Computer-Vision-Projects

Contains the following:

1. Computer Vision Projects:

   1.1 Leaf Classification: A fine grained classification problem on the Leafsnap datset. The dataset consists of
   images of 23147 Lab images belonging to 185 tree species.
   Link: http://leafsnap.com/
   Reference: Leafsnap: A Computer Vision System for Automatic Plant Species Identification.
   

2. Helper functions:

   2.1 copy_paste script: For directory level manipulations to get images from the main dataset and putting them
   into test, validation and specimen folders. The script reads from 'test.csv' which has the same format as
   https://github.com/srijan-mishra/Computer-Vision-Projects/blob/master/Projects/Leafsnap%20Dataset/species%20used.csv .
   
   2.2 generate_string.py: To generate the two strings to put in primary_train.py for different number of train/test
   cases.
   
   2.3 top_5.py: Script to use saved weights to get top 5 accuracy on a test dataset.