deep-symbolic-mathematics/Multimodal-Math-Pretraining
[ICLR 2024 Spotlight] This is the official code for the paper "SNIP: Bridging Mathematical Symbolic and Numeric Realms with Unified Pre-training"
SNIP helps researchers and data scientists predict properties of mathematical equations or discover the underlying equations from numerical data. It takes either symbolic math expressions or numerical datasets as input and can output predictions for properties like non-convexity or generate symbolic equations. This is for professionals working with complex mathematical models and data analysis.
No commits in the last 6 months.
Use this if you need to understand the characteristics of a mathematical function from its symbolic form or numerical observations, or to reverse-engineer an equation from a dataset.
Not ideal if your primary goal is basic arithmetic calculations or simple data fitting with known function types.
Stars
58
Forks
7
Language
Python
License
MIT
Category
Last pushed
Oct 22, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/deep-symbolic-mathematics/Multimodal-Math-Pretraining"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
galilai-group/stable-pretraining
Reliable, minimal and scalable library for pretraining foundation and world models
CognitiveAISystems/MAPF-GPT
[AAAI-2025] This repository contains MAPF-GPT, a deep learning-based model for solving MAPF...
UKPLab/gpl
Powerful unsupervised domain adaptation method for dense retrieval. Requires only unlabeled...
larslorch/avici
Amortized Inference for Causal Structure Learning, NeurIPS 2022
svdrecbd/mhc-mlx
MLX + Metal implementation of mHC: Manifold-Constrained Hyper-Connections by DeepSeek-AI.