Optical character recognition software written in C.
If you plan to train the network you will also need:
- Generate the custom dataset used to train the neural network
cd dataset # This can take a while ./generate_dataset.sh
- (optional) Adjust training parameters in
- Use the
--trainoption when launching the OCR
This will output the neural network in
output/ocr_network_eX after each epoch.
You can save the pre-computed dataset to avoid wasting time before each
training (see details about this in