• Stars
    star
    99
  • Rank 343,315 (Top 7 %)
  • Language
    Jupyter Notebook
  • Created over 8 years ago
  • Updated over 6 years ago

Reviews

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Repository Details

Fined grained classification On Car dataset

Fine_Grained_Classification

Task : Classification of 196 classes of cars with less than 9k images for training. It's intend for replication of the work :

Monza: Image Classification of Vehicle Make and Model Using Convolutional Neural Networks and Transfer Learning
  1. use pre-trained googlenet
  2. train with normal rates
  3. Achieved 87% top1 accuracy and 97% top5 acc with 10000 iterations

Lessons learned :

1.  Training from scratch is very hard with less data and more class.
2.  Bounding box of the car really helps
3.  fine tuning need to select a good lr and make the bottom layers learn slow, but the last few layers learn fast.