Skip to content

● The project deals with Air Pressure System (APS) which generates datasets (Scania Truck) when class corresponds to component failure. The problem is to the cost due to unnecessary repairs. So, it is required to minimize the false predictions. ● Achieved the best Model is XG-Boost Classifier with 99.6% accuracy and minimal cost of 2950.

Notifications You must be signed in to change notification settings

p2kalita/APS-sensor_fault_Detection

Repository files navigation

APS-sensor_fault_Detection

Step 1 - Install the requirements

pip install -r requirements.txt

Step 2 - Run main.py file

python main.py

To download your dataset

wget https://raw.githubusercontent.com/avnyadav/sensor-fault-detection/main/aps_failure_training_set1.csv

This is changes made in neuro lab

Git commands

If you are starting a project and you want to use git in your project

git init

Note: This is going to initalize git in your source code.

OR

You can clone exiting github repo

git clone <github_url>

Note: Clone/ Downlaod github repo in your system

Add your changes made in file to git stagging are

git add file_name

Note: You can given file_name to add specific file or use "." to add everything to staging are

Create commits

git commit -m "message"
git push origin main

Note: origin--> contains url to your github repo main--> is your branch name

To push your changes forcefully.

git push origin main -f

To pull changes from github repo

git pull origin main

Note: origin--> contains url to your github repo main--> is your branch name

.env file has

MONGO_DB_URL="mongodb://localhost:27017"
AWS_ACCESS_KEY_ID="aagswdiquyawvdiu"
AWS_SECRET_ACCESS_KEY="sadoiuabnswodihabosdbn"
curl -fsSL https://get.docker.com -o get-docker.sh
sudo sh get-docker.sh
sudo usermod -aG docker ubuntu
newgrp docker

AWS_ACCESS_KEY_ID=
AWS_SECRET_ACCESS_KEY=
AWS_REGION=
AWS_ECR_LOGIN_URI=
ECR_REPOSITORY_NAME=
BUCKET_NAME=
MONGO_DB_URL=

About

● The project deals with Air Pressure System (APS) which generates datasets (Scania Truck) when class corresponds to component failure. The problem is to the cost due to unnecessary repairs. So, it is required to minimize the false predictions. ● Achieved the best Model is XG-Boost Classifier with 99.6% accuracy and minimal cost of 2950.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published