search

nisargptl / vehicle-recognition

This enables users to gather information about any vehicle and they can see the details in a matter of seconds. Snap a quick photo in the app and it will provide the details about the vehicle make, model, year, user ratings and starting retail price. In addition, it will provide this same information for the 2 closest competitors so a user can do some quick comparison shopping.
https://viris.herokuapp.com/
0 stars 1 forks source link

Discrepancy in testing image dataset #42

Closed sjdhola closed 3 years ago

sjdhola commented 3 years ago

While testing the model we are getting very less accuracy for predicting car class "Chevrolet Silverado 1500 Classic Extended Cab 2007". In image dataset there is no class "Chevrolet Silverado 1500 Classic Extended Cab 2007" available in testing dataset but instead there is class present named "Chevrolet Silverado 1500 Classic Extended Cab 2012". Both are the different class. So, model is giving less accuracy for given class. So, change in image dataset is required to solve this issue.

201801196 commented 3 years ago

we have changed the car dataset to Chevrolet Silverado 1500 Classic Extended Cab 2007 in testing dataset now we are getting 100% validation accuracy for the given class And we are getting 96% validation accuracy on entire 100 classes dataset.

sjdhola commented 3 years ago

Noted thank you for the help