Author: Ioannis Gatopoulos, 2020
PyTorch implementation of research paper:
SuperTML: Two-Dimensional Word Embedding for the Precognition on Structured Tabular Data
Abstrack: Projects tabular data to images creating 2-dimentional embenddings. Then, with the use of a pretrained model, it is capable of performing regression and classification tasks.
Install all the dependencies:
pip3 install -r requirements.txtcd super_tml
python main.py --dataset iris --model densenet121
- Baohua Sun.
SuperTML: Two-Dimensional Word Embedding for the Precognition on Structured Tabular Data.
CVPR Workshop Paper, 2019.