The repository contains the code for an image classifier that identifies different species of flowers. The model was submitted as the final project for Facebook Udacity PyTorch Scholaship Challenge.
The data set contains images of flowers from 102 different species, split into a training set and a validation set. You can download the images from here as a zipped archive. Just uncompress it and you should be good to go.
- Click here to open Google Colab
- Paste the notebook link under the Github tab, or alternatively, download the notebook and upload it to Colab.
- Enable GPU: Go to Edit -> Notebook settings and select GPU as the hardware accelarator.
- Run all the code cells
Initially, the model chosen was VGG19, which was later shifted to Resnet 152 for better accuracy. Hence, the two files, Image Classifier Project — VGG19.ipynb and Image Classifier Project — ResNet-152.ipynb.
Further details can be found in the notebook itself.