Release v0.0.1

Author: radi-cho

README.md

@@ -14,13 +14,24 @@ Possible use cases may include:
 
 ## Installation
 
-Under active development. Coming to `PyPI` soon. Currently you can run `datasetGPT` by cloning this repository as described [here](#contributing).
+```
+pip install datasetGPT
+```
+
+Most of the generation features rely on third-party APIs. Install their respective packages:
+
+```
+pip install openai cohere petals
+```
 
 ## Usage examples
 
 ### Inference LLMs at scale
 
 ```bash
+export OPENAI_API_KEY="..."
+export COHERE_API_KEY="..."
+
 datasetGPT texts \             
     --prompt "If {country} was a planet in the Star Wars universe it would be called" \
     --backend "openai|text-davinci-003" \
@@ -140,6 +151,8 @@ for conversation in conversations_generator:
 
 ## Contributing
 
+> Still under active development.
+
 Contributions will be highly appreciated. Currently these features are under development:
 - [x] `datasetGPT conversations` - Make two ChatGPT agents talk with one another and record the conversation history.
 - [x] `datasetGPT texts` - Inference different LLMs with a given input prompt and generate multiple outputs by varying parameters.

pyproject.toml

@@ -0,0 +1,3 @@
+[build-system]
+requires = ['setuptools>=61.0']
+build-backend = 'setuptools.build_meta'

setup.py

@@ -1,19 +1,25 @@
 from distutils.core import setup
+from setuptools import find_packages
+
+with open("README.md", "r", encoding = "utf-8") as readme:
+    long_description = readme.read()
 
 setup(
     name="datasetGPT",
-    packages=["datasetGPT"],
     version="0.0.1",
     description="Generate textual and conversational datasets with LLMs.",
+    long_description = long_description,
+    long_description_content_type = "text/markdown",
     author="Radostin Cholakov",
     author_email="[email protected]",
     url="https://github.com/radi-cho/datasetGPT",
     # download_url="https://github.com/radi-cho/datasetGPT/archive/v0.0.1.tar.gz",
     keywords=["dataset", "llm", "langchain", "openai"],
+    package_dir={"": "src"},
+    packages = find_packages(where="src"),
     install_requires=[
-        "langchain",
-        "openai",
-        "click"
+        "langchain>=0.0.113",
+        "click>=8.1"
     ],
     entry_points={
         "console_scripts": [