Single2Multi_LOMP
Single to Multi: Data-Driven High Resolution Calibration Method for Piezoresistive Sensor Array
Code
This repository contains the code for model introduced in IEEE Robotics and Automation Letters [paper] (https://ieeexplore.ieee.org/document/9394718).
Prerequisites
It requires that you download the data from [(coming soon)]. or contact authors for multi-touch dataset. (We append a sampled single touch dataset that can be used to train our model)
conda create -n LoMP python=3.6
pip3 install torch torchvision torchaudio
conda install
pip install -U scikit-learn
pip install pandas
pip install matplotlib
Next, make sure to update the path to where you sotred the data in train_test_.py file
Run training
With running train_test_passing_single_new.py file, you can simply type
python train_test_passing_single_new.py
This loads a single touch dataset to train and test our dataset.
testing for multi touch dataset
Before you run the code, please make sure that you uncomment double touch loading code in train_test_passing_single_new.py file. Since the volume for multi touch dataset is huge, please make a contact or access NMAIL database to download the multi-touch dataset.