.constants.prompt import PromptConstants
+from allms.defaults.general_defaults import GeneralDefaults
+from allms.domain.enumerables import AggregationLogicForLongInputData, LanguageModelTask
+
+from allms.defaults.long_text_chain import LongTextChainDefaults
+from allms.domain.input_data import InputData
+from allms.domain.prompt_dto import SummaryOutputClass, KeywordsOutputClass
+from allms.domain.response import ResponseData
+import allms.exceptions.validation_input_data_exceptions as input_exception_message
+from allms.utils.long_text_processing_utils import get_max_allowed_number_of_tokens
 
 logger = logging.getLogger(__name__)
 

llm_wrapper/models/azure_llama2.py → allms/models/azure_llama2.py

@@ -1,11 +1,11 @@
 import typing
 from asyncio import AbstractEventLoop
 
-from llm_wrapper.defaults.azure_defaults import AzureLlama2Defaults
-from llm_wrapper.defaults.general_defaults import GeneralDefaults
-from llm_wrapper.domain.configuration import AzureSelfDeployedConfiguration
-from llm_wrapper.models.abstract import AbstractModel
-from llm_wrapper.models.azure_base import AzureMLOnlineEndpointAsync, ChatModelContentFormatter
+from allms.defaults.azure_defaults import AzureLlama2Defaults
+from allms.defaults.general_defaults import GeneralDefaults
+from allms.domain.configuration import AzureSelfDeployedConfiguration
+from allms.models.abstract import AbstractModel
+from allms.models.azure_base import AzureMLOnlineEndpointAsync, ChatModelContentFormatter
 
 
 class AzureLlama2Model(AbstractModel):

llm_wrapper/models/azure_mistral.py → allms/models/azure_mistral.py

@@ -1,11 +1,11 @@
 import typing
 from asyncio import AbstractEventLoop
 
-from llm_wrapper.defaults.azure_defaults import AzureMistralAIDefaults
-from llm_wrapper.defaults.general_defaults import GeneralDefaults
-from llm_wrapper.domain.configuration import AzureSelfDeployedConfiguration
-from llm_wrapper.models.abstract import AbstractModel
-from llm_wrapper.models.azure_base import AzureMLOnlineEndpointAsync, ChatModelContentFormatter
+from allms.defaults.azure_defaults import AzureMistralAIDefaults
+from allms.defaults.general_defaults import GeneralDefaults
+from allms.domain.configuration import AzureSelfDeployedConfiguration
+from allms.models.abstract import AbstractModel
+from allms.models.azure_base import AzureMLOnlineEndpointAsync, ChatModelContentFormatter
 
 
 class AzureMistralModel(AbstractModel):

llm_wrapper/models/azure_openai.py → allms/models/azure_openai.py

@@ -3,10 +3,10 @@
 
 from langchain.chat_models import AzureChatOpenAI
 
-from llm_wrapper.defaults.azure_defaults import AzureGptTurboDefaults
-from llm_wrapper.defaults.general_defaults import GeneralDefaults
-from llm_wrapper.domain.configuration import AzureOpenAIConfiguration
-from llm_wrapper.models.abstract import AbstractModel
+from allms.defaults.azure_defaults import AzureGptTurboDefaults
+from allms.defaults.general_defaults import GeneralDefaults
+from allms.domain.configuration import AzureOpenAIConfiguration
+from allms.models.abstract import AbstractModel
 
 
 class AzureOpenAIModel(AbstractModel):

llm_wrapper/models/vertexai_base.py → allms/models/vertexai_base.py

@@ -5,7 +5,7 @@
 from langchain_core.outputs import LLMResult, Generation
 from pydash import chain
 
-from llm_wrapper.constants.vertex_ai import VertexModelConstants
+from allms.constants.vertex_ai import VertexModelConstants
 
 class CustomVertexAI(VertexAI):
     async def _agenerate(

llm_wrapper/models/vertexai_gemini.py → allms/models/vertexai_gemini.py

@@ -2,11 +2,11 @@
 from langchain_community.llms.vertexai import VertexAI
 from typing import Optional
 
-from llm_wrapper.defaults.general_defaults import GeneralDefaults
-from llm_wrapper.defaults.vertex_ai import GeminiModelDefaults
-from llm_wrapper.domain.configuration import VertexAIConfiguration
-from llm_wrapper.models.vertexai_base import CustomVertexAI
-from llm_wrapper.models.abstract import AbstractModel
+from allms.defaults.general_defaults import GeneralDefaults
+from allms.defaults.vertex_ai import GeminiModelDefaults
+from allms.domain.configuration import VertexAIConfiguration
+from allms.models.vertexai_base import CustomVertexAI
+from allms.models.abstract import AbstractModel
 
 
 class VertexAIGeminiModel(AbstractModel):

llm_wrapper/models/vertexai_palm.py → allms/models/vertexai_palm.py

@@ -2,11 +2,11 @@
 from langchain_community.llms.vertexai import VertexAI
 from typing import Optional
 
-from llm_wrapper.defaults.general_defaults import GeneralDefaults
-from llm_wrapper.defaults.vertex_ai import PalmModelDefaults
-from llm_wrapper.domain.configuration import VertexAIConfiguration
-from llm_wrapper.models.vertexai_base import CustomVertexAI
-from llm_wrapper.models.abstract import AbstractModel
+from allms.defaults.general_defaults import GeneralDefaults
+from allms.defaults.vertex_ai import PalmModelDefaults
+from allms.domain.configuration import VertexAIConfiguration
+from allms.models.vertexai_base import CustomVertexAI
+from allms.models.abstract import AbstractModel
 
 
 class VertexAIPalmModel(AbstractModel):

llm_wrapper/utils/io_utils.py → allms/utils/io_utils.py

@@ -5,8 +5,8 @@
 import fsspec
 import pandas as pd
 
-from llm_wrapper.constants.input_data import IODataConstants
-from llm_wrapper.domain.input_data import InputData
+from allms.constants.input_data import IODataConstants
+from allms.domain.input_data import InputData
 
 logger = logging.getLogger(__name__)
 

llm_wrapper/utils/long_text_processing_utils.py → allms/utils/long_text_processing_utils.py

@@ -4,7 +4,7 @@
 from langchain.base_language import BaseLanguageModel
 from langchain.schema import Document
 
-from llm_wrapper.defaults.long_text_chain import LongTextChainDefaults
+from allms.defaults.long_text_chain import LongTextChainDefaults
 
 
 def truncate_text_to_max_size(

docs/api/input_output_dataclasses.md

@@ -1,4 +1,4 @@
-## `class llm_wrapper.domain.input_data.InputData` dataclass
+## `class allms.domain.input_data.InputData` dataclass
 ```python
 @dataclass
 class InputData:
@@ -12,7 +12,7 @@ class InputData:
 - `id` (`str`): Unique identifier. Requests are done in an async mode, so the order of the responses is not the same
    as the order of the input data, so this field can be used to identify them.
 
-## `class llm_wrapper.domain.response.ResponseData` dataclass
+## `class allms.domain.response.ResponseData` dataclass
 ```python
 @dataclass
 class ResponseData:

docs/api/models/azure_llama2_model.md

@@ -1,4 +1,4 @@
-## `class llm_wrapper.models.AzureLlama2Model` API
+## `class allms.models.AzureLlama2Model` API
 ### Methods
 ```python
 __init__(
@@ -48,6 +48,7 @@ is not provided, the length of this list is equal 1, and the first element is th
 
 ---
 
+## `class allms.domain.configuration.AzureSelfDeployedConfiguration` API
 ```python
 AzureSelfDeployedConfiguration(
     api_key: str,
@@ -63,9 +64,10 @@ AzureSelfDeployedConfiguration(
 ---
 
 ### Example usage
+
 ```python
-from llm_wrapper.models import AzureLlama2Model 
-from llm_wrapper.domain.configuration import AzureSelfDeployedConfiguration
+from allms.models import AzureLlama2Model
+from allms.domain.configuration import AzureSelfDeployedConfiguration
 
 configuration = AzureSelfDeployedConfiguration(
     api_key="<AZURE_API_KEY>",

docs/api/models/azure_mistral_model.md

@@ -1,4 +1,4 @@
-## `class llm_wrapper.models.AzureMistralModel` API
+## `class allms.models.AzureMistralModel` API
 ### Methods
 ```python
 __init__(
@@ -48,6 +48,7 @@ is not provided, the length of this list is equal 1, and the first element is th
 
 ---
 
+## `class allms.domain.configuration.AzureSelfDeployedConfiguration` API
 ```python
 AzureSelfDeployedConfiguration(
     api_key: str,
@@ -63,9 +64,10 @@ AzureSelfDeployedConfiguration(
 ---
 
 ### Example usage
+
 ```python
-from llm_wrapper.models import AzureMistralModel 
-from llm_wrapper.domain.configuration import AzureSelfDeployedConfiguration
+from allms.models import AzureMistralModel
+from allms.domain.configuration import AzureSelfDeployedConfiguration
 
 configuration = AzureSelfDeployedConfiguration(
     api_key="<AZURE_API_KEY>",