Skip to content

Latest commit

 

History

History
20 lines (15 loc) · 696 Bytes

README.md

File metadata and controls

20 lines (15 loc) · 696 Bytes
Raw

Image-Classification-CNN

  1. Delete all the checkpoints from the "model" directory before training a new model from scratch.

  2. Run the following command. This will read all the images from the dataset folder and split them into training and testing set and pickle them. This is done to avoid loading the images multiple times, so skip this step if you have already done this before.

python prepareDataSetFromImages.py
  1. Run the following command to train the models.
python convNetTrain.py
  1. Run the following command to test the model.
python convNetTest.py

The file config.py contains the various parameters/flags that can be set.