Unleash the Potential of Datasets with Intelligent Tasks, Tutorials, and Algorithm Recommendations.
Project description
🚀 ProTaska-GPT
Your AI-powered data companion 🤖
Specify your dataset of choice, and ProTaska-GPT generates a tailored codebase, empowering you to visualize and understand the dataset with tasks, tutorials, and actionable insights. Accelerate your data science journey with ease and efficiency!
🖊️ Key Features:
- Dataset Ingestion: ProTaska-GPT seamlessly integrates with dataset sources like Kaggle and Hugging Face (for now), allowing users to easily import and work with diverse datasets.
- Task Recommendations: Leveraging its GPT-backbone, it generates a customized set of tasks tailored to each dataset, providing users with valuable project ideas and challenges.
- Algorithm Suggestions: Based on the dataset characteristics, it suggests suitable machine learning algorithms, enabling users to make informed decisions during their project journey.
- Conversational Chatbot: Allow user to discuss about different techniques and scrape information from Wikipedia to give relevant responses.
🔎 Objectives:
- Beginner-Friendly Tutorials: ProTaska-GPT aims to offer automated generation of a collection of beginner-friendly tutorials that guide users through common data science workflows, step-by-step, fostering practical learning and skill development.
Installation:
pip install ProTaska-GPT --upgrade
📈 Usage:
Importing base descriptors and ideation bot:
from protaska.describer import describe_dataset
from protaska.ideate import main as chatbot
Providing meta-data about the dataset to be used:
openai_key = '**open-ai-secret-key**'
importer_type = "HuggingFaceDatasetImporter"
dataset_key = 'mteb/tweet_sentiment_extraction'
destination_path = './downloaded_data/'
Getting automated data descriptions:
description, dataloader_obj = describe_dataset(openai_key, importer_type, destination_path, dataset_key)
description
Running an interactive ChatBot for ideation and base-code building:
chatbot(openai_key, description, dataloader_obj.superficial_meta_data, agent_verbose=False)
💁 Contributing
This is an open-source project and we would be really grateful to any contributions.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for ProTaska_GPT-0.0.11-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | 5cd25ee1d825925ceb739d4f6eb50b5e7615b03387fcca37bc50c0ca50484c2f |
|
MD5 | cee2b00373984a699320a58744420eca |
|
BLAKE2b-256 | 814a49d0ff6a95008ad37d773cfc9b4094a2ae567bfd5bff732e6cf590502584 |