Using Language Model to Bootstrap Human Activity Recognition Ambient Sensors Based in Smart Homes

Last update: Oct 18, 2021

Overview

Using Language Model to Bootstrap Human Activity Recognition Ambient Sensors Based in Smart Homes

This repository is the official implementation of Using Language Model to Bootstrap Human Activity Recognition Ambient Sensors Based in Smart Homes.

Requirements

To install requirements:

To use this repository you should download and install SmartHomeHARLib package

git clone [email protected]:dbouchabou/SmartHomeHARLib.git
pip install -r requirements.txt
cd SmartHomeHARLib
python setup.py develop

Embeddings Training

To train Embedding model(s) of the paper, run this command:

To train a Word2Vec model on a dataset, run this command:

python Word2vecEmbeddingExperimentations.py --d cairo

To train a ELMo model on a dataset, run this command:

python ELMoEmbeddingExperimentations.py --d cairo

Activity Sequences Classification Training And Evaluation

To train Classifier(s) model(s) of the paper, run this command:

python PretrainEmbeddingExperimentations.py --d cairo --e bi_lstm --c config/no_embedding_bi_lstm.json

python PretrainEmbeddingExperimentations.py --d cairo --e liciotti_bi_lstm --c config/liciotti_bi_lstm.json

python PretrainEmbeddingExperimentations.py --d cairo --e w2v_bi_lstm --c config/cairo_bi_lstm_w2v.json

python PretrainEmbeddingExperimentations.py --d cairo --e elmo_bi_lstm --c config/cairo_bi_lstm_elmo_concat.json

Results

Our model achieves the following performance on :

Three CASAS datasets

	Aruba	Aruba	Aruba	Aruba	Milan	Milan	Milan	Milan	Cairo	Cairo	Cairo	Cairo
	No Embedding	Liciotti	W2V	ELMo	No Embedding	Liciotti	W2V	ELMo	No Embedding	Liciotti	W2V	ELMo
Accuracy	95.01	96.52	96.59	96.76	82.24	90.54	88.33	90.14	81.68	84.99	82.27	90.12
Precision	94.69	96.11	96.23	96.43	82.28	90.08	88.28	90.20	80.22	83.17	82.04	88.41
Recall	95.01	96.50	96.59	96.69	82.24	90.45	88.33	90.31	81.68	82.98	82.27	87.59
F1 score	94.74	96.22	96.32	96.42	81.97	90.02	87.98	90.10	80.49	82.18	81.14	87.48
Balance Accuracy	77.73	79.96	81.06	79.98	67.77	74.31	73.61	78.25	70.09	77.52	69.38	87.00
Weighted Precision	79.75	82.30	82.97	88.64	79.6	82.03	84.42	87.56	68.45	80.03	77.56	86.83
Weighted Recall	77.73	80.71	81.06	79.17	67.77	75.51	73.62	78.75	70.09	73.82	69.38	84.78
Weighted F1 score	77.92	81.21	81.43	82.93	71.81	77.74	76.59	82.26	68.47	74.84	70.95	84.71

You might also like...

HiPAL: A Deep Framework for Physician Burnout Prediction Using Activity Logs in Electronic Health Records

HiPAL Code for KDD'22 Applied Data Science Track submission -- HiPAL: A Deep Framework for Physician Burnout Prediction Using Activity Logs in Electro

4 Aug 8, 2022

Generic template to bootstrap your PyTorch project with PyTorch Lightning, Hydra, W&B, and DVC.

NN Template Generic template to bootstrap your PyTorch project. Click on Use this Template and avoid writing boilerplate code for: PyTorch Lightning,

520 Dec 30, 2022

Usable Implementation of "Bootstrap Your Own Latent" self-supervised learning, from Deepmind, in Pytorch

Bootstrap Your Own Latent (BYOL), in Pytorch Practical implementation of an astoundingly simple method for self-supervised learning that achieves a ne

1.4k Dec 29, 2022

Official implementation of our paper "Learning to Bootstrap for Combating Label Noise"

Learning to Bootstrap for Combating Label Noise This repo is the official implementation of our paper "Learning to Bootstrap for Combating Label Noise

21 Apr 9, 2022

One line to host them all. Bootstrap your image search case in minutes.

One line to host them all. Bootstrap your image search case in minutes. Survey NOW gives the world access to customized neural image search in just on

403 Dec 30, 2022

[CVPR2021] UAV-Human: A Large Benchmark for Human Behavior Understanding with Unmanned Aerial Vehicles

UAV-Human Official repository for CVPR2021: UAV-Human: A Large Benchmark for Human Behavior Understanding with Unmanned Aerial Vehicle Paper arXiv Res

129 Jan 4, 2023

Human Action Controller - A human action controller running on different platforms.

Human Action Controller (HAC) Goal A human action controller running on different platforms. Fun Easy-to-use Accurate Anywhere Fun Examples Mouse Cont

27 Jul 20, 2022

StyleGAN-Human: A Data-Centric Odyssey of Human Generation

StyleGAN-Human: A Data-Centric Odyssey of Human Generation Abstract: Unconditional human image generation is an important task in vision and graphics,

762 Jan 8, 2023

PyZebrascope - an open-source Python platform for brain-wide neural activity imaging in behaving zebrafish

1 May 31, 2022

Comments

Unable to find requirements.txt file
Hello, Unable to find the requirements.txt file to follow the below mentioned steps:

To use this repository you should download and install SmartHomeHARLib package

**```setup git clone [email protected]:dbouchabou/SmartHomeHARLib.git pip install -r requirements.txt cd SmartHomeHARLib python setup.py develop

Please do update it. I'm facing issues due to differences in versions.
opened by rkchaitu87 0

Using Language Model to Bootstrap Human Activity Recognition Ambient Sensors Based in Smart Homes

Related tags

Overview

Using Language Model to Bootstrap Human Activity Recognition Ambient Sensors Based in Smart Homes

Requirements

Embeddings Training

Activity Sequences Classification Training And Evaluation

Results

Three CASAS datasets

You might also like...

HiPAL: A Deep Framework for Physician Burnout Prediction Using Activity Logs in Electronic Health Records

Generic template to bootstrap your PyTorch project with PyTorch Lightning, Hydra, W&B, and DVC.

Usable Implementation of "Bootstrap Your Own Latent" self-supervised learning, from Deepmind, in Pytorch

Official implementation of our paper "Learning to Bootstrap for Combating Label Noise"

One line to host them all. Bootstrap your image search case in minutes.

[CVPR2021] UAV-Human: A Large Benchmark for Human Behavior Understanding with Unmanned Aerial Vehicles

Human Action Controller - A human action controller running on different platforms.

StyleGAN-Human: A Data-Centric Odyssey of Human Generation

PyZebrascope - an open-source Python platform for brain-wide neural activity imaging in behaving zebrafish

Comments

Unable to find requirements.txt file

Owner

Damien Bouchabou

TACTO: A Fast, Flexible and Open-source Simulator for High-Resolution Vision-based Tactile Sensors

Python scripts for performing 3D human pose estimation using the Mobile Human Pose model in ONNX.

This is the research repository for Vid2Doppler: Synthesizing Doppler Radar Data from Videos for Training Privacy-Preserving Activity Recognition.

Group Activity Recognition with Clustered Spatial Temporal Transformer

Face Recognition & AI Based Smart Attendance Monitoring System.

The official TensorFlow implementation of the paper Action Transformer: A Self-Attention Model for Short-Time Pose-Based Human Action Recognition

Point Cloud Denoising input segmentation output raw point-cloud valid/clear fog rain de-noised Abstract Lidar sensors are frequently used in environme

Official respository for "Modeling Defocus-Disparity in Dual-Pixel Sensors", ICCP 2020

Step by Step on how to create an vision recognition model using LOBE.ai, export the model and run the model in an Azure Function

Activity image-based video retrieval