kprohith / porsche-classifier Public

Notifications You must be signed in to change notification settings
Fork 2
Star 5

Identify Porsche models(718, 911, Taycan, Macan, Cayenne, Panamera) with 95% accuracy

porsche-classifier.azurewebsites.net/

5 stars 2 forks Branches Tags Activity

Notifications

Name		Name	Last commit message	Last commit date
Latest commit History 11 Commits
app		app
Dockerfile		Dockerfile
README.md		README.md
netlify.toml		netlify.toml
porsche_classifier.ipynb		porsche_classifier.ipynb
requirements.txt		requirements.txt

Repository files navigation

Porsche Classifier

Identify Porsche models(718, 911, Taycan, Macan, Cayenne, Panamera) with 95% accuracy.

Trained using fastai-v3, pytorch and Gradient.

Uses resnet50 and trained on a Nvidia Quadro P5000.

Built on docker and is hosted on Microsoft Azure Web Services.

Trained on a dataset of publicly sourced images containing 30000 Porsche car models of varying degree of quality.

Porsche cars, specially the latest generations of the Panamera/Taycan,Macan/Cayenne & 911/718 can be pretty tricky to tell apart for a layman who isn't paying very close attention, which is why I wanted to test out what kind of features this deep learning model would pick up.

About

Identify Porsche models(718, 911, Taycan, Macan, Cayenne, Panamera) with 95% accuracy

porsche-classifier.azurewebsites.net/

Report repository

Releases

No releases published

Packages

No packages published

Languages

Jupyter Notebook 98.7%
Other 1.3%