-
Chemically Motivated Solutions
- Australia
-
22:15
- 10h ahead - replit.com/@canstralian
- https://orcid.org/0009-0006-5150-9546
- dotcomhunters
- https://replit.com/@canstralian
Highlights
- Pro
Lists (6)
Sort Name ascending (A-Z)
Starred repositories
GitQL is a extensible SQL-like query language and SDK to perform queries on various data sources such .git files with supports of most of SQL features such as grouping, ordering and aggregation andβ¦
Dataset Migration and Analysis Tool
A powerful Streamlit-based tool for analyzing GitHub repositories
A composite GitHub Action to login to the HuggingFace Hub
Finetune Llama-3-8b on the MathInstruct dataset
RAGFlow is an open-source RAG (Retrieval-Augmented Generation) engine based on deep document understanding.
Inspects Python source files and provides information about type and location of classes, methods etc
canstralian / mycoder
Forked from drivecore/mycoderSimple to install, powerful command-line based AI agent system for coding.
Simple to install, powerful command-line based AI agent system for coding.
A curated list of awesome places to learn and/or practice algorithms.
warner-benjamin / nbdev
Forked from AnswerDotAI/nbdevCreate delightful python projects using Jupyter Notebooks
NLP with Transformers Study Group Materials & Resources
warner-benjamin / vllm
Forked from vllm-project/vllmA high-throughput and memory-efficient inference and serving engine for LLMs
Examples and recipes for Llama 2 model
Examples and recipes for Llama 2 model
A GitHub Action that installs and executes flake8 Python source linting during continuous integration testing. Supports flake8 configuration and plugin installation in the GitHub Action settings.
This is the reproduction repository for my π€ Hugging Face blog post on synthetic data
Hunt for C2 servers and phishing web sites using VirusTotal API , you can modify code to kill the malicious process
π Papers & tech blogs by companies sharing their work on data science & machine learning in production.
Find the best OSS coding LLMs by watching them battle
A collection of notebooks/recipes showcasing usecases of open-source models with Together AI.