The requirements of the project are numpy, torch, and curtsies.
The datasets to train the project are located here. hands_25.npy contains all 25000 images (25 emojis at 1000 images each). hands_10.npy contains the first 10000 images, noted by the commented OUTPUTS list at the beginning of cnn.py.
After training, test the network with an image in the working directory.
DISCLAIMER: hands_25.npy causes issues with the network's ouput when providing images beyond love-you_gesture in the output list. I will debug this soon.
To use hands_10.npy (or hands_25.npy), verify two things:
- The output of the final linear layer of the network needs to be changed to
10(or25respectively). - The dataset filename needs to be changed appropriately. The dataset is loaded one time at the head of
loadTrainingData.
To run the script, type python cnn.py.
The actual dataset (1080p PNGs) will be uploaded to the save drive link provided above soon (it needs remaking for more varied training data). I will look into publishing the dataset as well.